FEDERAL COURT OF AUSTRALIA

Seiko Epson Corporation v Calidad Pty Ltd [2017] FCA 1403

ORDERS

THE COURT ORDERS THAT:

1.    The parties confer, and by no later than 13 December 2017, provide to my associate joint draft short minutes of order reflecting the findings set out in these reasons, including on the question of costs and tracking any areas of disagreement.

2.    On or before 19 December 2017, the parties provide written submissions (of not more than 5 pages) setting out the basis on which they contend their position is to be preferred.

3.    The matter be listed for case management at 9.30 am on 20 December 2017.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

BURLEY J:

1.    INTRODUCTION

1.1    Overview

1    In the fiercely competitive world of computer printers and ink refills for those printers, the first applicant, Seiko Epson Corporation (Seiko) is a global player. It is a manufacturing company resident in Japan which sells printer products, including printer cartridges, under or by reference to the trade mark EPSON (Epson trade mark). It sells or authorises the sale of printer cartridges (original Espon cartridges) worldwide. The original Epson cartridges embody the invention claimed in the 2 patents in suit; patent no. 2009233643 (643 patent) and patent no. 2013219239 (239 patent). After their sale, the original Epson cartridges are used, and some are obtained by a third party to the proceedings, Ninestar Image (Malaysia) SDN. BHD (Ninestar). Ninestar then refills them with ink. In doing so, it makes modifications to them and then, relevantly to the present proceedings, sells them to the third respondent, Calidad Distributors Pty Ltd (CDP) which imports them into Australia and sells them.

2    The central dispute in these proceedings concerns the right of a patentee to control or limit what may be done with a patented product after it has been sold. This gives rise to consideration of the intersection of the general rights of property ownership in a chattel once sold, and the monopoly rights conferred on a patentee under the Patents Act 1990 (Cth) (Act). When a patentee sells a chattel that embodies an invention claimed in a patent, can the patentee restrain the subsequent use made of it by a purchaser or a successor in title to the purchaser? This case explores that question. The case also raises a question under the Trade Marks Act 1995 (Cth) (TM Act) as to whether the registered owner of a trade mark is able to bring civil proceedings for the breach of a criminal offence section under the TM Act. Other causes of action that arise include trade mark infringement, an action for misleading or deceptive conduct pursuant to s 18 of the Australian Consumer Law (Schedule 2 of the Competition and Consumer Act 2010 (Cth)) (ACL) and breach of contract. Seiko and its exclusive distributor in Australia, Epson Australia Pty Ltd (EAP) (referred to collectively as Seiko), bring these proceedings against CDP and three related companies (collectively, Calidad) for each of these causes of action. They contend that the first respondent, Calidad Pty Ltd (CPL), the second respondent, Calidad Holdings Pty Ltd (CHP) and the fourth respondent, Bushta Trust Reg (Bushta) are liable together with CDP as joint tortfeasors.

1.2    Summary of the issues and conclusions

3    The original Epson cartridges are ink cartridges that are sold by or with the permission of Seiko. Eleven different types of these cartridges are relevant to these proceedings. Each is compatible with 1 or more printers. When the ink in an ink cartridge runs out, it is necessary for the cartridge to be replaced. Seiko has a lively, worldwide business that supplies replacement ink cartridges for its printers. Calidad operates a business that competes with Seiko in the supply of replacement ink cartridges for Seiko’s printers in Australia. The Calidad products are sold under 5 different Calidad product numbers (250, 253, 258, 260H and 260S) (Calidad products). Eleven different original Epson cartridges (T0711, 125, 126, 18, ICLM50, 98, 73N, 133, 138, T200 and T200XL) were modified by Ninestar to produce one or other of the Calidad products. To complicate matters, different modifications were made at different times.

4    In relation to its claim for patent infringement, Seiko contends that Calidad infringes the claims of its patents by importing and selling the Calidad products. Calidad accepts that its products fall within the relevant claims of the patents, but submits that it has a complete answer to the allegation of patent infringement based on the fact that Seiko released the original Epson products on the market and, upon their sale, authorised any purchaser and subsequent owner of the cartridges to treat them as an ordinary chattel. More particularly, Calidad submits that Seiko implicitly authorised Ninestar to refill and modify the original Epson cartridges, after which Ninestar was free to sell them to Calidad who was then free to import them into Australia and sell them. In the United States, upon the first sale of a product that embodies a claimed invention, the patent rights in that product are said to be exhausted, which means that the patentee can no longer exert control over the product; Impression Products, Inc. v Lexmark Intern., Inc. 137 SCt 1523 (2017) (Lexmark). In Australia, a different principle has developed which arises from National Phonograph Co of Australia Ltd v Menck (1911) 12 CLR 15 (Privy Council) (National Phonograph). Applying the principles set out in that case, I have found that Seiko’s infringement claim succeeds for Calidad’s past range of products, but not in respect of its current products. In order to explain how this conclusion has been reached, a fairly detailed exegesis is required of the following topics; the relevant law, the patents in suit, the respective original Epson products and Calidad products and, finally, analysis of the modifications made to create the Calidad products. These matters are addressed in sections 2 – 6.

5    Seiko next contends that the Epson trade mark has been used by Calidad on the integrated circuit chip of 2 sample Calidad products, and as a consequence, Calidad has infringed that trade mark pursuant to s 120(1) of the TM Act. In section 7 below, I conclude that the impugned use was not use of the word Epson in a trade mark sense and that the cause of action is not made out.

6    Seiko thirdly contends that Calidad has acted in breach of statutory duty by dealing with the Calidad products in circumstances where the Epson trade mark has been obscured or removed from them. This, it submits, amounts to a breach of s 148 of the TM Act. That section falls within Part 14 of the TM Act, which deals with criminal offences. As a threshold matter, I have determined in section 8 of these reasons that this section of the TM Act does not confer upon Seiko a separate entitlement to sue for breach of statutory duty. In the event that it were so entitled, I have concluded that upon a proper construction of s 148 Calidad has, in any event, not acted in breach.

7    Fourthly, Seiko contends that by the use of the word “new” in the phrase “Status: New product, now available” as it appears on a Calidad website, Calidad has engaged in misleading or deceptive conduct in breach of ss 18(1), 29(1)(a) and/or 29(1)(c) of the ACL. In section 9 I find that that cause of action is not made out.

8    The fifth cause of action is for breach of a settlement deed entered between 3 of the respondents and Seiko in order to resolve a prior dispute. Calidad admits that CDP acted in breach of a provision of the deed. The only issues requiring resolution are the remedy, if any, in respect of that breach, whether a clause proscribing infringement of the 643 patent has also been breached, and the liability of the other respondents for it. In that regard in section 10 I find that CPL, CHP and CDP have acted in breach of the deed.

9    Finally, in respect of the question of liability of the non-CDP Calidad parties for infringement of the patents, I find in section 11 that each is a joint tortfeasor together with CDP.

10    At this point it is worthwhile to note that in this hotly contested litigation the parties cooperated successfully to limit and refine the issues between them so as to minimise the areas of dispute that require determination. They are to be congratulated. Part of this cooperation resulted in the settlement of a cross claim brought by Calidad. Another led to an agreement between the parties that it was only necessary for the Court to consider one claim of one patent in order to determine the patent infringement issues (being claim 1 of the 643 patent). A further aspect of cooperation was to cut down the number of categories of products that required specific consideration as much as is reasonably possible. That still left some 9 categories of Calidad products requiring consideration, but the task would have been much more cumbersome had the parties not largely agreed on those categories.

1.3    The witnesses

11    A number of witnesses were called to give evidence and several of them were cross-examined. No submissions were made to the effect that any witness lacked credibility. In general I found that all of the witnesses endeavoured to assist the Court in their evidence. Except where otherwise indicated below, I have accepted their evidence.

The Seiko witnesses

12    Keisuke Nishida is a member of the senior staff of Seiko. He has worked for Seiko since April 1998 in roles including head of development of the printhead of inkjet printers from 1999 to 2003 and the management of Seiko’s intellectual property. He gives evidence about the original Epson cartridges and the equivalent Calidad cartridges, providing photographs of each. In his second affidavit he gives evidence that the memory types of the integrated circuit chips (or memory chips) used in the original Epson cartridges are either Electrically Erasable Programmable Read Only Memory (EEPROM) or Ferroelectric Random Access Memory (FRAM). Mr Nishida was cross-examined with the assistance of a Japanese interpreter.

13    Diane Karam, is the Business Manager for Consumables, Moverio and LabelWorks employed by EAP. She gives evidence about the purchase of samples of the Calidad products in Australia. She was not cross-examined.

14    Yi Lu is a solicitor employed by Allens, the firm representing Seiko in these proceedings. She gives formal evidence about the patents in suit, the Settlement Deed, the website hosted at the domain name calidad.com.au and samples of cartridges and their components relevant to the proceedings. She was not cross-examined.

15    William Greg Hawkins is a recently retired senior engineer who completed a Bachelor of Science in Chemistry with honours from Union College, Schenectady, New York in 1976, a Masters in Science in physical chemistry from Cornell University in 1978 and was awarded a PhD in physical chemistry from Cornell University in 1980. He has held a number of senior research and development positions in various corporations involved, amongst other things, in inkjet technology and his engineering and management experience covers a number of fields relevant to the current dispute, including; device and circuit design, process, circuit and fluidic simulation, packaging, ink development and integration into printheads/maintainability and print driver development. He gives evidence in relation to the process for remanufacturing cartridges, including matters relevant to memory chips, the chip replacement process, the testing of chips, the programming of memory chips, integrating chips with cartridges, refilling ink and other similar and related matters. He also responds to the evidence of Mr Li, Mr Shen and Mr Knight. Dr Hawkins was cross-examined.

16    Fady Anton is a Senior Product Manager employed by EAP. He gives evidence about some sample Calidad products that he purchased in Australia. He was not cross-examined.

17    Vince Di-Mento is the Service Network Manager of EAP. He gives evidence going to the provenance of certain parts of the original Epson cartridges that were put into evidence. He was not cross-examined.

The Calidad witnesses

18    Bo Li is a Supervisor Software Engineer within the Research and Development department at Apex Microelectronics Co Ltd (AMC). He received a Bachelor’s Degree in Automation from Hebei University of Engineering in 2011 and since that date has been employed as a software engineer at AMC. In January 2016 he was promoted to his current role. AMC is a manufacturer and supplier of equipment that rewrites the data stored on original equipment manufacturer (OEM) integrated circuit chips that are attached to integrated circuit boards used in OEM printer cartridges. Mr Li gave two affidavits that explain the role of AMC in taking the original Epson cartridges and producing the Calidad cartridges. Mr Li was cross-examined with the assistance of an interpreter.

19    Wei Shen is a production engineer employed by Apex Technology Co Ltd who works at Ninestar in Selangor, Malaysia. He explains the procedure that Ninestar uses to restore used Epson cartridges to working condition for the purpose of their supply to Calidad. He gives evidence that Ninestar is one of the world’s largest manufacturers of generic printer consumables and that it remanufactures used OEM toner and inkjet printer cartridges. Mr Shen has been employed by Ninestar as a production engineer since 2007 and since 2014 he has been in charge of managing the restoration process at Ninestar’s facility in Malaysia. Mr Shen gave two affidavits in the proceedings and was cross-examined with the assistance of an interpreter.

20    David Gordon Gibbons is a director of Recycling Times Media Corporation (China) and has been active in the Australian industry involved in the remanufacturing of used printer cartridges since 1992. He has been a member of the Australasian Cartridge Remanufacturers Association since 1994 and since then has been involved in the provision of educational and consulting services to the global remanufacturing industry. Mr Gibbons was not cross-examined.

21    James Kenyon is a sales manager employed by CDP and the son of the founder of Calidad, Robin Kenyon. He has an Advanced Diploma in Business and Marketing and commenced working at Calidad in about 2004. He gives evidence about the activities of Calidad in relation to the Calidad products. He was cross-examined.

22    Paul Charles Knight is a consultant engineer employed by Marinov Consulting Services. He obtained a Bachelor of Science majoring in Computer Science from the University of New South Wales in 1972 and in 1974 obtained a Bachelor of Engineering (with honours), majoring in electronic design, from the same university. He completed a Master of Engineering Science in 1976. Since 1973 Mr Knight has worked in the field of software engineering. Since 2000 he has worked for Memjet Australia (or its predecessor) which is involved in technologies and components relevant to the computer printing industry and in that year became involved in an ongoing project focusing on the improvement of the speed, efficiency and quality of colour inkjet printing. Mr Knight gives evidence of some background aspects of the technology concerning the component parts of inkjet printers. Much of that evidence was uncontested and forms the basis of the background technology in section 2.1 below. He also gave evidence of his understanding of the disclosure of the patents in suit in the proceedings. He was provided with copies of the affidavit evidence of Mr Li and Mr Shen and gave certain opinions as to the processes involved in converting the original Epson cartridges into the Calidad cartridges. Mr Knight was cross-examined.

23    Philip Kerr is a solicitor employed by Allens. He gave an affidavit proving formal matters relevant to the claim for breach of statutory duty under s 148 of the TM Act. He was not cross-examined.

2.    BACKGROUND FACTS

2.1    Background technology

24    A printed circuit board or integrated circuit board is a non-conductive material (such as plastic or fibre glass) that has electronic components mounted onto it. The electronic components are connected together to form one or more working circuits or assemblies via strips of a conducting material (such as copper), which are printed or etched onto the board (also known as "traces"). A printed circuit board can have multiple layers with traces on both external and internal layers, and internal planes, which are layers of copper internal to the printed circuit board, used to distribute power to the components.

25    The electrical components on a printed circuit board generally include:

(a)    traces and (in some cases) internal planes;

(b)    pads;

(c)    one or more integrated circuits (described further below), which connect to the board via leads; and

(d)    discrete electronic components (such as, for example, transistors, capacitors, inductors, light emitting diodes ("LEDs")).

26    Integrated circuits (also known as "chips") are a collection of electronic components arranged on a piece of semi-conductor material (such as silicon). The components are connected in such a way to enable the circuit to control the flow of one or more electrical currents through the circuit in order to achieve some useful function. The components of an integrated circuit are commonly microscopic.

27    The feature that distinguishes an integrated circuit from an electrical circuit more generally is the fact that the circuit is 'integrated', meaning it is a single device containing multiple components rather than an electronic circuit that is built out of discrete components.

28    Piezoelectric devices are devices that are made of certain materials (for instance ceramics or quartz) which mechanically respond when the voltage is applied and removed. They generally require high voltages and both the high voltage requirement and the fact that they respond mechanically when a voltage is applied means that piezoelectric devices are not generally built into integrated circuits.

