Queensland Judgments
Authorised Reports & Unreported Judgments
Exit Distraction Free Reading Mode
  • Unreported Judgment

Madritsch KG v Thales Australia Ltd[2021] QSC 170

Madritsch KG v Thales Australia Ltd[2021] QSC 170

SUPREME COURT OF QUEENSLAND

CITATION:

Madritsch KG & Anor v Thales Australia Ltd [2021] QSC 170

PARTIES:

MADRITSCH KG 

(first plaintiff)

NIOA NOMINEES PTY LTD (AS TRUSTEE FOR THE

BILL NIOA FAMILY TRUST)

(second plaintiff)

v

THALES AUSTRALIA LIMITED

ACN 008 642 751

(defendant)

FILE NO/S:

BS 9497 of 2017

DIVISION:

Trial Division

PROCEEDING:

Civil

ORIGINATING COURT:

Supreme Court of Queensland

DELIVERED ON:

26 July 2021

DELIVERED AT:

Brisbane

HEARING DATES:

1 – 8 March 2021

JUDGE:

Bradley J

ORDER:

The parties are to confer and submit an agreed draft order, or separate draft orders, consistent with the findings in the reasons.  

CATCHWORDS:

INTELLECTUAL PROPERTY – CONFIDENTIAL INFORMATION – INFORMATION PROTECTED – where the first plaintiff designed and developed a technical solution (the Madritsch Solution) to the ‘bolt unlocking problem’ in the firing mechanism of the Austeyr assault rifle – where the defendant had been engaged by the Australian Department of Defence to investigate the manufacturability of a solution to the bolt unlocking problem – where the defendant became aware of the Madritsch Solution and expressed its interest in entering discussions to manufacture the solution under licence – where the parties executed a non-disclosure agreement (NDA) to protect information disclosed during those discussions – where upon execution of the NDA, the defendant was provided with the Madritsch Solution for testing and appraisal – whether the information contained in the Madritsch Solution was confidential information under the terms of the NDA

INTELLECTUAL PROPERTY – CONFIDENTIAL INFORMATION – USE OF INFORMATION – where the components of the Madritsch Solution were examined by representatives of the defendant – where in examining the Madritsch Solution the defendant discovered the AFL spring was in the form of a double torsion spring with significantly lower torque – where one of the persons that examined the Madritsch Solution later supervised the defendant’s own project (HMA1 V6) to solve the bolt unlocking problem – where HMA1 V6 involved a double torsion AFL spring with significantly lower torque – whether the defendant used the information contained in the Madritsch Solution to develop HMA1 V6

INTELLECTUAL PROPERTY – CONFIDENTIAL INFORMATION – OBLIGATION OF CONFIDENTIALITY – where the plaintiffs submit the information contained in the Madritsch Solution attracted an obligation of confidentiality under the terms of the NDA – where to maintain an obligation of confidentiality the information must be necessarily confidential – where the Madritsch Solution was provided to personnel of the Austrian Armed Forces to test for the purpose of deciding whether to adopt the solution – where the defendant contends that any confidentiality of the Madritsch Solution was lost as a consequence – where the court heard expert evidence on a piece of Austrian federal legislation that binds public servants to a duty of confidentiality in respect of information received in an official capacity – whether the Madritsch Solution was brought into the public domain and thereby lost its necessary quality of confidence 

TRADE AND COMMERCE – OTHER REGULATION OF TRADE OR COMMERCE – RESTRAINTS OF TRADE – VALIDITY AND REASONABLENESS – PARTICULAR CASES – COMMERCIAL TRANSACTIONS – GENERALLY– where the defendant contends that the clauses of the NDA prohibiting the use of the relevant information are unenforceable as unreasonable restraints of trade – where pursuant to section 4 of the Restraints of Trade Act 1976 (NSW) a restraint of trade is valid to the extent to which it is not contrary to public policy – where whether it is contrary to public policy is determined with regard to the legitimate interests of the parties and the interests of the public – whether, in light of the defendant’s breach, the relevant clauses of the NDA were reasonably necessary to protect the interests of the plaintiffs in the relevant information 

TRADE AND COMMERCE – COMPETITION, FAIR TRADING AND CONSUMER PROTECTION LEGISLATION – CONSUMER PROTECTION – MISLEADING OR DECEPTIVE CONDUCT OR FALSE REPRESENTATIONS – MISLEADING OR DECEPTIVE CONDUCT GENERALLY – MISLEADING OR DECEPTIVE: WHAT CONSTITUTES ––  where the defendant expressed its interest in producing the Madritsch Solution under licence from the second plaintiff – where the defendant received a draft sub-licence agreement and quickly formed the view it intended to reject it – where the defendant took steps that led the plaintiffs to believe it was considering the offer with a view to accept it or make a genuine counter offer – whether the defendant engaged in misleading or deceptive conduct in contravention of s 18 of the Australian Consumer Law

CONTRACTS – GENERAL CONTRACTUAL PRINCIPLES – FORMATION OF CONTRACTUAL RELATIONS – AGREEMENTS CONTEMPLATING EXECUTION OF FORMAL DOCUMENT – WHETHER CONCLUDED CONTRACT – where to alleviate the first plaintiff’s intellectual property protection concerns, the defendant provided assurances by way of an ‘Intentions Letter’ – where the plaintiffs contend a distinct ‘side agreement’ was entered into by way of exchange of the Intentions Letter for disclosure of the Madritsch Solution – whether the parties intended to be legally bound by a separate side agreement

Competition and Consumer Act 2010 (Cth) sch 2 (Australian Consumer Law, s 18 ).

Restraints of Trade Act 1976 (NSW) s 4

Ansell Rubber Co Pty Ltd v Allied Rubber Industries Pt Ltd [1967] VR 37, considered

Coco v AN Clark (Engineers) Ltd [1969] RPC 41, cited

Hampton Court Ltd v Crooks (1957) 97 CLR 367, cited

Industrial Rollformers Pty Ltd v Ingersoll-Rand (Australia) Ltd [2001] NSWCA 111, followed

Interfirm Comparison (Australia) Pty Ltd v Law Society of New South Wales [1975] 2 NSWLR 104, cited

Isaac v Dargan Financial Pty Ltd (2018) 98 NSWLR 343, applied

Optus Networks Pty Ltd v Telstra Corporation Ltd (2010) 265 ALR 281, applied

Racing Partnership Ltd v Done Bros Ltd (CA) [2021] 2 WLR 469, considered

Saltman Engineering Co Ltd v Campbell Engineering Co Ltd (1948) 65 RPC 203, considered

Seager v Copydex Ltd [1967] 1 WLR 923, cited

Sidameneo (No 456) Pty Ltd v Alexander [2011] NSWCA 418, followed

Thomas Marshall (Exports) Ltd v Guinle [1979] Ch 227, considered

Tullett Prebon (Aust) Pty Ltd v Purcell (2008) 175 IR 414, considered

COUNSEL:

S Cooper QC, with D Tay, for the plaintiffs

C Johnstone, with A Psaltis, for the defendant

SOLICITORS:

Carswell and Company for the plaintiffs

Minter Ellison for the defendant

  1. [1]
    This decision concerns a dispute about the firing mechanism in the standard service rifle (Austeyr)[1]used by the Australian Defence Force (ADF) since about 1985.  The Austeyr is based on the Steyr Armee-Universal-Gewehr rifle (Steyr AUG), developed by a company from the Austrian town of Steyr.  The Bundesheer (the Austrian army) has used the Steyr AUG as its standard service rifle since about 1978.     
  2. [2]
    All three parties are in the firearms trade. The first plaintiff (Madritsch) is based in Villach, Austria.  Among other things, it manufactures rifle components and accessories for the Steyr AUG.  
  3. [3]
    The second plaintiff (Nioa) is the largest privately owned supplier of arms and ammunition in Australia.  It is an importer and supplier of various components, including some produced by Madritsch.  Since 2007, Nioa has represented Madritsch in all its dealings with Australian customers, including all those with the defendant (Thales).  
  4. [4]
    Since about 2006, Thales has been a wholly owned subsidiary of a large Europe-based aerospace, transport, defence and security corporation.  It owns and operates the longstanding small arms factory at Lithgow, New South Wales.  There it manufactures Austeyr rifles and supplies them to the Commonwealth of Australia for the use of the ADF.[2]  
  5. [5]
    The plaintiffs allege that Thales used confidential information about a solution to a problem affecting the Steyr AUG family of weapons, including the Austeyr, in breach of a written non-disclosure agreement, a “side agreement”, and an equitable duty of confidence.  They also allege Thales engaged in misleading or deceptive conduct in contravention of s 18 of the Australian Consumer Law (ACL).  This decision is about whether Thales is liable to Madritsch or Nioa in any of these ways.  If Thales is liable, then another hearing may determine the amount of any damages or compensation payable or other relief to which the plaintiffs may be entitled.[3]  

Background – the Steyr AUG and Austeyr rifles 

  1. [6]
    Both the Steyr AUG and the Austeyr are gas-piston operated automatic bullpup-type assault rifles that fire from a closed bolt (or firing chamber).  In each model, the rifle’s firing action is facilitated by a component called the gun lock group and sometimes called the slide group.  The following is a simplified description of how this component operates in automatic mode.    
  2. [7]
    When the trigger is depressed and held down, the gun lock group slides back and forth over a part called the automatic firing lever (the AFL).  When the gun lock group is as far forward as possible, the bolt is closed, and the hammer may be released.  When the hammer is released, two springs (the hammer springs) drive the hammer to strike the firing pin, firing the cartridge.  The expanding gases generated from the fired cartridge propel the ammunition down the rifle barrel towards its target.  The gases also drive the gun lock group rearward and back over the AFL.  When the gun lock group is behind the AFL, the next cartridge can enter the opened bolt from the magazine.  As the gun lock group moves rearward, the hammer is pushed back into position, and return springs in the rifle butt at the rear of the gun lock group are compressed.  When the return springs are fully compressed, they release and push the gun lock group forward again and over the AFL, closing the bolt, and allowing the hammer to be released again, striking the firing pin and firing the next cartridge.  In automatic firing mode, the rifle is intended to repeat this cycle until the operator stops it by releasing the trigger or the rifle has fired all the ammunition in the magazine.  
  3. [8]
    When the gun lock group is not fully forward, and the bolt is partially open, the hammer cannot be released and a cartridge in the bolt cannot be fired.  
  4. [9]
    It is convenient at this point to note that the AFL, the hammer, and the hammer springs are assembled as a unit within the rifle called the hammer mechanism and sometimes referred to as the trigger mechanism.

The bolt unlocking problem 

  1. [10]
    When first manufactured, the Steyr AUG and the Austeyr were not designed to incorporate a grenade launcher attachment.  Over time, whether or not associated with the use of an M203 grenade launcher in Rambo III (1988), both rifles have been manufactured with an accessory rail below the barrel on which a grenade launcher may be mounted.  Thales produces a grenade launcher attachment for use with the Austeyr it manufactures at the Lithgow factory for the ADF.  Madritsch also produces such an item. 
  2. [11]
    Perhaps as early as 2006, staff at the Commonwealth’s Defence Material Organisation (DMO) identified “an operational interface issue” with the Austeyr when fitted with a grenade launcher.  
  3. [12]
    In February 2007, a fault was formally reported.  The Austeyr would fail to fire a round of ammunition immediately after the grenade launcher attached to the rifle had been fired.  The operator had to stop and “re-cock” the rifle before it could be fired.   This created problems for soldiers in the field and, to some extent, defeated the purpose of equipping them with an automatic assault rifle and attached grenade launcher.
  4. [13]
    In May 2007, a report verified the existence of the fault, based on an investigative trial by DMO’s Armament Systems Project Office (ARMTSPO).  The report concluded the fault was caused by the recoil force generated when the attached grenade launcher was fired.  Some of the recoil force was absorbed by the gun lock group in the rifle.  This could cause the gun lock group to move rearward, but not as far as when the rifle is fired.  This slight rearward movement opened the bolt, slightly.  While the bolt remained open, the rifle would not fire.  It is convenient to refer to this fault as the bolt unlocking problem.  In some instances, it has been referred to as the Austeyr being in an “out of battery” condition or a failure to “return to battery”.  
  5. [14]
    The report identified three possible solutions: a device preventing the gun lock group from moving backwards when a grenade is fired; a redesign of the return springs to provide increased spring load; and a design that would dampen the recoil forces from the grenade launcher. 
  6. [15]
    After further investigations and reports, a desktop assessment of these possible solutions concluded they would be complex, affect other components, and would not address the root cause of the bolt unlocking problem.  

JGCM solution

  1. [16]
    Between March and October 2008, ARMTSPO proposed, developed and oversaw the manufacture of a prototype for a possible engineering solution to the bolt unlocking problem.  It was called the “Joule Gain Cycle Modified Trigger Mechanism” (the JGCM).    
  2. [17]
    The assumption underlying the JGCM was that the pressure from the hammer springs acting on the AFL exerted a vertical force on the gun lock group causing a resistance that stopped the gun lock group moving over the AFL to return to a fully forward position, after it was dislodged by the recoil from grenade fire.  In the standard hammer mechanism, two single torsion springs apply torque (rotational force) to the hammer and to the AFL.  The JGCM introduced a new separate single torsion spring for the AFL (the AFL spring), so that the hammer springs applied torque to the hammer only.  The AFL spring applied torque to the AFL.  The AFL spring could be designed so it caused the AFL to apply a lower vertical force to the gun lock group passing over it,[4] and so reduced the resistance.  The JGCM also reworked the hammer springs, and redesigned the AFL to reduce resistance.  
  3. [18]
    ARMTSPO carried out a proof of concept trial, a two-stage performance and endurance trial and an environmental trial of the JGCM.  It was satisfied the JGCM, with either a lighter 1.25mm diameter or a heavier 1.4mm diameter AFL spring, was effective to resolve the bolt unlocking problem.  ARMTSPO elected to design the JGCM with the heavier spring.  The selection appears to have reflected a preference for a component that would be more robust in service.
  4. [19]
    An October 2008 report by ARMTSPO concluded that the JGCM had “solved” the bolt unlocking problem, allowing the Austeyr “to be fired reliably” in conjunction with an attached grenade launcher.  The report recommended the manufacture of production-ready prototypes of the JGCM, with some minor changes to the materials for some elements.  It also recommended that, once the manufactured prototypes were validated for compliance with the relevant ADF specification, it be adopted and introduced into service with all variants of the Austeyr.  The author of the report was Joseph Grima, Senior Technical Adviser – Special Forces.  It was reviewed by a Senior Technical Adviser – Small Arms and approved by a Design Acceptance Authority Representative.  

Thales work October 2008 to December 2009

  1. [20]
    In October 2008, the same month as the report, ARMTSPO engaged Thales to investigate the manufacturability of the JGCM to be “retro-fitted” to the ADF’s existing Austeyr rifles.  

The work of Mr Williams

  1. [21]
    Much of the work by Thales on this engagement was performed by Martin Williams, a technical officer.  
  2. [22]
    In 1981, Mr Williams had completed a four-year apprenticeship as a fitter machinist toolmaker at the Lithgow factory.  From 1981 to 1997, Mr Williams was a tradesman.  From 1997 to 2003, he was a technical officer or technical adviser for the factory’s indoor firing range.  In 2000 or 2001, he completed an advanced diploma of mechanical engineering from TAFE New South Wales.  In 2003, he became supervisor or manager of the indoor firing range. In about 2007, Mr Williams moved from the firing range to be a technical officer working for Thales on the upgrade of the Austeyr.  
  3. [23]
    From October 2008 to late 2009, Mr Williams did some work on what he called “developments to the hammer mechanism” of the Austeyr.  This was amongst much other work Mr Williams did and many other responsibilities he had over that period.  
  4. [24]
    In a prepared statement, Mr Williams described his relevant work in this way: 

“My recollection is that I also learned about this time that Joe Grima[5]  had developed a solution to the [bolt unlocking] problem and some money had been allocated by the Commonwealth to try to implement that solution in a way that could be retrofitted to the F88(SA1) and F88(SA2) [models of the Austeyr]. 

… 

It was not part of my role to investigate whether or not the theory of the JGCM Solution was correct.  It was also not part of my role to identify whether there was another solution available.  My role was to deal directly with Joe Grima on behalf of Thales and to identify and design the necessary changes so that the JGCM Solution would be able to be mass produced.  By mass produced, I mean, to make it so that it could be manufactured and assembled in a cost effective manner having regard to manufacturing, manufacturing tolerances and human factors.”

  1. [25]
    At the hearing, Mr Williams had a limited recollection of the detail of his work on the HMA1 Project.  This is not surprising.  The project was a small part of his work over the period.  Many years passed before he was asked to give evidence.  It was apparent from his oral evidence that his written statement had been prepared from documents and not directly from Mr William’s own recollection.  Some of those documents were not created by Mr Williams and so, again unsurprisingly, he was unable to confirm or corroborate the written record.  He was a credible witness, but with little independent recollection and limited recollection when refreshed by documents.  
  2. [26]
    Mr Williams produced drawings and digital models of the components of the JGCM prototype supplied by ARMTSPO.  He recorded his measurements of the forces applied by each of the two hammer springs and by the AFL spring in the JGCM.  He then created additional copies of the JGCM prototype for testing.  He did this by altering standard Austeyr hammer mechanism parts and obtaining alternative springs with forces approximating those he had measured in the JGCM prototype.  Mr

Williams called the new copies of the prototype the “Hammer Mechanism Assembly No 1” (or HMA1).  Within Thales, the work on this mechanism was referred to as the HMA1 Project

  1. [27]
    Mr Williams quickly formed an opinion that the JGCM prototype could not be mass produced unless certain manufacturing issues were addressed.  He described the three main manufacturing issues in his witness statement: 

“(a)the mounting points for the hammer springs on the inside walls of the hammer body were fiddly and it would be difficult to manufacture and assemble these intricate parts;

  1. (b)
    the shape of the AFL spring (with asymmetrically offset legs) would be difficult to reliably manufacture and also difficult to install;
  1. (c)
    the intricate nature of the JGCM Solution would make it difficult for an armourer in the field to work on the amended design.”
  1. [28]
    Most of his work on the HMA1 Project involved producing design drawings and two and three-dimensional digital models.  As his worked progressed, he referred to the first completed model as HMA1 version 1 (V1).  It was as close to an exact reproduction of the JGCM prototype that Mr Williams could manage.  
  2. [29]
    By December 2008, Mr Williams had prepared designs for a second version of the HMA1 (V2), and, by February 2009, a third version (V3). The main changes were modifications to the mounting points and the shape and length of the legs for the hammer springs, but he made some changes to the design of the AFL spring.  
  3. [30]
    In a progress report dated 5 March 2009, he explained he was working to two requirements: that the hammer springs would exert a force plus or minus 2.5% of the force exerted on the hammer in the standard mechanism; and that the force exerted by the AFL spring would be plus or minus 2.5% of the force exerted on the AFL in the JGCM prototype.  These requirements were based on Mr Grima’s ARMTSPO report of October 2008.  
  4. [31]
    On 12 May 2009, he completed drawing U-20343 of the AFL spring for V3.  It showed the spring at rest (with its legs at an angle of 135⁰ to each other).[6]   The drawing included these “spring specifications”: 

“@ 76⁰ ± 0.5⁰ = 283 Nmm ± 30 Nmm 

@ 68⁰ ± 0.5⁰ = 315 Nmm ± 30 Nmm”

  1. [32]
    In these specifications, Nmm (Newton-millimetres) is a measure of the torque.  A torsion spring can be twisted along its axis.  When twisted, the spring exerts a torque (or rotational force) in the opposite direction to the twist.  The torque is proportional to the angle to which the spring is twisted.  The greater the angle (the more it is twisted) the greater the torque (in the opposite direction).  As a rotational force, torque may be measured in Newton-millimetres (Nmm), as Mr Williams did.  This is because torque depends not only on the force applied (measured in Newtons), but also on the perpendicular distance from the axis of rotation to the point where the force is applied (here measured in millimetres).   
  2. [33]
    The actual torque exerted by the AFL spring in V3 (or in the JGCM or any other version of the device produced by Thales) was not proved at the hearing.  Nor was the vertical force exerted by the AFL on the gun lock group in each such device.  The various measurements made by Mr Williams (and later by others) may be relevant to findings to be made about the state of Thales’ knowledge and the reasons or explanations for design changes.  It is not necessary to determine the actual torque or force for that purpose.
  3. [34]
    In May 2009, Thales arranged a “checklist” meeting with Commonwealth representatives to consider V3 in anticipation of a Critical Design Review (CDR).  CDR was a required step before Commonwealth acceptance of a new or modified component.  The minutes of the meeting record three concerns of the Commonwealth representatives.  

“In reviewing the details of the design the Commonwealth presented the following concerns: a. the reliability of the steel to plastic interfaces, b. Auto Fire Lever, one spring leg is encased and the other has an angle bend at the plastic interface and the lever is not sitting square, c. Interoperability outcomes have resulted in very toleranced design specifications; and the Joule Gain design changes likely to effect the operation of the weapon and put interoperability development work at risk?”

  1. [35]
    The minutes record three relevant action items for Thales: 

Thales to review the design with respect to the steel to plastic interfacing and to develop a life cycle testing trial to evaluate these concerns. 

Thales to review the design with respect to the Auto Fire Lever spring arrangement, and that this lever does not appear to sit squarely.  …

Thales to confirm the effect of the Joule Gain design on the Interoperability Trial outcomes.”   

  1. [36]
    The second and third concerns and action items have some continuing relevance.  I will return to the second later in these reasons.  The third indicate a concern that any changes to the JGCM prototype could alter the effectiveness of the mechanism as a solution to the bolt unlocking problem, as found by Mr Grima’s ARMTSPO report.  It helps explain later evidence about Thales’ frequent checks, measures and comparisons between HMA1 versions and the JGCM prototype. 
  2. [37]
    On 1 June 2009, according to Mr Williams’ progress report, the “simplest manufactureable [sic] version of the mechanism [was] ready for critical design review.”  It was V3.  He noted that the “Specification drawings” for the AFL spring had yet to be done and that “forward progress has stopped due to no critical design review acceptance.”
  3. [38]
    On 3 June 2009, Thales measured the vertical force exerted by the AFL on the gun lock group in the JGCM prototype and the same force in the V3.  Thales also measured the force exerted by the hammer springs.  Mr Williams explained that: 

“one goal of the HMA-1 project was to ensure that any development to allow for mass production ensured that the forces being exerted by the hammer springs and the AFL spring were the same as those in the JGCM Solution.  In order to achieve this I had to arrange for those forces to be measured on some assembled HMA1 mechanisms.”

  1. [39]
    Across ten test measures of each of five JGCMs, the average vertical force exerted by the AFL was 26.45N.  Ten test measures of another five JGCMs produced an average of 26.47N.  Across ten test measures of each of ten V3s, the average force was 28.38787N.
  2. [40]
    On 19 June 2009, Mr Williams produced a changed issue of drawing U20-343-0 of the V3 AFL spring, showing it at rest (135⁰) and in two rotated angles (68⁰ and 76⁰).  The spring specifications for the two angles remained the same. 
  3. [41]
    At this time, a draft Thales’ Development Report noted Thales considered the “spring force” in V3 was very close to that in the JGCM.  
  4. [42]
    On 14 July 2009, Thales conducted live-fire testing of four Austeyr rifles fitted with standard hammer mechanism, four fitted with V1 and four fitted with V3.  Mr Williams had arranged for springs for the V3 to be manufactured by Gibbens Industries Pty Ltd (Gibbens), who made the standard hammer mechanism springs for the Austeyr manufactured by Thales.  Three operators were selected to fire each rifle.  Each operator fired two 5.56mm rounds from each rifle, followed by a 40mm practice grenade from the grenade launcher attachment (GLA), and then attempted to fire another two 5.56mm rounds from the rifle.  The results were reported in a later position paper.  

Hammer mechanism

Pass/Fail[7]

% Pass

Standard mechanism

1/12

8%

V1

9/12

75%

V3

12/12

100%

  1. [43]
    On 16 July 2009, Thales gave a CDR presentation to the Commonwealth about V3.  It was presented as “a mass producible version” of the JGCM.  The V3 did not pass the review presentation.  Amongst other issues, the Commonwealth’s concern about the AFL not sitting squarely, raised at the May 2009 checklist meeting, had not been addressed.
  2. [44]
    On 7 October 2009, a Thales R&D monthly progress meeting was told ARMTSPO wanted to have a contract for HMA1, amongst other things, in place with Thales by the end of 2009.  Concern was expressed that Mr Williams was fully occupied on a range of projects and more resources were required. 