29    The electrical currents that flow through an integrated circuit and between integrated circuits and/or other discrete components on one or more printed circuit boards often constitute digital information in the form of bits, which are the smallest unit of data, comprising single binary values, either 0 or 1.

30    Integrated circuits (and the boards on which they sit) can be designed to perform a number of different functions. For example, in relation to printers, they can include memory, which stores digital information, and communication devices between the printer and the computer (for example in a Universal Serial Bus (USB), or Ethernet).

31    Short circuiting is a problem for any electrical circuit. Short circuiting occurs when an accidental path is created in an electrical circuit, which results in an electrical current moving down an unintended path. For example, short circuiting may occur if a conductive material (such as metal or liquid) connects two conductors together, resulting in the electrical current flowing from one conductor through the conductive material to the other conductor. Short circuiting can interrupt the electrical signal in the circuits, thereby preventing the circuit from functioning properly. In cases where a high voltage is directed onto a low voltage circuit, this can also cause the components in the low voltage circuit to fuse, thereby rendering it (and consequently the electrical device) unusable. This is because modern integrated circuits, which tend to be powered at quite low voltages, are susceptible to damage from higher than specified voltages.

32    An electronic printer is a device that takes an electronic representation of an image and produces an impression of that image on a medium (such as paper). Printers in the Australian household consumer market generally include inkjet and laser printers. Inkjet printers have a number of very small nozzles through which liquid ink is forced, in the form of small droplets, onto the paper. Laser printers produce an image electrostatically on an intermediate surface. That electrostatic image attracts toner and applies it to the paper, using heat to set the toner onto the paper.

33    Inkjet printers (as well as most other printers) comprise hardware and firmware. Hardware is the physical machinery that exists inside the printer. Firmware is the software or set of instructions programmed onto the hardware. Electronic printers generally also need to be associated with a printer driver, which is a piece of software commonly installed on the computer that is connected to the printer.

34    Inkjet printers can differ substantially depending on the make and model of the printers and there are many different ways of achieving the same result. However, the key aspects that inkjet printers commonly include are:

(a)    a power source, which includes a converter that converts the mains electricity (which in Australia is 240 volts) to a lower voltage that is useable by the printer hardware;

(b)    a communications interface which provides an electrical or wireless interconnection with a computer (this could be via, for example, USB, Ethernet, Wireless network ("wifi") or Bluetooth technology);

(c)    a main processor, which would typically be an integrated circuit on a printed circuit board of the type described above, that handles all the basic system functions that control the printer mechanism and convert the image data into ink ejection control signals.

(d)    the main processor will have some memory which may be on the same integrated circuit or an external integrated circuit on the printed circuit board. A memory is an integrated circuit that commonly contains cells that store electrical charge (which represents the binary data that I refer to above) and circuitry to read and write the electrical charge to the cells. The main processor addresses cells to read and write binary data. The pattern of the bits represents the digital data;

(e)    one or more ink cartridges (colour inkjet printers commonly utilise four colours, cyan, magenta, yellow and black; the colours will commonly be housed in four separate cartridges, however, others have multi-colour cartridges with separate ink orifices for each colour);

(f)    a printhead assembly, which contains the components (including the nozzles referred to above and the heating element or piezoelectric device). The printhead assembly ejects the ink from the ink cartridges onto the page (in some cases the printheads are located in the ink cartridges but in other cases the printheads are located on the printer itself). The printhead assembly, if separate from the ink cartridge, commonly includes a needle designed to pierce the ink cartridge to enable ink to flow, via tubes and sometimes a filter, down to the nozzles, where a heating element or piezoelectric device cause droplets of the ink to be pushed out of the nozzles onto the paper;

(g)    a carriage, which holds the ink cartridges and printhead assembly, that is attached to a mechanism to move the carriage (including the ink cartridge and the printhead assembly) along the width of the paper;

(h)    a feed paper mechanism, which commonly includes paper input and output trays and one or more rollers and motors that interact to feed the paper through the printer; and

(i)    a control panel to allow the user to interact with the operation of the printer.

35    Depending on the sophistication of the inkjet printer, it may also have a number of sensors to detect (for example):

(a)    whether the ink cartridges are installed in the printer;

(b)    the ink levels in the printer cartridges;

(c)    whether the lid of the printer is open or closed;

(d)    whether there is paper in the input tray;

(e)    whether there is too much paper in the output tray; and

(f)    whether there is a paper jam in the printer.

36    There are various types of sensors used in relation to inkjet printers such as optical, mechanical and piezoelectric device sensors.

37    There are many ways in which the information about the remaining ink levels can be stored. However, a common way as at December 2005 (priority date) and today was to store the remaining ink level information on an integrated circuit on the printer cartridge itself. These types of integrated circuits are called EEPROM, and form part of a non-volatile memory family. The data stored in the memory in this family can be retained by the integrated circuit even when it is not powered (another example of a type of memory that falls into this class is a flash drive, also known as a USB memory stick).

38    By storing the information about the ink levels in a memory on the printer cartridge, the information about how much ink is in the printer cartridge is transportable with the cartridge, so the ink information is retained even if the cartridge is removed and replaced or moved to another printer.

39    Having memory on the ink cartridge itself also allows physically identical ink cartridges containing different amounts of ink (for instance small, medium and large volume cartridges) to be sold and used in the same printer. This is because the available ink volume can be determined from information written to the memory on the cartridge.

40    Storing the information concerning the ink levels on an integrated circuit on the printer cartridge requires a communication protocol to be established between the memory on the cartridge and the main processor in the printer. This is because it is the main processor that contains the firmware capable of receiving, processing and re-writing the data onto the memory on the cartridge. The only function of the memory on the cartridge (at least in this context) is to store information and communicate it back to the main processor when prompted. The main processor may also have the ability to re-write the information on the memory. The information on the memory is stored as bits, i.e. as a series of binary numbers (0s and 1s), which is described above.

41    Generally, the communication protocol involves the following:

(a)    to read data from the memory on the cartridge, the main processor sends a request to the memory on the cartridge over a data line. The memory responds by sending the requested data back to the main processor over the same data line, which enables the main processor to determine the binary values that are stored on the memory on the cartridge;

(b)    for the main processor to write data to the memory on the cartridge, the main processor sends a request which includes the data to be written to the memory on the cartridge over the same data line referred to above. This causes the hardware within the memory to alter the charge pattern written into the memory cells, i.e. the binary values that are stored on the memory on the cartridge.

42    Estimating the amount of ink that remains in the cartridge in the manner described above can involve the main processor tallying the number of (for example) pages printed or ink drops ejected intermittently (for example at the end of the print job or at the end of each page). The main processor then sends a communication to the memory on the cartridge, which re-writes the binary value that is stored in the memory, representing the new estimated ink level.

43    Estimating the amount of ink that remains in the cartridge using a sensor can involve the main processor intermittently interrogating the sensor on the printer cartridge to determine if the ink level is above the threshold.

44    In some cases, once the ink levels reach a particular threshold level, the firmware in the main processor will recognise this and prevent the print job from occurring. The main processor will not allow the print job to occur until an ink cartridge is installed whose memory indicates an adequate level of ink. This means that, even if the used ink cartridge is refilled, if the data stored on the memory on the cartridge still indicates an inadequate ink level, the main processor will recognise the ink cartridge as empty and will prevent the print job from occurring.

45    In addition to storing binary code about the ink levels, the memory on the cartridge may also have fields that relate to other information that is communicated to and interpreted by the main processor in a similar manner to that describe above. This could include, for example, the type of ink, the colour of the ink, the brand of the cartridge, the date of manufacture etc. The memory could also contain a field relating to the serial number of the cartridge, which would enable the main processor to determine when a cartridge has been changed.

2.2    The original Epson products

46    The original Epson cartridges are cartridges sold by or with the permission of Seiko and have the Epson model numbers T0711, 125, 126, 18, ICLM50, 98, 73N, 133, 138, T200 and T200XL. They contain an integrated circuit chip mounted on or connected to a printed circuit board (or integrated circuit board). Different models of the original Epson cartridges have different types of memory chips located on the rear of the integrated circuit boards, and have differing compatibility with Epson printers.

47    Original Epson cartridges are designed so that a printer with which they are compatible is able to read and process the data in the memory chip. The memory chips store information regarding the printer cartridge, such as the volume of ink, the status of the cartridge, its date of manufacture and model number. Each category of information is stored in its own memory address on the memory chip.

48    The functions of the integrated circuit chip on the original Epson cartridges attracted significant attention during the course of argument. Dr Hawkins’ affidavit evidence addressed those functions. His opinions on the subject were largely uncontested, and are summarised in part below.

49    A memory chip is one of the integrated circuit chips in a computer or other piece of electronics whose function is to store information. It is comprised of an array of individual memory cells, each of which stores a single bit of information. It also has peripheral circuitry surrounding the memory cells. The original Epson cartridges use either EEPROM or FRAM memory chips. These represent non-volatile memory, which means that (unlike volatile memory chips) data stored in the memory is retained even if power is removed from the chip.

50    Information is stored on a memory chip as binary data in the form of a series of 1s and 0s. Each such 1 or 0 is known as a “bit”, which is the smallest unit used for data. Each bit is stored on a unique physical location in the memory chip, that is, each memory cell contains one bit. One particular electrical state at a physical location would be interpreted as a 1 and another electrical state at that location would be interpreted as a 0. With EEPROMs, the presence of electrons at a physical location on the memory chip would be at interpreted as a 0, while the absence of electrons at that location would be interpreted as a 1. For FRAMs, data is stored on a capacitor (that is, a stack that has a lower electrode, a layer of piezoelectric material such as PZT (lead zirconate titanate), and an upper electrode) that has a ferroelectric insulator between the capacitor plates. The ferroelectric insulator material contains many charged irons that are present in 1 of 2 locations within the ferroelectric material, and as a consequence, the capacitor is polarised in 1 of 2 directions to represent either a 1 or a 0. The polarisation is retained when no voltage is applied to the capacitor.

51    The memory chips in the Epson cartridges have two different modes, “normal” mode and “test” mode. When in operation in the printer, a printer cartridge’s chip will be in normal mode. The “test mode” is accessed when testing the chips after manufacture and when certain types of data in certain types of Calidad products is re-written.

52    The primary function of a memory chip in an ink cartridge is to keep track of ink in the ink cartridge. It is important for the printer to know how much ink remains in the ink cartridge. There is a high risk that some of the inkjet printhead channels will not be re-primed by the printer’s maintenance station if the ink is depleted to the point that the inkjet channels are emptied. While it is sometimes possible to recover such a de-primed printhead, the recovery takes a lot of ink and is not always successful. If the printhead cannot be recovered, then a new printer or printhead would need to be purchased or print defects would have to be accepted by the user.

53    Inkjet printers can print either a single drop size or a small number of different drop sizes. The printer’s electronics count the drops fired to determine how much ink has been used. The amount of ink used is monitored and then periodically entered by the printer into the memory chip on the ink cartridge. The printer prevents printing from taking place after the printer determines from the memory chip in the ink cartridge that the ink remaining on the cartridge has fallen below a threshold amount.

54    There are several additional benefits attached to tracking the quantity of ink used on the cartridge integrated circuit chip. One is to alert the printer user that the ink cartridge is getting close to empty. Another is to allow information regarding ink quantity to be transferred to any compatible printer in which the cartridge is used. Further, memory chips on ink cartridges often contain information about the original size of the ink supply (for example, whether it is a regular or high-capacity cartridge), the expiration date of the cartridge, and the length of time since the cartridge was first inserted into the printer, which may assist in determining whether the printer has been idle for a long period of time, which would affect reliability.

55    The description that follows is taken from the evidence of Mr Shen, who in his first affidavit includes some photographs of an Epson cartridge bearing the number “126” in order to illustrate the processes that Ninestar uses to refill those cartridges with ink. The figures below were included in Mr Shen’s evidence to illustrate the general physical features of an original Epson cartridge which was subsequently the subject of the Ninestar remanufacturing process which led to the Calidad product.

56    The cartridge is rectangular in shape and is roughly 6.5 cm long, 1.1 cm wide and 4.7 cm high. Figure 1 includes the words “Injection Port Site” and an arrow pointing to the left hand side of the cartridge. An injection port is added by Ninestar as part of its refilling the process, but is not present in the original Epson cartridge. In the original Epson cartridge there is an outlet hole towards the right of the base of the cartridge that protrudes about 7 millimetres, as the arrow notes.

57    Figure 2 depicts the base of the cartridge. One can see an air inlet hole on the left and an outlet hole on the right.

58    The lower portion of the right hand side of the cartridge is depicted in Figure 3. The photograph depicts the facing surface of an integrated circuit board. The circuit board contains two rows of contact points, with 4 in the upper row, and 5 contact points in the lower row. It may be seen that the integrated circuit board is a single unit that has been attached to the cartridge. Figure 4 shows the reverse side of the integrated circuit board after it has been removed from the cartridge. It may be seen that the integrated circuit chip (also known as “memory”) is located on the reverse side of the integrated circuit board.

59    Figure 5 (below) depicts an “interface pattern” located on the left hand side (looking at Figure 1) of the cartridge which takes the form of 3 vertical ribs that project downwards on the side and one horizontal rib joining the first 2 of the ribs. The image on the left hand side of Figure 5 shows the pattern as it originally appeared. The image on the right hand side of Figure 5 shows a cartridge after Ninestar has removed this pattern.

60    All of the memory chips in the original Epson cartridges are connected to the integrated circuit board via terminals. The terminals include; a 3.3 volt power supply, a ground, a reset terminal which enables the printer, via a communications agreement, to return the memory to its idle state, and clock and data terminals, which enable the transfer of the communication signals between the chip in the printer. Many of the cartridges (except the Epson T200 cartridges) also have a piezoelectric sensor which is used to detect the ink level in the cartridge. The piezoelectric sensor requires 2 terminals which are applied with 36 volts to drive the piezoelectric sensor. The arrangement of terminals differs slightly from one Epson cartridge to another, but not in a manner which is material to the issues in dispute. The terminal arrangement for the Epson 133 and 138 cartridges is set out in the evidence of Mr Li as follows:

61    In the case of the Epson T200 cartridge, in lieu of the piezoelectric sensor is a prism at the base of the cartridge to determine when the ink level has dropped below a particular level. Instead of having 2 terminals relating to an ink sensor, the terminals relate to a common resistor with 62K resistance. However, the arrangement of terminals is not materially different to that of the other Epson cartridges.

2.3    The Calidad products

2.3.1    Introduction

62    The process of modifying or converting the original Epson cartridges to Calidad products is described in the evidence of Mr Shen and Mr Li, who were cross-examined in some detail, the result of which was to provide a good picture of the work that was performed. The evidence of Dr Hawkins addresses a number of steps that would be necessary in order for the Calidad products to be produced, focusing in particular on the memory chips. His evidence particularly addressed first, the research and development work that would be necessary to convert a used original Epson cartridge into a Calidad product. This involves research and development and resetting or rewriting of the software contained in the memory chips. Secondly, the work that would be necessary in order to reset or rewrite the information contained in the memory chips in order to permit converted original Epson products to operate as Calidad products. Although Dr Hawkins evidence was stated primarily at a theoretical level, Mr Li and Mr Shen, insofar as either had knowledge of the matters to which Dr Hawkins evidence related, frankly and openly confirmed most of the matters that he had surmised. Ultimately, the parties were not in substantial dispute about the steps taken by Ninestar to modify the Epson products. In section 6 below I return to make findings as to the particular steps taken in respect of each original Epson cartridge converted into a Calidad product.

2.3.2    Ninestar and Apex

63    Ninestar, and AMC are related companies that are collectively owned by Apex Technology Co Ltd (Apex). Apex recently acquired Lexmark, a global technology company that provides imaging and output solutions.

64    Ninestar is one of the world’s largest manufacturers of generic printer consumables, which include toner and inkjet printer cartridges. It has been operating for over 13 years and supplies products to over 100 countries. Ninestar’s products include Epson printer cartridges that were originally sold by Seiko and which, once used and discarded by the customer, are collected by third parties and restored by Ninestar to working condition. Cartridges of this type are the subject of the present proceeding. In addition to restoring Seiko printer cartridges, Ninestar also restores cartridges originally made by other original equipment manufacturers including HP, Canon, Brother, Lexmark and Dell.

65    AMC is a manufacturer and supplier of equipment that rewrites the data stored on OEM integrated circuit chips that are attached to the integrated circuit boards used in OEM printer cartridges at the time of first sale. The equipment made and supplied by AMC can be used to rewrite the data stored on different OEM chips, including those that are attached to integrated circuit boards on Canon, HP, Samsung and Epson printer cartridges. AMC also manufactures and supplies replacement integrated circuit chips that are compatible with OEM printer cartridges and may be substituted for the OEM chips on integrated circuit boards. These Mr Li refers to as “compatible chips”. Mr Li gives evidence that AMC is one of the largest manufacturers of compatible chips for printer cartridges in the world. He gives evidence that he supervises a team of 5 people who are responsible for all software engineering research and development work related to both compatible chips and the equipment used to rewrite the data stored on OEM chips.

66    There is no dispute that the original Epson cartridges have been sold by or with the permission of Seiko.

67    Ninestar, for a fee, obtains used original Epson cartridges from third party suppliers who have collected them. It is unclear on the evidence whether the third party suppliers acquired the cartridges from the original consumers of the original Epson cartridges or otherwise obtained them, perhaps by collecting them from recycling bins. I find that it is likely that a portion of the original Epson cartridges have been sold or given by the consumers to the third party suppliers, and that otherwise they have come into the hands of the third party suppliers by a variety of means, including, perhaps, collection from recycling facilities.

68    At the time when Ninestar acquires the original Epson cartridges, the information on the memory chip records that the cartridge is “used” such that when it is inserted into a compatible printer, the printer will not work because the information conveyed from the memory chip to the printer is that the cartridge is empty.

69    In order to be refilled and made available for sale the steps set out in detail in the table at [73] below must be taken. Broadly, for a re-filled cartridge to work, it is necessary to reconfigure or rewrite the information on the memory chip so that when it is inserted into a compatible printer, the information in the memory chip does not indicate that the cartridge is used or empty.

2.4    Summary of steps taken to modify the original Epson cartridges

70    The Calidad parties admit that CDP has imported, offered to sell, sold and/or otherwise disposed of and kept for the purpose of doing those acts in Australia the following, which are the Calidad products:

(a)    Calidad 250 cartridges;

(b)    Calidad 253 cartridges;

(c)    Calidad 258 cartridges;

(d)    Calidad 260H cartridges (high capacity volume); and

(e)    Calidad 260S cartridges (standard capacity volume).

71    Calidad products are supplied to Calidad by Ninestar and promoted in Australia as remanufactured Epson cartridges. They are compatible with certain Epson printers which are sold by or with the permission of Seiko (compatible Epson printers). The parties accept that in broad terms there are 4 steps undertaken by or with the approval of Ninestar in the modification of the original Calidad cartridges to make each Calidad product. However, Calidad contends that two significant aspects of the modification work performed are irrelevant. The first is the preparatory research and development work conducted by or on behalf of Ninestar in order to determine how to make modifications to the contents of the memory chip. The second is the change brought about to the contents of the memory chip. The 4 broad steps are:

(a)    the preparation of the cartridge for ink refill (preparation);

(b)    the refilling of the printer cartridge with ink not supplied by Seiko or with Seiko’s approval (refilling processes);

(c)    the replacement, reprogramming or resetting of the memory chip (memory replacement/reprogramming);

(d)    the research and development work in order to make alterations to the contents of the memory chip when it is in either “normal mode” or “test mode”.

72    Some of the steps have changed over time for each model of the Calidad cartridges and there are some variations on the steps which are applied to different models. As a consequence, there are 9 different categories of Calidad product reflecting, in broad terms, the types of work performed leading to a Calidad product.

73    Set out below is a table listing each Calidad product, the original Epson product from which it was derived and, in summary form, the steps taken by Ninestar to produce the Calidad product. To a substantial degree there was no dispute between the parties as to the modifications conducted. The only exception lies in Category 2 step (3), which is addressed further in section 6.4 below. A difference of principle emerges between the parties as to the relevance of the research and development processes. That is not a dispute as to fact, but rather as to the legal significance of those processes.

74    It is necessary later to consider in more detail the modifications made by Ninestar, but before doing so I turn to the legal context in which the modifications must be considered.

3.    PATENT INFRINGEMENT – THE RELEVANT PRINCIPLES AND APPROACH

3.1    Introduction

75    Seiko places the original Epson cartridges on the open market and sells them. They are then used for the purpose for which they are intended, and subsequently, Ninestar acquires some of the used cartridges. Ninestar then refills them with ink. In the process of doing so, it makes modifications whereupon they become what has been defined in the proceedings as the Calidad products. CDP then acquires those products, imports them into Australia, offers them for sale and sells them.

76    There is no dispute that the original Epson cartridges and the Calidad products possess all of the essential features of claim 1 of the 643 patent. However, Calidad contends that it does not infringe the patents because it is the beneficiary of a licence. It submits that after selling the original Epson products, Seiko’s rights to control the subsequent use of them is, in effect, exhausted. This, it submits, arises from National Phonograph. It submits that a presumption arises from a sale of products that embody an invention that, upon sale, full rights of ownership are vested in the purchaser unless conditions are imposed by the patentee. It submits that in the present case no such conditions have been established and, as a result, Calidad does not infringe. In the alternative, Calidad submits that a patentee’s exclusive rights under s 13(1) of the Act do not include the right to prevent the owner repairing or refurbishing a patented product, or have subsequent dealings in that repaired or refurbished product (including importation). The relevant question under s 13 is whether a “new” product, as distinct from a repaired or refurbished product, has been manufactured and dealt with by the alleged infringer. This will be determined as a matter of fact and degree. In the present case, Calidad submits that the work done by Ninestar does not amount to a “making”. In this respect, Calidad relies on United Kingdom authority, and in particular Schütz (UK) Ltd v Werit (UK) Ltd [2013] UKSC 16; (2013) 100 IPR 583 (Schütz) and United Wire Ltd v Screen Repair Services (Scotland) Ltd [2001] RPC 24 (House of Lords) (United Wire).

77    Seiko’s response is that Calidad infringes the patent because it has, without authorisation, imported and sold the Calidad products, not because it has “made” them. The only defence potentially available to Calidad lies in an implied licence. Two implied licences are potentially available to a defendant who has acquired a patented product. The first arises from a sale sub modo - a sale made in the absence of restrictive conditions within National Phonograph. It submits that such licences extend only to the “use” or “re-sale” of a product and do not extend to remanufacturing of the type done by Ninestar or the subsequent importing of that product by Calidad. Seiko further contends that even if there were potentially an implied licence that extends to a making or importation, the original Epson cartridges contain certain “in-built restrictive conditions” in the form of limitations on the manner in which the memory chip operates which act to limit the licence, and these limitations put Ninestar and Calidad on notice that the work done by Ninestar is not authorised. Seiko also contends that any implied licence does not travel with the goods once they have been discarded by the user and collected by persons who subsequently sold them to Ninestar. In this context, Ninestar is not a purchaser, and cannot benefit from any licence.

78    The second implied licence, Seiko submits, is an implied licence to repair a patented article. That licence does not emerge from Australian authority, but may be understood by reference to the United Kingdom decision in Solar Thomson Engineering Co Ltd v Barton [1977] RPC 537 (Solar Thomson). In Australia, the Schütz and United Wire decisions have no application. Further, Seiko submits that to the extent that the Court decides to follow the line of authority represented by United Wire and Schütz then, in any event, the modifications made to the original Epson products to result in the Calidad products do not involve a “repair”.

79    Before turning to consider factual matters, it is necessary first to consider the principles relevant to the consequence of a patentee selling a product that embodies a claimed invention that is subsequently modified and then imported into Australia and sold.

3.2    A patentee’s rights to control the use of patented goods after sale

80    The exclusive rights of the patentee pursuant to s 13(1) of the Act are, during the term of the patent, to exploit the invention and to authorise another person to do so. The definition of “exploit” in the Dictionary (Schedule 1 to the Act) is:

exploit, in relation to an invention, includes:

(a)    where the invention is a product – make, hire, sell or otherwise dispose of the product, offer to make, sell, hire or otherwise dispose of it, use or import it, or keep it for the purpose of doing any of those things; or

(b)    where the invention is a method or process – use the method or process or do any act mentioned in paragraph (a) in respect of a product resulting from such use.

81    The exclusive right to “use” a product the subject of a patent means that upon sale or resale of a product the patentee will have continuing control of the use. The purchaser of a chattel such as a garden hose or a fishing rod which is the subject of a patent can never consider himself or herself to be entitled to all of the usual incidents of ownership, because the patentee could impose limitations on his or her “use”. However, there is a level of impracticality about this arrangement. It creates a tension between the notion that the purchaser of a chattel is its owner who can deal with it as he or she wishes, and the notion that a patentee can continue to have control over any use to which the chattel may be put after sale. Historically, 2 competing theories were developed to address this tension, both of which operated to confer on the purchaser of the fishing rod untrammelled rights to use it as its owner.

82    One arises from the notion that upon a first authorised sale of a chattel by or with the approval of a patentee, the patentee exhausts all patent rights in the subject matter of its invention. This is often called the “exhaustion of rights” doctrine. This is the approach that has been adopted for many years in the United States and which was recently confirmed by the United States Supreme Court in Lexmark. The other is that there exists an implied licence to use a chattel in any manner in which the purchaser desires upon purchase, unless the patentee imposes conditions at the time of sale. Both of these theories were considered in the influential decision of National Phonograph. The majority in the High Court (National Phonograph Co of Australia Ltd v Menck (1908) 7 CLR 481) favoured an approach that broadly adopted an exhaustion of rights approach. On appeal to the Privy Council ((1911) 12 CLR 15), this was overturned and an implied licence approach was adopted, in the manner set out below.

83    In National Phonograph the National Phonograph Company (NPC) sold Edison records that were made “in accordance with, and under the protection of, the letters patent” (at 17). That is, the records embodied the invention claimed. NPC’s business model was to sell its records to intermediaries (“jobbers”) under contracts that enabled them in turn to sell them to dealers. The dealers were themselves subject to contracts with NPC which imposed restrictions upon how they could sell the records. It was a term of the dealer contract that if the dealer was removed from the list of authorised dealers, then he could no longer sell, deal in or use, either directly or indirectly, the records. The defendant had his name removed from the list of dealers and yet declared that he was free to dispose of the goods as he wished. He was found not to have acted in breach of any contract, but was nevertheless bound not to make use of the patented products in any way inconsistent with restrictions imposed by the patentee of which he was given notice.

84    The relevant part of the dispute concerned whether the rights conferred upon a patentee pursuant to s 62 of the Patents Act 1903 (Cth), which gave the patentee full power, sole privilege and authority to “make, use, exercise, and vend the invention”, entitled the patentee to constrain Mr Menck (the question of importation did not arise).

85    In the High Court, the majority found that a patentee is not entitled to impose conditions upon the sale or use of patented articles separately from any contract that it might have with a subsequent purchaser. This conclusion arose from a survey of the law of England and the United States as it stood at the time of the decision (that is, 1908), and the conclusion was that neither demonstrated a clear line of authority (7 CLR at 518 per Griffiths CJ). The majority favoured the view that it is an elementary principle of the law of personal property that the owner of chattels has an absolute right to use and dispose of them as he or she thinks fit, and that no restrictions can be imposed upon this right, except by positive law or by contract. The right asserted by NPC depended on whether the words “use” and “vend” in s 62 of the Patents Act 1903 (Cth) imposed such a restriction. Chief Justice Griffiths, as one of the judges in the majority, found that neither term could be construed, on the authorities, in such a way as to warrant a deviation from the elementary principle and broadly found that a patentee exhausts all patent rights upon the first sale of the goods.

86    NPC appealed to the Privy Council ((1911) 12 CLR 15), which overturned the decision of the High Court, Lord Shaw delivering the judgment. The Privy Council found that it was not in question that the elementary principle that an owner may use and dispose of ordinary goods as he or she thinks fit remains correct. It recognised that there exists a tension (referred to as a “difficulty” at page 22) between that principle, and the right of property granted under the Patents Act by way of monopoly “to make, use, exercise, and vend the invention … in such manner as to him seems meet”. The tension was reflected in the two separate and conflicting strands of reasoning between the majority in the High Court (Griffiths CJ, Barton, and O’Connor JJ) and the dissentient judges (who, on this point were Isaacs J and possibly Higgins J). The former found that upon sale of a chattel the subject of a patent, the product “passed out of the limit of the monopoly”. The latter preferred an analysis whereby any conditions imposed by the sale by the patentee simply ran with the goods.