Madritsch KG v Thales Australia Ltd [2021] QSC 170

Figure 1: Single-torsion AFL spring drawn by Mr Williams for HMA1 Version 3

The role of Mr Eyles

  1. [45]
    From 14 October 2009, Michael Eyles became engineering manager for the Thales production and design group at the Lithgow factory.  In this role, he had responsibility for managing the HMA1 Project.  Mr Williams reported to him.  
  2. [46]
    Mr Eyles held a Bachelor of Engineering from the University of Technology, Sydney (1992).  He had commenced working at the Lithgow factory in 1989-90, before graduating, and continued there as an employee until 2004 or 2005.  By then he was an engineering manager.  In October 2009, after four or five years working elsewhere in the same field, Mr Eyles had returned to work for Thales as an engineering manager.
  1. [47]
    Mr Eyles explained that his “specialty within the field of engineering” is in:

“the manufacture of design solutions.  By this I mean that my experience enables me to look at particular products which designers have produced to meet a customer’s requirements and be able to manage the process of taking that design from a proven solution, to one which is manufacturable.”

  1. [48]
    He further explained:

“Manufacturability is the key to the commercial production of any design.  If it cannot be manufactured in a cost effective manner and in a way that ensures consistency in quality and therefore functionality, I do not consider a design to be complete.” 

  1. [49]
    By 29 October 2009, Mr Eyles had prepared and reviewed a position paper.  It confirms he was aware Mr Williams had used the spring forces measured from the JGCM prototype to create V3.  He noted the successful small-scale live fire tests of V1 and V3 in July 2009 and some slide-velocity tests on 16 October.[8]   
  2. [50]
    Mr Eyles reported Thales had determined that the individual mass and technique of the operator had been shown to have an effect “on the extent of the recoil/inertial forces transferred to the gun lock and hammer mechanism.” 
  3. [51]
    On 5 November 2009, Mr Eyles reviewed a technical report on the HMA1 Project prepared by Thales mechanical engineer, Michael Johnston.  It might be called a

“desktop review”.  No original testing or research was undertaken.  After considering the available information about the bolt unlocking problem, the report concluded: 

“The fail scenario is the result of the operator’s response to GLA recoil being transmitted through the [Austeyr] F88.  Due to the response movement, the gun lock will dislodge from battery and [fail] to return to battery.  As the operator plays a significant role producing the outcome, an undefined failure variable exists within the problem space. 

The existence of the [bolt unlocking problem] is an inherent design feature of the F88.  Within the F88 design, the gun lock is not mechanically restrained.  Mounted through a series of springs, the movement … of the gun lock is defined by the movement of the main body of the F88.  A design solution will require the F88 to [perform] above the original F88 performance expectations.” 

  1. [52]
    The report identified five design options with a “realistic validity” to solve the bolt unlocking problem.  The preferred design solution was the JGCM.  It was “not ideal” and was selected for “its minimal impact to the existing design and its simplicity to implement.”  Its disadvantages were put this way: 

“Not 100% solution.  Operator may still influence an out of battery condition.  

System may not be able to be held in a fail safe position.”

  1. [53]
    The report’s final conclusion was that a reduction in the “resistive force” of the AFL was the “optimal compromise above all other possible solutions.”

“Although it may not eradicate the issue completely it can be design[ed] to provide an improvement for the majority of users in the majority of situations.”  

  1. [54]
    On 26 November 2009, a draft technical report was prepared on earlier live-fire testing using the standard hammer mechanism and the V3.[9]  There was preliminary live firing and then secondary live firing.  In the preliminary test, Austeyr rifles with the standard mechanism that had been used in service did not show the bolt unlocking problem.  When the rifles were fitted with new butts, they did show the bolt unlocking problem.[10]  The secondary test used the rifles with new butts.  Various measures and records were made using a high-speed camera and other devices. There were three reported outcomes.  
    1. (a)
      First, the Austeyr rifles with the “used” butts successfully fired 100% of the time after the attached GLA was fired.  The result was the same whether a standard mechanism or a V3 was fitted.  With new butts, the Austeyr rifles with standard mechanisms successfully fired 21% of the time, while those with V3 successfully fired 100% of the time.  The report concluded that the “age or condition of the butt contributes to the existence” of the bolt unlocking problem. 
    2. (b)
      Second, contrary to Mr Eyles’ position paper and Mr Johnston’s technical report, this report concluded that the operator has a “minimal” effect on the recoil absorption “with respect to body type and weapon inclination.”  
    3. (c)
      Third, in Austeyr rifles fitted with V3, the gun lock group travelled on average slightly less far rearward when the GLA was fired than in those fitted with the standard mechanism.  This “unexpectedly contradicts prior views” that the reduced disengagement force from the AFL would allow the gun lock group to move further rearward before returning.  

Madritsch KG v Thales Australia Ltd [2021] QSC 170

Figure 2: Automatic Firing Lever with spring for HMA1 Version 3 drawn by Mr Williams

  1. [55]
    Before the end of 2009, Mr Williams did some further work on the HMA1 Project.  He prepared drawings of HMA1 version 4 (V4).  On 24 November 2009, he completed drawing SK-107556 of the AFL spring for HMA1 version 5 (V5).  The V5 AFL spring was to be of the same material as V3.  Amongst other changes, at rest the legs were at an angle of 126⁰ rather than 135⁰.  The torque specifications were yet to be determined. 
  2. [56]
    On the final drawings of the AFL spring for V4 and for V5, Mr Williams reduced the torque specified for the AFL spring.  He did so to reduce the disengagement force exerted by the AFL on the gun lock group.  He intended to make it closer to the force he had measured in the JGCM prototype.  
  3. [57]
    On 10 December 2009, Mr Williams took ten measurements of the disengagement force exerted by the AFL in four V5s he had assembled.  According to his recorded measurements, the average force was respectively 26.70N, 29.49N, 26.30N and 27.03N.  These figures indicate some success in reducing the disengagement force to the 26.45N he had measured in the JGCM prototype. 
  4. [58]
    Mr Williams explained V4 and V5 were “not progressions from” V3.  He gave them successive version numbers to distinguish them from V3, because it had been live fire tested.  He assembled some V5s in physical form. 
  5. [59]
    According to Mr Eyles, towards the end of December 2009, it became evident that the test outcomes Thales could obtain from using the HMA1 versions in the Austeyr were neither consistent nor reliable.  As a result, according to Mr Eyles, Thales stopped all testing prior to Christmas 2009.  This is consistent with the evidence of Mr Williams that, by the end of 2009, he had ceased any active work on the HMA1 Project and that concerns remained about reliability and performance of the HMA1 design.  There was not yet an HMA1 version that had passed CDR and was suitable for mass production.
  6. [60]
    Over the period of his work on the project, Mr Williams had drawn five versions of the HMA1 (V1, V2, V3, V4 and V5) and assembled three in physical form (V1, V3 and V5).  In each instance, his aim was to address the manufacturing issues he had identified, which concerned the making, assembly, and maintenance of the mechanism.  See [27] above.  The versions he developed were otherwise consistent with the JGCM.  Each used a single torsion AFL spring, as the JGCM had done.  Each aimed to reproduce the 26.45N disengagement force of the AFL he had measured in the JGCM prototype.  

Madritsch KG v Thales Australia Ltd [2021] QSC 170

Figure 3: Single-torsion AFL spring drawn by Mr Williams for HMA1 Version 5

  1. [61]
    In about December 2009 or January 2010, the Commonwealth raised with Thales other performance issues about the Austeyr.  Mr Eyles took the view that until these other performance issues were resolved, it would not be clear whether the HMA1 Project would be necessary at all.  The limited documentary evidence is consistent with Mr Eyles’ oral evidence that during 2010 Thales had “no resource” and “did not work on [the project] until very late in 2010.”
  2. [62]
    In late 2010 and early 2011, Graham Evenden, then the Testing and Evaluation Manager at Lithgow, considered whether to continue work on the HMA1 Project, which had ceased in 2009, “but at that time Thales did not have anyone to do that work.”  

Further Thales tests – December 2010 to March 2011

  1. [63]
    Mr Eyles gave some evidence about three further tests. [11]  Like much of his evidence, it was at times confusing and unsatisfactory.  See [257] to [267] below.  
  2. [64]
    The first test was a pendulum test.[12][13]  It was probably conducted between December 2010 and January 2011.  The test compared the Austeyr13 with the Steyr AUG.  Thales did not finalise the report of the testing.  The hammer mechanism assembly used is unclear.  It was likely the standard mechanism.  
  3. [65]
    The second test was also a pendulum test.  It compared the frequency of the bolt unlocking problem of the Austeyr[14] in three different configurations: 
    1. (a)
      fitted with a standard hammer mechanism; 
    2. (b)
      fitted with V5; and 
    3. (c)
      fitted with a standard mechanism, from which the AFL had been removed.  
  4. [66]
    In all cases, when fitted with V5, the bolt unlocking problem did not occur.  The comparison of results with V5 and those with the AFL removed reportedly showed the V5 to be “as effective as no AFL/Slide interface friction at all.” 
  5. [67]
    The third test was in about March 2011.  It was a live-fire test, using both an individual operator and mounted rifles.  The object was to validate the results of two earlier pendulum tests. According to Mr Eyles, the third test showed that neither of the pendulum test results reflected the individual operator fired test results.  Nor did the individual operator fired test results reflect the mount-fired test results.  This was consistent with Mr Eyles’ view in the October 2009 position paper and Mr Johnston’s view in the November 2009 technical report.  See [49] to [51] above.   It contradicted the first of the reported outcomes in the November 2009 draft report on earlier test firing.  See [54](a) above.  
  6. [68]
    Mr Eyles described the limited purpose of the three tests, consistently with the other contemporaneous evidence, in this way: 

“At that point we did further studies into it just to reassure the current in-service rifle – as we refer to it the 2010 version – could operate … We were checking [whether] the changes that we made to the rifle system in 2010 may have invalidated the requirement to have a HMA1 solution. So [the 2010 Austeyr] may work without doing [the HMA1 change to the hammer mechanism assembly] because of other design changes in the rifle [over] that period.”

  1. [69]
    Mr Eyles later described it as “re-testing to prove the 2010 build rifle ... in comparison to the current [model] in service.”  As he put it, “no development was done” on the HMA1 at that time.   This confirms Thales’ limited understanding of the cause or causes of the bolt unlocking problem and the possible solutions.  In effect, Thales was testing whether the problem had been solved without any conscious intervention or change to the hammer mechanism or gun lock group.
  2. [70]
    This is not surprising.  ARMTSPO had done the investigation and testing work on developing a solution.  The results of Thales’ small-scale tests were neither consistent nor reliable.  Thales did not fully understand the extent that operator influence, age, wear and tear on the rifle, and weapon inclination contributed to the bolt unlocking problem.  Thales had only sought to make a manufacturable prototype in which the AFL applied a force to the gun lock group close to that Mr Williams had measured in the JGCM and maintained the torque applied to the hammer in the JGCM and standard mechanism.  
  3. [71]
    By March 2011, it was accepted by the ADF, and by Thales, that the bolt unlocking problem was separate from the other performance issues with the Austeyr.  It would have to be solved separately.  However, Thales’ work on the HMA1 Project would not resume for nearly another 12 months, at the end of February 2012.  
  4. [72]
    It is convenient to consider the relevant work by Madritsch before turning to the events that led Thales to resume work on the HMA1 Project.  

Madritsch work to January 2011

  1. [73]
    Since 2008, Madritsch has designed and developed a grenade launcher attachment (GLA) for the Steyr AUG and supplied it to the Bundesheer.  While developing and testing the GLA, Madritsch became aware of an operation interface issue with the Steyr AUG that, for relevant purposes, is the same as the bolt unlocking problem with the Austeyr identified by the DMO.  
  2. [74]
    Madritsch set out to develop a solution in the form of a kit that could be “retro-fitted” to the Steyr AUG.  In this way Madritsch could supply both the GLA and the associated rifle modification kit.  The development of what became the Madritsch Solution may be summarised in this way.  
  3. [75]
    On 16 September 2008, at a Bundesheer facility in Felixdorf, Stefan Madritsch and Thomas Frohnwieser conducted live fire testing of the Steyr AUG with hammer mechanism assemblies from different production batches.  These were recorded using a high-speed camera.  
  4. [76]
    Mr Madritsch had been the director of Madritsch since 1995 and is the chief executive of the company.  He holds a degree in mechanical engineering from Höhere Technische Lehranstalt für Waffentechnik und Maschinenbau[15] in Ferlach, Austria and a licence for the production and repair of military weapons and ammunition in Austria.  Mr Frohnwieser, also an engineer, was the chief technical officer of Madritsch.  
  5. [77]
    They analysed the data from the Felixdorf test and observed that, when the recoil from the GLA caused the gun lock group to move rearwards and open the bolt, the force exerted by the AFL against the gun lock group prevented the gun lock group from moving forwards and returning the bolt to a closed position.  Mr Madritsch called this the trigger force.  He explained that the trigger force included the force applied by the AFL through the hammer springs, friction between the surfaces of the AFL and the gun lock group, and a leverage effect.   Mr Madritsch decided to explore two possible solutions.  One involved preventing the gun lock group from moving rearward when the GLA was fired.  The other involved allowing the gun lock group to return more easily to its forward position after the GLA was fired.  
  6. [78]
    Mr Madritsch and Mr Frohnwieser conducted tests to measure the trigger force exerted by the AFL in the standard Steyr AUG and found it to be about 40 Newtons (N). 
  7. [79]
    Madritsch ordered the production of testing springs able to cause the AFL to exert trigger forces greater than 40N and conducted firing tests using these in place of the standard hammer springs in the Steyr AUG.  Despite the increased trigger forces, the gun lock group still moved rearward when the GLA was fired.  So, the increased hammer spring torque did not solve the bolt unlocking problem. 
  8. [80]
    In early October 2008, Madritsch produced springs that caused the AFL to exert trigger forces below 40N, used them to replace the standard hammer springs and conducted live firing tests.   Despite the decreased trigger forces, the gun lock group did not always return to its fully forward closed position.  
  9. [81]
    In the second week of October 2008, Madritsch conducted further live firing tests using only one hammer spring, rather than the usual two.  The gun lock group remained in the unlocked position after the GLA was fired.  At this point, Madritsch began redesigning the hammer mechanism so that modified hammer springs would drive the hammer and a new separate spring would apply torque to the AFL.  This redesign allowed Madritsch to reduce the trigger force from the AFL against the gun lock group without affecting the torque exerted on the hammer, which was important for the normal operation of the rifle.  Mr Madritsch and Mr Frohnwieser worked on the redesign.  It took about one and a half months to redesign the hammer mechanism.    
  10. [82]
    In about November 2008, Madritsch conduced live firing tests with the redesigned hammer mechanism. The tests showed the trigger force was still too high and the redesigned hammer springs were interfering with the function of the AFL.  
  11. [83]
    Madritsch further modified the design, including by reducing the trigger force from the AFL to a range of about 8 to 10N.[16]  It was tested in January 2009 and was observed to be effective to solve the bolt unlocking problem, in that the gun lock group consistently returned to its locked position after the GLA was fired. 
  12. [84]
    In February and March 2009, the Bundesheer conducted live fire testing on the further modified design in accordance with the standardised NATO AC/225 weapons testing procedure.  An operational issue was observed during extremely cold weather.  This led to a further modification to the AFL spring.  
  13. [85]
    In March 2009, the results of live firing tests at normal temperature were positive.  Additional modified components were produced and tested, including in extreme cold temperatures.  This also yielded positive results, consistently solving the bolt unlocking problem. 
  14. [86]
    Between late May 2009 and early July 2009, the parts for the further modified design (Madritsch Solution) were manufactured.  Between then and 31 August 2009, Madritsch designed three assembly aids for installing the Madritsch Solution into existing Steyr AUG rifles, together with technical instructions and a maintenance manual.  
  15. [87]
    On 31 August 2009, Madritsch provided four Madritsch Solution kits to the Bundesheer for the sole purpose of testing to confirm their safety and functionality. Oberst Wolfgang Weinseiꞵ,[17] an engineer and commander of the Systemreferat Leichte Waffen[18] (SLW), passed them to Fachinspektor[19] Stefan Stamm, a Bundesheer specialised instructor, for testing. The only persons who were able to access the kits were members of the SLW.  They were not accessible to and were not provided to any person outside the Bundesheer.  They were tested and approved.  When the testing was completed, the four kits were returned to Madritsch.  The Head of the Department for Technical Testing of Small Arms and Obst Weinseiꞵ later signed a certificate confirming the Madritsch Solution kits had been tested and approved in accordance with the procedures in NATO AC/225.  
  1. [88]
    Following the successful tests, Madritsch began planning to manufacture Madritsch Solution kits to supply to the Bundesheer for retrofitting to the Steyr AUG. 

The worlds of Thales and Madritsch intersect

  1. [89]
    In late 2010 or early 2011, the Commonwealth selected Thales to develop a new model of the Austeyr for use as the standard service rifle for the ADF.  The project was referred to as Land 125 program Phase 3C (L125-3C).  According to Thales: 

“…  The requirements of that new rifle were manifold, however, the requirements most relevant for the purposes of this proceeding were that … it must be capable of having a GLA fitted and the rifle to be able to be fired immediately after firing a grenade (i.e. without recocking the weapon). 

In other words, the new rifle … must not suffer from the boltunlocking problem.” 

  1. [90]
    Patrick Whitty is the General Manager – Government Business of Nioa. He was primarily responsible for managing the day to day relationship between Nioa and Madritsch.  Between 2007 and 2010, on behalf of Madritsch he negotiated with Thales about the potential supply of the GLA manufactured by Madritsch for use by the ADF with the Austeyr.  
  2. [91]
    In 2010, Mr Whitty during discussions with Thales, Major Simon Johns, Acquisition Project Manager of ARMTSPO, was present.  Major Johns had responsibility for L125-3C.  He told Mr Whitty that the ADF was having problems with the bolt mechanism of the Austeyr becoming unlocked when an attached grenade launcher was fired. Within a few months, Mr Whitty met with Mr Madritsch in Austria.  From him he learned that Madritsch had a technological solution to the bolt unlocking problem for the Steyr AUG.  
  3. [92]
    Mr Whitty arranged for Mr Madritsch and Mr Frohnwieser to attend the Australian Land Warfare Conference in January 2011.  There, during a live firing event near Brisbane, Mr Whitty introduced the two Madritsch executives to Major Johns.  
  4. [93]
    Mr Madritsch told Major Johns that the Madritsch Solution had been successfully tested by the Bundesheer and found to solve the bolt unlocking problem.  Major Johns told Mr Madritsch that the ADF would take a very keen interest in any demonstrated solution to the bolt unlocking problem with the Austeyr.  Mr Madritsch said he would prefer to supply the modification kits directly to the ADF, as he intended with the Bundesheer.  Major Johns said the ADF had no capacity to modify weapons independently of Thales.  He said Madritsch or Nioa would have to liaise with Thales about the supply and installation of modification kits.  
  5. [94]
    On 24 January 2011, shortly after this encounter, Mr Whitty spoke by telephone with Major Johns and with Darren Christopher, the chief weapons engineer for the ADF.  In this conversation, Major Johns recommended that Mr Whitty approach Thales about supplying the Madritsch Solution.  He explained that DMO was concerned that Thales would not honour the warranty on the Austeyr rifles supplied for the ADF if someone other than Thales were to modify the weapons.  Major Johns suggested Nioa and Madritsch make a proposal to Thales about the Madritsch Solution.  Mr Whitty took up this suggestion with Madritsch.  
  6. [95]
    On 24 February 2011, Mr Whitty exchanged emails with Mr Alex Wootton, Procurement Manager – Operations for Thales, about supplying the Madritsch Solution to the ADF through Thales.  Mr Whitty also sent him a draft non-disclosure agreement.  
  7. [96]
    The draft agreement was accompanied by a letter from Mr Frohnwieser, dated 22 January 2011.  In it, he explained that the third-party testing information on the Madritsch Solution “must be kept strictly confidential” until an Austrian Government tender was completed.  Mr Frohnwieser’s letter continued: 

“We wish to offer our solution to the Steyr AUG unlocking bolt issue to Thales, so that Thales can solve the issue in the F88 family of weapons.  We therefore propose establishing another three way NDA between Madritsch, Thales Australia and NIOA to allow discussion of the solution, and keep Madritsch’s IP rights secured.

Please find attached a draft NDA for your consideration.  The NDA allows for listing of any pre-existing IP developed by Thales in relation to this issue of the unlocking of the bolt.” 

  1. [97]
    About three hours after the email to Mr Wootton, Mr Whitty received a telephone call from Major Johns, who congratulated him on offering the solution to the ADF via Thales.  Major Johns told Mr Whitty he was “pushing Thales very hard” to accept and assess the Madritsch technology.
  1. [98]
    On 28 February 2011, Mr Wootton forwarded Mr Whitty’s email to Nicholas Flynn, Thales’ Purchasing Manager for Soldier Systems.  Within minutes, Mr Flynn sent it on to Mr Evenden and James Fulford-Talbot, asking for instructions.  Mr FulfordTalbot was Product Line Manager for Armaments in charge of Thales’ small arms weapons program and the manager of Thales’ relationship with the Commonwealth about the contract for L125-3C.  Later that morning, Mr Flynn called Mr Whitty asking to be shown independent evidence that the Madritsch Solution solved the bolt unlocking problem.  
  2. [99]
    On 1 March 2011, after some enquiries, Mr Whitty informed Mr Flynn that the Austrian authorities would not release test data on the Madritsch Solution, but were prepared to release a letter stating the scope and scale of the testing and confirmation that the testing demonstrated the Madritsch Solution solved the bolt unlocking problem.  
  3. [100]
    On 14 March 2011, the Austrian Federal Ministry of Defence and Sport produced a Prüfzertifikat,[20] signed by Hermann Dorninger, Head of Department for Technical Testing of Small Arms, and Obst Weinseiꞵ, confirming the Madritsch Solution had been tested and approved in accordance with the procedures in NATO AC/225.  On 17 and 18 March 2011, Mr Whitty sent Mr Flynn two copies, one in English and the other in German.

The non-disclosure agreements and related matters

  1. [101]
    In April 2011, Mr Flynn requested a list of the tests the Bundesheer had conducted on the Madritsch Solution.  Mr Whitty responded, advising that Madritsch required Thales to enter into the non-disclosure agreement before it would release any additional information about the Madritsch Solution.  
  2. [102]
    On 4 May 2011, Mr Flynn explained that Mr Fulford-Talbot would like to meet “the Madritsch team” in Europe to:

“learn more about the Madritsch solution to the bolt unlocking issue.  The ability to resolve this issue is a key factor in our customer taking this [L125-3C] program forward.  James’ main interest is the testing done to date and the evidence collected (rather than the details of the specific technical solution).” 

  1. [103]
    On 9 May 2011, Mr Fulford-Talbot and Mr Evenden met Mr Whitty.  Mr FulfordTalbot asked to visit Madritsch in Austria to learn more about the testing and effectiveness of the Madritsch Solution.     
  2. [104]
    Madritsch insisted on a non-disclosure agreement before such a visit.  
  3. [105]
    On 8 June 2011, Thales and each of the plaintiffs executed a non-disclosure agreement.  Thales had modified the terms proposed by Madritsch.  Rather than listing any pre-existing intellectual property developed by Thales on the bolt unlocking problem, it had deleted the schedule for each party to list their “prior intellectual property”.[21]   
  4. [106]
    Between 13 and 15 June 2011, Mr Fulford-Talbot visited Mr Madritsch and Mr Frohnwieser in Vienna.  They took him to a Bundesheer defence testing facility for a demonstration firing of the Steyr AUG fitted with the Madritsch Solution.  In discussions with them, Mr Fulford-Talbot said Thales was interested in being able to build and integrate the Madritsch Solution into the Austeyr rifles it was making for the ADF under a licence from Madritsch.  He said he wanted to move forward with pursuing the Madritsch Solution as quickly as possible.

The Thales brief

  1. [107]
    On 24 June 2011, after Mr Fulford-Taylor had returned from Vienna, he sent Mr Whitty an email, in which he wrote:

“We want to follow up on our discussions regarding their [Madritsch’s] solution to the GLA issue, which appeared to be validated.  Our aim is to get to a position [where] we fully understand the solution and have our engineers confirm its validity. 

In order to alleviate fears of IP breaches, under the existing NDA, I have included a short brief regarding our concepts which we have developed to a degree.  However, if a tested solution exists on the market, we do not see the value in developing these further.  We hope that by exposing this to Madritsch we can establish a recognised Thales development position.  If the Madritsch design is similar to this, we should perhaps both register drawings somehow as we move forward.  

Concurrently, we would like to discuss commercial arrangements.  Whilst this will be subject to negotiation we would like to state that we do not forsee [sic] a solution which involved an Austrian third party manufacturing any modified parts, rather some type of licensing arrangement.

We would like to move this forward as rapidly as possible.”  

  1. [108]
    The enclosed “short brief” was two pages (the Thales brief).  It explained: 

1.Background:

  • Thales determined a number of concepts to eradicate the problem of the rifle ‘unlocking’ when firing the Grenade Launcher such that a rifle stoppage was incurred. 
  • The 2 main concepts that were taken to prototype stage consisted of modifications [to] the Hammer Springs and Auto Fire Lever.
  • These concepts are explained below and although limited testing has been conducted limited time and resource prevents further exploration, refining and qualification.
  1. Modified Auto Fire lever:
    • Auto Fire lever (AFL) notched as shown in Figure 1.
    • Concept was to increase to engage with Slide’s ‘Cocking Piece’ once locked in battery
    • Resist Slide’s initial rearward movement because of increased friction due to steep contact incline between AFL & Cocking piece
    • Not to increase required Slide energy for complete mechanical lock.
    • Still disengage Hammer at appropriate time, during normal single shot / auto fire usage.
  2. HMA1-Hammer Group:
    • Change in Hammer Springs function so that they no longer have dual purpose of supplying force to Hammer and AFL.
    • AFL has own spring that is lower in force rating.
    • Concept was to reduce friction between AFL & Slide Cocking Piece and thereby increase likelihood of Slide lock-up after 40mm shot recoil.” 
  1. [109]
    Figures 1 and 2 in the Thales brief were un-dimensioned perspective drawings of an AFL and a hammer mechanism assembly.    There was no drawing of any spring.  No force or torque rating or specification was stated for any part. 
  2. [110]
    Thales’ first concept was reshaping the AFL to resist the rearward movement caused by the GLA recoil.  Mr Eyles says he “documented this concept in early to mid-2011”.  He did not recall who came up with it. He said it was “a late throw-in by one of our engineers”.  It was not part of the HMA1 Project.[22]  The distinction between it and the HMA1 Project is apparent from the way Mr Eyles explained the concept: 

“It worked on the theory that an alternative shape to the ‘horns’ of the AFL could resolve the Bolt Unlocking Problem without the need for an AFL Spring or to change the shape and placement of the hammer springs.” 

  1. [111]
    Mr Eyles described it as having “little bumps” on the AFL, intended to resist the rearward movement of the gun lock group.  
  2. [112]
    In late 2008, as noted above, Madritsch had investigated the concept of increased resistance to rearward movement of the gun lock group.  Madritsch had applied increased hammer spring force, rather than increased surface friction between the AFL and the gun lock group.  Based on live-fire testing in October 2008, Madritsch had concluded that increasing resistance did not solve the bolt unlocking problem.  
  3. [113]
    Thales had no data or test results from which to conclude that the first concept would be effective or reliable to solve the bolt unlocking problem.[23]  Although Mr Eyles decided to include it in the Thales brief in June 2011, he said Thales had no intention of progressing it to mass production at that time.  I infer it was included to give the impression that Thales had been doing more development work on the bolt unlocking problem that was in fact the case. 
  4. [114]
    The second concept in the Thales brief was to reduce the resistance to forward movement of the gun lock group by confining the hammer springs to operating on the hammer and adding a separate AFL spring that was “lower in force rating”[24] reducing the vertical or trigger force the AFL exerted on the gun lock group.  The Thales brief did not disclose the vertical force exerted by the AFL in the second concept.  Mr Williams had measured it in V3 and calculated the average force as 28.38787N.  He had measured the same force in V5 and calculated the average to be between 26.30N and 29.49N.  Later measurements would record a disengagement force of about 43.2N in V3 and about 42.5N in V5. See: [206] and [211] below  
  5. [115]
    As at 24 June 2011, Thales had not determined either concept to any degree.  ARMTSPO had developed and tested the second concept and concluded it was a suitable solution to the bolt unlocking problem.  Thales had assembled physical prototypes of the concept as V1, V3 and V5.  It had conducted a small scale live-fire test, pendulum tests and another live fire test of V3 and V5.  
  6. [116]
    The small scale live-fire testing Thales had undertaken in July 2009 had produced apparently successful results for V3.  The further testing in March 2011 had apparently shown the V5 to be as effective as entirely removing the AFL from the Austeyr.  There was no indication in Thales’ reports that better performance might be gained by further reducing the vertical force exerted by the AFL on the gun lock group from that in the V3 and V5.  Thales had shown no interest in improving the design, only in making it more suitable for manufacturing.  
  1. [117]
    According to Mr Eyles, by early 2011, he considered the HMA1 Project would be an easy design to take to manufacturability stage:

“We already had the HMA1 prototype as a proven engineering solution already worked out and tested (many times).  I did not consider that the amount of effort it should take to ‘clean up’ the design to make it able to be mass produced was going to be much. … I actually considered that the bulk of the work was going to be in the verification testing and the set up to manufacture the parts (eg dies, tooling, engaging third party manufacturers to supply quality materials).”

  1. [118]
    As Mr Eyles put it, “The only thing that remained was to design the way in which the altered parts could be industrialised.” Of course, that was the task Thales had been engaged to perform since October 2008, without apparent success.
  2. [119]
    As noted above, through live-fire testing undertaken in November 2008 and January 2009, Madritsch had concluded that the trigger force exerted by the AFL needed to be reduced as low as 8 to 10N to be reliable and effective to resolve the bolt unlocking problem across the range of conditions required by NATO AC/225.  In June 2011, Thales was not aware that any further significant force reduction might be required; that the 26.45N force Mr Williams had measured and sought to reproduce might be about three times too high; and that the disengagement force exerted by the AFL in V3 and V5 might be measured at more than four times the disengagement force measured in the successfully tested and approved Madritsch Solution. 
  3. [120]
    For Thales to move from its position in June 2011 to producing a mechanism as effective and reliable as the Madritsch Solution, Thales would have had to undertake sufficient testing to identify the shortcomings in the JGCM, V3 and V5.  Then it would have to design, make and test new HMA1 versions until it had sufficient data to determine the level of disengagement force at which a new version was sufficiently effective and reliable to solve the bolt unlocking problem and still allow the weapon to operate in automatic mode.  There is no evidence that Thales was aware of any shortcoming in JGCM, V3 or V5 or that it planned any further work that would bring about such an awareness or lead to significant modification.      
  4. [121]
    In these circumstances, on 27 June 2011, Mr Fulford-Talbot asked Mr Whitty to supply six kits of the Madritsch Solution for Thales to fit in the Austeyr and test.  
  5. [122]
    When Mr Fulford-Talbot pressed for the kits, Mr Whitty told him that “Madritsch was seeking stronger IP protection for the solution.”  There had been some discontent by Madritsch about Thales’ changes to its earlier draft non-disclosure agreement.  Mr Whitty and Mr Fulford-Talbot exchanged emails about the terms of another agreement and the terms of a letter about Thales’ future intentions in respect of the Madritsch Solution and associated topics, to give comfort to Madritsch.  

The Intentions letter 

  1. [123]
    On 17 August 2011, Mr Fulford-Talbot sent Madritsch the letter (the Intentions Letter).  From the earlier exchanges between Mr Fulford-Talbot and Mr Whitty and those between Mr Whitty and Mr Frohnwieser, I am satisfied that the content of the Intentions Letter and the provision of it by Thales to Madritsch had been the subject of consideration by Thales, Nioa, and Madritsch and negotiations between them, before it was signed and sent.  The parties’ efforts to settle the letter confirm its importance to the plaintiffs, and Thales’ knowledge of that importance.  
  2. [124]
    In the Intentions Letter, Thales stated: 

“If, upon testing of the modified weapons, the Madritsch Solution is technically and commercially suitable, then it is Thales Australia’s intention to enter into a technology licence arrangement with NIOA/Madritsch for access to and commercialisation of the Madritsch IP. If the Madritsch Solution is not suitable for our purposes, then Thales agrees to return all components to Madritsch, and to not use the Madritsch IP for any purpose whatsoever”.

  1. [125]
    The Intentions Letter also included the following statements about Mr FulfordTalbot’s letter of 24 June 2011 and the enclosed Thales brief:

“In order to address your concerns over IP protection, under the terms of the NDA[25] Thales has disclosed the IP internally generated to date to find a solution to this issue.  This represents the current extent of development within Thales.”

The second non-disclosure agreement

  1. [126]
    By 11 August 2011, the further non-disclosure terms were settled in the form of a revised agreement.  By 19 August 2011, this “Non Disclosure Agreement” (NDA) was executed by Thales, Nioa and Madritsch.    
  2. [127]
    The parties described the general purpose of the NDA as:

“to protect confidential information disclosed among the [parties] regarding the issue of the bolt unlocking on the Steyr AUG family of weapons when firing an underslung grenade launcher.”

  1. [128]
    The parties gave a more specific purpose in these words in recital A of the NDA: 

“The Parties wish to enter into discussions in relation to modification of the Steyr AUG family of weapons in order to overcome the problem of the bolt unlocking upon firing of an underslung grenade launcher (the “Purpose”).”

  1. [129]
    By cl 1, the parties agreed that in the NDA, unless the context required otherwise, the following definitions would relevantly apply: 

“‘Proprietary Information’ means any information or data in whatever form, nature or media, including, but not limited to, all technical … and/or commercial information, … test results, discoveries, samples, improvements, models, specifications, drawings, designs, … intellectual property and intellectual property rights of any description which the Parties disclose to one another for the Purpose …

Effective Date’ means the date on which this [NDA] is executed by all of the Parties, or if executed on separate days[,] the date on which the last of the Parties executed the [NDA].

Purpose’ has the meaning given to that term in Recital A.”  

  1. [130]
    By clauses 2 and 3 of the NDA, the parties relevantly agreed and warranted as follows: 

 “2.  Confidentiality Obligation

  1. (a)
    Notwithstanding the Effective Date, the Parties agree and warrant that all data characterised as Proprietary Information and which has been disclosed to effect the Purpose of this [NDA], prior to the Effective Date of this [NDA], had been and will be protected under this [NDA]’s terms and conditions.”
  2. (b)
    Subject to legislative requirements, the receiving Party undertakes that such Proprietary Information shall for a period of FIVE (5) years following the date of receipt of Proprietary Information: 

  1. (ii)
    be only disclosed to and used by its Representatives who have a need to know and solely for the Purpose, …; 
  2. (iii)
    not be used in whole or in part for any purpose other than the Purpose; 
  3. (iv)
    neither be disclosed nor caused to be disclosed whether directly or indirectly to any third party or persons other than those mentioned in sub-paragraph (ii) above and the Parties to this [NDA]; and 
  4. (v)
    neither be copied, nor otherwise reproduced nor duplicated in whole or in part where such copying, reproduction or duplication has not been specifically authorised in writing by the disclosing Party. 

 3.  Restrictions

  1. (a)
    Each Party agrees that it shall not itself, or cause or enable any other person to reverse-engineer, decompile, or disassemble any software, process or tangible items, including refraining from any activities, examinations, performance or qualitative analyses, or other investigations that seek to discover the contents, algorithms, composition or manufacturing formulae of the Proprietary Information.” 
  1. [131]
    In cl 5, the parties listed their respective employees “authorised to receive and/or transmit Proprietary Information” under the NDA on their behalf.  The Thales employees were Warwick Spencer, the L125-3C design manager for Thales, Mr Fulford-Talbot, Mr Flynn and Mr Evenden.  
  1. [132]
    By cl 6, they agreed the NDA was not to be construed as compelling any party to disclose any Proprietary Information to any other party, or “to enter into any further contractual relationships.”  
  2. [133]
    By cl 7 the parties agreed: 

 7.  Ownership

All Proprietary Information and any copies of that information, including without limitation to such other documents, materials or media containing, summarising or derived from any

Proprietary Information, to the extent that any of them contains, summarises or is derived from any Proprietary Information, shall remain the property of the disclosing Party and shall be returned or destroyed by the receiving Party to the disclosing Party immediately upon request in writing or termination of this Agreement.”

  1. [134]
    By cl 8, each party had represented that it had the rights to disclose the Proprietary Information.
  2. [135]
    By cl 12(b), the parties agreed: “

 12.  Breach

  1. (b)
    The receiving Party must notify the disclosing Party immediately upon becoming aware of any disclosure or use of the Proprietary Information of the disclosing Party other than for the approved Purpose or otherwise than as permitted by this [NDA].”
  1. [136]
    By cl 13, the parties agreed the NDA did not exclude “the operation of any principle of law or equity intended to protect and preserve the confidentiality of the Proprietary Information”, and that rights and remedies under the NDA were cumulative and not exclusive of any other right or remedy.    
  2. [137]
    By cl 15, the parties relevantly agreed:  “

 15.  Termination/Expiry

  1. (a)
    This Agreement may be terminated by any Party at any time upon thirty (30) days prior written notice to the other Parties, and unless so terminated shall expire THREE (3) years after the Effective Date.
  1. (b)
    The termination or expiry of this agreement shall not relieve each receiving Party of the obligations imposed by this Agreement in respect of Proprietary Information received prior to the date of the termination or expiry.  Such obligations shall continue for the period applicable as set out in Clause 2(b) of this Agreement …” 
  1. [138]
    By cl 16, the parties agreed that the NDA constituted the entire agreement between them and superseded and cancelled, amongst other things, any prior “agreements among the Parties with respect to the subject matter” of the NDA.  It is common ground that, by executing the NDA, the parties cancelled the earlier non-disclosure agreement executed on 8 June 2011.  
  2. [139]
    By cl 19, the parties agreed that the NDA would be governed by and interpreted in accordance with the laws of New South Wales (NSW).  This was Thales’ home jurisdiction, and not that of Madritsch (Austria) or Nioa (Queensland).  
  3. [140]
    There was some delay in Thales sending a full executed copy of the NDA to Nioa.  By 1 September 2011, it was received.

Madritsch discloses its solution to Thales – September 2011

  1. [141]
    After the Intentions Letter was delivered and the executed NDA exchanged, Madritsch arranged for the kits and associated documents to be sent to Australia.  
  2. [142]
    Very early on 5 September 2011, Mr Frohnwieser emailed Mr Whitty the encrypted Madritsch manual for the installation and maintenance of Madritsch Solution kits in the Steyr AUG (Manual).  He sent the link to the decryption program later that day.  Also on 5 September 2011, Madritsch sent nine of the kits and two sets of installation tools to Nioa by Fedex.  

Madritsch manual disclosed

  1. [143]
    On 6 September 2011, under the NDA terms, Mr Whitty sent Mr Fulford-Talbot the Manual.  He asked Mr Fulford-Talbot for the correct delivery address and recipient for the kits.  Mr Fulford-Talbot advised the correct delivery address was the Lithgow factory and the correct recipient was Mr Eyles.
  2. [144]
    On 6 September 2011, within five minutes of receiving it from Mr Whitty, Mr Fulford-Talbot forwarded the Manual to Mr Eyles.  Mr Fulford-Talbot also forwarded it to Mr Evenden, Stuart Higgins, Thales’ L125-3C Project Manager - Lethality, and Mr Spencer.  His covering comment was: 

“Parts on the way, but the attached should give you a clear idea of the hammer mech modification.” 

  1. [145]
    From reviewing the Manual, Mr Eyles was aware that the Madritsch Solution used a double torsion spring as a separate AFL spring. The spring was illustrated several times in the Manual.  This was unlike the JGCM and V1, V2, V3, V4 and V5, each of which had a single torsion AFL spring. 

Madritsch Solution kits disclosed

  1. [146]
    On 13 September 2011, the nine kits and two sets of tools were delivered to Nioa in Brisbane.  From there, six kits and one tool set were delivered to Thales at Lithgow, addressed to Mr Eyles as Engineering Manager.  These arrived at Thales on about 14 September 2011.
  2. [147]
    I accept the evidence of Richard Basladynski, then Thales’ Testing and Evaluation Manager, as to what occurred when the kits arrived at the Lithgow factory.  It accords with the recollection of Mr Evenden.  I reject the contrary account of the same events given by Mr Eyles.  
  3. [148]
    When the kits arrived, Mr Basladynski and Mr Eyles spent some time examining them.  The individual parts were in an unassembled form.  They compared the parts in the kits with those in the standard hammer mechanism in the Austeyr to identify the changes Madritsch had made.  They identified that Madritsch had altered the mechanism so that the hammer springs no longer exerted any force on the AFL and instead the AFL was operated by a separate AFL spring applying significantly lower force.  In examining and handling the Madritsch Solution parts, it is more likely than not that Mr Eyles and Mr Basladynski identified that the torque exerted by the AFL spring was significantly lower than that exerted by the AFL spring in JGCM, V3 and V5.[26]  It is similarly likely that both Thales employees noticed the Madritsch AFL spring was a double torsion spring.  
  4. [149]
    Immediately the kits were examined and assembled, Mr Eyles instructed Mr Basladynski to organise and run live-fire tests. Mr Basladynski did so. He did so even though this was a particularly busy period for him and the other engineers at Thales.  
  5. [150]
    On 15 September 2011, Mr Fulford-Talbot told Mr Whitty the tests of the Madritsch Solution would happen about 29 September 2011
  6. [151]
    On about 29 September 2011, Thales conducted the testing at Lithgow.  According to a later draft report, there were two stages to the testing.[27]  In Stage 1, Thales tested and identified Austeyr rifles with the highest probability of exhibiting the bolt unlocking problem.  There were four rifles that consistently showed the problem: two Austeyr SA1s and two SA2s.  In Stage 2, Thales took rifle models from the production batches with “the highest probability of the failure mode” and “increased the sample size for greater statistical confidence”.  20 rounds were fired from each of the “high failure” rifles.  The number of rifles is not clear, as the detailed test results were not included in the draft report and, it appears, the report was never finalised.  
  7. [152]
    In the testing, the Madritsch Solution “proved to prevent the failure” in “all cases” and “the Madritsch hammer modifications seem to alleviate the bolt unlock failure mode.”  Based on the results of the tests, Mr Basladynski concluded that the Madritsch Solution “overcame” the bolt unlocking problem.   
  8. [153]
    An observer from Nioa, Greg Toms, attended the trial.  He reported to Mr Frohnwieser, Mr Whitty, Mr Madritsch and Mr Nioa that the two rifle models (Austeyr SA1 and SA2) were selected “for their routine failure” to fire the standard 5.56mm calibre cartridges after a 40mm grenade had been fired from an attached grenade launcher.  Each was fitted with a Madritsch kit, which had been randomly selected from five assembled kits.  Each was then loaded with a full magazine of 5.56mm cartridges. Ten 40mm grenade rounds were loaded to the grenade launcher.  
  1. [154]
    According to Mr Tom’s report, after each 40mm grenade was fired, “the operator would without any change of grip etc then attempt to fire” the 5.56 cartridge round from the rifle.  Each rifle was tested in this way from two different positions: one firing at a 50m target; then at 400m elevation angle. Every time the rifle fired its cartridge “without an issue”.  He noted: 

“The high speed video captured the bolt carrier sometimes moving back approx 1-2mm but every time the carrier closed fully on its own accord.” 

  1. [155]
    When the testing was completed, Mr Basladynski returned the assembled kits to Mr Eyles.   
  2. [156]
    On 3 November 2011, Mr Fulford-Talbot told Mr Whitty that: 

“early indications are that the solution proposed by Madritsch to solve the issue of the GLA followed by firing seems to have potential but we need to do more work to verify and qualify.”  

  1. [157]
    Two emails were exchanged that day about the preparation of a draft licence agreement for Thales to manufacture the Madritsch Solution.  
  2. [158]
    On 17 November 2011, Mr Fulford-Talbot told Mr Whitty that Thales had a “target date” of 18 January 2012 to “have an agreement established” for the Madritsch Solution.  Mr Fulford-Talbot described this as “a reasonably demanding timetable.”    
  3. [159]
    In his communications with Mr Whitty, Mr Fulford-Talbot made no reference to any similarity between the design of the Madritsch Solution and any of HMA1 versions then in existence or the “concept” of the JGCM and HMA1.  Although the kits had been examined and assembled by Mr Eyles and Mr Basladynski and subjected to livefire testing, Mr Fulford-Talbot did not propose that Madritsch and Thales “both register drawings” as he had mentioned in the Intentions Letter.  
  4. [160]
    Nor did Thales convey to Madritsch that Thales would not require a licence to manufacture the Madritsch Solution or that Thales had itself developed a sufficiently similar mechanism to solve the bolt unlocking problem that it could finalise and manufacture without the use of any information gained by examining and testing the Madritsch Solution kits. 

Thales reacts to the draft licence agreement – December 2011 to February 2012

  1. [161]
    The draft licence agreement did not emerge until 11:00 am on 19 December 2011, when Mr Whitty sent it to Mr Evenden.  By this time, Mr Evenden had taken over from Mr Fulford-Talbot as the Thales person responsible for dealing with Nioa about the Madritsch Solution and the Madritsch GLA in relation to L125-3C.  In form the draft was a “technology sub-licence agreement” between Nioa and Thales.  It assumed Madritsch would grant Nioa a licence with a right to sub-licence.  The technology the subject of the sub-licence comprised the five parts and three special tools listed in the schedule.  These were the parts in the Madritsch Solution kit and the special tool kit.  In an email to Mr Madritsch, Mr Frohnwieser, Mr Nioa and others, Mr Whitty predicted Thales “will probably choke on the price”.  This was a reference to the €700,000 licence fee and €20 per article production royalty in the draft sub-licence agreement.  
  2. [162]
    Before 2:30 pm that day, Mr Evenden had prepared a short analysis of the draft sublicence agreement and Thales’ options.  On his analysis, the draft agreement was “farcical”.  He raised other matters in the note, beginning with some background history: 

“We needed a fix for the issue of the rifle unlocking when a GLA is fired leading to a rifle stoppage. 

We had done some PV work to determine a fix via design change but the outcomes were not 100%.

L125 Ph3C Stage 2A had the rifle and GLA as a new requirement, we did not have the time nor budget to pursue a fix within schedule and price. 

Decision was to pursue Madritsch solution.

Under terms of NDA we tested Madritsch solution with 100% success, way forward was: 

  1. a.
    Qualify solution under L125 Ph3C Stage 2A using programmed S3 testing.
  1. b.
    Seek licence agreement with Madritsch to manufacture in Australia.
  1. c.
    NIOA is Madritsch in-country agent and Madritsch asked them to lead negotiations.”
  1. [163]
    Then, under the headings “THE BIGGER PICTURE” and “OTHER FACTORS”, he posed some questions, including: 

“If (when) we say no how do we prevent NIOA making a direct approach to Defence with solution? …

Time to market for alternative solution critical to prevent NIOA making an approach.  We (DA) must be the preferred provider but must also be timely.”

  1. [164]
    Lastly, Mr Evenden made some recommendations.  The first was “No reply to NIOA this year.” 
  1. [165]
    On 20 December 2011, shortly after 8:00 am, Josh Annells, the Thales Program Manager – Soldier Systems, circulated an email asking Mr Eyles to confirm that the AFL and three springs were the only parts of the hammer mechanism to be replaced under the Madritsch Solution.  Mr Eyles replied, identifying the four parts required for the “Madritsch Hammer Mech solution”.  He accepted that he probably identified these from the Madritsch Solution kits or the Manual.  
  2. [166]
    On 21 December 2001, Mr Whitty, responding to a question from Mr Evenden, informed him that Nioa had executed a head licence agreement with Madritsch.  The Technology Licence Agreement is dated 21 December 2011.  It gave Nioa an express right to sublicence the relevant technology to Thales.  
  3. [167]
    On 9 January 2012, at about 10:51 am, Mr Evenden had a short instant messenger conversation with Mr Eyles, in which he asked three consecutive questions: 

“Can you develop a solution instead of the Madritsch one for HMA in L125 timescales and budget?  

Can we speak briefly?  

Now?”