87    The Privy Council rejected both approaches. Lord Shaw said (at 23 to 24, emphasis added):

There is no doubt that, if the doctrine contended for by the appellants and affirmed by the dissentient Judges in the Court below were to be given effect to, namely, that the conditions imposed by the patentee run with the goods, a radical change in the law of personal property would have been made. But if that latter view be an extreme view, and if the restriction upon alienation, use or otherwise of the chattel purchased, be a restriction arising from the fact that the person who has become owner has done so with the knowledge brought home to him of the limitation of his rights of alienation or otherwise, then there seems to be no radical change whatever.… These limitations are merely the respect paid and the effect given to those conditions of transfer of the patented article which the law, laid down by Statute, gave the original patentee power to impose.

88    It is relevant to one of the present arguments that in this passage the Privy Council emphatically rejected that conditions imposed by a patentee “run with the goods”. The Court adopted the language of licence. It said (at 24):

It may be added that where a patented article has been acquired by sale, much, if not all, may be implied as to the consent of the licensee to an undisturbed and unrestricted use thereof. In short, such a sale negatives in the ordinary case the imposition of conditions and the bringing home to the knowledge of the owner of the patented goods that restrictions are laid upon him.

89    It was in this way that the Privy Council was able to reconcile the tension that it had earlier identified. It did so by adjusting “the incidence of ownership of ordinary goods with the incidence of ownership of patented goods in such a manner as to avoid any collision of principle” (at 22). In its conclusion (at 28) it said (emphasis added):

In their Lordships’ opinion, it is thus demonstrated by a clear course of authority, first, that it is open to the licensee [of the patentee], by virtue of his statutory monopoly, to make a sale sub modo, or accompanied by restrictive conditions which would not apply in the case of ordinary chattels; secondly, that the imposition of these conditions in the case of a sale is not presumed, but, on the contrary, a sale having occurred, the presumption is that the full right of ownership was meant to be vested in the purchaser; while thirdly, the owner’s rights in a patented chattel will be limited if there is brought home to him the knowledge of conditions imposed, by the patentee or those representing the patentee, upon him at the time of sale. It will be observed that these propositions do not support the principles relied upon in their absolute sense by any of the Judges of the Court below. On the one hand, the patented goods are not, simply because of their nature as chattels, sold free from restriction. Whether that restriction affects the purchaser is in most cases assumed in the negative from the fact of sale, but depends upon whether it entered the conditions upon which the owner acquired the goods. On the other hand, restrictive conditions do not, in the extreme sense put, run with the goods, because the goods are patented.

90    It is to be noted that in this and other passages, the focus is upon the owner’s rights in a “patented chattel”. That is, the physical embodiment of the invention, as defined by the claims.

91    Contrary to the view of Griffiths CJ in the High Court, their Lordships did not consider that the law so stated was at all novel. They considered that it arose from a clear lineage that could at least be traced to statements of Lord Hatherley LC in Betts v Willmott (1871) LR 6 Ch App 239 (Betts v Willmott). In that case it was held that if the owner of a patent manufactures and sells the patented article in France, the sale carries with it a licence to use the article in England. Lord Hatherley LC said (at 245):

…unless it can be shewn, not that there is some clear injunction to his agents, but that there is some clear communication to the party to whom the article is sold, I apprehend that, inasmuch as he had the right of vending the goods in France or Belgium or England, or in any other quarter of the globe, he transfers with the goods necessarily the license to use them wherever the purchaser pleases. When a man has purchased an article he expects to have control of it, and there must be some clear and explicit agreement to the contrary to justify the vendor in saying that he has not given the purchaser his license to sell the article, or to use it whereever he pleases against himself.

92    It is particularly relevant to the present case that this passage identifies that the implied licence arising upon the sale includes a licence upon the part of a person who has acquired a patented product abroad to import it into the country. Here, Ninestar acquired the original Epson cartridges abroad, modified them and then CDP acquired those products from Ninestar and imported them into Australia for subsequent sale.

93    As to the nature of the right of the patentee to limit subsequent use, the Privy Council endorsed the following reasoning of Farwell J in British Mutoscope and Biograph Co Ltd v Homer (1901) 1 Ch 671 (British Mutoscope) at 673 – 674 (emphasis added):

… it has recently been held in Incandescent Gas Light Co. Ltd. v Brogden that a purchaser who buys with knowledge of the conditions under which his vendor is authorised to use the patented invention is bound by such conditions, and that such conditions are not contractual, but are incident to and a limitation of the grant of the license to use, so that if the conditions are broken there is no grant at all.

94    This passage provides an indication that whilst the patentee may impose restrictions upon a purchaser or subsequent title holder of the patented product, the licence does not arise strictly from contract but rather as an incident of the nature of the patent rights.

95    The Privy Council in National Phonograph moved on to approve the observations of Cozens-Hardy LJ in McGruther v Pitcher (1904) 2 Ch 306 at 312 who remarked that in an action by a patentee claiming an injunction to restrain an infringement of patent it is open to the defendant to plead a licence by the plaintiff. His Lordship said:

That license may be express, or it may be implied from the sale by the patentee of the patented article, but, if the defendant pleads a license, then it is competent for the plaintiff to reply, “The license which I granted is a limited licence, and you, the person who has now got the patented article, were aware it was only a limited licence, and you cannot therefore defend yourself against my claim for an infringement of my patent, because you are going outside the license which to your knowledge I gave with reference to this article”.

96    It is apparent from this passage, as well as from the passage quoted from page 28 of National Phonograph in [89] above that the onus lies upon the patentee to establish that upon sale, conditions limiting the use that may be made of the patented product are notified to the owner, and that in the case of any subsequent owner, they too have notice prior to coming into ownership.

97    On the facts of National Phonograph the defendant was not free to deal with the goods as he wished, because the NPC had given him notice of restrictions that it wished to impose upon any dealings with goods. Despite having acquired the patented records, that notice was sufficient to constrain the defendant. The conclusion of the Privy Council was as follows (at 28-29, emphasis added):

Applying these principles to the present case, the result is this: the respondent, Mr Menck, has been acquitted of every charge of violation of contract which was laid against him by the appellants. He has also succeeded in showing that the claim made by the appellants as patentees was in its nature extreme and unsound in law. But he made this mistake: he assumed that, being guiltless of violation of contract, he was as free as an ordinary member of the public who had acquired possession of articles embodying the appellants’ patent. His misfortune, however, consists in this, that by the very fact that he entered into contractual relations with the appellants, he has become seized with the knowledge of the conditions on which they dispose of their goods, and he is not free to propone the plea that such conditions have not been brought home to him. When he therefore announced his intention to deal in these articles as ordinary articles of commerce, he must be held to have pursued a mistaken course, the course of treating himself as an unrestricted instead of a restricted trader. In this particular case the result may involve some hardship to him, but their Lordships cannot see their way to a departure from the principle that a restriction rests upon a purchaser of goods which are covered by a grant of patent, and which have come into the possession of a purchaser in the full knowledge of the restrictions imposed by the patentee upon their disposal.

98    Accordingly, there is a presumption that a sale sub modo (sale without restriction) confers an unrestricted right of use upon the purchaser or subsequent owner of the patented product, which covers at least the rights to import, use and dispose of the product. To avoid this consequence, a patentee must make a sale subject to limitations which are effective to restrict the importation, use or disposal of the product by the purchaser or successor in title to the purchaser. It must show that notice of any restriction was brought to the knowledge of the purchaser at the time of sale.

99    In T. A. Blanco White’s Patents for Inventions, 4th ed (1974) at 3-219 (Blanco White) the learned authors say (emphasis added):

A sale of a patented article made by a patentee gives to the purchaser, in the absence of notice to the contrary, license under the patent to exercise in relation to that article all the normal rights of an owner, including the right to resell. It follows that in order to establish infringement the patentee must prove that alleged infringing articles were not made by him, either here or abroad [citing Betts v Willmott at 242, 244]. The position is the same where the article has been made in the United Kingdom under license from the patentee, or (it would seem) where in any other way it comes upon the United Kingdom market with his consent;… The implied license extends to repair of the article bought - to the “prolonging of its life” - but not to renewal under pretence of repair; for repair is one of the normal activities of an owner, but sale of one patented article gives the purchaser no license to make himself others.

100    In Interstate Parcel Express Co Pty Ltd v Time-Life International (Netherlands) BV (1977) 138 CLR 534 (Interstate Parcel) the High Court referred to National Phonograph in obiter dicta. In that case, a bookseller imported books from an American wholesaler and sold them in Australia. In response to an allegation of copyright infringement, the bookseller submitted that it took the books free of any restrictions, relying on National Phonograph. Justice Gibbs quoted (at 540) the first sentence of the passage in Blanco White set out above, and also the following passage from 10-104, as summarising the decided case law in relation to implied licences upon sale of patented products:

In the absence of any express term to the contrary, when a patented article is sold by or with the consent of the patentee (or the proprietor of a registered design) the purchaser will take it together with a full license to deal with it as if it were not patented. Further, any person into whose hands it may later come is entitled to assume that such a full license has been given with it; it makes no difference that he may later discover that this was not so, if he was ignorant of it at the time of purchase.

101    After setting out aspects of the reasoning of the Privy Council in National Phonograph, Gibbs J expressed doubt that the conditions that may be imposed by a patentee conformed with normal principles of licences, but nevertheless, noting that the long line of distinguished judges had concluded that it was so, accepted that it was indeed the consent or licence of the patentee to use the article that might be implied from the sale (at 542). Justice Stephen considered that such a licence arises from the “quite special case” of the sale by a patentee of patented goods, which turned upon the unique ability which the law confers upon patentees of imposing restrictions upon what use may, after sale, be made of those goods (at 549). His Honour said (at 549 - 550);

But, to ensure that consequence despite the existence, albeit in the instance unexercised, of this power on the patentee’s part, the law treats the sale without express restriction as involving the grant of a license from the patentee authorising such future use of the goods as the owner for the time being sees fit. The law does this because, without such a license, any use or dealing with the goods would constitute an infringement of the patentee’s monopoly in respect of the use, exercise and vending of the patent. A sale of goods manufactured under patent is thus a transaction of a unique kind because of the special nature of the monopoly accorded to a patentee; the license, whether absolute or qualified, which arises upon such a sale is attributable to the existence and character of that monopoly.

102    It follows from the principles set out in National Phonograph and as subsequently interpreted that, if a restriction of the right to use was not imposed as a condition of the purchaser’s acquisition of the goods, but at another time or by some other means is sought to be imposed by the patentee, it is not effective.

103    This was the view taken by Jacob J (sitting in the United Kingdom Patents Court) in Roussel-Uclaf v Hockley International Ltd [1996] RPC 441 (Roussel). In that case, the patentee, Roussel, made a pharmaceutical product in France and supplied it to a Chinese joint venture company. The defendant, Hockley International, an English company, acquired the product directly or indirectly from the Chinese company. The European law of patent exhaustion did not apply, because there was no relevant sale inside the European economic zone. Hockley International relied upon the National Phonograph principles in asserting that it took the product free of any restrictions. Justice Jacob found that it did, and stated the applicable principle as follows (at 443):

It is the law that where the patentee supplies his product and at the time of the supply informs the person supplied (normally via the contract) that there are limitations as to what may be done with the product supplied then, provided those terms are brought home first to the person originally supplied and, second, to subsequent dealers in the product, no license to carry out or do any act outside the terms of the license runs with the goods. If no limited license is imposed on them at the time of the first supply no amount of notice thereafter either to the original supplyee (if that is the appropriate word) or persons who derive title from him and turn the general license into a limited licence.

104    That reasoning was adopted by Arnold J in HTC Corporation v Nokia Corporation [2013] EWHC 3247; [2014] RPC 19 at [156].

105    Justice Jacob confirmed that Roussel, as the patentee, must establish the relevant restriction (at 443), and accordingly, bore the onus. In relation to the supply by Roussel to the Chinese company, Roussel relied upon a practice of labelling the goods with restrictions as to re-export. However, the Court held that it had not established that the particular products in question bore those labels, or that a restriction was imposed or brought home by any other means (at 443 - 444). Accordingly, no conditions were imposed on that subsequent supply.

106    In relation to supply to Hockley International, Roussel relied upon a letter sent directly by Roussel to Hockley asserting that the products were subject to conditions as to sale and use. Justice Jacob considered that this was not a restriction imposed as a condition of Hockley’s acquisition of the products; the time for imposing such condition having passed with the making of the contract of supply (at 444):

Next I turn to the position of the defendants as subsequent supplyees. Had it been brought home to them that there are such conditions? In a sense it is virtually impossible to do this once it has not been established that it was not brought home to the original purchasers. What is relied upon is indeed the letter which I have just quoted. That says in the previous sentence: “it is not correct that our client’s technical grade material is freely available to the world market without conditions as to its use”. That is a pretty opaque way, if it be intended, to say that there is a positive imposition of a limited permission to use. … The question is not whether the market is grey or any other colour or shade, but whether or not conditions of a limited licence were imposed when the product was first supplied by Roussel. This letter of 24 December 1993 does not convey to the defendants that the product was originally supplied subject to a limited licence.

107    The essence of the licence is that it arises automatically upon the unconditional sale of the patented product. The patentee must act to negative the licence if limitations are to apply.

108    National Phonograph was recently considered and applied by Dowsett J in Austshade Pty Ltd v Boss Shade Pty Ltd [2016] FCA 287; (2016) 118 IPR 93 (Austshade). In that case, at [80], his Honour drew the following conclusions in relation to the application of National Phonograph:

80. The decision in National Phonograph establishes three primary propositions:

•    First, the principle applicable to ordinary goods bought and sold is that the owner may use and dispose of those goods as he or she thinks fit. Notwithstanding any agreement which such owner may have made with the person from whom he or she bought the goods (by which he or she is bound contractually), he or she is not bound, in his or her capacity as owner, by any restrictions concerning the use or sale of the goods. Hence it is, “out of the question” to suggest that any such restrictive conditions run with the goods. The rights of an intermediate owner in that “capacity” are to be distinguished from the contractual arrangements which may bind him or her in dealing with the person from whom the goods were acquired. The point is that the intermediate owner may give to his or her purchaser, good title to the goods, notwithstanding any contractual restriction imposed upon the intermediate owner at the time of his or her acquisition of the goods.

•    Second, the general doctrine of absolute freedom in the disposal of chattels of an ordinary kind is, in the case of patented chattels, subject to the restriction that the person purchasing them, having knowledge of the conditions attached by the patentee, which knowledge is clearly brought home to him at the time of sale, is bound by that knowledge and accepts the situation of ownership subject to the limitations. In other words, a patentee may bind the power of a purchaser to deal with the goods, but such limitation will only bind a subsequent purchaser from that purchaser if the former has, at the time or purchase, actual knowledge of the limitations.