  1. [168]
    Mr Eyles recalled Mr Evenden speaking with him after this email.  Mr Eyles said, “I think he told me in that conversation that the terms Madritsch was asking for were not acceptable.”  In cross-examination, Mr Evenden denied this.  In an email sent that same day, Mr Evenden described the sub-licence agreement as “ridiculous” and “not acceptable”.  I reject Mr Evenden denial, and accept Mr Eyles’ account of the conversation because it is consistent with the contemporaneous documents.[28]  
  2. [169]
    18 January 2012 is the date of the first draft report by Thales about the 29 September 2011 live fire testing of Austeyr rifles fitted with the Madritsch Solution kits.  It was prepared, at least in part, by Mr Basladynski.  He said he knew from his own recollection that he drafted his part of the report relatively shortly after he undertook the testing at the end of September 2011.  He had no explanation for the absence of any earlier draft, or for the later date of these drafts.  Nor did he know whether the report was ever put in a final form.  
  3. [170]
    The draft report noted that the Madritsch Solution was fitted in Austeyr rifles selected as consistently repeating the bolt unlocking problem.  When these were tested, none of the weapons had the bolt unlocking problem and “In all cases the Madritsch hammer mechanism proved to alleviate the failure.”  The draft report included a recommendation that:

“Extensive trials on the hammer mechanism be carried out to ensure that the reduction of force on the AFL does not adversely affect the function of the rifle through environmental [changes?] to prove that this is an enduring solution.”

  1. [171]
    This confirms that Thales, in the person of Mr Basladynski, had observed and noted the “reduction of force” exerted on the AFL by the AFL spring before, at the time of the 29 September 2011 testing or shortly afterwards.  The reduction was significant enough for him to propose a recommendation for extensive trials.  
  1. [172]
    According to the draft, the report was to be approved by Mr Eyles.  It is not clear whether he did so.  

Thales work on V6 – February to June 2012

  1. [173]
    On 6 February 2012, Mr Eyles asked Richard Banks and Mr Higgins to add eight tasks under a subheading in a plan relating to L125-3C.  Mr Banks was Thales’ Manager, Project Management Office, Armaments and Mr Higgins L125-3C Project Manager–Lethality.  One of these new tasks was “HMA1 Revised Start 27/3/12 - Finish 26/3/12 – Undefined Resource-40hrs, Material $800”.  From the sequence in which this task is listed (between tasks commencing 3/2/12 and 12/3/12), and the oddity of a finish date before the start, I conclude that the “Start” date is a typographical error and should be read as “27/2/12”.  Responding to the request, Mr Higgins identified that this task was “NOT L125” and rather “PV”.  This reply seems to indicate that the Madritsch Solution had not yet been replaced by the HMA1 as the device to be included in the Austeyr to be manufactured under the L125-3C program.  That decision may have been about two weeks’ away.  See [178] below.  
  1. [174]
    On 15 February 2012, at about 6:50 pm, Mr Eyles sent an email to Mr Basladynski asking: 

“In your travels can you please pull together all the HMA1 .. latest version.  They are all marked with stickers on the side.  

I know we made a few and I need a minimum of 6 possibly 12 HMA1 for DATL.[[29]]  I only wish to modify minor parts in these.” 

  1. [175]
    In this instruction, Mr Eyles did not ask Mr Basladynski to collect any documents about the HMA1 Project and he said he did not do so.  He collected the V5, and perhaps some V3, mechanisms that had been assembled by Mr Williams in 2009.  As is clear from the instruction, Mr Eyles had already decided to modify parts of Mr Williams’ earlier versions.  
  2. [176]
    On 16 February 2012, Mr Eyles had an instant messenger conversation with another Thales employee, Colin Steele, in which he asked Mr Steele to let him know “before you get low on work”.  He said he had “some BD work and a trip or two to sunny Lithgow” for him.  The work was two tasks.  The first was “to take the lead on finalising the [HMA1] Hammer Mech.”  Mr Eyles said, “This means a couple of two day trips to Lithgow.”  
  3. [177]
    Mr Eyles asked Mr Steele to be on site at the Lithgow factory to see him the following Wednesday (22 February 2012) so he could “fill you in on the jobs.” 
  4. [178]
    On 20 February 2012, before Mr Steele’s first on site briefing, Thales made the decision to “Change to HMA1” for the hammer mechanism.  This is recorded in a note of the L125 Rifle Design Update, dated that day. 

The work of Mr Steele

  1. [179]
    Mr Steele worked for Thales as a technical officer/draftsperson from about April 2008 until about 2015.  He had a diploma of engineering from Regency TAFE in Adelaide (2001) and an Advanced diploma from RMIT (2009).  At the beginning of 2012, Mr Steele was working for Thales at its Melbourne office in the armaments division.  He was doing work relating to bullets and bombs at the time Mr Eyles selected him.
  1. [180]
    Mr Steele described his role at Thales as a “minion” or a “grunt” on the shop floor.  He explained that a “grunt” is “a term used in the Army to describe a soldier on the bottom rung of the ladder.”  He was not an engineer.  He had no professional or work responsibility for the implementation of any of the things on which he worked.  
  1. [181]
    Once seconded by Mr Eyles, Mr Steele worked either by remote access or by flyin/fly-out arrangement for the Lithgow factory.  He explained:

“The most common way in which I carried out my work in the 2012 to 2015 period was that the engineer or developer responsible for a project, or part of a project, informed me about design changes to an existing part or about new parts that were required.  These instructions were not a formal process.  Sometimes they were provided by email but other times they were provided in a conversation, or by telephone or in meetings. 

Generally, when Michael Eyles and I spoke in person, it was late in the day because I was staying in Lithgow and didn’t need to be anywhere.  Michael was also really busy and this allowed us more time as there were less interruptions. 

… Once I received the instructions I then prepared drawings and other necessary documentation to reflect the parameters I was given.  There were two purposes to that needed to be fulfilled by the drawings and documents that I created.  First, the drawings needed to document a design that met the purpose or needs of the engineer or developer.  The second purpose was to provide documentation so that the part could be manufactured.”

  1. [182]
    He reported to Mr Eyles for his work on the HMA1 Project.  As Mr Steele put it, Mr Eyles “looked at the drawings from a technical perspective (ie did it meet the design brief or solve the problem).”[30]  
  2. [183]
    At the time, Mr Steele had a general understanding that the ADF had identified a problem with the Austeyr caused by the launching of a grenade and had proposed a solution.  He had no intricate knowledge of the causes of the problem.  Mr Steele’s role was like that of Mr Williams in 2009.  He was to produce drawings of the parts specified by Mr Eyles that could be commercially produced.  
  3. [184]
    Mr Steele had little recollection of the work he did on the HMA1 Project.  He said, “it was not very significant to me.” He recalled being given “a pile of materials from the previous HMA1 work in progress that was to be the starting point for the work I was doing.”  These were physical documents, but he may have had access to electronic versions, he says, “where I had need of them.”  According to Mr Steele:

“I only ever received the information necessary for me to undertake the specific task assigned to me.  Obviously, I could and would ask for more information if I thought it was necessary and to the extent I did need to know, I would be given it, but I also can recall several times being told that I did not need to know what it is I asked for.”

  1. [185]
    On 27 February 2012, at about 5:10 pm, Mr Eyles sent a spreadsheet to Mr Steele.  It set out each item required to build the proposed F90 model of the Austeyr for L1253C, assigning an item description, drawing number, and quantity, and identifying it as a “Make” or a “Buy” with a “Vendor”.  For the AFL and the two hammer springs (left and right), the vendor was “HMA1”.  The AFL was a “make” with a work program between about 5 and 19 March 2012.  The hammer springs were a “buy”.  The AFL spring was a “buy” and the vendor was “Gibbins” with a work period from 20 February and 19 March 2012.  None of these items had a drawing number.
  1. [186]
    Mr Eyles asked Mr Steele to be in Lithgow on Wednesday 22 February 2012 to be briefed on the HMA1 work.  Mr Eyles had asked the HMA1 work to be scheduled to start on 27 February 2012.  He sent the spreadsheet to Mr Steele after 5:00 pm on Monday 27 February 2012.  The first document created by Mr Steele is an email he sent at 10:15 am on Tuesday 28 February 2012.  Mr Steele recalled that he took apart the assembled V3 and V5 “packs” to see how they worked.  It is likely he reviewed some documents to ascertain the state that work on the HMA1 Project had reached in Mr Williams’ time.  From these facts, I infer that he started work on the HMA1 Project between 22 and 28 February 2012.  It is more likely he started work on the drawings towards the end of this period.  It makes no difference to the determination of the present dispute between the parties whether he began work on 22 or on 28 February 2012.  
  1. [187]
    As Mr Eyles observed, Mr Steele’s experience was with compression springs and he did not understand torsion springs.  Mr Eyles gave Mr Steele access to an online spring calculator by the Institute of Spring Technology (IST) and a hard copy of an “Encyclopedia of Spring Design” by the Spring Manufacturers Institute.  He recommended Mr Steele use them to help with his work on the HMA1 Project. 
  2. [188]
    Mr Steele had access to the prior work of Mr Williams, including the drawings and assembled test models of V5.[31]  Mr Steele recalls looking at the models Mr Williams had produced in 2009.  He noted the AFL moved slightly off its plane because pressure was not being evenly distributed through the AFL by the AFL spring.  
  3. [189]
    The first drawing to show a double torsion AFL spring is SK-108250 dated 28 February 2012.[32]  Mr Steele sent the drawing to Jason Egan of Gibbens at 11:43am on 28 February 2012.  At about 10:15am that day, in his first contact with Mr Egan, Mr Steele told him that “my boss” Mr Eyles had provided Mr Egan’s details.  

Madritsch KG v Thales Australia Ltd [2021] QSC 170

Figure 4: Double-torsion AFL spring drawn by Mr Steele for HMA1 V6

  1. [190]
    Mr Steele was keen to take credit for the double torsion spring in the design for what would become HMA1 version 6 (V6).  His initial explanation proved inaccurate.[33]  In a second statement, he recalled being prompted to revise the design while hanging out washing when he saw a spring inside a clothes peg.  He did not know where this happened or, it seems, when, save it must have been before 28 February 2012.  
  2. [191]
    At the hearing, the more qualified engineering witnesses[34] identified a double torsion spring as an obvious solution to the concern the Commonwealth had raised about the AFL almost three years earlier in May 2009.  Any of the qualified and experienced engineers in Thales would have been able to identify this solution, without looking to laundry duties for inspiration.  Mr Steele was not among them.  In fact, he did not have an understanding of torsion springs, single or double.
  3. [192]
    In Mr Eyles, Mr Steele had an available source of advice, who could readily identify that he should replace the single- with a double-torsion spring for the AFL.  Both Mr Eyles and Mr Basladynski were aware that Madritsch had used a double-torsion AFL spring.  Mr Eyles had recently identified it for Mr Annells, from the Madritsch Solution kit or the Manual. 
  4. [193]
    There were illustrations of double torsion springs in the encyclopaedia Mr Eyles gave Mr Steele, but Mr Steele disavowed it as a source of the idea.  Mr Eyles denied using the book to point out a double torsion spring to Mr Steele.  
  5. [194]
    In the documents tendered at the hearing, the first reference to a similarity between the V6 AFL spring and a clothes peg is in Mr Eyles’ email to Mr Evenden of 12 June 2012. This was sent late at night and before it had been discussed with Mr Steele.  (See [248] to [251] below).  In none of the contemporaneous documents did Mr Steele make any such reference. 
  6. [195]
    Mr Steele is unlikely to have been inspired by a common clothes peg, which has a single torsion spring, not a double.  Having observed his generally poor recollection of the detail of his work in early 2012, I conclude that his revised recollection is more likely reconstructed to defend or reinforce his initial erroneous recollection, and  not an actual memory of something that occurred.  
  1. [196]
    In this first issue of drawing SK-108250, Mr Steele specified the dimensions, material,[35] final condition and “applied force” for the AFL spring in V6.  The last of these was as follows: 

“APPLIED FORCE: 22N @ 47⁰ NORMAL CONDITION”.  

  1. [197]
    In instructing him to prepare a quote for manufacture, Mr Steele told Mr Egan: 

“The only fixed conditions I have at the moment is the Force in its normal condition and the external size.  Wire Dia, # of coils, material is all flexible and can be changed to suit to meet the force requirement.”[36]

  1. [198]
    In other words, within a day or a few days of commencing work, Mr Steele specified an applied force measure of 22N as one of only two fixed conditions for the AFL spring in V6.  He also described it as the one requirement that the other relevant parameters (save for external size) could change to accommodate.  
  2. [199]
    Mr William’s prototype models, parts and working papers were not the source of the 22N value.  In seeking to replicate the JGCM, Mr Williams had fixed on a force of about the 26.45N he had measured as exerted by the AFL in the JGCM.  He had not measured or specified any “force” for the AFL spring.  He had specified a torque for the spring, as was appropriate, which he calculated or estimated would result in the required disengagement force.  
  3. [200]
    There is an oddity about Mr Steele specifying an “applied force” of 22N in his first drawing of the V6 AFL spring.  As a torsion spring, it does not have such a property.  As noted above, when twisted along its axis, a torsion spring exerts a torque (or rotational force) in the opposite direction to the twist.  Torque may be measured in Newton-millimetres (Nmm).  It is not a force, such as the force exerted by a compression spring or an extension spring, which may be measured in Newtons (N).[37]    
  4. [201]
    In his first statement, signed on 27 February 2020, Mr Steele explained: 

“As my role was as a draftsperson, rather than an engineer, to the extent that engineering work regarding force calculations was done, my role was to ensure that the parts I drew performed to the forces required. … I was not the person who performed those initial force calculations.  I believe that the person who was responsible for those calculations was Michael Eyles.”  

  1. [202]
    On the eve of the hearing, Mr Steele signed a third statement.  In it, he dealt with something he thought he “had not remembered” when gave his first statement.  It was that he tested and recorded the “vertical force results” for V3 and V5 “probably within a few days after receiving them.”  Mr Steele says the record of these forces is in a table (Table 1) in part 4.1 of a draft report he created more than two months later, on 1 May 2012.  Thales produced no earlier record of them to the court.  
  1. [203]
    The 1 May 2012 draft report was prepared by Mr Steele to be reviewed by Mr Basladynski and approved by Mr Eyles.  According to Mr Basladynski, it would then be forwarded to the ADF for consideration.  He did not recall if it was finalised and, if it was, he did not recall seeing a final report.  
  2. [204]
    In its introduction, Mr Steele described the “scope” of the draft report in this way: 

“This report details the results of the Certified Grenade Firing Hammer Mechanism trial (TI-12L016) conducted at the Lithgow Small Arms Test Facility on April 19 & 20 2012.” 

  1. [205]
    Section 4 of the draft is entitled “TRIAL RESULTS”.  Within it, in part 4.1 headed “Spring Force”, Table 1 appears.  In explaining the table, Mr Steele wrote: 

“The proposed solution was to reduce the amount of force required to disengage the AFL [from] the hammer.[38]  The vertical force on the

Hammer Group Assemblies HMA1 V3 No 9 and HMA1 V5 No 21

(developed in earlier studies) were used as a baseline, along with a new design introducing a “lighter” spring into the AFL Drawing No SK-108204 Rev A.  The results are tabulated below.” 

  1. [206]
    Here is Table 1:

 Table 1:  Force Comparison

Sequence

V3

V5

SK-108204

1

44.2

44.3

14.5

2

45.4

43.6

15.6

3

43.8

44.2

16.6

4

43.9

43.4

14.8

5

42.6

44.2

14.8

6

43.6

42.8

15.2

7

44.2

41.4

14.9

8

43.4

42.4

15.1

9

44.2

42.4

15.6

10

42.4

41.4

14.8

11

41.8

41.6

14.9

12

42.8

42.2

15.1

13

43.1

41.4

15.0

14

42.4

42.4

14.7

15

43.0

40.4

14.8

16

41.2

42.5

14.9

17

42.2

42.7

14.8

18

42.4

41.6

14.9

19

43.6

42.1

15.0

20

43.1

42.8

14.9

  1. [207]
    It is a sequence of 20 measurements in Newtons for each of three versions of HMA1.  Mr Steele’s more recent recollection was that the first two measures in each sequence were made in the first few days he worked on the HMA1 Project, i.e. before or about 28 February 2012.  The third in each sequence is a measurement of the same force in V6, which cannot have been made until after 16 April 2012, when the springs for V6 had been manufactured by Gibbens and delivered to Thales.  See [235] below.
  2. [208]
    Mr Steele was not sure whether testing the vertical force of the AFL in V3, V5 and V6 was his idea or someone else’s.  He did not measure the torque of the AFL spring.  He said he did not have immediate access to specialist machinery “to test a spring just for its torque value”.  
  1. [209]
    The position is not clarified by Mr Eyles.  In a second statement, he said: 

“It was not part of my design verification process to measure the upward force exerted by the AFL in the Thales hammer mechanism in the F90 [model of the Austeyr rifle] so I am not aware of what upward force the current Thales AFL spring produces when assembled in the hammer mechanism.” 

  1. [210]
    Mr Eyles was Mr Steele’s direct supervisor.  He was the person to approve the test report on the V6.  This comment by Mr Eyles is an indication of the unsatisfactory nature of his evidence, considered at [257] to [267] below.  
  2. [211]
    At the hearing, it was common ground that Mr Steele’s “vertical force” in Table 1 was more accurately described as the disengagement force[39] applied to the gun lock group.  Mr Wachsberger examined the figures in Table 1.  He calculated the average disengagement forces measured by Mr Steele for the three versions of HMA1, as follows:

Mechanism

Average disengagement force at AFL spring (N)

V3

43.2

V5

42.5

V6

15.0

  1. [212]
    Based on Table 1, Mr Wachsberger concluded the AFL spring used in V6 was “considerably weaker” than the springs in V3 and V5.  I accept that conclusion.  It matters not whether the measurements were made in late February or any time before 1 May 2012.  It matters not whether the measurements were accurate as to the actual disengagement forces.  Importantly, they show that, from his first drawing of V6, Mr Steele set out to significantly reduce the disengagement force exerted by the AFL on the gun lock group from that in V3 and V5.  Mr Steele thought he had reduced the disengagement force by about 65% from that in V3 and V5.  Although the specifications in Mr Williams’ drawings of the AFL spring in V3[40] and in Mr Steele’s later drawings of the AFL spring in V6[41] might indicate a significant reduction in torque, the different angles for each torque specification make direct comparison inappropriate.[42] 
  1. [213]
    In his third statement, Mr Steele also recalled: 

“When I was tasked with refining the HMA1 model I recall a conversation with Michael [Eyles] who said something like ‘this still needs fixing’.  The specifics of why he had that view he didn’t say and was actually not my concern.  My task was to refine the HMA1 prototypes so that they could be productionised (meaning easily manufactured).” 

  1. [214]
    Mr Steele sought to take responsibility for the lower disengagement force in V6.  He said: 

“I saw no reason to keep the same AFL disengagement force that was in the HMA1 versions [V3 and V5], if reducing the force would solve whatever was wrong with the AFL.

… Once I had a number [from the test], I could halve it and have something to work towards in the spring design.

Once I had taken those measurements I took 22N as the ‘known force’ on my 28 February 2012 drawing which was a rough half the average of all the vertical results from V3 and V5, which was about 44N (rounded). 

… Because there was no particular force required to be on the AFL to solve the bolt unlocking problem, I just proposed a force of half of what I had measured as a starting point to test my new design.” 

  1. [215]
    Mr Steele explained that he knew “no particular force” was required because he had read the report on the March 2011 test (see [67] above), which he thought “proved that reducing the force in the AFL by including a separate spring solved the [bolt unlocking] problem” as well as “removing the AFL altogether.”  Yet, he said he was trying to “solve whatever was wrong with the AFL”, and doing so by further significantly reducing the disengagement force. 
  2. [216]
    During cross-examination, unprompted, Mr Steele explained: 

“I’m just trying to recall when I was told they thought the forces were … still too high, … which led to a lot of the work that I did, otherwise I wouldn’t have done it at all.” 

  1. [217]
    For the plaintiffs, Mr Cooper QC pursued this explanation in subsequent questions: 

“MR COOPER QC: So do I understand that answer to be that at some point you were told that the forces were still too high?

MR STEELE: The – that’s – that was my interpretation of – my feel for the – the designs that I were given, being V3 and V5 that I was handed, that I refer back to in my statement, yes.

MR COOPER QC: Was that something that Mr Eyles discussed with you, the forces still being too high?

MR STEELE: I have – if I have a – a very vague recollection of – my feel the AFL in it, that was – it’s – if it’s the slide or the breech, or something, that’s hitting it and it’s stopping it from closing, but is it the forces being too high.  And that’s, sort of, what got us looking back through the – through the springs again.  Yeah, it’s sort of how it all came about. 

MR COOPER QC: … My question was - do you recall Mr Eyles discussing that with you, the forces still being too high?

MR STEELE: I recall talking to Mr Eyles about the forces being too high.  I can’t recall whether it was specifically Mr Eyles who raised it in the first instance, or whether it was just a, ‘Is this the reason?’  Like, is it a question that I asked him. 

MR COOPER QC: Do you agree with me that what you say you did in your third statement is not consistent with the process you describe in paragraphs 41 and 42 of your first statement? …

MR STEELE: The decision to halve … the value, yes.  And I didn’t do the force calculations as I stated in statement paragraph … 41 and 42.  Yep.  … My role wasn’t specifically to design.  My role was to primarily take designs done by the engineers to productionise them. 

MR COOPER QC: Yes - - - ?

MR STEELE: To make them. 

MR COOPER QC: … what I’m suggesting to you is that you wouldn’t have taken a step of reducing the force of the AFL spring you were designing from - from where it was at previous versions without guidance from Mr Eyles as the engineer?

MR STEELE: Most of the time, yes.  Yep. 

MR COOPER QC: Well. Mr Eyles was the person setting the parameters which your design had to meet; correct?

MR STEELE: Yeah correct. 

MR COOPER QC: All right.  So, while you might not recall the specifics of a conversation, I suggest that Mr Eyles must have been involved in the decision to reduce the spring force of this – the AFL spring you were – you were working on.  Do you accept that?

MR STEELE: Yeah, he ultimately approved the change. 

MR COOPER QC: Well, what I’m suggesting is that he was involved in the decision, not – not simply approving the change after it had been made.  He was involved in the decision?

MR STEELE: Yes.” 

  1. [218]
    On 29 February 2012, between 10:30 and 11:20 am, Mr Eyles and Mr Steele exchanged emails.  Mr Steele told Mr Eyles that he had completed the design of the AFL spring.  He attached the calculations from the IST online spring calculator and drawing SK-108250 of the AFL spring with the 22N applied force specification.  Mr Eyles responded:  “I need to spend a few minutes with you later this afternoon 5:00PM?? For about 20 min.  Is this OK.  I just want to get an overall appreciation of what you have done.” 
  2. [219]
    On 6 March 2012, Mr Steele was again working on the HMA1 Project.  He told Mr Egan that: 

“Part of this project of mine is to re-evaluate the Hammer Springs.  Part of this current design has a bend in one of the legs, that now is in a different position.  I have attached a PDF detailing the design I’m thinking of reverting back too [sic].  However I need to confirm that this meets the current spring forces of the existing spring.

The 1.4 wire meets the current 450Nmm however I would like to use 1.5mm wire, and just reduce the total number of coils???

Can you please confirm what the spring rate is on our current Hammer Springs?  And if we can use a ‘straight leg’ design when we have nearly 170 deg of movement?” 

  1. [220]
    Mr Egan responded: 

“If we increase the wire and reduce the number of coils the rate will increase significantly.”  

  1. [221]
    Mr Steele then sent Mr Egan “the IST calcs” for 1.4mm Japan steel.  He explained that the design he was working on was “close to the current design of 450Nmm”.  Mr Steele asked: 

“Being that my knowledge in spring design is limited, can you please tell me if the ‘bent leg’ we currently have in our hammer spring lowers the torque value when the spring is compressed (when compare to a spring without a bent leg)”?

  1. [222]
    Mr Egan replied: 

“The main issue with loss of theoretical torque load in this spring is due to the spring distorting / buckling under the load as there is considerable preload.  We test the spring the same way as you have it installed so the loads are suitable for your end application.”

  1. [223]
    Although these communications are about the hammer springs, not the AFL spring, they indicate the limited knowledge and understanding Mr Steele was able to apply to the task and the extent of his reliance on input from others.  This was consistent with his own assessment of his role as within Thales as a draftsperson, not an engineer, and his lack of understanding of torsion springs. 
  2. [224]
    On 6 March 2012, Mr Steele produced three drawings: SK-108255 of the AFL; SK108256 of the left hammer spring; and SK-108257 of the right hammer spring.  Mr Eyles approved all three.  Mr Steele added two torque specifications on each of the hammer spring drawings.  They were the same for left and right: 

“TORQUE: @ 92⁰ ± 0.5⁰ = 280 Nmm ± 25 Nmm

TORQUE: @ 181⁰ ± 0.5⁰ = 450 Nmm ± 25 Nmm”

  1. [225]
    From this it seems, by 6 March 2012, Mr Steele had grasped that torsion spring specifications should be torque in Nmm and not “applied force” in N. 
  2. [226]
    On 13 March 2012, Mr Steele asked Mr Eyles if he could come to see him “regarding spring calcs”.  This was the day Mr Steele completed another revision of drawing SK108250 of the AFL spring.  Mr Eyles approved the changed drawing.  On this drawing Mr Steele deleted the previous reference to the “Applied Force” of “22N @ 47⁰ NORMAL CONDITION”.  Instead, he included the following under the spring specification heading: 

“TORQUE: @ 35⁰ ± 0.5⁰ = 40 Nmm ± 5 Nmm

TORQUE: @ 25⁰ ± 0.5⁰ = 80 Nmm ± 5 Nmm”

  1. [227]
    Sending this drawing (and drawings of the two hammer springs) to Mr Egan, he explained that the “design requirement has changed slightly and I have had to redesign them with a different wire.”  In fact, the specified wire was unchanged.[43]  
  2. [228]
    The significant change in the drawing was from a specified “applied force” of 22N to a torque of 40Nmm and 80Nmm at particular degrees of torsion.  Mr Steele had no recollection of when he made this change.  He recalled that “Gibbens actually told me what the final torque values would be of the springs they would manufacture by making these changes.”  The final torque values were the same as the spring specifications in this 13 March 2012 edition of SK-108250.  
  3. [229]
    On 20 March 2012, Mr Eyles engaged Mr Steele in an instant messenger conversation about progress with the hammer springs and AFL spring. During the conversation, Mr Eyles spoke by telephone with Gibbens to expedite a quote to manufacture the test springs and, in the event of a delay with Gibbens, tried to arrange for the Lithgow factory toolroom to manufacture the springs.[44]  
  4. [230]
    That evening, Mr Steele asked Mr Eyles to confirm when the HMA1 Project needed to be completed, including testing.  His email was copied to Mr Evenden.[45]  The next day, Mr Eyles responded: 

“HAM1 [sic]…Simple…ASAP…Assembled by mid April (Say 16th) for testing in the week 16 to 20 April.  This permits installation in DATL weapons by end April.  Testing???? Graham [Evenden], Richard [Basladynski] Comments please.” 

  1. [231]
    On 21 March 2012, Mr Evenden replied copying his email to Mr Basladynski: 

“Rich, have a look at loading and let me know what’s possible.  HMA1 testing content is to be repeat of what you did with Madritsch solution this then puts us in the same place we would have been had we taken their solution forward (assuming HMA1 works).”

  1. [232]
    Mr Eyles replied, “It will work…the same as Madritsch”. He expressed he was “Confident” and that “Repeatability of the solution will be now more related to the rifle Build quality.” 
  1. [233]
    On 21 March 2012, Mr Steele produced another revised drawing SK-108250 of the AFL spring. The specifications were for torque at new angles:

“TORQUE: @ 145⁰ ± 2⁰ = 40 Nmm ± 5 Nmm

@ 155⁰ ± 0.2⁰ = 80 Nmm ± 5 Nmm”

  1. [234]
    Mr Eyles approved the changes to the AFL spring and hammer springs designs prepared by Mr Steele.  These required Mr Eyles’ approval that “the new torque values have been verified and are suitable for the design intent.”  The changes were related to manufacturability.  Some had been suggested by Mr Egan.  Mr Eyles identified one explanation in these terms: 

“The increase in tolerance for torque angles, along with the increase in the bend radius of the wire allow for a more achievable manufacturing process with a lower rate of rejection (thus lower cost/unit value).”   

  1. [235]
    Gibbens manufactured the springs.  They arrived at Thales Lithgow in two batches.  The AFL springs arrived on about 16 April 2012 and the Hammer springs later.  Mr Steele arranged for a small number of AFLs to be manufactured by a Victorian foundry and brought them with him to Lithgow. 
  2. [236]
    On 19 and 20 April 2012, Mr Basladynski conducted a small-scale test of the Austeyr fitted with the V6 mechanism.[46]   He was assisted by Mr Steele.  
  3. [237]
    On 1 May 2012, Mr Basladynski sent Mr Steele a draft technical instruction for the 19 and 29 April 2012 testing.  It bears a preparation date of 18 April 2012.  It was prepared using a draft for the 29 September 2011 Madritsch Solution testing.  
  4. [238]
    Mr Steele revised the draft.  As noted at [203] to [207] above, this draft contains the first detailed information about the disengagement force exerted by the AFL in V6 on the gun lock group compared to the same force exerted by the AFL in V3 and V5.  Oddly, the draft report does not include detailed results of the live-fire test undertaken on 19 and 20 April 2012. 
  5. [239]
    On 9 May 2012, Mr Eyles told Mr Higgins that V6 would become the “new baseline” for the L125-3C model of the Austeyr.  In response to Mr Higgins’ question, “What design reviews have there been for it?” Mr Eyles replied: 

“My self.  Colin [Steele].  Have you met the force requirements, for the hammer and selected AFL force, manufacturable, Assemblable …   All the rest is the same”

  1. [240]
    In cross-examination, Mr Eyles explained that Mr Higgins was “looking for a formal design review” and “there really wasn’t one done.” 
  2. [241]
    On 8 June 2012, Mr Eyles gave Mr Steele and Mr Higgins a “Heads up” warning: 

“We will have to allocate time to developing a report to clearly formalise the start point and the design rational used to achieve the end result.  We will have to get the information tabled from [Graham Evenden][47] that covers [us] commercially as well as review the work carried out previously.” 

  1. [242]
    This topic arose during an email exchange about whether to make further revisions to Mr Steele’s drawings to accurately record the Victorian-cast AFLs used in the 19-20 April testing or whether to create wholly new drawings with new drawing numbers.  
  2. [243]
    It was a timely consideration. 

Paris trade show June 2012

  1. [244]
    Between 11 and 15 June 2012, Thales displayed an Austeyr, fitted with the V6 hammer group assembly, at its trade stand for the Eurosatory International Defence and Security Trade Show in Paris.  
  2. [245]
    Mr Madritsch and Mr Frohnwieser attended the show, where they saw the Austeyr in the Thales display.  A Thales representative at the show disassembled the rifle in their presence.   Mr Madritsch and Mr Frohnwieser became concerned that Thales had incorporated the Madritsch Solution in the rifle’s hammer mechanism.  They contacted Mr Whitty and, through him, Mr Nioa.  
  3. [246]
    On 12 June 2012, Mr Nioa asked Mr Evenden and Kevin Wall, who were also attending the show, to meet with him the next day.  Mr Wall was Thales Vice President – Armaments.
  4. [247]
    At that time, Thales had not returned the six Madritsch kits it had received ten months earlier in September 2011.  Thales would not return them for another four months.[48]  
  5. [248]
    Mr Madritsch’s concern about Thales copying from Madritsch technology without permission drew a “HIGHLY SENSITIVE” email from Mr Eyles to Mr Evenden at about 11:15 pm on 12 June 2012.  I infer it was prepared and sent in advance of Mr Evenden’s meeting with Mr Madritsch and Mr Nioa scheduled for the next day.  
  1. [249]
    Mr Eyles began his email by stating, “I do not know what details have been given over to secure our position.”[49]  Next, he cautioned that the bulk of the balance of his email was a “Draft only” and would be discussed with Mr Steele and Mr Williams “to assure facts.”  On that basis, Mr Eyles then set out his account of what he called “Evidence on why we reduced AFL spring load??”  
  1. [250]
    It is interesting, from the first challenge by Mr Madritsch, that Mr Eyles identified the reduction in the force exerted by the AFL spring as the key issue.  The topic does not seem to have been raised by Mr Madritsch or Mr Nioa before or at this time.  
  2. [251]
    In his abbreviated summary, Mr Eyles makes the following points: 

“1.In late 2008, early 2009 ?? DMO tabled HMA1 (V1) design to Thales.  …  The Auto Fire Lever (AFL) was independently sprung by an additional single torsion spring reducing the force to acuate by approximately 60% …

  1. Thales redesigned the attachment methods and spring form (HMA (V3) to improve the manufacturability.  …  The modification to the AFL and spring remained unchanged. 
  2. October 2009 HMA (V5) was developed to overcome the potential of the hammer spring legs rubbing on the side of the AFL.  …  No changes to the AFL and AFL springs were made.  Somewhere in the vicinity of 12 prototypes were made for design testing. 
  3. Late 2009 testing was conducted on HMA (V5).  …  The results indicated F88 CAS2 weapon worked but the SA2 2009 build weapons proved inconsistent and the testing was ceased 23rd December 2009 with the SA2 weapons failing to operate correctly.  …
  4. During 2010 further pendulum impact tests showed that the only method to assure greater probability of return to battery (short of locking the slide mechanically) was to further reduce friction in the system and AFL.  No further testing was conducted using SA2 2010 weapons as Madritsch solution was to be tested.

In regards to the current design we have reverted to HMA1 V3 location of the hammer springs, and changed to a double legged torsion spring (similar to a clothes peg) inside the AFL.  …

In principle, the method used is the same for all designs i.e. to reduce the force applied by the AFL to the slide during the final stages of closure.”

  1. [252]
    Point 5 of Mr Eyles’ summary is curious.  
  2. [253]
    Mr Eyles gave evidence about the pendulum test results in his statement.  The tests did not show the only method to assure greater probability of solving the bolt unlocking problem was a “further” reduction in the “friction in the system”.  Mr Eyles did not know which version of the HMA1 was used in the first pendulum test but said the Austeyr rifles tested had no bolt unlocking problem failures during the test.  According to Mr Eyles, in the second pendulum test the Austeyr with V5 showed improved performance over the Austeyr with a standard hammer mechanism “by allowing the slide to return faster.”  The March 2011 live fire testing, according to Mr Eyles, did not reflect the results of the earlier pendulum tests and showed the then current model of the Austeyr “still had the Bolt Unlocking Problem because it still had out of battery instances when man fired.”[50]  Elsewhere in his statement, Mr Eyles maintained that the tests on V3 and V5 proved that they solved the bolt unlocking problem and needed only minor manufacturing-related changes.  
  3. [254]
    Considering this evidence and the contemporaneous documents, I am satisfied that Mr Eyles’ point 5 had no factual basis.  In cross-examination Mr Eyles resisted that proposition, quite ineffectively.  In the end he said, “I still don’t know how to answer it.”  The only satisfactory answer is that point 5 was a false explanation for the deliberate and significant reduction in the torque exerted by the AFL spring with the purpose of significantly reducing the disengagement force applied by the AFL from that in V5 to that in V6.  
  4. [255]
    On 13 June 2011, at the Paris meeting, Mr Evenden told Mr Nioa and Mr Madritsch that for some time Thales had been developing its own “modified hammer pack” as a solution to the bolt unlocking problem.  Mr Evenden said the solution to the problem would always be the same.  
  5. [256]
    He showed them a Powerpoint® presentation explaining that although the Madritsch Solution had been “assessed with good outcome”, it was “cost prohibitive” with the result that “Thales reverts back to own solutions (V3).”  This was a misleading statement, at least to the extent that it concealed the fact that the AFL spring in V6 was significantly different to that in V3 in that it was a double torsion spring that caused the AFL to exert half or less the disengagement force in V3.  In these respects, it was much closer to the spring in the Madritsch Solution.  It might have been more accurate for Mr Evenden’s presentation to read “Thales reverts to Madritsch Solution.”  

Conclusion on Mr Eyles’ evidence

  1. [257]
    Overall, Mr Eyles was a dissembling, obstructive and evasive witness.  
  2. [258]
    At the outset of his oral evidence, Mr Eyles swore the contents of his two written statements to be true.  His cross-examination lasted about three hours.  Its length was due to his manner as a witness.
  3. [259]
    On several occasions, when asked a question, he paused for a long time while he flicked through and sight-read parts of his first, longer statement, before asking for the question to be repeated.  When it was repeated, he read out a paragraph, sentence or part of a sentence from his statement.  Often, he read passages that were not responsive to the question.  
  4. [260]
    For much of the cross-examination, he declined to agree with any propositions put to him.  When, after such an answer, Mr Cooper QC took Mr Eyles to a passage in his statement to the same effect as propositions he had denied – sometimes in the exactly the same words – Mr Eyles disavowed his own statement rather than agree with the cross-examiner.  Mr Eyles sought to justify these departures on the basis he could not check what was in his statement, so he could not be certain that it was correct.  
  5. [261]
    At other times, he was reluctant to depart from his statement even when shown that, for example, a date or time period was incorrect.  Having agreed that the pendulum tests and live-fire test on V3 and V5 were between December 2010 and March 2011, and not “During December 2011 to March 2012” as he said in the introduction to paragraph 7.5 of his statement, Mr Eyles was asked to accept that the different dates in his statement were a mistake.  He answered, “I honestly don’t know how to answer that one.”  He was not a stupid person.  He knew exactly how to answer.  He was being obstructive. 
  6. [262]
    Mr Eyles’ written statements were false in a number of respects.  Among the most important were those parts in which he sought to minimise or deny his involvement with the Madritsch Solution kits.  He said he “had very little to do” with the Madritsch kit parts, which “I think came across my desk” and saw “a brief picture of the Madritsch Solution that indicated that Madritsch had put a different spring into an Austeyr rifle, which reduced the load.”  When pressed, he said he could not recall whether he saw the kits at all.  He said he was “not involved in the assembly” of the kits, had “nothing to do with the testing” and “was at arm’s length of all of that.”  He said he “was not interested in how the Madritsch Solution worked”.  I am satisfied, from the whole of the evidence, that each of these statements by Mr Eyles was false or deliberately misleading.  
  1. [263]
    He also sought to minimise his involvement with Mr Steele’s work on V6.  He said he did not know whether Mr Steele performed the force and torque calculations for V6.  He said he was expecting Mr Steele, who had no prior experience and did not understand torsion springs, to “learn off” the spring encyclopaedia how to do force and torque calculations.  He described Mr Steele as a “detail designer” who was “capable of all those sorts of things.”  He denied that he continued to give Mr Steele direction on the design of V6, including the springs.  He denied responsibility for the force and torque specifications in the drawings he approved.  He avoided answering to the suggestion he was the responsible engineer for the HMA1 Project, describing himself as the “engineering manager.”  In Thales’ very flat workforce structure, with Mr Steel reporting directly to him, Mr Eyles described his role as merely “the manager, directing all the traffic, on all the projects.”  
  2. [264]
    His denial and minimisation of knowledge of the Madritsch Solution and of his direct involvement in Mr Steele’s work, was the foundation of a more significant falsehood.  It is in the first sentence of paragraph 11.2 of his first written statement: 

“None of the people who had access to the Madritsch Solution were involved in the development work of the HMA1 Project after this time [16 February 2012].”

  1. [265]
    Although it was obviously untrue, Mr Eyles was visibly reluctant to accept that conclusion.  After a lengthy period of cross-examination, Mr Eyles eventually accepted that he had access to the Madritsch Solution kits and the Manual and that he continued to be involved in the development work on the HMA1 Project after his 16 February 2012 instant messenger conversation with Mr Steele.  
  2. [266]
    Frankly, Mr Eyles’ attempts to minimise his knowledge and role were without credibility.  If accepted, it would follow that a major supplier of armaments for the ADF left the design of a solution to a significant operational issue in the standard service rifle to a draftsperson without professional engineering qualifications.  
  3. [267]
    His persistent evasiveness in answering questions, his refusal, at times, to adopt parts of his own written evidence, and his reluctance to concede when parts of his statements were obviously wrong, rendered Mr Eyles’ testimony confusing, to say the least.  His false or deliberately misleading passages on the most important matters in issue render his written statements broadly unreliable.  Owing to these matters and his evident uncertainty as to the correctness of parts of his statements, I have not accepted Mr Eyles’ prepared evidence where it is uncorroborated by a contemporaneous record or the evidence of another witness.  I have been similarly circumspect with his oral evidence. 

The expert witnesses

  1. [268]
    At the hearing, the parties called two expert witnesses on Thales’ development of V6 (and its predecessors) and its relationship to the Madritsch Solution.  Each party tendered their expert’s written reports.  A joint report was also tendered.[51]
  2. [269]
    The plaintiffs also tendered a report on a matter of Austrian law.  .  It is convenient to consider that report when dealing with an issue about whether the Madritsch Solution was confidential when it was disclosed to Thales.

Mr Wachsberger  

  1. [270]
    The plaintiffs’ relied on four reports of Christian Wachsberger.[52] He holds Bachelor of Applied Physics from the University of South Australia (1981)[53] and is a member of the Australian Institute of Physics.  Mr Wachsberger worked for the Defence Science and Technology Organisation in its various forms for 33 years (June 1983 August 2016) specialising in gun systems.  He served as small arms subject matter expert during the early stages of the L125-3C program.  He was not involved in identification of or solution to the bolt unlocking problem.  
  2. [271]
    Mr Wachsberger accepted that an experienced design team should be able to conceive several possible solutions to the bolt unlocking problem and grade them to identify the most suitable within a relatively short time.  This could be done by such a team within the time taken by Mr Steele to finalise his drawings of V6, i.e. between about 22 February and 20 April 2012.  
  3. [272]
    However, there were two important impediments to this being the explanation for Mr Steele’s progress with the V6 in 2012. 
  4. [273]
    The first, and most obvious, was that Mr Steele, by himself, did not comprise an experienced design team.  His work was explicable only if he was directed by others with superior knowledge, qualifications, and experience, who were checking and taking professional responsibility for it.  Principal among them Mr Eyles.  
  5. [274]
    Second, in 2012, Thales’ knowledge, from the limited testing it had performed, was that V5 worked as well as removing the AFL from the standard mechanism.  There is no logical explanation for the decision to specify for the V6 AFL spring an applied force one half of that in V3 and V5. Mr Wachsberger reasoned, “if you know that historically the system works well for a higher force, then it’s better to stick with something that is similar to that rather than [make] a major change.”  In his opinion, there was nothing in the material produced by Thales that would have caused Thales to reduce the applied force further, certainly not to the extent of a reduction by 50% or more.  
  6. [275]
    In Mr Wachsberger’s opinion, the production of engineering ready technical drawings and collaboration with relevant personnel to ensure manufacturing viability of various concepts would not be completed within lonely a few weeks.  In his view the consolidation process requires confirmation of the validity of the concept, including the manufacture of items to be evaluated for suitability.  In Mr Wachsberger’s experience, a satisfactory product seldom occurs after the first build.  Several adjustments are usually required before it becomes a suitable solution.  The process, in his experience, tends to take months rather than weeks to complete.  This conventional process of research, development, design, testing and production is reflected in the iterative approach applied to create the Madritsch Solution.  
  7. [276]
    To a lesser extent it is shown by the work Thales undertook between October 2008 and September 2001 in its attempt to produce a mass-manufacturable version of the JGCM.  Based on his background and experience, Mr Wachsberger identified the following steps as “line with expectations”: 
    1. (a)
      developing a spring test methodology;
    2. (b)
      taking relative spring force measurements;
    3. (c)
      producing prototypes and subjecting them to both non-live fire and liver fire experiments;
    4. (d)
      undertaking of safety assessments; and
    5. (e)
      conducting critical design reviews. 
  8. [277]
    The Thales process in 2012 was abridged in most of these respects.  There appears to have been no safety assessment.  A design review was not done.  
  9. [278]
    Mr Wachsberger gave his oral evidence as a measured, detached and thoughtful witness.  He accepted that views different from his were open.  He made reasonable concessions.  He readily confirmed mistaken assumptions he had made, when they

were pointed out, and was able to identify where he had misunderstood other evidence.  

Dr Pratt

  1. [279]
    Thales relied on three reports prepared by Heath Pratt.[54]  Dr Pratt is a chartered professional engineer with over 30 years’ experience in key roles across a broad spectrum of complex, high value projects.  He has a PhD in Aerospace Engineering from the University of New South Wales (2017), from where he graduated with a Bachelor of Mechanical Engineering (1995).  
  2. [280]
    A significant part of Dr Pratt’s evidence was his conclusion that Thales produced V6 without any reference to the Madritsch Solution.  Dr Pratt based this conclusion on an exercise he did to identify a solution to the bolt unlocking problem from information provided to him by Thales’ solicitors; said to be the material available to Thales before it was given the Madritsch kits and Manual.  Dr Pratt said he came up with a solution like V6, based solely on this information.  He concluded Thales had done the same.  
  3. [281]
    Leaving aside some obvious shortcomings in Dr Pratt’s reasoning, such as the absence of any person with Dr Pratt’s qualifications and experience working on the V6 at the relevant time, his opinion turned out to be fundamentally wrong.  
  4. [282]
    Under cross-examination, Dr Pratt accepted that the task he undertook – designing a solution to the bolt unlocking problem – was not the task any Thales staff were undertaking at the relevant times.  In fact, in October 2008 the ADF had identified the JGCM solution and had asked Thales to produce a version of it that could be mass manufactured.  This same task was before Thales in the L125-3C program.  Dr Pratt had not undertaken that task.  He was not asked to do so.  
  5. [283]
    There was nothing in the instructions Thales had received to authorise and nothing in the limited testing Thales had done on JGCM, V1, V3 and V5 to indicate any significant alteration to the mechanism, such as that Dr Pratt formulated or, indeed, that manifested in V6.   
  6. [284]
    Dr Pratt’s reports proceeded on an erroneous basis that: 

“Thales was engaged to conduct an independent assessment of the failure scenario and seek to rectify the issue.  In response Thales conducted a comprehensive investigation …”  

  1. [285]
    According to Dr Pratt, with an experienced design team, Thales would have produced V6 in the time it took Mr Steele to do so.  It would have done so, Dr Pratt opined, by specifically addressing the Commonwealth’s concern in the May 2009 CDR checklist meeting minutes.  That concern was about the AFL not sitting squarely.  The change from a single to a double torsion spring addressed that concern.   [286] Dr Pratt’s narrow focus on this single concern overlooked two other relevant matters.  First, the minutes show no concern was raised about the disengagement force applied by the AFL to the gun lock group.  Second, they record the Commonwealth’s concern that any changes to the JGCM might alter its effectiveness to solve the bolt unlocking problem.  There is no logical connection between addressing the Commonwealth’s concerns in those minutes and reducing the disengagement force from the average over 40N Mr Steele had measured in V3 and V5 to the 22N Mr Steele specified in his first drawing the AFL spring for V6. 
  1. [287]
    Dr Pratt’s evidence was unsatisfactory in other respects.  His written reports contain speculative comments plainly intended to help Thales in the case.  He sought to fill gaps in the factual evidence from Mr Steele by speculating on the possible sources of the design changes, forces and torque specifications Mr Steele adopted.  He offered an “explanation” for the significant reduction in disengagement force between that measured in the V3 and V5 models and that specified in V6[55] for which there was no evidentiary basis.  In doing so, he moved beyond the role of independent expert towards acting as an advocate for his client.  He manifested the same engaged demeanour in his oral testimony.  He tried to avoid answering challenging questions and was visibly saddened when forced to concede points that were unfavourable to Thales’ case.
  2. [288]
    Both Dr Pratt and Mr Wachsberger agreed that in the evidence produced by Thales “there is no one definitive source” that can be identified as that used by Thales “to derive the AFL force value adopted [in] the [V6] design.”  The situation is somewhat complicated by Mr Steele specifying an applied force of 22N in the 28 February 2012 issue of drawing SK-108250 of the AFL spring and then a torque value of 40-80Nmm in the 13 and 21 March 2012 revisions.  
  3. [289]
    Dr Pratt concluded the source of the technical information used to derive the AFL applied force and torque values in the specifications in these drawings “is either missing or has not been included in the brief of materials.”  The absence or exclusion of expected records for Mr Steele’s work contrasts with the records and suitable reporting documentation, both formal and informal, covering Mr Williams’ work on the HMA1 Project.
  4. [290]
    In Dr Pratt’s opinion:

“the reduction in spring force from 22 Newtons to the torque values described above [40-80Nmm] is most likely the result of Mr Steele collaborating with a Thales engineer (…) [who] interpreted data from Thales experimental reports to adjust the 22 Newtons to a more meaningful design parameter, such as torque.  Since the new torque parameters resulted in a much lower applied force, considering the mechanical advantage of the AFL moulding, it is likely that the Thales engineers used experimental data to do an energy balance that resulted in the final torque values adopted by Thales design.  This would likely explain the reduction in the AFL spring force from HMA1 version 5 to 6.”   