•    Third, in the case of such a resale, it is not presumed that conditions imposed on the vendor by the patentee will bind the subsequent purchaser. Rather, the sale will be presumed to have vested the full right of ownership in the subsequent purchaser unless the conditions imposed by the patentee have been brought home to him or her at the time of sale. Hence it is not sufficient that the ultimate purchaser may know of the patent and its application to the goods acquired. There must be actual knowledge of the limitations imposed by the patentee at the time of the first sale.

109    The following propositions are relevant to the consideration of the present dispute, and arise from the cases above.

110    First, it is important to be clear about references to “a patented product” or a “patented chattel” or similar language. In each instance this is a reference to a product that embodies the invention as claimed in the patent in suit. It was in this manner that the cases referred above used these and similar terms.

111    Secondly, any restriction or limitation notified by the patentee does not run with the patent and goods in rem, but rather applies in personam as an obligation on the owner of the goods from time to time, contingent upon the provision to the owner of notice prior to or at the time of taking title to the patented product; National Phonograph at 23 – 24 and 28 – 29; Betts v Willmott at 245; Austshade at [80].

112    Thirdly, the onus lies on the patentee to establish that the first and subsequent dispositions were subject to restrictive conditions; National Phonograph at 28; Interstate Parcel at 540; Rousel at [443]. Although Seiko contends that an evidentiary onus lies on the Calidad parties to establish their entitlement to re-sell repaired or modified original Epson cartridges, there is no dispute between the parties that Seiko bears the legal onus in respect of the imposition of any restrictive conditions.

113    Fourthly, the rights conferred by an acquisition of patented goods without notice of any restriction from the patentee are in the nature of a somewhat unusual implied licence that arises from the intersection of patent rights and the “elementary principle” of the law of personal property that the owner of chattels has an absolute right to use and dispose of them as he or she thinks fit; National Phonograph at 22, 28; British Mutoscope at 673 – 674; Interstate Parcel at 542 (Gibbs J), and 549 (Stephen J). Seiko submits that an aspect of the implied licence is that it arises in order to give “business efficacy” to a transaction involving a patented article in line with the purchaser’s commercial expectations, citing BP Refinery (Westernport) Pty Ltd v Hastings Shire Council (1977) 180 CLR 266 at 283 (BP Refinery), and the reference by Gibbs J in Interstate Parcel to “business efficacy” (at 542). In this context, it submits that whilst upon a sale of a patented product the implied licence would arise, on the facts of the present case an uncertain number of the original Epson cartridges were discarded by the first purchaser (in recycling bins and the like). Upon that occurring, it is no longer necessary to give business efficacy to the initial or any subsequent sale. To infer the grant of an implied licence to an entity which obtains patented goods after they have been discarded would, it submits, be contrary to the purpose of granting an implied licence, which is giving effect to a purchaser’s commercial expectations so as to give business efficacy to the sale.

114    However, Seiko’s submission tends to load the use of the words “business efficacy” by Gibbs J in Interstate Parcel with freight that his Honour did not intend. His Honour had already noted (at 541) the unique nature of the implied licence that arises from the intersection of patent rights and the “elementary principle”. There is no suggestion in the authorities, including Interstate Parcel, that the implied licence arises only in circumstances where business efficacy was required. Indeed, Seiko’s submission tends to isolate but 1 of the 5 conditions that the Privy Council found in BP Refinery (at 283) must be satisfied in order for a term of a contract to be implied.

115    Furthermore, a first owner of patented goods obtained without restriction may do with them as he or she thinks fit; National Phonograph at 24. That includes selling the goods, giving them away or throwing them away. As Blanco White noted at 10-104, any person in whose hands the patented product subsequently falls is entitled to assume that such a full licence has been given with it. Unless the patentee makes it clear to all concerned that it is not giving the usual licence, then no restriction is applicable.

116    It is irrelevant whether a person acquires title from the first purchaser by a subsequent sale or by other means such as a gift. In this context, once title has passed from the patentee to the first purchaser without a restrictive condition of sale, the patentee is unlikely to have any ability to impose such a condition upon subsequent title holders; see Interstate Parcel at 549 (Stephen J); Austshade at [80] (third bullet point); Rousel at 444.

117    Fifthly, the authorities demonstrate that there is no difficulty implying the licence where the patented goods are first sold outside of the patent area (that is, broadly, outside Australia); Betts v Willmott at 245; National Phonograph at 24; Rousel. This has direct application in the present case, where Ninestar purchased the original Epson cartridges from a third party overseas, who in turn appears to have obtained those cartridges from users overseas. CDP in turn acquired the Calidad products from Ninestar and imported those products into Australia.

3.3    The issues arising concerning patent infringement

118    Calidad contends that it is the beneficiary of a licence of the type recognised in National Phonograph because Seiko sold the original Epson products without restriction (I refer to this below as the implied licence). Seiko raises several objections to this contention, which are conveniently addressed in the following order.

119    First, Seiko contends that implied licences do not travel with patented goods that have been discarded by their owners. It submits that to infer the grant of an implied licence to an entity which obtains patented goods after they have been discarded would be contrary to the purpose of granting an implied licence, which is to give effect to a purchaser’s commercial expectations so as to give business efficacy to the sale. I have addressed and rejected that submission in [113] – [116] above.

120    Secondly, Seiko contends that each of the original Epson cartridges was sold subject to restrictive conditions in the form of conditions that are built into each original Epson cartridge’s memory chip. The condition that it submits was imposed is that the original Epson cartridge is not to be used in a relevant Epson branded printer where at least one of the following circumstances applied: (a) the ink in the original Epson cartridge has dropped below a threshold level; (b) the cartridge has already been “used” and is “empty”; and (c) where applicable, the original Epson cartridge is not compatible with the Epsom branded printer. Seiko submits that these restrictive conditions are programmed into each memory chip to give effect to the restrictive conditions and physically embody them. A reasonable person would, Seiko submits, appreciate that the memory, either in normal mode or test mode, operated as an inbuilt restriction (inbuilt restriction argument). This is particularly obvious with respect to the reverse engineering required to operate the category 2 and 3 original Epson cartridges, which must be operated in test mode if they have been refilled with ink. The extensive research and development required to enable the memory to operate in that way demonstrates the limitation to any user endeavouring to reprogram the chip. However, as the presence of a restriction does not depend on the technical difficulty of the reverse engineering process, a reasonable person would also appreciate that the memory in normal mode operates as an inbuilt condition.

121    Seiko submits that Ninestar is on notice of the inbuilt technical restrictions and that various inferences should be made to the effect that Calidad is similarly on notice of those restrictions.

122    It is necessary to consider the inbuilt restrictions argument as it applies to Ninestar and Calidad. These arguments are considered in section 3.4 below.

123    Thirdly, Seiko contends that the implied licence cannot apply to the importation of goods. This argument must be rejected. The authorities set out above, commencing with Betts v Willmott, make plain that an implied licence may arise in respect of products imported from abroad.

124    Fourthly, Seiko submits that even if the Court finds that the implied licence applies to the original Epson cartridges in the hands of Ninestar, Calidad (or, more specifically, CDP), cannot benefit from it. This is because the modifications were made by Ninestar prior to the importation of the Calidad products. The modifications made cannot constitute conduct falling within any implied licence made. The act of importation into Australia of a product so altered from the original Epson cartridge is not licensed.

125    Fifthly, and related to the fourth argument, Seiko contends that even if the implied licence applied to Calidad products upon their importation and sale, the modifications made extend beyond any permissible right to repair conferred on the owner of a patented product, and so infringe. In this regard, Seiko relies particularly on the analysis of Solar Thomson at 554. Calidad responds by submitting that it is no infringement of the exclusive rights of the patentee to make modifications to a patented product that fall short of a “making”, and that when Ninestar modified the original Epson cartridges no fresh “making” occurred. In this regard, Calidad relies on the later United Kingdom decisions of United Wire and Schütz.

126    The fourth and fifth arguments broadly give rise to question of whether the implied licence can exist on the peculiar facts of this case, which involve modifications to the patented product made abroad and their subsequent importation into Australia. They are considered in section 3.5 below.

3.4    The inbuilt restrictions argument

Ninestar

127    I am not satisfied that the inbuilt restrictions to which Seiko draws attention can properly be characterised as conditions which restrict the manner in which Ninestar may use the original Epson cartridges.

128    First, Seiko submits that the “real question” is whether the condition or restriction is brought to the notice of the alleged infringer. In the present case, that notice, it submits, was conveyed by the inbuilt technical limitations imposed by the memory chip. I have above summarised the principles relevant to the provision of notice. The learned authors of Blanco White (after summarising the principles arising from National Phonograph in the paragraph approved by Gibbs J) said at 10-105 (emphasis added):

It follows, from the principle set out in the last paragraph, that a limitation placed by a patentee upon the implied license given on sale of an article is effective if and only if the person against whom it is desired to enforce the restriction acquired the article concerned knowing that some limitation existed. …

…

It would appear to be sufficient in an action to enforce license restrictions against an owner of patented goods, to show that he acquired the goods with knowledge of restrictions, without also showing that his predecessors in title knew of them too.

129    The inbuilt restriction argument depends upon notification of limitations to Ninestar arising implicitly from the subject matter of the product itself. Seiko asks the Court to draw an inference that Seiko’s own intentions and reasons for programming its memory chips were for a particular purpose and that Ninestar would or should have knowledge of this fact. However, I do not accept that the particular requirements necessary to alter the functionality of the memory chip (and thereby the printer cartridge) amount to notice of conditions on the use to which the product may be put. They are digital attributes of the goods themselves: data stored on a memory chip. None amounts to notice of a relevant limitation.

130    Secondly, the evidence of Mr Gibbons is that there has since the late 1990s existed technology whereby OEMs could “lock out” or prevent any further use being made of the ink cartridge. Mr Gibbons’ evidence is that since the late 1990s, manufacturers of “aftermarket” cartridges have developed technology for resetting or reprogramming machines which are able to restore the data on the existing OEM chips to their original settings, thereby enabling the cartridges to be refilled and resold.

131    The fact that the information on the memory chip has to be changed in order to enable the cartridge to be refilled does not of itself amount to notice to the owner that the cartridge may not be changed, or that it would be a violation of the patentee’s patent rights to do so. Indeed, the evidence of Dr Hawkins is that it is entirely possible for an OEM such as Seiko to arrange that the information be stored in a memory chip in such a way that it cannot be rewritten. Seiko has chosen not to do so.

132    As for category 2 and category 3 cartridges, in order for the cartridge to work after it has run out of ink the chip must be put into test mode. Seiko submits that the purpose of “test mode” was to prevent subsequent owners from altering the memory of the chip. But no Seiko witness gave evidence that test mode was for that purpose. Dr Hawkins explained in his evidence that the purpose of test mode was to enable a manufacturer to test the functionality of the memory chip, in circumstances where after manufacture a significant fraction of the chips are defective. He explains that the existence of and access to test mode is “closely guarded” by memory chip manufacturers “primarily because the test mode is only intended to be used in testing … this means that the test mode is not intended to be used when the ink cartridge is being used in the printer”. Dr Hawkins’ opinion that the test mode is not intended to be used cannot serve as a proxy for notice of a restriction. No doubt many manufacturers would prefer their products not to be reverse engineered, and would take steps to thwart competitors in their attempts to do so. However, presenting a technical challenge to competitors is too opaque to amount to notice.

133    The position in relation to “normal mode” is weaker still, because, it seems, that the content of the memory chip can be rewritten without having to first place it into “test mode”.

134    Thirdly, Seiko has at all times had the ability to state in words, written on or in relation to the cartridges, the manner in which the owner is able to deal with those cartridges. For its own reasons, it has chosen not to.

135    Mr Nishida gives evidence that included amongst the Calidad products that were supplied to him by EAL was an Epson ICLM50 cartridge that had been modified by Ninestar and sold by Calidad as a Calidad 258 cartridge. When he peeled a Calidad sticker off the side of the cartridge he found an Epson sticker that included the words, in Japanese, “for Japanese market only”. Seiko does not in its submissions rely on this writing as evidence of a restriction. The evidence discloses that these original Epson cartridges, sold in Japan, have not been restored for sale to Calidad since prior to April 2016. In my view, these words are not sufficient. They are likely to be understood as applicable only upon the first sale of the cartridges, not to subsequent sales after refilling.

136    Fourthly, it is to be noted that Ninestar would only become aware of any inbuilt restrictions after obtaining the used original Epson cartridge, and at the point in time when it sought access to content of the memory chip. That is, after the point of Ninestar acquiring the product. Such notice would be too late. See, for instance, Incandescent Gas Light Co Ltd v Cantelo (1895) 12 RPC 262 at 264; Roussel at 444.

137    Ultimately, these matters do not support the proposition that Seiko has deliberately placed “inbuilt restrictions” for the purpose of ensuring that subsequent owners of the original Epson cartridges may not configure them for further use. I am not satisfied that they do so for that purpose. More importantly, I am also not satisfied that, even if that were to be imputed as Seiko’s intention, the inclusion of such restrictions would sufficiently amount to “notice” for the purpose of the implied licence arising upon a sale by the patentee, or subsequent dealings in the original Epson cartridges. In this regard a further and separate factor for consideration is that the original Epson cartridges are products that are capable of re-use (or recycling). It is to be expected that products of this type are purchased in the expectation that they would or might be so used and that a manufacturer who desires to prevent such use would indicate with reasonable clarity that it intends to do so. A court would be slow to infer an intention to restrict a subsequent owner of the product from re-using it and slow to find that notice is adequately given to purchasers where the patentee did not attempt to do so in express terms.

Calidad

138    Seiko’s case concerns the alleged infringement of patents by Calidad, not Ninestar. It contends that Calidad was, when it acquired the Calidad products from Ninestar, aware of the conditions of use arising by reason of the inbuilt restrictions applicable to Ninestar. That is, it contends that Calidad was aware of the same restrictions as Ninestar.

139    Seiko was unable to point to any direct evidence on this subject, but it asks the Court to infer that because Mr Kenyon (the only Calidad witness) was unable to assist on the subject, that other employees or directors of Calidad, such as Mr Robin Kenyon (the Chief Executive Officer and father of Mr Kenyon) would have been able to assist the Court but failed to do so. Seiko also submits that such knowledge may be inferred to have arisen since no later than October 2013 when the Settlement Deed was entered, or at least since these proceedings commenced and the first version of the Statement of Agreed Facts was finalised (27 May 2016), or, as a final alternative, when Mr Li and Mr Shen gave evidence in Court.

140    The difficulty that Seiko faces is that I have rejected the inbuilt restriction argument. It follows that the other arguments, which rely on Calidad being aware of the inbuilt restrictions must also fail.

141    Seiko separately advances the argument, in one sentence in its closing submissions, that “in any event” since these proceedings commenced, and since at least the initial version of the statement of agreed facts (or alternatively the evidence of Mr Li and Mr Shen), the Calidad parties are now intimately aware of the steps necessary to stop the original Epson cartridge’s memory chip from preventing the cartridge from working. These factors are said to amount to sufficient notice of the inbuilt limitation.

142    As articulated by Seiko, this argument continues to rely upon the fact of the inbuilt restrictions as the basis for the limitation. I reject this argument for substantially the same reasons advanced in relation to the inbuilt restrictions argument insofar as it applied to Ninestar. Further, notice of restrictions sought to be imposed or notified after ownership is transferred cannot apply; see [136] above.

143    Accordingly, I am not satisfied that Seiko has established that Calidad acquired the patented products subject to limitation.

3.5    The legal relevance of the modifications to the original Epson cartridges

3.5.1    Introduction

144    I have found that neither Ninestar nor Calidad took the products in issue subject to notice of restrictions imposed by Seiko. However, that finding is not determinative of the matter in favour of Calidad. The elephant remaining in the room is the fact that Ninestar made modifications to the original Epson cartridges, and Calidad imported the modified products (ie the Calidad products) into Australia. To what extent do the modifications affect the implied licence? Seiko contends that the modifications so made are such that the implied licence cannot survive them.

145    No Australian law directly touches upon this subject. The parties called in aid United Kingdom authority in support of their respective positions. Seiko pointed to differences between the original Seiko cartridges and the Calidad products to contend that no implied licence could apply. Calidad drew on United Wire and Schütz to contend that there was no infringement of rights. In the sections below I first consider the United Kingdom cases relevant to these arguments. I then turn to the question of the broad circumstances in which under Australian law an implied licence might survive modifications made to a patented product in section 3.5.3. In section 6 I consider the question in relation to the facts of the present case.

3.5.2    The United Kingdom law

146    Seiko relies upon the United Kingdom authorities concerning repairs. In Sirdar Rubber Co v Wallington Weston & Co (1907) 24 RPC 539 (Sirdar Rubber) at 543 Lord Halsbury said:

The principle is quite clear although its application is sometimes difficult; you may prolong the life of a licensed article but you must not make a new one under the cover of repair.

147    This statement of principle was endorsed by Lord Buckley (Orr and Goff LJJ agreeing) in Solar Thomson at 554, who went on to approve (at 555) the reasoning of the trial judge in Sirdar Rubber (Swinfen Eady J) to the effect that the purchaser of a patented article has a right to prolong its life by fair repair, but does not have a right to obtain, without licence from the patentee, a substantially new article made in accordance with the invention, retaining only some subordinate part of the old article so that it may be said that the combination is not an entirely new one. In Solar Thomson, in concluding that the repair defence was made out, the Court identified that the “cardinal question” must be whether what has been done can fairly be termed a repair, having regard to the nature of the patented article. If it is, any purchaser of such an article is impliedly licensed to carry it out or to contract with someone else to carry it out for him.

148    In the United Kingdom, the right to repair a patented product is no longer characterised as arising from an implied licence.

149    In Canon Kabushiki Kaisha v Green Cartridge Co (Hong Kong) Ltd [1997] AC 728 (Canon) Lord Hoffman (with whom their Lordships agreed) said that the concept of a licence, which is something which “makes an action lawful which without it had been unlawful” (citing Thomas v Sorrell (1674) Vaughan 330, 351) is “not really” applicable to the repair of a patented article. His Lordship said at 822:

Because repair is by definition something which does not amount to the manufacture of the patented article, it is not an infringement of the monopoly conferred by the patent. It cannot therefore be an unlawful act and needs no special licence to make it lawful, unless as part of a general implied licence to use the patented product at all, which is sometimes used to explain why mere user does not infringe the patentee’s monopoly. But this is perhaps better regarded as a consequence of the exhaustion of the patentee’s rights in respect of the particular article when it is sold. In Solar Thomson, however, (at pages 560-561) the Court of Appeal extended the concept into the field of copyright by holding that a purchaser of an article is impliedly licensed to infringe the owner’s copyright in drawings of parts of that article so far as is necessary to enable repairs to be carried out. This is a genuine implied licence since, without it, the reproduction of the parts would be unlawful.

150    The shift of ground from Solar Thomson portended in Canon was reinforced in the decision of United Wire. In that case the decision of Lord Hoffman (Lords Steyn, Cooke and Bingham and Hutton agreeing) found that the proper analysis did not lie in considering whether there was an implied licence to repair a patented product once it had been sold, but whether in conducting work (whether defined as “repair”, “modification” or “adaptation”) there was a “making” of the product within the meaning of that term under s 60(1)(a) of the Patents Act 1977 (UK) (UK Patents Act). That section provides that a person infringes a patent if (emphasis added), “where the invention is a product, he makes, disposes of, offers to dispose of, uses or imports the product or keeps it whether for disposal or otherwise”, without the consent of the patentee. Their Lordships also set to one side an analysis that might arise from the doctrine of exhaustion of rights as it is applied in “European” patent systems.

151    His Lordship said (emphasis added):

70.     Where however it is alleged that the defendant has infringed by making the patented product, the concepts of an implied license or exhaustion of rights can have no part to play. The sale of a patented article cannot confer an implied license to make another or exhaust the right of the patentee to prevent others from being made. A repair of the patented product is by definition an act which does not amount to making it: as Lord Halsbury LC said of the old law in Sirdar Rubber Co Ltd v Wallington, Weston & Co (1907) 24 RPC 539 at 543:

you may prolong the life of a licensed article but you must not make a new one under the cover of repair.

71.     Repair is one of the concepts (like modifying or adapting) which shares a boundary with “making” that does not trespass upon its territory. I therefore agree with the Court of Appeal that in an action for infringement by making, the notion of an implied license to repair is superfluous and possibly even confusing. It distracts attention from the question raised by section 60(1)(a), which is whether the defendant has made the patented product. As a matter of ordinary language, the notions of making and repair may well overlap. But for the purposes of the statute, they are mutually exclusive. The owner’s right to repair is not an independent right conferred upon him by license, express or implied. It is a residual right, forming part of the right to do whatever does not amount to making the product.

152    On the facts, the House of Lords upheld the decision of the Court of Appeal. The patents in suit related to the construction of mesh screens used in vibrating sifting machines sold to the oil exploration industry. When drilling fluid, known as drilling mud, has been pumped down a wellbore it returns to the surface carrying drill bit cuttings and other solids which must be removed in order to recycle the drilling fluid. Claim 1 was for a sifting screen assembly for use in a vibratory sifting machine comprising a frame and two screens superimposed on one another which are adhesively secured to the edges of the frame, the upper screen having a relatively fine mesh and the lower screen having a relatively course mesh, both screens being tensioned and the upper screen being at a lower tension than the lower screen so that when the frame is vibrated the screens strike each other so as to affect or assist in effecting clearing of blockages or to prevent or reduce the possibility of blockage of the screens.

153    The defendant operated a business acquiring frames from the patentee’s customers, stripping them down to the bare metal, recoating them and attaching 2 layers of mesh, course below and fine above and which satisfy the requirements of the claim. Lord Hoffman, in rejecting that the defence applied on the facts, accepted that the frame was an important part of the assembly but said at [73]:

… It is quite true that the defendants prolonged the useful life of the frame. It would otherwise presumably have been scrapped. But the screen was the combination of frame and meshes pre-tensioned by attachment with adhesive according to the invention. That product ceased to exist when the meshes were removed and the frame stripped down to the bare metal. What remained at that stage was merely an important component, a skeleton or chassis, from which a new screen could be made.

154    In Schütz the UK Supreme Court (Lord Neuberger P, Lord Walker, Lady Hale, Lord Mance and Lord Kerr SCJJ agreeing) followed Lord Hoffman’s reasoning in United Wire and developed it somewhat.

155    The subject of that case was a patent for an intermediate bulk container (IBC) which is a large container of two-part construction used for the transport of liquids consisting of a metal cage into which a large plastic bottle is fitted. The claim in suit was for a container for transporting and storing liquids, having an exchangeable container made of plastic and an outer sleeve consisting of vertical and horizontal lattice bars made of metal surrounding the and container and supporting it. The claim focussed on the way that lattice bars of the cage were configured and went on by describing the contact points of the lattice bars being welded together at intersection points which were “characterised” in a certain, detailed way so as to add to the strength and durability of the joins.

156    The defendant, Werit, supplied plastic bottles to replace those used in used IBCs sold by the patent licensee, Schütz. A third party, Delta, acquired discarded IBCs put on the market by Schütz and replaced the original bottles with the bottles supplied by Werit. Delta then offered these cross-bottled IBCs for sale in the market in competition with the original Schütz IBCs.

157    Lord Neuberger applied a multifactorial approach in determining that the work performed in replacing the plastic bottles in the IBCs did not amount to a “making”. He concluded at [78] saying:

… However, in this case, it appears to me that it is a classic example of identifying the various factors which apply on the particular facts, and, after weighing them all up, concluding, as a matter of judgement, whether the alleged infringer does or does not “make” the patented article. In the present case, given that (a) the bottle (i) is a freestanding, replaceable component of the patented article, (ii) has no connection with the claimed inventive concept, (iii) has a much shorter life expectancy than the other, inventive, component, (iv) cannot be described as the main component of the article, and (b) apart from replacing it, Delta does no additional work to the article beyond routine repairs, I am of the view that, in carrying out this work, Delta does not “make” the patent that article.

158    The decisions in these cases provide some guidance on the question of modifications made to patented products, but are not directly on point in the present case. The older “repair” cases ask the “cardinal” question of whether a repair has been done having regard to the nature of the patented article. That assumes that the owner performing the repairs has the benefit of an implied licence. But in the present case, the anterior question must be asked as to whether the implied licence survives the modifications made to the original Epson cartridges.

159    A similar difficulty arises in the “making” cases more recently adopted in the United Kingdom. As Lord Hoffman said in United Wire at [70], where the allegation is that the defendant has infringed by making, the concept of implied licence (to repair) or exhaustion of rights can have no part to play. That is because if there is no “making” there is no need for a licence. But that assumes that making in the jurisdiction is the relevant infringement alleged.

160    In none of these cases did the court consider whether the implied licence arising from a sale sub modo was extinguished by modifications made to the patented product and, if so, what (if any) modifications could be made. That question arises on the facts of the present case because modifications were made prior to importing the Calidad product. None of these cases, nor the authorities stemming from National Phonograph consider this question.

3.5.3    When modifications extinguish the implied licence

161    In the present case, CDP imported the products after Ninestar had modified them. The only defence available to Calidad is that the importation of the Calidad products falls within the umbrella of the implied licence.

162    A distinction must be drawn between two sets of rights that may be available to an owner of a patented product who takes it without restriction. First, the rights available to deal with the product without restriction. These rights concern the importation, sale and use of the patented product. Secondly, rights to repair the product (under an implied licence to repair) or otherwise perform acts which fall short of a “making” (pursuant to a residual right to do acts less than a “making”). The present case concerns only the first of these.

163    The proper question is accordingly not based on a repair or making analysis, but rather whether the implied licence can be said to apply in circumstances where one owner of the product, Ninestar, has modified it and then passed title in it to CDP. The relevant question then becomes; at what point has an owner of the embodiment made changes to it such that the implied licence arising from a sale sub modo is extinguished? No authority appears to have directly addressed this topic.

164    For the reasons developed below, in my view the threshold question to consider is whether or not the modified product is materially the same embodiment of the invention as claimed as the product that the patentee sold without restriction. That is not an inquiry that is at large. The right of the patentee to impose restrictions upon sale arises because the product is an embodiment of an invention as claimed. The question is not whether or not the product was altered or repaired, but whether the product, insofar as it is an embodiment of the invention as claimed, was materially altered, such that the implied licence can no longer sensibly be said to apply.

165    The implied licence arises from the disposition by a patentee of a particular embodiment of its invention. As noted in National Phonograph (at 22, 28), the patented goods are not free of restriction simply because they are chattels, but because the patentee has chosen not to exercise its right to impose restrictions on their use arising under the relevant Patents Act.

166    The basis for the implied licence is because of the sale of the particular patented product. That is, a physical product that embodies the features claimed in the patent. When such a product is sold by or with the authority of the patentee, the authority passed on to subsequent owners of the product is a permission relating to that embodiment of the features of the claimed invention. Accordingly, while the sale sub modo will remove the ability of the patentee to restrict a purchaser from importing, selling, using or disposing of the product, the licence is necessarily confined to such rights insofar as they apply to the particular embodiment sold with the approval of the patentee.

167    The extent to which modifications are to be regarded as relevant must be examined in the context of the exclusive right of a patentee to “exploit” the invention as conferred by s 13 of the Act. It is constrained by the extent to which any alleged infringing product falls within the claims of the patent. This is seen by the language of the definition of “exploit”, which in the chapeau identifies that the exclusive rights of the patentee exist “in relation to an invention”, being the invention as claimed.

168    The plurality noted in D’Arcy v Myriad Genetics Inc [2015] HCA 35; (2015) 258 CLR 334 at [14] (French CJ, Kiefel, Bell and Keane JJ, citations omitted):

… The function of the claim was described by Lord Russell of Killowen in 1938 as “to define clearly and with precision the monopoly claimed, so that others may know the exact boundaries of the area within which they will be trespassers.” Its limiting role was emphasised:

It and it alone [defines] the monopoly; and the patentee is under a statutory obligation to state in the claims clearly and distinctly what is the invention which he desires to protect.

Lord Russell’s observations have stood the test of time in the United Kingdom as “[t]he best-known statement of the status of the claims in UK law”. They also describe the function of the claim mandated by s 40(2)(b) of the Act. As succinctly, but accurately, stated in a recent Australian text that function is “to define what it is that the patentee has exclusive right to, during the term of the patent.”

169    In short, the exclusive rights of a patentee to constrain the conduct of the owner of a patented product can only arise insofar as that product is an embodiment of the invention as claimed. Otherwise the scope of the patentee’s rights would exceed the rights conferred by the Act.

170    This approach is supported by the long-established approach to the question of infringement of patent rights. Whilst the Act itself contains no definition of infringement, the fundamental rule is that there will be no infringement unless the alleged infringer has taken all of the essential features of the patentee’s claim; Rodi and Wienenberger AG v Henry Showell Ltd [1969] RPC 367; Olin Corporation v Super Cartridge Co Pty Ltd [1977] HCA 23; (1977) 180 CLR 236 at 246; 14 ALR 149 at 157; Fresenius Medical Care Australia Pty Ltd v Gambro Pty Ltd [2005] FCAFC 220; (2005) 67 IPR 230 at [49], [50]. An act of alleged infringement that falls short of possessing all of the integers of the claim will not (leaving aside the application of s 117 of the Act) amount to infringement; Coretell Pty Ltd v Australian Mud Company Pty Ltd [2017] FCAFC 54 at [56].