  1. [291]
    Thales engineers could not have done as Dr Pratt speculated by interpreting the data from the small-scale September 2009 tests of V3 or the late 2010 and early 2011 tests of V5 or by measuring the properties of the V3 and V5 components.  They might have done so by interpreting data from the September 2011 tests of the Madritsch Solution or by examining the Madritsch Solution kit components.
  1. [292]
    This speculative opinion by Dr Pratt concerns the change from a 22N applied force specification to a 40-80Nmm torque.[56]  Dr Pratt offered no possible reason for Mr Steele’s initial decision to reduce the applied force from about 44N (he thought was the average he had measured in V3 and V5) to 22N in his first drawing on 28 February 2012. 

Conclusion on the use of information from the Madritsch Solution kits and Manual

  1. [293]
    From 6 September 2001, Mr Eyles had access to the Manual, which Mr FulfordTalbot sent to give him “a clear idea of the hammer mech modification.”  As noted at [147] above, I accept Mr Basladynski’s recollection[57] that he and Mr Eyles spent time examining the Madritsch Solution kit components and identified the changes Madritsch had made.  Mr Eyles directed Mr Basladynski to organise testing.  Mr Basladynski’s draft report on the testing notes a significant reduction in force of the AFL spring.  I infer this was observed when the kits were examined on arrival.
  1. [294]
    There is no evidence that Thales measured the exact torque exerted by the AFL spring or the vertical force exerted by the AFL on the gun lock group when the Madritsch Solution kit was fitted in the Austeyr.  There was no cause to do so when Thales was testing and confirming that the Madritsch Solution solved the bolt unlocking problem in the Austeyr as well as it had in the Steyr AUG, with a view to negotiating a licence to manufacture it.  I do not find that Thales measured or knew precise extent of the observed significant reduction in torque and force Madritsch had made.
  2. [295]
    From the contemporaneous documents and Mr Steele’s evidence, I am satisfied Mr Eyles was Mr Steele’s direct report and supervisor with professional responsibility for Mr Steele’s work on the HMA1 Project.  
  3. [296]
    It is likely that Mr Egan from Gibbens assisted Mr Steele to understand some of the properties of torsion springs.  He may have calculated the torque of some springs drawn by Mr Steele.  He may have suggested to Mr Steele how the design could be changed so that the springs would exert the torque Mr Steele wished to specify.  I regard it as improbable that Mr Egan made any design decisions that resulted in the specifications Mr Steele included in the drawings.  
  4. [297]
    Mr Steele’s evidence is that Mr Eyles told him something needed to be fixed in the existing V3 and V5 mechanisms as the applied force was still too high.  He used a gauge to “test” the disengagement force exerted by the AFL in V3 and V5.  Mr Steele said he adopted 22N as roughly half the disengagement force he had measured in Mr Williams’ mechanisms.  In December 2009, Mr Williams had measured the disengagement force in four V5 mechanisms as 26.70N, 29.49N, 26.30N and 27.03N.  I do not accept Thales’ submission that “the best evidence before the court” is Mr Williams’ measurements of these forces.  Mr Williams’ and Mr Steele’s measurements are evidence only of what each person measured and therefore thought the relevant force to be.  For the purposes of finding the facts necessary to determine this dispute, it matters only that Mr Steele thought the average disengagement force was about 44N in V3 and V5 and initially specified an applied force in V6 that he thought was half that disengagement force.
  1. [298]
    Mr Steele’s use of an “applied force” of 22N as a specification for the AFL spring on 28 February 2012, confirms Mr Steele did not understand the appropriate specifications for a torsion spring, as Mr Williams had done.[58] This is consistent with Mr Eyles’ assessment that Mr Steele had no understanding of torsion springs. 
  2. [299]
    I infer that Mr Steele was (ineptly) seeking to specify the force the AFL would apply to the gun lock group, and not the torque the AFL spring would apply to the AFL. From the specification and the accompanying email to Mr Egan I infer it was important to Mr Steele to draw components so, when manufactured and assembled, the AFL exerted a disengagement force of 22N on the gun lock group.  It was so important that all other properties of the spring were flexible and could be changed to meet this applied force requirement.
  3. [300]
    Mr Steele’s explanations for specifying a 22N applied force given in his written statements were ultimately unsatisfactory.  
  4. [301]
    In cross-examination, Mr Eyles accepted that nothing in Mr Williams’ work would have required Mr Steele to make any change to the torque specified for the AFL spring or the estimated disengagement force applied by the AFL.  
  5. [302]
    Apart from Mr Williams’ earlier work and whatever Mr Eyles had told him, Mr Steele had no background in the HMA1 Project.  I accept the submission put for Thales that there is no evidence Mr Steele had any direct knowledge of the Madritsch Solution.  
  6. [303]
    I am satisfied from Mr Steele’s communications with Mr Egan of Gibbens, that Mr Egan would not have been the source of an “applied force” of 22N as a torsion spring specification.  Their communications about the hammer springs indicate Mr Egan knew such a specification was inappropriate.[59]  
  1. [304]
    I reject the submission by Thales that the values Mr Steele specified in the drawings were “simply” the “products of the variables of the spring he had designed” and “not some pre-determined force which he had converted to torque values”.  It is contrary to Mr Steele’s evidence.  It is inherently unlikely that a major weapons manufacturer would approach the design of this mechanism in such a way.  I reject as unlikely the submissions that the change from a single- to a double-torsion spring explains the specification of a 22N applied force.  I similarly reject the submission that the “ultimate spring force” was in some way not attributable to the decisions made by Thales about its design but due to “the design of the spring itself”.    
  2. [305]
    In the five working days between 22 and 28 February 2012, Mr Steele redesigned the AFL spring with a significantly lower torque with the purpose of reducing by half the applied force exerted by the AFL on the gun lock group.[60]  In the same period, he resolved the issue of the AFL not “sitting squarely” by changing the single torsion spring to a double torsion spring: a concern not resolved in the previous 34 months.  
  3. [306]
    I accept Mr Wachsberger’s opinion that there was a further significant reduction in the disengagement force applied by the AFL to the gun lock group as a result of the changes Mr Steele made to the specifications for the AFL spring between 28 February and 13 March 2012.  In Mr Wachsberger’s opinion the reduction was probably greater than 65%.[61]
  4. [307]
    Over the 20 calendar days between 22 February and 13 March 2012, Mr Steele had finalised the torque specifications for the AFL spring.  As the plaintiffs submitted, this is surprising, given his lack of experience and understanding of torsion springs.  In this time, Mr Steele was commuting from Melbourne to work a couple of days a week at Lithgow on HMA1 and on another project.    
  5. [308]
    It is more surprising, given Mr Evenden’s assessment on 19 December 2011, that Thales had pursued the Madritsch Solution during 2011 because Thales “did not have the time nor the budget to pursue a fix within schedule and price.”  
  1. [309]
    Given his role, training and level of responsibility, it is unlikely Mr Steele on his own initiative set out to halve the disengagement force exerted by the AFL on the gun lock group or fixed the 22N figure for the AFL spring.  
  2. [310]
    A fixed force condition different from that used by Mr Williams is more likely than not to have originated from Mr Eyles, as Mr Steele’s supervisor, in discussions on or before 28 February 2012.  This is consistent with the effect of Mr Steele’s evidence during cross-examination at [216] to [217] above.  
  3. [311]
    It is more likely than not that Mr Eyles instructed Mr Steele to do this to fix the “something” (unknown to Mr Steele) that Mr Eyles told him “still needs fixing” in V5.  A person of Mr Steele’s position, training and experience might well have responded to the instruction by measuring the disengagement force in V3 and V5, as Mr Steele did, and adopting 22N (approximately half that measured force) as the “applied force” in drawing SK-108250 of the AFL spring and telling Mr Egan of Gibbens that it was a “fixed condition” of the spring.  
  1. [312]
    I also conclude it is more likely than not that, Mr Eyles explained to Mr Steele how to specify the torque for the AFL spring in V6 to achieve or maintain this significant reduction in the 13 and 21 March 2012 drawings of the AFL spring.[62]  
  2. [313]
    I reject Mr Eyles’ denial that he was the source of the specifications for a significantly lower disengagement force.  Given his close and regular supervision of Mr Steele, I also reject his claim that he only discovered the applied force was significantly lower in V6 “well after the fact.”  He approved the drawings with the relevant specifications before the components were manufactured.
  3. [314]
    Leaving aside the information Mr Eyles and Mr Basladynski obtained from examining the Madritsch kit components in September 2011, there was no other reason for Thales to seek to make any significant change to the disengagement force applied by the AFL in V3 and V5.  The absence of a logical explanation for seeking to halve the measured applied force supports an inference that, in instructing Mr Steele to make that change, Mr Eyles used the information he derived from examining the Madritsch Solution kit components.    
  4. [315]
    Consideration of these matters and the evidence lying behind them, leads me to conclude that it is more likely than not that Mr Eyles directed Mr Steele to reduce the applied force of the AFL on the gun lock group significantly from that in V3 and V5 by significantly reducing the torque applied by the AFL spring.  Acting on this direction, Mr Steele changed the AFL spring specification.  He thought he was reducing it by half in specifying an applied force of 22N.  
  5. [316]
    I reject the submission that Mr Eyles’ use of his knowledge of the Madritsch Solution was no more than the “merely taking of comfort” in the fact that another manufacturer had successfully designed and tested a solution involving a disengagement force “somewhere between a range of possible forces”.
  6. [317]
    I accept Dr Pratt’s opinion, as summarised by Thales, that “there is no evidence that there is an optimal AFL spring force to both ensure that the bolt unlocking problem does not occur and ensure that the rifles perform as automatic weapons.”  I reject Thales’ submission that the court should infer from this opinion that both goals may be achieved at any point on the spectrum from zero to 44N of applied disengagement force.  Dr Pratt did not express such an opinion.  It would be plainly wrong to do so.  
  7. [318]
    I also reject Thales’ submission that “the solution to the bolt unlocking problem does not require any particular spring force in the AFL spring.”  It is contrary to the evidence.  
  8. [319]
    The Madritsch Solution was developed through an iterative process of designing and testing mechanisms in which the disengagement force was changed.  It had been tested and approved across a range of conditions (as required by NATO AC/225).  It incorporated an AFL spring with a torque low enough to reliably prevent the bolt unlocking problem and high enough to allow the rifle to fire reliably in automatic mode.  Thales had not “tested and proved” such a device in a comparable manner prior to the disclosure of the Madritsch Solution.
  9. [320]
    Without using that information, it is unlikely Thales would have made a change to the design from V5 to V6 to bring about any significant reduction in the applied force of the AFL by reducing the torque exerted by the AFL spring.  Only after undertaking further testing and review would Thales have been able to identify that the force exerted by the AFL on the gun lock group in V3 and V5 was “still too high” to solve the bolt unlocking problem reliably and effectively.  Thales had done no further testing in the period since V5 was designed, save for its testing of the Madritsch Solution.  As Mr Eyles explained to Mr Higgins, Thales had done no formal design review for V6.  Thales would have had to undertake further, more extensive testing and review of its final design, had it not relied on the successful testing and acceptance of the Madritsch Solution.  
  10. [321]
    Although Thales may have been able to use the ARMTSPO test information, there is no reason to assume Thales could have undertaken and completed the further testing and design work required in a period much shorter than the 12 months Madritsch took to bring the Madritsch Solution to a tested and approved state.  Thales would have had to apply similar human resources and technology.  In short, Thales would not have been able to achieve the goal identified by Mr Evenden of bringing to market an alternative to the Madritsch Solution in a timely way, so as to prevent Nioa making a direct approach to the Commonwealth with the Madritsch Solution.  
  11. [322]
    These conclusions are consistent with Mr Eyles’ “Confident” advice to Mr Evenden on 21 March 2012 that V6 “will work…the same as Madritsch”.  He could be so confident because the changes made to the AFL spring in V6 were derived from his knowledge of the Madritsch Solution, which had already been tested and approved.  
  12. [323]
    I also conclude that Mr Eyles was the source of Mr Steele’s “inspiration” to draw the AFL spring in V6 as a double torsion spring.  This is a change of design that Thales could have made without further testing.  It merely required the application of professional engineering experience with torsion springs, which Mr Steele lacked.  It would not have taken Thales very long to do so, had such experience been applied.     
  13. [324]
    Thales offered no plausible explanation for the significant reduction in the 22N disengagement force specified by Mr Steele in his first drawing of the AFL spring (SK-108250) on 28 February 2012 or for the torque specified in his later issues of the drawing on 13 and 21 March 2012.  The facts about how this occurred must be peculiarly within the knowledge of Thales.  
  14. [325]
    By 8 June 2012, Mr Eyles was aware of the need to formally record the starting point and the design rationale “used to achieve the end result.”  See [241] above.  As early as 12 June 2012, senior officers of Madritsch and Nioa had raised allegations of copying or misuse of confidential information.  The events in the first six months of 2012 were then quite recent.  They could have been recalled and relevant documents identified and retained.  Instead, Mr Eyles supplied Mr Evenden with an “explanation” that was false in a material respect. 
  15. [326]
    Considering all the evidence, I am satisfied that Thales did not proceed through an independent design sequence to produce the V6.  Instead, Thales used the information about the properties of the AFL spring in the Madritsch Solution kits as a springboard to advance V6 without having to undertake the necessary further work itself.  
  16. [327]
    Weighing the evidence according to the power or ability of a party to produce it,[63] the available credible evidence, which leads to the conclusion that Mr Eyles directed Mr Steele to include specifications for the AFL spring in V6 that would have the effect of significantly reducing the force exerted by the AFL on the gun lock group and inspired Mr Steele’s decision to use a double torsion spring, is sufficient to discharge the plaintiffs’ onus of proof in this respect.  This is so notwithstanding the seriousness of the allegations made in the plaintiffs’ case.

Breach of the NDA and duty of confidence

  1. [328]
    The plaintiffs contend that confidential information was “embodied within each of the components of the Madritsch Solution” that were delivered to Thales on 13 September 2011.  They point to two specific features as confidential information embodied in those components.  
    1. (a)
      The first is that the torque exerted by the AFL spring in the Madritsch Solution was significantly lower than the torque exerted by the AFL springs in V3 and V5 and, in turn, the lower torque caused the AFL to exert a significantly lower disengagement force against the gun lock group when the two came into contact.  
    2. (b)
      The second was that the AFL spring in the Madritsch Solution was in the form of a double torsion spring, unlike the single torsion in each of V3 and V5.  This second feature was also information contained in the Manual where the AFL spring was illustrated.  
  2. [329]
    Each of these two features was information or data disclosed by the plaintiffs to Thales for the purpose of entering into discussions in relation to the modification of the Austeyr to overcome the bolt unlocking problem.  This information was about a component in a device that, when assembled and fitted in rifle in the Steyr AUG family of weapons, had been tested and approved under NATO AC/225 as consistently solving the bolt unlocking problem without affecting the operation of the weapon.  

Did Thales breach the NDA?

  1. [330]
    The plaintiffs contend that Thales breached clauses 2(b)(ii), (iii) and (v), 3(a) and 12(b) of the NDA. 

Clause 2(b)(ii) - “be only disclosed to and used by its Representatives who have a need to know and solely for the Purpose” 

  1. [331]
    In September 2011, Thales, in the persons of Mr Eyles and Mr Basladynski, examined the components in the Madritsch kits to seek to discover information or data about the Madritsch Solution.  In doing so, Mr Eyles discovered the relevant two features of the AFL spring.  Neither Mr Eyles nor Mr Basladynski was involved or intended to be involved in discussions with Madritsch or Nioa in relation to modification of the Austeyr to overcome the bolt unlocking problem.  However, they might have provided advice to others within Thales about such discussions and so they might have been employees who had a need to know of the information solely for that purpose within the scope of cl 5 of the NDA.  
  2. [332]
    In the circumstances, I am not satisfied that Thales breached cl 2(b)(ii) of the NDA.  

Clause 2(b)(iii) – “not be used in whole or in part for any purpose other than the Purpose”

  1. [333]
    Between 22 February and 11 June 2012, Thales, in the persons of Mr Eyles and Mr Steele,[64] used that information or data, obtained by Mr Eyles and Mr Basladynski in their September 2011 examination, to design a revised AFL spring in V6 as a double torsion spring with a significantly lower torque than that exerted by the AFL springs in V3 and V5 so that the AFL in V6 applied a significantly lower disengagement force on the gun lock group than did the AFL in V3 and V5.  
  2. [334]
    This was not done for the purpose of discussions with Madritsch or Nioa.
  3. [335]
    I am satisfied this was a breach by Thales of cl 2(b)(iii) of the NDA. 

Clause 2(b)(v) – “neither be copied, nor otherwise reproduced nor duplicated in whole or in part where such copying, reproduction or duplication has not been specifically authorised in writing by the disclosing Party”

  1. [336]
    Although I am satisfied Thales used the information about the Madritsch Solution (which Mr Eyles and Mr Basladynski gained from examining the kits), I am not persuaded that Thales copied, reproduced or duplicated the information.  Thales did not copy, reproduce or duplicate the AFL spring.  Rather, it used information about the spring’s form and properties to advance the design of the comparable part in V6. [337] In the circumstances, I am not satisfied that Thales breached cl 2(b)(v) of the NDA. 

Clause 3(a) – “Each Party agrees that it shall not itself, or cause or enable any other person to reverse-engineer, decompile, or disassemble any software, process or tangible items, including refraining from any activities, examinations, performance or qualitative analyses, or other investigations that seek to discover the contents, algorithms, composition or manufacturing formulae of the Proprietary Information”

  1. [338]
    The plaintiffs contend that between 22 February and 11 June 2012, Thales, in the person of Mr Eyles, caused or enabled itself, in the person of Mr Steele, to reverseengineer the AFL spring in the Madritsch Solution.  If this occurred, it was possible because, in September 2011, Thales, in the persons of Mr Eyles and Mr Basladynski, did not refrain from examining the AFL springs to discover the two relevant features.  In seeking to identify the changes Madritsch had made, it is more likely than not he and Mr Eyles sought “to discover the contents” of the Madritsch Solution, including those two features.  It may be that the torque specification is a manufacturing formulae, but the language of cl 3(a) seems directed to something more complicated than the observation, handling and copying of physical components.   
  2. [339]
    Thales’ attempt to “reverse-engineer” the Madritsch AFL spring, if was what occurred, was not entirely successful.  The force exerted by the AFL in V6 may be twice or more that exerted by the equivalent part in the Madritsch Solution.  I infer that the torque exerted by the V6 AFL spring might also substantially higher than its Madritsch equivalent.  The prohibition in cl 3(a) is not framed to apply expressly to attempts at reverse-engineering.  Although, as a commercial agreement, it might be given that effect.  
  3. [340]
    Construing the NDA as a whole, the conduct of Thales through Mr Eyles and Mr Steele is plainly within the scope of the prohibition in cl 2(b)(iii).  It is not necessary to adopt a broader meaning of cl 3(a) to permit the parties’s agreement to regulate such conduct.  The plaintiffs did not identify anything they might gain from a construction that, by the same conduct, Thales breached cl 3(a) as well as cl 2(b)(iii). I decline to construe cl 3(a) to encompass it. 
  4. [341]
    I am not satisfied Thales breached cl 3(a) of the NDA. 

Clause 12(b) – “the receiving Party must notify the disclosing Party immediately upon becoming aware of any disclosure or use of the Proprietary Information of the disclosing Party other than for the approved Purpose or otherwise than as permitted by [the NDA]”

  1. [342]
    Thales, in the person of Mr Eyles, was aware of the use of the information about the Madritsch Solution by himself and by Mr Steele. He was aware it was not used for the purpose of discussions with Madritsch or Nioa.  So he new facts sufficient to establish a breach of cl 2(b)(iii).  
  2. [343]
    Thales did not notify Madritsch or Nioa of that breach.  This failure was a breach by Thales of cl 12(b) of the NDA.  

Thales’ defences to NDA breaches

  1. [344]
    Thales sought to avoid the consequences of its breaches of the NDA in three ways. 
  2. [345]
    First, there was a submission that if Mr Eyles had directed Mr Steele to specify a significantly lower applied force for the AFL and to use a double torsion spring for the AFL spring, then this entailed no breach of the NDA because both these things were known to Mr Eyles without reference to the information gained from the Madritsch Solution kits and Manual. 
  3. [346]
    Second, related to the first but quite specifically propounded, Thales asserted that the information was not confidential.  
  4. [347]
    Thirdly, Thales contended that the covenants in the NDA were otherwise unenforceable as restraints of trade.  

Did Thales already know the information?

  1. [348]
    Thales’ first point misunderstands the plaintiffs’ case.  The information in question is specific attributes of a component of the Madritsch Solution.  It is important because, as an assembled mechanism, it had been tested and approved as a solution to the bolt unlocking problem. It was not the information alone, but the information as aspects or elements of a successful mechanism, that was important.  Mr Eyles (and Mr Basladynski) did not know this until Madritsch disclosed it in September 2011.

Was the information confidential?