171    As Calidad submits (in the context of a “making” analysis), it is the relationship between the features of the product and the invention claimed that is relevant. It is those aspects of the product that benefit from monopoly rights. Attention must focus upon the patent and product as sold insofar as it is a representation of the invention as claimed. As Lord Neuberger said in Schütz at [25], it is hard to see what else could be the subject matter of the “making”. The same applies to the implied licence.

172    That is not to say that the invention as claimed is not to be considered as a whole, or that particular integers are to be taken out of context or divorced from the combination of integers which make up the invention, but rather where modifications have no bearing on the invention as claimed, they are unlikely to amount to a material modification of the product. This was the approach taken in Schütz at [61], [63] and is applicable in the present case.

173    However, factors such as whether or not the modification concerns a freestanding, replaceable component of the patented article, whether modifications are made to something described as a subsidiary component of the article (each of which was identified in Schütz as relevant) or factors such as whether the part modified or replaced is durable and so forth (identified in Solar Thomson as relevant) have no role to play in considering the modifications for present purposes.

174    Calidad submits that in considering modifications the Court should take into account the inventive step or inventive concept identified in the patent and reflected in the claims. To the extent that modifications do not bear upon the inventive concept then they are less relevant and less likely to have a bearing on the question. This, they submit, was the approach taken in Schütz. Furthermore, they submit that the language of inventive concept has been adopted in Australian courts in several contexts, including questions of entitlement pursuant to s 15 of the Act (Polwood Pty Ltd v Foxworth Pty Ltd [2008] FCAFC 9; (2008) 165 FCR 527 at [61]), sufficiency pursuant to s 40 of the Act (Apotex Pty Ltd v Warner-Lambert Company LLC (No 2) [2016] FCA 1238; (2016) 122 IPR 17 at [258]) and lack of inventive step (Apotex Pty Ltd v Sanofi-Aventis [2009] FCAFC 134; (2009) 82 IPR 416 at [159].

175    However, the approach taken in Schütz in recognising the potential relevance of the inventive concept to analysing modifications arose in the context of the “making” analysis, which is not the present inquiry. It was also heavily influenced by the approach of European courts, including the German Bundesgerichtsof, which takes into account the extent to which the “technical effects of the invention” (that is, the inventive concept) are reflected by the replaced parts. The admonition made by the High Court to be cautious of the direct application of post 1977 United Kingdom law, influenced as it is by concepts of European law not presently applicable in Australia, suggests that a single judge would be cautious in adopting that approach in the present case; Lockwood Security Products Pty Ltd v Doric Products Pty Ltd [2004] HCA 58; (2004) 217 CLR 274 at [66], [67]. The Australian cases cited by Calidad are, in my view, not analogous.

176    Seiko urges that for the present purposes the “correct comparator” is not between the invention as claimed and reflected in the original Epson cartridges on the one hand, and the modified Calidad products on the other, but rather between the original Epson cartridges as products and the Calidad products. This approach is to diminish the role of the invention as claimed and embodied in the original Epson cartridges. For the reasons set out above, I do not consider that approach to be appropriate.

177    Seiko also submits that in considering the modifications made, the Court would also consider the research and development effort required by Ninestar in order to enable the modifications to take place. In this respect, Seiko particularly emphasises research and development work conducted in order to make changes to the memory on the memory chip of the original Epson cartridges. That consideration is also irrelevant to consideration of the modifications. I address this further in the context of reviewing the invention as claimed in the patent.

178    This process of reasoning leads to the conclusion that to assess whether or not the implied licence continues after modifications are made one must consider the significance of modification work done on a product and how the modifications in question relate to the features of the patented product that are defined by the claim. Where that work done, or alteration made, does not concern a claimed feature, then the work is irrelevant to the analysis. That is because the patentee’s rights to limit the use of the patented product arise because the product represents an embodiment of the claimed features. This involves the 3-staged factual inquiry undertaken below; first, what is the scope of the invention as claimed (which is addressed in section 4), secondly, what is the manner in which the patentee’s product is an embodiment of the invention as claimed (section 5) and thirdly, to what extent do the modifications made affect the patented product insofar as it represents an embodiment of the claims (section 6).

4.    THE PATENT

4.1    Background

179    The parties agree that there was no material difference between the disclosure in the body of the specification of the 643 patent and the 239 patent. They also draw no material distinction, for the purpose of the present dispute, between the content of the 2 asserted claims, being claim 1 of the 643 patent and the content of claim 1 of the 239 patent. In these reasons, following the approach of the parties, attention focusses entirely upon the disclosure of the 643 patent, and claim 1 of that patent.

180    Each of Mr Knight and Dr Hawkins addresses the disclosure of the specification. There is no dispute that each have the qualifications of persons skilled in the art able to assist in the construction of the patent.

181    To the extent that any background information was necessary to understand the disclosure of the patent specification, the parties agreed to the tender of a primer that contains background information known and understood to those skilled in the art as at the priority date of the patents. As such, they agreed that it reflects the state of the common general knowledge as at the priority date. The primer records some slight areas of disagreement between the experts, but for present purposes it is not necessary to record those differences. Section 2.1 above (“Background Technology”) sets out the relevant parts of the agreed primer.

4.2    The disclosure of the specification

182    The patent is entitled “Printing material container and board” and relates in general to a printing material container (which includes an ink cartridge) and/or a board adapted to be attached to or connectable to a printing apparatus.

183    In the section entitled “Background Art” the patent describes that it has been common practice in the field to equip ink cartridges used in inkjet printers with a device, such as a memory, for storing information relating to the ink, together with another device such as a high voltage circuit which is used to detect the remaining ink level via a piezoelectric element. This second device applies a higher voltage than the driving voltage of the memory. In such cases, the ink cartridge and the printing apparatus may be electrically connected through terminals. The patent proposes “a structure for preventing the information storage medium [or memory] from shorting and becoming damaged due to a drop of liquid being deposited on the terminals connecting the printing apparatus with the storage medium”.

184    The section goes on (at [3]):

However, the technologies mentioned above do not contemplate an ink cartridge having [been] equipped with a plurality of devices, for example, a memory and a high voltage circuit, with terminals for one device and the terminals for another device. With this kind of cartridge, there was a risk that shorting could occur between a terminal for the one device and the terminal for the [sic] another device. Such shorting caused the problem of possible damage to the ink cartridge or to the printing apparatus in which the ink cartridge is attached. …

185    It may immediately be seen that the specification directs attention to an ink cartridge that has 2 or more devices such as a memory and a high voltage piezoelectric device for detecting ink levels, each with electric terminals. A risk of shorting between the terminals arises. The patent proposes a solution to prevent the information storage medium (or “memory”) from being damaged as a result of shorting between terminals.

186    The specification continues under the heading “Disclosure of the Invention” with a series of statements of aspects of the invention. The section commences at [4] with the observation that the invention, at least in a preferred embodiment, provides a printing material container and/or a board (which the evidence indicates may be a printed circuit board or an integrated circuit board) adapted to be attached to or connected to a printing apparatus, wherein damage caused by shorting between terminals can be prevented or reduced.

187    A number of aspects of the invention are then described. The first (at [5]) is said to provide a printing material container adapted to be inserted into a printing apparatus. The printing apparatus has a print head and a plurality of terminals. The printing material container includes a memory driven by a memory driving voltage and an “electronic device” driven by a higher voltage. It has terminals including a plurality of memory terminals electrically connected to the memory and two terminals (a first and a second terminal) electrically connected to the electronic device. The terminals include a contact portion (or pad) for contacting a corresponding terminal on the printing apparatus. The contact portions are arranged in two rows which are generally orthogonal to the direction of insertion of the ink cartridge, one above the other. The lower, or first row, is longer than the second, or higher row. A contact portion of the first electronic device terminal is at one end of the contact portion and a contact portion of the second electronic device terminal is disposed at the opposite end of the first row.

188    It may be seen that the combination described in this aspect of the invention provides for a spatial arrangement that ensures separation between the 2 contact portions of the higher voltage electronic device. This first aspect is a statement of the invention, or consistory clause corresponding to claim 1 of the 643 patent.

189    A number of further aspects are then described, many of which address and describe different arrangements for contact portions to ensure that there is a separation between terminals of the electronic device. In [5D] a different form of the invention is described where the plurality of terminals includes at least one “short detection terminal” provided to detect shorting between it and the first electronic device terminal. No short detection terminal appears in claim 1.

190    The specification then provides a “Brief Description of the Drawings” and then a lengthy section entitled “Best Modes of Carrying Out the Invention”, which describes embodiments of the invention by reference to the figures. It commences with a description of the arrangement of the printing apparatus and ink cartridge. Figure 1 shows a perspective view of the construction of a printing apparatus. It is set out below.

191    With reference to figure 1, the specification describes that printing apparatus 1000 has a sub-scan feed mechanism, a main scan feed mechanism and a head drive mechanism. The sub-scan feed mechanism carries the printing paper P in the sub-scanning direction using a paper feed roller 10 powered by a paper feed motor (not shown). The main scan feed mechanism uses the power of a carriage motor 2 to reciprocate in the main scanning direction a carriage 3 connected to a drive belt. The head drive mechanism drives a printhead 5 mounted on the carriage 3, to eject ink and form dots. The printing apparatus additionally comprises a main control circuit 40, for controlling the various mechanisms, which are connected to the carriage 3 via a flexible cable 37.

192    The ink cartridge of 1 embodiment is depicted in figure 2 as follows:

193    The specification explains that the ink cartridge, which is attached to holder 4 (figure 1), comprises a housing 101 containing ink, a lid 102, a board 200, and a device in the form of a sensor 104. On the bottom face of the housing 101 there is an ink supply orifice 110 into which the ink supply needle 6 (depicted in figure 1) inserts when ink cartridge 100 is attached to the holder 4. The board 200 fits between ribs 106 and 107.

194    The “board” 200 is of particular relevance to the present dispute, as it is the arrangement of the terminals on it that provides an example of an embodiment within claim 1. This may be explained by reference to figures 3A, 3B and 3C, which are depicted below:

195    The arrow R indicates the insertion direction of the ink cartridge. The board 200 comprises a “device” in the form of a memory 203 disposed on its back face, and a terminal group composed of 9 terminals 210-290 disposed on its front face. The memory 203 stores information relating to the ink contained in the ink cartridge 100.

196    The specification explains the electrical arrangement set out in figures 3A, 3B and 3C as follows:

[42] … the ground terminal 220, the power supply terminal 230, the reset terminal 260, the clock terminal 270 and the data terminal 280 are electrically connected to the memory 203. The memory 203 is, for example, EEPROM comprising serially accessed memory cells, and performing data read/write operations in sync with a clock signal. …

197    In [74], [75] the specification provides additional explanation (emphasis added):

[74] As mentioned previously with reference to FIG. 3A, in the ink cartridge 100 pertaining to the embodiment the first sensor drive terminal 250 and the second sensor drive terminal 290 which apply the driving voltage to the sensor are arranged at the two ends of the terminal group, so the number of adjacent terminals is small. As a result, the likelihood of the first sensor drive terminal 250 and the second sensor drive terminal 290 shorting to other terminals is low. Fig. 3C shows another example of the board. This board 200’ is different from the board 200 shown in Fig. 3A in that the board 200’ has seven terminals 220, 230, 250, 260, 270, 280, and 290 excluding the first and second short detection terminals 210, 240. The upper row consists of the ground terminal 220 and the power supply terminal 230. The lower row consists of the first sensor drive terminal 250, the reset terminal 260, the clock terminal 270, the data terminal 280, and the second sensor drive terminal 290. This board 200’ also has the effect that the likelihood of the first sensor drive terminal 250 in the second sensor drive terminal 290 shorting to other terminals is low, as is the case with the board 200 shown in Fig. 3A.

[75] On the board 200, if the first sensor drive terminal 250 should short to the adjacent first short detection terminal 210, the shorting will be detected by the aforementioned cartridge detection/short detection circuit 502. …

198    The specification notes that damage to the circuits of the memory 203 and the printing apparatus 1000 caused by shorting or by infiltration of foreign matter can be prevented or reduced.

199    The specification describes 2 broad forms that provide a solution to the problem of shorting. The first is to place the terminals (or pads) on the printed circuit board in a particular way to reduce or prevent the likelihood of shorting by reason of the special arrangement. The second is to use one or more “short detection circuits” to detect shorting and take action to prevent damage to the printer or cartridge. These 2 aspects are combined in some embodiments.

4.3    The Claims

200    The 643 patent has 40 claims. Claim 1 provides as follows (for convenience, integer numbers have been added):

[1] A printing material container adapted to be attached to a printing apparatus by being inserted in an insertion direction, the printing apparatus having a print head and a plurality of apparatus-side terminals, the printing material container including:

[2] a memory driven by a memory driving voltage;

[3] an electronic device driven by a higher voltage than the memory driving voltage;

[4] a plurality of terminals including a plurality of memory terminals electrically connected to the memory, and a first electronic device terminal and a second electronic device terminal electrically connected to the electronic device, wherein:

[5] the plurality of terminals each include a contact portion for contacting a corresponding terminal of the plurality of apparatus-side terminals,

[6] the contact portions are arranged in a first row of contact portions and in a second row of contact portions, the first row of contact portions and the second row of contact portions extending in a row direction which is generally orthogonal to the insertion direction,

[7] the first row of contact portions is disposed at a location that is further in the insertion direction than the second row of contact portions,

[8] the first row of contact portions is longer than the second row of contact portions, and,

[9] the first row of contact portions has a first end position and a second end position at opposite ends thereof,

[10] a contact portion of the first electronic device terminal is disposed at the first end position in the first row of contact portions and

[11] a contact portion of the second electronic device terminal is disposed at the second end position in the first row of contact portions.

201    It may be seen that the features of the product claimed, broadly, are that it is an ink cartridge that can be attached to a printer by inserting it, with both the ink cartridge and printer having a plurality of terminals. The ink cartridge must have a memory driven by a memory driving voltage (integer [2]), an electronic device driven by a higher voltage (integer [3]), terminals electrically connected to the memory and terminals electrically connected to the electronic device (integer [4]), and contact portions for the terminals which are arranged in a particular way (integers [5]-[11]).