  1. [349]
    The limits of confidential information were considered by Lord Green MR in Saltman Engineering Co Ltd v Campbell Engineering Co Ltd:  

“The information, to be confidential, must, I apprehend, apart from contract, have the necessary quality of confidence about it, namely, it must not be something which is public property and public knowledge. On the other hand, it is perfectly possible to have a confidential document, be it a formula, a plan, a sketch, or something of that kind, which is the result of work done by the maker on materials which may be available for the use of anybody; but what makes it confidential is the fact that the maker of the document has used his brain and thus produced a result which can only be produced by somebody who goes through the same process.”[65]

  1. [350]
    Although different scrutiny has been applied to the protection of an employer from the disclosure or use of confidential information by a former employee, the analysis of Sir Robert Megarry V-C in Thomas Marshall (Exports) Ltd v Guinle is, as his Lordship suggested, of assistance in considering the extent of a claim of confidentiality: 

“… four elements may be discerned which may be of some assistance in identifying confidential information or trade secrets which the court will protect.  I speak of such information or secrets only in an industrial or trade setting.  First, I think that the information must be information the release of which the owner believes would be injurious to him or of advantage to his rivals or others.  Second, I think the owner must believe that the information is confidential or secret, i.e., that it is not already in the public domain.  It may be that some or all of his rivals already have the information: but as long as the owner believes it to be confidential I think he is entitled to try and protect it.  Further, I think that the owner’s belief under the two previous heads must be reasonable.  Fourth, I think that the information must be judged in the light of the usage and practices of the particular industry or trade concerned.  It may be that information which does not satisfy all these requirements may be entitled to protection as confidential information or trade secrets: but I think that any information which does satisfy them must be of a type which is entitled to protection.”[66]

  1. [351]
    Confidentiality is a relative and not an absolute concept.  Information must be inaccessible before it can be considered confidential.  It follows that confidentiality is context- and fact-sensitive.[67] 
  2. [352]
    The following context and facts are relevant to the question of whether the information about the form and the comparative torque exerted by the AFL spring in the Madritsch Solution was confidential. 
    1. (a)
      Between September 2008 and September 2009, Madritsch had researched the bolt unlocking problem, applied professional engineering knowledge and skills to propose solutions, expended time and resources designing and testing proposed solutions and, through an iterative process, obtained the information about the disengagement force that the AFL should apply to the gun lock group (and so the torque that should be exerted by the AFL spring) to consistently and reliably solve the bolt unlocking problem in the Steyr AUG while allowing the rifle to operate as an auto oat ice weapon.  It had also applied professional engineering knowledge and skills to select the form of AFL spring as a double torsion spring.  From September 2009 to August 2011, Madritsch had done further work on the mass-manufacture of the Madritsch Solution in a kit form.
  1. (b)
    The Madritsch Solution, including an AFL spring with the relevant properties, was sought by the Bundesheer and was of interest to senior adviser to the ADF.  It was useful and valuable.  It was reasonable to recognise the information to be the property of Madritsch, as Thales did when it executed the NDA.[68]
  2. (c)
    On 19 August 2011, when Thales executed the NDA, it did so to obtain access to the kits to test the Madritsch Solution and the associated additional information.  This information was not otherwise accessible to Thales.  Thales was aware that Madritsch prized the information and set out to protect it from disclosure and use by effectively restricting disclosure and limiting use by the 8 June 2011 non-disclosure agreement, the Intentions letter and then the NDA.  A reasonable person in the position of Thales would have realised the information being given by the provision of the Manual and the kits was given in confidence.[69] Thales impliedly assumed an obligation to respect the confidentiality of the information.[70]  
  3. (d)
    Anyone, including Thales, could only have obtained the information independently, by doing the same type and extent of work as Madritsch in researching, designing, developing and testing the Madritsch Solution.  This does not alter the confidential nature of the information derived by Madritsch’s own work.[71]  The extent of work required is indicated by the fact that the ARMTSPO, which had done some development and testing work on the JGCM mechanism between May 2007 and October 2008, and Thales, which had done some limited testing of the JGCM, V3 and V5 between October 2008 and December 2009, were not aware of the information Madritsch had derived from its more extensive efforts between September 2008 and September 2009.[72]  
  4. (e)
    Between 22 February and 20 April 2012, by using the information about the form and the torque of the AFL spring in the Madritsch Solution kits and in the Manual, Thales, in the person of Mr Eyles, was able to direct the revision of Thales’ earlier versions and with Mr Steele, create V6.  In this way, Thales avoided doing its own more extensive iterative research, development, design and testing.  It did so at a time when Thales considered it important to bring to market an alternative to the Madritsch Solution quickly to prevent Nioa (as agent for Madritsch) making a direct approach to the Commonwealth to supply the Madritsch Solution for the use of the ADF.  
  1. (f)
    In this way, Thales was able to produce V6 as its alternative solution at a lower cost than it would otherwise have incurred.  Thales was able to produce V6 and propose it as a solution to the bolt unlocking problem with no professional engineering research, design or development work, by seconding a draftsperson to do the “design”, and with minimal testing.  Thales could do so because it knew the V6 would “work” because it was the “same as Madritsch” in material respects, and it knew the Madritsch Solution had been successfully tested and approved by the Bundesheer and by Thales in September 2011.  
  2. (g)
    In this way, Thales was able to produce V6 in a shorter time and at a lower cost than it would otherwise have been able, had it had to undertake its own further research, development, design and testing.  
  1. [353]
    Thales knew from the second pendulum test that V5 was “as effective as no AFL/Slide interface friction at all.”  See [65] - [66] above.  Thales submitted this knowledge meant that the properties of an AFL spring that caused the AFL to exert a force on the gun lock group between zero and the 42.5N (exerted by V5) could not be confidential.  I reject that submission.  Thales did not know the disengagement force exerted by the components in the successfully tested and approved Madritsch Solution was significantly lower than those in V3 and V5.  The information was confidential, notwithstanding that the force was within a known range.  
  2. [354]
    At no time did Madritsch make the kits or Manual generally available.  The provision of the kits to the SLW for the limited purpose of testing was not done with a view to general publication.  This disclosure “falls far short of the disclosure to the world, which was held to destroy confidentiality”.[73] There is no evidence that it became publicly known in any other way, perhaps until the hearing.
  3. [355]
    However, for Thales, Mr Johnstone and Mr Psaltis submitted that the Madritsch Solution had entered the public domain by its disclosure to “an unidentified number of people within the Austrian army” in the testing done before the kits were disclosed to Thales.  
  4. [356]
    On this issue, the plaintiffs relied on the expert evidence of Dr Schwank. Dr Schwank
  5. [357]
    Dr Schwank holds a Doctorate of Law from the University of Innsbruck, Austria.[74] He was admitted as a lawyer and advocate in 1976 and has been in practice continuously since that time, in Innsbruck (1976-1985) and Vienna (since 1986).  He is a Fellow of the Chartered Institute of Arbitrators, London and a board member of the German-Australian-Pacific Lawyers Association.    
  6. [358]
    He was asked to provide an opinion of whether s 46(1) of the Beamten- Dienstrechtsgesetz 1979[75] (BD 1979) applied to the provision by Mr Madritsch of four kits to Obst Weinseiꞵ of the Bundesheer for testing of the Madritsch Solution. No other expert opinion on Austrian law was tendered.
  1. [359]
    According to Dr Schwank, BD 1979 is a piece of federal legislation passed by both houses of the Austrian Parliament and signed into law by the Austrian President. 
  2. [360]
    Dr Schwank explained that, by s 46(1), the BD 1979 relevantly provides: 

“The public servant is bound by confidentiality about all facts of which he has gained knowledge exclusively as a result of his official duty, the secrecy of which is required for the sake of comprehensive defence of the country, .. for the preparation of a decision or for the sake of the predominant interest of parties.  The duty of confidentiality has to be observed in respect to everybody unless he is bound to make an official notice about such facts (official confidentiality).” 

  1. [361]
    I accept Dr Schwank’s opinion that, by s 1, the BD 1979 applies to all public servants employed by federal authorities in Austria and, by ss 146 to 152d, it applies specifically to army personnel.  By reason of his employment in the Bundesheer, Obst Weinseiꞵ is subject to s 46(1).  I infer it applies to Fachinspektor Stamm and any other Bundesheer personnel involved in testing the Madritsch Solution.  
  2. [362]
    The unchallenged evidence of Mr Madritsch is that he provided four kits to Obst Weinseiꞵ to test for the purpose of the Bundesheer deciding whether to adopt the Madritsch Solution for the Steyr AUG.  The Bundesheer SLW conducted the testing in accordance with NATO AC/225 criteria.  This was confirmed by the Prüfzertifikat.  The testing was relevant to the possible use of the Madritsch Solution by the Bundesheer and so relevant to the defence of Austria.  The Bundesheer had a direct economic interest in having access to the kits and testing them for the purpose of determining whether to adopt and purchase the Madritsch Solution.
  1. [363]
    There is no evidence that Obst Weinseiꞵ gained any knowledge of confidential information about the Madritsch Solution otherwise than as a result of his official duty.  The position is the same with respect to any other Bundesheer members involved in the testing.  
  2. [364]
    It is more likely than not that maintaining the secrecy of the information that could be obtained by dismantling and examining the kit components and from the testing of them was relevant for the purpose of the Austrian federal authorities making a decision on whether to adopt the Madritsch Solution.  The Austrian authorities obviously had a vested economic interest in that decision.
  3. [365]
    I accept Dr Schwank’s opinion that, in the circumstances, s 46(1) applied to the provision of the kits to Obst Weinseiꞵ.  He and the other members of the Bundesheer involved in the testing were bound by s 46(1) of the BD 1979.  
  4. [366]
    I reject Thales’ contention that any confidentiality in the Madritsch Solution was lost as a consequence of Madritsch permitting the Bundesheer to complete verification and testing of it and that, in this way, the Madritsch Solution was brought into the public domain.  
  5. [367]
    In the circumstances, the relevant information Mr Eyles and Mr Basladynski derived from examining the kit components and the Manual had the necessary quality of confidence when it was disclosed to Thales in September 2011.  
  6. [368]
    The court is not faced with considering whether cl 2(b)(iii) prevents Thales from using information that is not and was not intended to remain confidential.    

Is the NDA enforceable?

  1. [369]
    Thales contends that, relevantly, clauses 2(b)(iii) and 12(b) of the NDA are unenforceable as restraints of trade. 
  2. [370]
    Only cl 2(b)(iii) is a prohibition on use of the Proprietary Information.  The obligation in cl 12(b) to notify Madritsch of a breach does not appear to be a restraint of trade, but it would operate to enhance the effect of the restraint in cl 2(b)(iii).  
  3. [371]
    The NDA is governed by the laws of New South Wales.  The parties chose to make it so.  The Lithgow factory is in New South Wales.  That is where, according to the plaintiffs, Thales engaged in the conduct that breached terms of the NDA.  The laws of that State include the Restraints of Trade Act 1976 (NSW) (RTA).  The RTA applies relevantly to restraints created by contract.[76]  It is a short enactment.  The relevant operative provision appears to be s 4(1).  It provides as follows: 

“4Extent to which restraint of trade valid

  1. (1)
    A restraint of trade is valid to the extent to which it is not against public policy, whether it is in severable terms or not.”
  1. [372]
    The extent to which a restraint is not against public policy is the extent to which it is reasonably necessary for the protection of the legitimate interests of the parties (or one or more of them) at the time the NDA was made and also reasonable in the interests of the public.[77]  
  2. [373]
    As Young JA expressed it in Sidameneo (No 456) Pty Ltd v Alexander,[78] while the interests a party must possess in order to protect itself by a restrictive covenant traditionally “were considered to be particular types of proprietary interests, such as goodwill”: 

“The true rule is that the plaintiff must have some legitimate interest which equity will protect.  The word ‘proprietary’ is used in a special sense and will include legitimate commercial interests.”[79]

  1. [374]
    Trade secrets and confidential information are “legitimate interests which may be the subject of protection by covenant”.[80]  The relevant information the plaintiffs sought to protect was confidential. See [349]-[367] above. 
  2. [375]
    In New South Wales, the validity of a restraint is determined in light of the actual (or apprehended) breach, and not “imaginary or potential breaches.”[81]  
  3. [376]
    Thales’ breach of cl 2(b)(iii) was the use of information, obtained by examination of the kit components and Manual, to design a revised AFL spring in V6.  The breach of cl 12(b) was the failure to notify the plaintiffs of the cl 2(b)(iii) breach.   
  4. [377]
    The following matters are relevant to whether these provisions of the NDA were reasonably necessary to protect the plaintiffs’ interest in the confidential information:
    1. (a)
      The information to be protected is limited to information that is “of a proprietary or confidential nature”.  The obligations in clause 2(b) are that the information “remain protected and kept in strict confidence”, and in doing so each party “use at least the same safeguards it uses to protect its own Proprietary Information”.  I accept the submission put by Mr Johnston and Mr Psaltis for Thales that, construing the NDA as a whole and consistently with its express purpose, the subject matter of the restraints in cl 2(b) are confined to information that is confidential and not already in the public domain.  If the information ceases to be confidential, otherwise than by some unlawful act or omission by a party, the covenants cease to apply to it. 
    2. (b)
      The restraints limited the further disclosure and use of information disclosed by the parties by confining its use to the agreed purpose of entering discussions. 
    3. (c)
      The restraints in cl 2(b) sought to protect the parties by limiting the further disclosure and use of confidential information for a limited time.  The restriction applied for five years from the date a party received the confidential information.  The notification obligation in cl 12(b) depended upon a breach, so it endured during the five years in which a breach could occur. 
    4. (d)
      The Steyr family of weapons was used by several national armies.  The value of confidential information about the bolt unlocking problem extended across national borders and was not confined to a specific geographic area or region.   
    5. (e)
      The restraint for this limited period after disclosure might be considered a ‘‘head start’’ handicap, reflecting the advantage a party might gain by access to the confidential information, while it remained confidential, over competitors who had not dealt with the other parties. 
    6. (f)
      The terms of the NDA were the subject of negotiation between the parties between 27 June and 11 August 2011.  Each of the parties is an established commercial enterprise in the same trade.  Each proposed and rejected provisions for the NDA.  There is no indication of any inequality in bargaining power or of any party bring overborne in the negotiation.   The obligations are mutual.[82]  They are for the benefit of each of the parties to the NDA.  
  5. [378]
    The following matters, known to all parties at the time they entered into the NDA, may also be relevant to whether the restrains were reasonably necessary to protect the interest of the plaintiffs in the confidential information:
    1. (a)
      The communications between Thales and Nioa make clear the discussions were to be about a licence for Thales to manufacture the Madritsch Solution in Australia and to manufacture the components as an integrated part of the new model Austeyr for the ADF under the L125-3C program.  
    2. (b)
      Thales had sought access to the kits and related information for the specific purpose of progressing those discussions, including by undertaking live fire testing of the effectiveness and reliability of the Madritsch Solution in the Austeyr.  The plaintiffs had insisted on limits on the disclosure and use of Madritsch’s confidential information.  Thales negotiated and agreed the limits.  
  1. (c)
    Madritsch had started work on developing its solution to the bolt unlocking problem well before the date of the NDA.  It had progressed the solution to approval, after testing, by the Bundesheer.  This must have involved the application of resources, time and effort.  The restraints would protect relevantly the value to Madritsch of the fruits of its work.
  2. (d)
    In January 2011, the ADF had expressed to Nioa its interest in the Madritsch Solution.  As the agent of Madritsch, Nioa also had an interest, if Thales did not take up the Madritsch Solution, in arranging to supply the kits for the use of the ADF.  The restraints would also protect Nioa’s interest.  
  3. (e)
    Madritsch had maintained the confidentiality of the detail of its solution during research, development, design and testing and thereafter.  As the NDA attests, Madritsch had no plan to put the information into the public domain in August 2011.  It appears Thales had no other legal means of obtaining the confidential information it desired to continue negotiations with the plaintiffs. 
  1. [379]
    In the circumstances, when the NDA was made, the restraint on Thales’ conduct in cl 2(b)(iii) was reasonable for the protection of the plaintiffs’ interests.  The covenant in cl 12(b), to the extent that may be characterised as a restraint of trade, was also reasonable as additional protection of the same interests of the plaintiffs, alerting them quickly to a breach so that they could take timely action to protect those interests.
  2. [380]
    The following matters may also be relevant to the reasonableness of the restraint.
  3. [381]
    Madritsch had started work on developing its solution to the bolt unlocking problem by 2008, conducting its first live-fire testing in September 2008.  It had tested its final version in September 2009.  The work was done by two senior officers, each a qualified engineer.  It had taken Madritsch almost another two years to be at the point of commercial production of the kits.  I accept this three-year period as a reasonable time in which an engineering organisation applying appropriately qualified and experience staff and sufficient financial resources could expect to research, develop, design, test and finalise for production a solution to the bolt unlocking problem.  The use of the confidential information could be expected to give a person in the position of Thales a “headstart” on a competitor. 
  4. [382]
    At the time the NDA was executed, Madritsch was embarking on manufacturing and supplying the Madritsch Solution to the Bundesheer.  Madritsch was pursuing an approval from the Austrian Government to export the kits.  The confidentiality of the detail of the Madritsch Solution had an economic value to Madritsch.  
  5. [383]
    From 19 December 2011, Thales wanted to prevent Nioa making a direct approach to the Commonwealth to supply the Madritsch Solution to the ADF, at least until Thales had its alternative solution in place and had secured the position of preferred provider of it.  The ADF’s interest in it, indicated confidentiality of the detail of the Madritsch Solution had an economic value to Nioa.  
  6. [384]
    There is a public interest in the development of better technologies.  It is promoted by reasonable restraints on copying work that is otherwise confidential, valuable, and the result of effort.  The covenants in cl 2(b)(iii) and 12(b) serve that public interest.
  7. [385]
    I am satisfied that the covenants in cl 2(b)(iii) and cl 12(b) are not against public policy within the meaning in RTA s 4(1).    
  8. [386]
    As Giles JA explained in Industrial Rollformers Pty Ltd v Ingersoll-Rand (Australia) Ltd

“The operation of s4(1) of the Act is now relatively well settled. It does not permit the Court to remake the contract or a covenant in it, and although sometimes it is said that it allows the covenant to be read down or redrafted that is really an inaccurate description. The provision looks to the postulated breach, and permits the Court to enforce a covenant otherwise invalid as against public policy if the restraint in the covenant so far as it applies to the postulated breach is not contrary to public policy. The Court is given the capacity to enforce a reasonable restraint of trade falling within the expressed restraint although the expressed restraint is too widely stated.[83]

  1. [387]
    Thales’ breach of these covenants commenced within six months of the disclosure of the Manual and the components of the Madritsch Solution kits.  It appears to have continued.  A restraint on Thales conduct within the scope of cl 2(b)(iii) and cl 12(b) applying for six months would not be contrary to public policy.  It is enforceable on that basis.  
  2. [388]
    In all the circumstances, I am satisfied the court should enforce the restraint against the breaches by Thales.  

Clause 3(a)

  1. [389]
    On the basis noted at [338] above, it is not necessary to consider the validity of the restraint in cl 3(a) of the NDA.  For completeness, I make the following observations.  
  2. [390]
    The restraint in cl 3(a) is different to those in cl 2(b).  It is an indefinite prohibition, relevantly on reverse engineering the Madritsch Solution by examining the kit components to seek to discover their contents and manufacturing formula.  No evidence was adduced about the period over which the Madritsch Solution was likely to remain in use and the information about it confidential and of value.    
  3. [391]
    As the plaintiffs’ claim illustrates, the same conduct appears to be capable of falling within the range of both cl 2(b)(iii) and cl 3(a).  The internal logic of the NDA indicates that the parties agreed the five-year restraint period for each of the covenants in cl 2(b) was sufficient to protect the plaintiffs’ interests.  
  4. [392]
    There is no apparent justification for an indefinite restraint of the kind in cl 3(a).  The covenant in cl 3(a) would appear to be against public policy by reason of a manifest failure by the person who created or joined in creating it to attempt to make the restraint a reasonable one.     
  5. [393]
    However, Thales’ breach of cl 3(a) commenced within six months of the disclosure.  Applying the law of New South Wales, s 4(1) of the RTA permits the court to enforce the covenant in cl 3(a) so far as its application to Thales’ breach is not contrary to public policy.  Although the restraint in cl 3(a) is too widely stated, being unlimited in time, a restraint for six months (or even five years), falling within the expressed restraint, is reasonable and so enforceable as not against public policy.    

Conclusion on breach of the NDA

  1. [394]
    The plaintiffs are entitled to damages for breaches of cl 2(b)(iii) and cl 12(b) of the NDA.  The quantification of those damages is a matter for another day.

Breach of “side agreement”

  1. [395]
    The plaintiffs contend that, on or about 11 August 2011, the parties entered into a “Side Agreement” that, in exchange for Thales providing the assurances in the Intentions letter, Madritsch would disclose the Madritsch Solution to Thales.  They rely on emails passing between Mr Whitty and Mr Fulford-Talbot.  The terms of the alleged side agreement are those found in the Intentions letter, which Thales signed and sent six days later.
  1. [396]
    As noted above, the parties negotiated the terms of the Intentions letter and the NDA were between 27 June and 11 August 2011.  In an email of 4 August 2011, Mr Whitty advised that Madritsch had: 

“agreed to release the samples to Thales for evaluation on the basis of [five things, including] Thales provide the letter of understanding and intent to Madritsch along the lines we discussed.”  

  1. [397]
    Mr Fulford-Talbot replied, rejecting one thing Mr Whitty had stated Madritsch required.  He did not object to or reject the Intentions letter.  The negotiations continued.  
  2. [398]
    By 11 August 2011, Thales had proposed or indicated its willingness to execute the NDA and sign the Intentions letter in the terms in which they were eventually executed and signed.  After that time, Thales and Nioa seemed only to be waiting for Madritsch to accept those proposed terms.  It did so on 18 August 2011.  
  3. [399]
    I have considered the communications between the parties, in particular those from Mr Frohnwieser to Mr Whitty on 17 August 2011,[84] and from Mr Whitty to Mr Fulford-Talbot on 18 August 2011.[85]  From these, I conclude that in all the exchanges, including the 4 August 2011 email from Mr Whitty to Mr Fulford-Talbot, the parties were negotiating the terms of the Intentions letter and what became the NDA to address concerns raised by Madritsch about disclosing its intellectual property to Thales.  The only agreement by which the parties expressed an intention to be legally bound was the NDA.[86]  I am not satisfied that there was a distinct “side agreement” that bound Madritsch to disclose the Madritsch Solution to Thales, if Thales executed the NDA and signed the Intentions letter.  
  4. [400]
    The claim by the plaintiffs for damages for breach of the side agreement fails.  

Breach of duty of confidence

  1. [401]
    The plaintiffs also contend that Thales breached an equitable duty of confidence by making use of the information that the Madritsch Solution involved a double torsion AFL spring that imparted a significantly lower torque than the equivalent spring in V3 and V5.  The “long-standing equitable principles relating to fair-dealing with the work of another” inform the basis of this duty.[87]  
  2. [402]
    As Arnold LJ noted recently in Racing Partnership Ltd v Done Bros (Cash Betting) Ltd

“The clearest statement of the elements necessary to found an action for breach of an equitable obligation of confidence remains that of Megarry J in Coco v AN Clark (Engineers) Ltd [1969] RPC 41, 47:

“First, the information itself … ‘must have the necessary quality of confidence about it’. Secondly, that information must have been imparted in circumstances importing an obligation of confidence. Thirdly, there must have been an unauthorised use of the information to the detriment of the party communicating it.”

This statement of the law has repeatedly been cited with approval at the highest level: see Lord Griffiths in Attorney General v Guardian Newspapers Ltd (No 2) [1990] 1 AC 109, 268, Lord Nicholls of Birkenhead in Campbell v MGN Ltd [2004] 2 AC 457, para 13 and Lord Hoffmann in Douglas v Hello! Ltd (No 3) [2008] AC 1, para 111. (It is not, however, a complete statement of the ingredients of a successful claim: there is a further requirement, namely that the unauthorised use of information was without lawful excuse.)”[88]

  1. [403]
    For the reasons set out in [349] to [368] above, the information that the AFL springs in the Madritsch Solution kits were double torsion springs and had a significantly lower torque than the AFL springs in V3 and V5 had the necessary quality of confidence, not being public property or public knowledge.[89] 
  2. [404]
    When Thales received that confidential information, it owed Madritsch a duty not to make any unauthorised use of it to the detriment of Madritsch.  
  3. [405]
    Thales made use of the confidential information to modify its own V3 and V5 designs to create V6 and incorporate it in the new model of the Austeyr supplied to the ADF as part of the L125-3C program.  This was to the detriment of Madritsch.  Madritsch did not authorise it.  
  4. [406]
    As Lord Denning MR put it in Seager v Copydex Ltd: when a person has received information in confidence: 

“He should not get a start over others by using the information which he received in confidence.  At any rate, he should not get a start without paying for it.  It may not be a case for injunction or even for an account, but only for damages, depending on the worth of the confidential information to him in saving him time and trouble.”[90]

  1. [407]
    Thales proffered no lawful excuse for its unauthorised use of the confidential information. 
  2. [408]
    In Equity, Thales is accountable for any profits it acquired by this improper use of the confidential information.  The plaintiffs may recover any equitable damages or compensation for loss suffered by reason of Thales’ breach.  The formal proof and quantification of those matters are to be left for another day.  

Misleading and deceptive conduct

  1. [409]
    The plaintiffs’ final claim is for a statutory remedy for contravention of s 18(1) of the ACL.  This well-known provision has been described as a “legislatively imposed standard of normative behaviour”.  It provides: 

“A person must not, in trade or commerce, engage in conduct that is misleading or deceptive or is likely to mislead or deceive.”  

  1. [410]
    It is common ground that Thales’ communications with Nioa and Madritsch relevantly between 19 January and 13 June 2012 was conduct in trade or commerce within the meaning in the ACL.     

Context for the alleged misleading or deceptive conduct

  1. [411]
    The conduct of Thales, through Mr Evenden, between 19 January and 13 June 2012 occurred in the context created by Thales sending the Intentions letter and executing the NDA in August 2011.  During the period, Thales retained the six Madritsch Solution kits, notwithstanding the statement in the Intensions Letter that “if the Madritsch Solution is not suitable for our purposes” Thales would “return all components to Madritsch, and not use the Madritsch IP for any purpose whatsoever”.[91]
  1. [412]
    On 19 December 2011, Mr Evenden read and understood the draft sub-licence agreement and decided to reject Nioa’s offer to sub-license the right to use and manufacture the Madritsch Solution on those terms.  The same day, he decided to delay providing any response to Nioa or Madritsch rejecting the offer or indicating that the offer would not be accepted.  
  2. [413]
    On 9 January 2012, Mr Evenden circulated within Thales an email about “Response to Nioa/Madritsch Proposed Licence Agreement”.  The email and attached flow chart showed Thales intended to reject the sub-licence agreement, which Mr Evenden had described as “ridiculous” and “not acceptable”.  The email also proposed Thales would prepare a “brief” for a direct approach to Madritsch proposing Thales acquire the intellectual property for the Madritsch Solution in a “one-off buy”.  
  3. [414]
    According to the email, Mr Evenden proposed that Thales respond to Nioa, rejecting the sub-licence agreement and telling Nioa of Thales’ intention to discuss the matter directly with Madritsch, doing these things before it proceeded to “Brief Madritsch in Austria with Thales Austria in attendance” on about 12 February 2012. 
  4. [415]
    Mr Evenden sent the email and flow chart to Mr Wall and copied it to Mr Wootton, Mr Fulford-Talbot, Mr Higgins, Paul McCarthy who was Thales’ commercial manager, and Brian O'Connell who was Thales’ senior commercial lawyer.  He later sent a copy to Mr Eyles. 
  5. [416]
    Perhaps Mr Wall had other instructions for Mr Evenden.  Perhaps other views, when canvassed, prevailed.  That is not known.  For whatever reason, Mr Evenden did not respond to Nioa about the sub-licence agreement or tell Nioa that Thales would directly approach Madritsch, as his email and flow chart had proposed.  Instead, he and others within Thales, took steps to lead Nioa and Madritsch to believe that Thales was considering the offer with a view to either accepting it or making a genuine counteroffer.  

Conduct alleged to be misleading or deceptive and it’s effect

  1. [417]
    On 19 January 2012, Mr Evenden sent Mr Madritsch and Mr Whitty an email proposing a meeting with them in Vienna on 16 February 2012, while Mr Evenden was travelling in Europe. He said he would be accompanied by Hannes Boyer from the Vienna office of Thales.  The meeting was “to discuss the way forward with the hammer pack modifications.” This was the first direct written communication between a Thales employee and Mr Madritsch.  All other communications had been through Madritsch’s agent, Nioa. 
  2. [418]
    This direct approach to Mr Madritsch for a discussion was misleading and deceptive.  When it was sent, Thales, in the person of Mr Evenden, had decided not to pursue a sub-licensing agreement for the Madritsch Solution. Thales did so perhaps as early as 9 January 2012, when Mr Evenden told Mr Eyles the terms of the draft sub-licence were not acceptable and instructed Mr Eyles to set about developing a plan to finalise and “industrialise” the HMA1 design.  See [168] above.  
  3. [419]
    On 23 January 2012, Mr Whitty began seeking a date for a negotiation with Thales about the draft sub-licence agreement, proposing they meet during the “Pacific 2012 conference (31 Jan – 02 Feb) at Darling Harbour”.  Mr Whitty’s request for a meeting date shows he had been misled and deceived by Mr Evenden’s direct approach to Mr Madritsch. 
  4. [420]
    The same day, Mr Evenden replied:

“As you know we received your proposed sub-licence agreement on 19 Dec and due to a number of key personnel being on leave etc, it has not been possible to conduct a full financial and legal assessment of your proposals.  Also, I’m afraid I won’t be able to commence negotiations at Pacific 2012 [31 January – 2 February 2012].”  

  1. [421]
    Thales conduct, by Mr Evenden, in sending this reply was likely to mislead or deceive.  As Mr Evenden accepted in cross-examination, by this time he had already assessed the Nioa sub-licence agreement as “farcical”.  Thales was not intending to conduct a full financial and legal assessment of it.  Thales had not been delayed in doing so by any key personnel being on leave.[92]  Thales was not intending to commence negotiations with Nioa about the sub-licence agreement.  
  2. [422]
    On 24 January 2012, Mr Whitty asked Mr Evenden to nominate “a suitable date for us to meet to initiate the discussions” about the licence agreement between Thales and Nioa.  Mr Whitty’s request shows that he was misled and deceived by Mr Evenden’s earlier communications on 19 and 23 January 2012. 
  3. [423]
    On 24 January 2012, Mr Evenden responded to Mr Whitty’s request, in this way:

“I will get back to you reference sub-licence discussions once we have finalised our assessment of your proposal.”

  1. [424]
    This was another instance of conduct by Thales likely to mislead or deceive.  Thales was not working on an assessment of the Nioa proposal.  That assessment had been made by Mr Evenden on 19 December 2011.  As noted above, Thales was not intending to initiate negotiations with Nioa about the sub-licence agreement.  
  2. [425]
    On 16 February 2012, Mr Evenden met with Mr Madritsch and Mr Frohnwieser in Vienna.  As Nioa was the representative of Madritsch in Australia and as the draft licencing agreement was framed as one between Nioa and Thales, the parties had earlier agreed in correspondence that the licensing agreement would not be discussed in Nioa’s absence. According to Mr Evenden, due to that agreement, he did not discuss the Madritsch Solution “at all” in the meeting.  
  3. [426]
    On 23 April 2012, Mr Evenden responded to Mr Whitty’s request for “an update on the status of Thales’ intentions” with respect to the Madritsch Solution: 

“We have considered NIOA’s proposal presented as the licence holder of the Madritsch bolt unlocking technology.  On receipt of your proposal it became clear that the terms offered would add significant costs to our rifle product line and in turn our price to our customers.  As a consequence we had to ensure that not only was the proposed solution appropriate from a technical perspective but also provided value for money.  As you know we had pursued a number of internal solutions to the problem and due to the significant costs associated with the NIOA proposal we will now only move to negotiate your proposal once we have confirmed that the Madritsch Solution represents a significant benefit over our internally designed options.  I can keep you appraised as necessary.” 

  1. [427]
    Thales conduct, by Mr Evenden sending this email, was likely to mislead or deceive Nioa. Thales was not taking and had no intention of taking any steps to confirm that the Madritsch Solution represented a significant benefit over Thales’ “internally designed options”.   As noted above, Thales was not intending to initiate negotiations with Nioa about the sub-licence agreement.  In fact, as noted above, Thales, in the person of Mr Evenden, had decided not to pursue a licensing agreement for the Madritsch Solution.  By this date, Thales had decided to market its alternative solution to the bolt unlocking problem to the Commonwealth.  Thales had taken steps to implement this decision on 6 February 2012, when Mr Eyles instructed Mr Banks and Mr Higgins to add the HMA1 task to the work plan for the L125-3C Project, with an intention of completing the task by about 27 March 2012.  See [173] above.  On 20 February 2012, Thales had changed to the HMA1 mechanism for the L125 rifle design, instead of the Madritsch Solution.  By this time, Thales had used Madritsch’s confidential information to advance its own solution to the bolt unlocking problem.  It had manufactured and tested the prototype of its new design V6. 
  2. [428]
    On 24 April 2012, Mr Whitty thanked Mr Evenden for his response.  He expressed confidence that Madritsch would consider negotiating the price of the technology and recommended Thales consider offering a valuation for Madritsch to consider.  Mr Whitty’s request for an update shows that he continued to be misled and deceived by Mr Evenden’s communications on 19, 23 and 24 January 2012.  His 24 April 2012 response shows Mr Whitty was also misled or deceived by Mr Evenden’s email of 23 April 2012. 
  1. [429]
    On 3 May 2012, Mr Evenden told Mr Whitty that Thales welcomed “the opportunity to offer a valuation” of the sub-licence for the Madritsch technology from Nioa.  He continued: 

“Once we have confirmed that it represents a significant benefit over our internal design options we will be happy to progress.”  

  1. [430]
    He said Thales would get back to Nioa about the sub-licence on completion of the “DV&V phase” of the L125-3C program.   
  2. [431]
    Thales conduct, by Mr Evenden sending this email, was likely to mislead or deceive. Thales was not taking and had no intention of taking any steps to confirm that the Madritsch Solution represented a significant benefit over Thales’ “internal design options”.  As noted above, Thales was not intending to initiate or progress negotiations with Nioa about the value of sub-licence agreement.
  1. [432]
    It was not until the 13 June 2012 meeting in Paris that Thales informed Madritsch and Nioa, through Mr Evenden’s presentation that the sub-licence agreement was “cost prohibitive” and that Thales had reverted “to own solutions (V3)”.  The effectiveness of this disclosure to remedy the effect of Thales’ earlier conduct might be gauged by the fact that, at the end of the meeting, Mr Nioa invited Thales, through Mr Evenden and Mr Wall, to respond to the offer to sub-licence Thales to manufacture the Madritsch Solution, which Nioa had made in December 2011.  
  2. [433]
    On 22 June 2022, Mr Evenden responded to Mr Nioa stating that Thales would give the matter of a response to the draft sub-licence agreement attention quickly.  
  3. [434]
    On 11 July 2022, in the absence of a response, Mr Nioa followed up the email again.  Mr Evenden informed him that a response was in progress.  
  4. [435]
    On 13 July 2022, Mr Wall responded to Mr Nioa by email, advising that Thales did not require the Madritsch solution and would not make any counteroffer for a licensing arrangement through Nioa.  This appears to have been the first honest communication from Thales on this topic in nearly six months.         

Conclusion on whether Thales breached the standard of normative behaviour

  1. [436]
    Considered objectively in the relevant context, Thales’ conduct in its communications with Madritsch and Nioa on each of 19, 23 and 24 January, 23 April, and 3 May 2012 was likely to mislead or deceive Nioa and Madritsch by leading them to believe that Thales had a continuing interest and even an intention to enter into a sub-licence agreement with Nioa for the Madritsch Solution.  By engaging in that conduct, Thales contravened s 18 of the ACL.      

Did the breaches cause loss or damage?

  1. [437]
    I am satisfied Thales engaged in this conduct with the purpose of misleading or deceiving the plaintiffs so that they (and Nioa in particular) would not make a direct approach to the Commonwealth about supplying the Madritsch Solution kits for the use of the ADF until such time as Thales was ready to market its alternative solution to the bolt unlocking problem to the Commonwealth in the Land125-3C program.
  2. [438]
    I am also satisfied that Thales’ conduct served its intended purpose.  It had the effect of misleading the plaintiffs to believe that Thales had a continuing intention to negotiate and enter into an agreement permitting it to manufacture the Madritsch Solution.  Holding that belief, Noia (for itself and as agent for Madritsch) did not make a direct approach to the Commonwealth about supplying the Madritsch Solution kits for the use of the ADF.  The effect of the contravening conduct endured until 13 July 2022 when Thales told Nioa it would not pursue a sub-licence agreement.  
  3. [439]
    I reject Thales’ submission that any loss the plaintiffs may have suffered by Thales’ contravention of s 18 of the ACL “must have crystallised upon the provision by the plaintiffs to Thales of the Madritsch kits.” This odd contention seems to imply that once Thales have Madritsch’s confidential information it’s fate was sealed. In any case, it misunderstands the plaintiffs’ s 18 claim. 
  4. [440]
    While the plaintiffs were misled, Thales completed the design and some testing for its V6 mechanism for inclusion in its L125-3C offering to the Commonwealth.  The plaintiffs lost the opportunity to renew a direct approach to the Commonwealth to supply the tested, accepted and manufacturable Madritsch Solution before Thales had an alternative ready.  They lost this opportunity because of Thales’ conduct that contravened s 18 of the ACL. The opportunity was valuable.  It is not necessary to quantify the value for present purposes.  

Costs

  1. [441]
    The plaintiffs have substantially succeeded in their case against Thales on liability.  In the absence of any relevant impediment, they should be entitled to recover their costs of the hearing.

Draft order to be prepared

  1. [442]
    The proceeding is at an unusual stage, as only the liability issues have been determined.  I will direct the parties to confer and submit an agreed draft order, or separate draft orders, consistent with the findings in these reasons.

Footnotes

[1]There are several models, including the F88(SA1), F88(SA2) and F90, but it is convenient to refer to them by the collective generic name Austeyr.

[2]The Lithgow factory was once under the control of the Commonwealth Department of Defence.  It passed to Australian Defence Industries or ADI.  In 2006, Thales’ parent acquired 100% ownership of ADI and, in time, it was renamed Thales Australia Limited.

[3]Pursuant to the Order of Martin J dated 13 May 2020. 

[4]Lower than the force exerted by the AFL in the standard hammer mechanism, where the hammer springs exerted torque on the AFL. 

[5]Mr Grima was the author of the October 2008 ARMTSPO report. 

[6]It was to be made of 1.4mm diameter 17-7 precipitation hardening stainless steel with specification AMS 5678B.

[7]It is clear from the context of the report that the fractions in this column are the number of successful test firings over the total number of test firings or attempts.  They are not the number of passes over the number of fails. 

[8]According to his position paper, Thales had adopted “a system engineering approach” to the HMA1 Project “to ensure a systematic approach was taken to all aspects of the project” and “All work in support of this project is being documented”.

[9]It does not state the date of the live fire tests.  These may have been those conducted on 14 July 2009. 

[10]The report does not elaborate, but it would appear the “butt” of the Austeyr may have included some components that interacted with the gun lock group or the hammer mechanism assembly.

[11]In cross-examination, Mr Eyles accepted Thales conducted the tests were between December 2010 and March 2011.  I have disregarded the contrary evidence in his first prepared statement.

[12]A test of the friction offered by a surface when it comes into contact with an object.  It is sometimes described as a measure of the slip potential or slip resistance of a surface. It does not involve firing a round from the rifle. 

[13]   F88 SA2 model.

[14]Again, the 2010 F88 SA2 model.

[15]Higher Technical School for Weapons and Mechanical Engineering.

[16]There was some contest about whether the range was 8 to 9N or 8 to 10N, as the plaintiffs’ evidence included both ranges.  I have adopted 8-10N to which Mr Madritsch swore, which include the narrower range.  This choice makes no difference to my conclusions in these treasons.  

[17]Oberst (Colonel) is the highest staff officer rank in the Bundesheer.

[18]The Light Weapons Section of the Bundesheer.

[19]19 Specialist inspector.

[20]Certificate of Examination.

[21]The Madritsch Solution was listed as Madritsch’s prior intellectual property in the schedule of the draft agreement Mr Whitty had sent to Mr Wootton on 24 February 2011.

[22]   Thales had left development of the HMA1 Project in abeyance since the end of 2009.  

[23]Although a draft test firing instruction was prepared, Thales had not undertaken any testing of it.  

[24] 24 Presumably meaning lower in torque than the hammer springs.

[25]This is a reference to the agreement executed on 8 June 2011.

[26]From his measurements, Mr Williams had calculated the average vertical force exerted by the AFL in V3 as about 28.4N.  Mr Steele’s later record of the disengagement force in the V3 models gives an average of 43.2N.  Mr Steele’s measures of the force in V5 produce an average disengagement force of 42.5N.  Each of these is significantly higher than the 8 to 10N in the Madritsch Solution.  As engineers, it is more likely than not that Mr Eyles and Mr Basladynski noticed that a significantly lower torque was exerted by the AFL springs in the Madritsch Solution kits compared with the AFL springs in V3 and V5.  Mr Basladynski noted a significant reduction in his later draft report on testing the Madritsch Solution kits. 

[27]This is the draft Technical Report TR-12007 Revision A, dated 18 January 2012.

[28]It is convenient to turn to this and related evidence when considering the plaintiffs’ claim under the ACL. See [409] to [440].  In the interim, the balance of this narrative is focussed on the work of Thales on what became a new version of the HMA1.

[29]DATL is an acronym for “design acceptance test lot”, which Thales distinguished from PATL or “production acceptance test lot”.  These are two of the tests Thales usually undertook before a part or a complete product was approved for production. 

[30]Another draftsperson, Bruce Kennedy, who was more senior than Mr Steele, was also involved to some extent in checking Mr Steele’s work. No evidence was adduced from Mr Kennedy.

[31]Mr Steele called his first design version 6 (V6).

[32]Mr Steele entitled the drawing “Spring – Lever Automatic Firing CGF”. It is the earliest drawing by Mr Steele in the trial bundle. 

[33]In his first statement Mr Steele said he was inspired by clothes pegs while hanging out washing at a rented house.  It was shown that he could not have stayed in that house until some months after he drew the double torsion AFL spring.  

[34]Mr Eyles, Mr Wachsberger, and Dr Pratt. 

[35]The material specified was “Steele to Japanese music wire spec: JIS G 3522 SWPB”.

[36]The external size of the AFL spring would be fixed by the dimensions of the AFL and the surrounding body of the Austeyr rifle into which it fitted. 

[37] The torque exerted by the spring when twisted may be expressed in Newton-metres (Nm) or Newtonmillimetres (Nmm).  The force required to twist a torsion spring through a full rotation (360⁰) is sometimes called the spring rate.  As a force, it may be expressed in Newtons (N).  There is no reason to think Mr Steele was embarking upon an examination of any such property of the AFL spring. 

[38]It seems Mr Steele intended to refer to the gun lock group.  This is another indication of his limited understanding of the technology he was drawing and modelling. 

[39]Mr Wachsberger explained that the disengagement force included the “no load” force and the additional resistance force (or frictional force) associated with disengaging the back of the AFL and the notch on the hammer. 

[40]Torque @ 76⁰ ± 0.5⁰ = 283 Nmm ± 30 Nmm; Torque @ 68⁰ ± 0.5⁰ = 315 Nmm ± 30 Nmm.

[41]Torque @ 35⁰ ± 0.5⁰ = 40 Nmm ± 5 Nmm; Torque @ 25⁰ ± 0.5⁰ = 80 Nmm ± 5 Nmm.

[42]Owing to Mr Steele’s use of Newtons as a measure, it was common ground at the hearing that the exact reduction in the torque exerted by the AFL spring could not be calculated.  Mr Wachsberger’s evidence, which I accept, was that the torque reduction was probably greater than 65%. In a 28 August 2012 instant messenger exchange with Mr Eyles, Mr Steele referred to the change to the AFL spring from V5 as showing “70% less force required for the Gun lock [group] to return ‘home’”.  

[43]It remained Japan steel music wire to JIS G 3522 SWPB.

[44]Gibbens supplied quotes for the three springs that evening.  Mr Steele followed up the prospect of Thales toolroom being able to modify AFLs for the HMA-1 Project.

[45]At this time, Mr Evenden was the Testing and Evaluation Manager at the Lithgow factory.

[46]This was assembled by Mr Steele using parts made according to the drawings he had completed. 

[47]This was a reference to the Thales brief sent by Mr Fulford-Talbot on 24 June 2012. 

[48] On 22 October 2012, Mr Basladynski directed Michael McGrath, a test and evaluation technician, to return them to Nioa.  Mr McGrath tried to have a colleague to arrange the return.  When this failed, Mr McGrath put the six kits in the post to Nioa on 23 October 2012.  It appears they were delivered on 25 October 2012.

[49] The only “details” Thales had “given over” were in the Thales brief. It appears Mr Eyles had prepared the Thales brief.  Mr Evenden’s email to Mr Fulford-Talbot attaching it had been copied to him. 

[50] i.e. fired by a human operator rather than from a mounted stand. 

[51] The approach adopted to expert evidence resulted in some unsatisfactory outcomes.  The joint expert report is one.  It is replete with statements that one or other expert “agrees” with a proposition, when it is apparent from their evidence – including parts of the joint report itself – that the proposition is not agreed.

[52] Expert Report (27 March 2020), Supplementary Report (15 May 2020), Response [to Dr Pratt’s reports of 7 May 2020 and 15 June 2020] (10 July 2020), and Reply to [Dr Pratt’s report of 15 July 2020] (16 July 2020).  

[53] Then the South Australian Institute of Technology.

[54] Expert Witness Report (7 May 2020), Expert Witness Report Phase II (15 June 2020), and Expert Witness Report Clarification Response (15 July 2020). 

[55] He suggested a “design optimisation” approach had been adopted to establish the lowest force required to consistently engage and operate the hammer mechanism.  There was no evidence such an approach had been adopted by Mr Steele or anyone else within Thales.

[56] The likely source of this change is Mr Eyles, in reviewing Mr Steel’s work, or Mr Egan in correcting the inappropriate force measure in Newtons and calculating (or showing Mr Steel how to calculate) the torque (or rotational force) in Newton-millimetres.

[57]Shared by Mr Evenden. 

[58] Mr Williams had measured the force applied by the AFL in his versions of the HMA1.  However, Mr Williams had not included specifications for any such force in his drawings, to which Mr Steele had access.  As early as 19 June 2009, in drawing U-20343-0, Mr Williams had shown the AFL spring with the spring specifications in torque: “@ 76⁰ ± 0.5⁰ = 283 Nmm ± 30 Nmm”; and “@ 68⁰ ± 0.5⁰ = Nmm ± 30 Nmm”

[59] Those communications also show Mr Steele’s limited knowledge and understanding and the extent of his reliance on input from others.  This was consistent with his own assessment of his role as within Thales as a draftsperson, not an engineer, and his lack of understanding of torsion springs.

[60] That is, reduced from the force exerted in V3 and V5. 

[61] As noted above, the different angles at which the torque is specified makes a direct comparison of Mr Steele’s figures on the two drawings inappropriate.  

[62] If the disengagement force had been reduced to 22N on 28 February, a further 65% reduction would bring it closer to the 8 to 10N force in the Madritsch Solution, which Mr Eyles had examined with Mr Basladynski in September 2011. 

[63] Hampton Court Ltd v Crooks (1957) 97 CLR 367, 371-372 (Dixon CJ).

[64] There is no evidence that Mr Steele knew of the source of the information he used or that he was aware of the terms of the NDA. Mr Eyles was aware of the NDA.  He said he had “no specific recollection of reading” it.  

[65] (1948) 65 RPC 203 at 215.

[66] [1979] Ch 227 at 248. 

[67] Racing Partnership Ltd v Done Bros Ltd (CA) [2021] 2 WLR 469 at 484 [48] (Arnold LJ)  

[68] See, for example, cl 7.

[69] Optus Networks Pty Ltd v Telstra Corporation Ltd (2010) 265 ALR 281 at 290. 

[70] Saltman Engineering Co Ltd v Campbell Engineering Co Ltd (1948) 65 RPC 203 at 211 (Lord Greene MR). 

[71] Ansell Rubber Co Pty Ltd v Allied Rubber Industries Pt Ltd [1967] VR 37 at 49 (Gowans J). 

[72] Prior to obtaining access to the Madritsch kits, Thales had not identified any need to significantly reduce the disengagement force of the AFL on the gun lock group by significantly reducing the torque exerted by the AFL spring.  Thales had no reason to determine the effects of doing so.  Nor had Thales (or ARMTSPO) identified that a double torsion spring would be preferable to a single torsion spring to drive the AFL.

[73]Interfirm Comparison (Australia) Pty Ltd v Law Society of New South Wales [1975] 2 NSWLR 104, 119C (Bowen CJ in Eq) (Interfirm Comparison)

[74]Leopold-Franzens-Universität Innsbruck.

[75]Civil Service Law 1979.

[76]RTA, s 2(2). 

[77]Isaac v Dargan Financial Pty Ltd (2018) 98 NSWLR 343 (Isaac v Dargan) at [59]-[61]. 

[78][2011] NSWCA 418.at [31]

[79]at [32], approved in Isaac v Dargan at [65].

[80]Tullett Prebon (Aust) Pty Ltd v Purcell (2008) 175 IR 414 at [65]. 

[81]RTA, s 4(1); Isaac v Dargan at [62].

[82]At the time they were agreed, each covenant could operate to protect Thales in respect of any Proprietary Information it disclosed to Madritsch as well any Madritsch disclosed to Thales.  

[83][2001] NSWCA 111 at [165].  His Honour referred to Kone Elevators Pty Ltd v McNay (1997) ATPR 41-564 at 43,833 and the cases considered by that decision, including Orton v Melman [1981] 1 NSWLR 583.

[84]Copied to Mr Madritsch, Mr Nioa and Mr Toms.

[85]Copied to Mr McCarthy, Mr Higgins, Mr Flynn, Mr Evenden, Mr Nioa and Mr Toms. 

[86]This is not to say that the plaintiffs did not rely upon the Intentions letter in entering into the further agreement and in disclosing the Madritsch Solution to Thales.

[87] Interfirm Comparison [1975] 2 NSWLR 104, 118. 

[88][2021] 2 WLR 469, 483 [44]-[45], CA (9 Oct 2020). Although Arnold LJ dissented in certain respects, Lewison LJ (and Phillips LJ) did not challenge this summary of the law. 

[89]Saltman Engineering Co Ltd v Campbell Engineering Co Ltd (1948) 65 RPC 203 at 215 (Lord Greene MR).

[90][1967] 1 WLR 923, 931-932 CA.

[91]Thales did not send the kits back to Nioa until 23 October 2012.  

[92]On 20 December 2011, Mr Annells had asked another Thales employee to “work up a ROM cost ASAP” for Mr Evenden.  Nothing more than this rough order of magnitude assessment was made.

Close

Editorial Notes

  • Published Case Name:

    Madritsch KG & Anor v Thales Australia Ltd

  • Shortened Case Name:

    Madritsch KG v Thales Australia Ltd

  • MNC:

    [2021] QSC 170

  • Court:

    QSC

  • Judge(s):

    Bradley J

  • Date:

    26 Jul 2021

Appeal Status

Please note, appeal data is presently unavailable for this judgment. This judgment may have been the subject of an appeal.

Require Technical Assistance?

Message sent!

Thanks for reaching out! Someone from our team will get back to you soon.

Message not sent!

Something went wrong. Please try again.