202    The ink cartridge of the claim accordingly incorporates a connector that provides an electrical connection between the printer’s electronics and both devices. Both the ink cartridge and the printer in which it is used have terminals which are corresponding. The printer and cartridge terminals are positioned in two rows. The first row of contact portions is further in the insertion direction than the second row of contact portions. The apparatus-side terminals that have a higher voltage are unable to make contact with the low voltage ink cartridge terminals as a consequence of their relative placement.. Two rows of contacts allow the adjacent contacts to be twice as far apart than one, while still allowing the cartridge to have a narrow width as opposed to placing all the contacts in one row.

203    Dr Hawkins explains in his evidence that he understands from the specification that it is the special arrangement of contact portions on the container as set out in claim 1 that reduces or eliminates the risk of shorting. His evidence is that the danger of damage to the device (including the memory, or integrated circuit) is avoided by that arrangement.

204    There is a potential for confusion arising from the use in the evidence of the word “memory”. It general terms it could mean a piece of hardware (as in a memory device), or alternatively information stored such as data providing information about the contents of an ink cartridge. In yet another context it could be understood as a reference to instructions within a device which may be regarded as software or firmware.

205    Mr Knight understands the reference in the claim to “memory” to be to an item of hardware, being a memory device. Dr Hawkins agrees, although in his affidavit evidence he gives his opinion that the word “memory” as described in claim 1 has 3 basic requirements. First, that it be a device that operates at a lower voltage than the second electronic device, secondly, that it contains at least 1 piece of information that it provides to the printer when a read instruction is issued to it, and thirdly, that it would retain its information when the cartridge is removed from the printer. The first of these is apparent from the language of claim 1 (integers [2], [3]), which makes clear, as the parties ultimately accepted in their closing submissions, that “memory” in integer [2] is an electronic device, that is, a piece of hardware. In integer [2] the memory is driven by a memory driving voltage, and in integer [4] it is connected to terminals. This conclusion is supported by the way in which “memory” is identified in the specification as a hardware device. A distinction is made between memory and the information that is stored in it. For instance, at [2] reference is made to “a device, for example, a memory for storing information relating to the ink”; at [30] where board 200 “comprises a device in the form of a memory 203 disposed on its back face”; and at [0042] where “[t]he memory 203 is, for example, EEPROM comprising serially accessed memory cells, and performing data read/write operations in sync with a clock signal”.

206    This construction of “memory” also accords with the normal usage of the term in the field as at the priority date, which Dr Hawkins accepted in cross-examination means “a device in which data can be entered, in which it can be held and from which it can be retrieved at a later time”.

207    The second and third aspects that Dr Hawkins contends form part of the “memory” within integer [2] do not represent aspects of limitation to be found in the claim. Rather, in my view, they amount to glosses that Dr Hawkins has added to the term which are not present in the language of the claim and which cannot sensibly be derived as forming part of the language of the claim having regard to the disclosure of the body of the specification; see Décor Corporation Pty Ltd v Dart Industries (1988) 13 IPR 385 at 400.

208    Accordingly, as a matter of construction integer [2] is to be understood as referring to a hardware device. Such a device is characterised by its ability to hold or store information but the information stored on the device is not the subject of any claim limitation. The reason for this is that the invention is directed towards a particular terminal layout arrangement for a printing material container that has a reduced likelihood of shorting. The particular data stored in the memory has no role to play in reducing the likelihood of shorting. The combination claimed is concerned with the particular layout of the features described. Put another way, the claim itself is indifferent as to the information or data stored on the memory.

5.    HOW DO THE ORIGINAL EPSON CARTRIDGES EMBODY THE INVENTION AS CLAIMED?

209    The original Epson cartridges are described in section 2.2 above. There is no dispute that they are embodiments of claim 1 of the 643 patent. The manner in which the cartridges embody the invention must be explained by reference to the integers of claim 1.

210    Integer [1] is for :

[1] A printing material container adapted to be attached to a printing apparatus by being inserted in an insertion direction, the printing apparatus having a print head and a plurality of apparatus-side terminals, the printing material container including:

211    Each original Epson cartridge is a printing material container adapted to be attached to a printing apparatus in conformity with integer [1]. Integer [1] does not define or limit the particular shape that the printing material container must take, although it requires that it be inserted in a particular direction and be adapted to be attached to a printing apparatus. That adaptation, in some models of the original Epson cartridges, is affected by an interface pattern (see figure 5 at [59] above), which, is relevant to the manner in which the ink cartridge is able to be inserted into and attached to the printing apparatus.

212    Integer [2] is for:

[2] a memory driven by a memory driving voltage;

213    The memory of integer [2] is a piece of hardware. The evidence of Dr Hawkins explains that the memory is a repository of information relevant to the interaction of the printer cartridge with the printing apparatus. There is no dispute that the memory chips of the original Epson cartridges perform that function.

214    Integer [3] is for:

[3] an electronic device driven by a higher voltage than the memory driving voltage;

215    The evidence summarised in [60] above indicates that a 3.3 volt power supply operates the memory chip in the original Epson cartridges and that many of the original Epson cartridges with the exception of the Epson T200 cartridge contain a piezoelectric sensor requiring 2 terminals which are driven by 36 volts. In the case of the Epson T200 cartridge, it has a prism at the base of the cartridge which detects the ink level with two resistor terminals with a common resistance of 62K.

216    Integers [4] – [11] are for:

[4] a plurality of terminals including a plurality of memory terminals electrically connected to the memory, and a first electronic device terminal and a second electronic device terminal electrically connected to the electronic device, wherein:

[5] the plurality of terminals each include a contact portion for contacting a corresponding terminal of the plurality of apparatus-side terminals,

[6] the contact portions are arranged in a first row of contact portions and in a second row of contact portions, the first row of contact portions and the second row of contact portions extending in a row direction which is generally orthogonal to the insertion direction,

[7] the first row of contact portions is disposed at a location that is further in the insertion direction than the second row of contact portions,

[8] the first row of contact portions is longer than the second row of contact portions, and,

[9] the first row of contact portions has a first end position and a second end position at opposite ends thereof,

[10] a contact portion of the first electronic device terminal is disposed at the first end position in the first row of contact portions and

[11] a contact portion of the second electronic device terminal is disposed at the second end position in the first row of contact portions.

217    The terminal layout diagrams set out in [60] above set out 2 examples of the terminal arrangements for the Epson 133 and 138 cartridges which fall within integers [4] – [11]. Each of the other original Epson cartridges have arrangements of rows of contact portions that similarly meet the requirements of the claims.

6.    DO THE MODIFICATIONS TERMINATE THE IMPLIED LICENCE?

6.1    Introduction

218    Section 3.5.3 above identifies that material modifications made to a patented product may serve to terminate an implied licence consequent upon a sale made without restriction. The relevant question is whether the product, insofar as it is an embodiment of the invention as claimed, was materially altered.

219    The parties each made extensive submissions going to the physical modifications that were made and the work done to achieve them. Three general points may be addressed before attending to the detail of the changes made.

220    The first concerns the relevance of the contents of the memory chip.

221    I have noted above (at [208]) that a distinction is to be drawn between the piece of hardware identified in integer [2] as the “memory”, and the data or information that is stored on a memory. The claim leaves open the content of the information that may be stored on the memory of the claim. A memory chip for a printer may store data as to how full an ink cartridge is, the colour of the ink or provide instructions that can communicate with the printer to determine whether the cartridge is compatible with the printer and so on. However, a memory could also be absent of these items of information and yet still fall within integer [2]. Accordingly, I regard as irrelevant to the current comparison the question of the changes made by Ninestar to the data contained in the memory chip.

222    The second general issue concerns the role of anterior research and development work conducted by the party modifying a patented product. Seiko contends that this is a factor to take into consideration, that Ninestar and its agents performed extensive and largely unexplained research and development work and that this should be taken into account and tend against a finding that the modifications made by Ninestar fell within the implied licence. Calidad contends that this work is irrelevant to the invention as claimed and ought to be set to one side and regarded as irrelevant.

223    In my view, there are 2 reasons why, on the facts of the present case, the research and development work relied upon by Seiko is irrelevant. The first is that the work is anterior to the making of changes to the embodiment of the claimed invention. It is like the preparation of a mould for producing a cast iron product, or conducting reverse engineering research. Neither such steps fall within the scope of the claimed invention and both are irrelevant to the question of whether the final form of the product falls within its claims. In the present context, the focus of the implied licence is upon the manner in which the product embodies the claims. That depends on the alternations made, not how they come to be made. The claim is for a product, not a process.

224    Quite separately, the research and development that is the focus of attention primarily concerns alterations made to the information contained in the memory chip. That content is not a feature of the invention as claimed. Although it no doubt took considerable time and effort on the part of Ninestar or Apex (or other unnamed companies or organisations) to work out how to re-set the contents of the memory chip in each of the original Epson cartridges, that work does not reflect a material change to the embodiment that was sold with the approval of Seiko.

225    Thirdly, Calidad contends that in considering the work done to the original Epson cartridges it is necessary and appropriate to consider what may be characterised as the “inventive concept” that lies within the claims. That submission arose in the context of Calidad’s contention that the United Kingdom law going to the question of the “making” of a product was applicable in the present circumstances (see Schütz at [78]). I have rejected that analysis as irrelevant to the present controversy. In my view, the present inquiry does not require consideration of the extent to which the original Epson products reflect an embodiment of the inventive concept (or, for that matter, the degree to which the Calidad products do). The relevant inquiry is whether the work done by Ninestar results in a material alteration to the product, insofar as it is an embodiment of the invention as claimed.

6.2    Analysis of the modifications to the original Epson cartridges

226    In the section below, my findings of fact are set out in relation to modifications made by or on behalf of Ninestar in relation to each of the categories listed in the table set out above in section 2.4. I also indicate my findings as to whether the changes are properly regarded to be material alterations to the embodiment as first sold on behalf of Seiko, such that the implied licence can no longer apply to them. This section cross-references the categories set out in the table in section 2.4 above at [73].

6.3    Category 1 – sold after April 2016

6.3.1    Summary of the modifications

227    Category 1 cartridges are the Calidad 260S and 260H cartridges that were originally the Epson T200 or T200XL cartridges and which have been modified in the manner set out below. They are current Calidad products.

228    Step (1) - preparation involves inspecting the original Epson cartridges to confirm that they are the correct type and in a suitable condition to be restored. The cartridges are then emptied and cleaned in preparation for refilling. Any residual ink inside the cartridge is removed by inserting a vacuum device using a needle and a tube into the cartridge and applying suction and then weighing the cartridge to ensure all of the ink has been removed. Where the original ink contains pigment (rather than dye), the cartridges are also washed with water. The outlet hole is prepared by removing the original seal covering it, which was broken when the Epson cartridge was first used.

229    Step (2) - refilling involves the creation of a second hole (injection port) in the middle of the main side of the cartridge with a needle which is then used to inject replacement ink into the cartridge whilst a vacuum, which is part of the refilling device, draws air out of the cartridge from its outlet hole. The injection port and the outlet hole are each sealed by placing pieces of thin clear plastic over them and applying force and heat to effect a seal.

230    In all category 1 cartridges Ninestar also makes a small slit in the gas membrane of the cartridge (using a knife) which is then resealed with glue. This is done in order to improve the gas exchange within the cartridge when it is used in the printer. A cut in the gas membrane is also made to approximately 5% of Calidad 250, 253 and 258 cartridges (when a pressure test indicates that the pressure value of the cartridge is outside a certain range). The additional 253 and 258 cartridges without their membrane cut are referred to in the table above as category A cartridges.

231    The relevant background to step (3) - reset in normal mode is as follows. In an Epson printer, each time there is a print job, the Epson printer will write a series of bits into the “ink level” memory address of the printer cartridge’s memory chip that represents the level of ink that the printer software estimates is remaining or will remain in the cartridge after the completion of the print job. When the ink level becomes depleted, the bits written into the ink level memory address represent to the printer that the ink volume is beyond a threshold level. When the Epson cartridge is refilled the data must be altered so that it is recorded as unused again. A similar exercise is required in relation to the cartridge status. The data stored in the memory address on the Epson chips that relates to cartridge status represents that the cartridge is either an unused cartridge, a cartridge that has been previously used but is not yet empty, or an empty cartridge. During operation, the software causes the data stored in this memory address to be rewritten to represent that the cartridge status is “used”. This must be changed upon refilling because the printer will not print unless the data stored in the chip indicates that the cartridge is an unused cartridge or that the cartridge has been used previously but is not yet empty.

232    Step (3) resetting is a process in which the data stored on the memory chip in the addresses that relate to the ink level and cartridge status is re-written from one series of binary bits to another to record that the ink volume is “full” and that the cartridge status is either “unused” or “has been used previously but is not yet empty”. In category 1 cartridges the Epson cartridge memory is reset in normal mode. The process of rewriting takes a matter of seconds and involves holding the terminals of the cartridges up to a connector to equipment that causes the rewriting to take place.

233    The software process executed by the equipment used causes the data in the memory address on the original Epson chip to be rewritten from one series of bits to another series of bits. Mr Li gives an example in his affidavit whereby the change from a used original Epson cartridge to a restored Calidad cartridge might involve a change from “01111110” to “00000000” or from “11100000” to “00000001” in order to change the memory relating to ink level. The same type of rewriting takes place with respect to changes in cartridge status.

234    Step (4) - normal mode research and development processes concern the steps required to be undertaken in order to understand the original Epson cartridges, and their interaction with the Epson printers, to enable the modifications identified above to take place. Although the parties dispute the relevance of this stage to the legal dispute, there was no dispute as a matter of fact about what was required.

235    It is first necessary to obtain samples of the Epson printers in which the Calidad cartridges are intended to be used and to obtain used and unused samples of the original Epson cartridges that are to be replaced. They must be inspected in order to understand the path that the ink follows inside the cartridge and to understand where the sensor is located.

236    In order to determine how the memory chips of the Epson cartridges operate it is necessary to understand the electrical circuitry by removing a black seal on top of the memory chip, removing protective layers to reveal circuitry around each transistor, photographing each block of circuitry including each transistor (and there may be hundreds) and conducting electrical testing. It is also necessary to understand the interface protocol by which the printer and chip communicate, which involves use of an oscilloscope, to determine the data stored on the chip by collecting the communication between the printer in the chip and determine what the data means and the memory addresses at which information needs to be read or written. Finally, it is necessary to program electronic equipment to “reset” the data concerning ink level and cartridge status. The evidence indicates that these steps were either undertaken by specialist third parties commissioned by Apex or within the Apex group.

237    I accept that the process of research and development described involves a reasonable degree of expertise upon the party performing it and that it involves an invasive examination of sample original Epson cartridges.

6.3.2    Consideration of the category 1 cartridges

238    In my view, the modifications described above do not have a material bearing on the manner in which the original Epson cartridges described embody the invention as claimed.

239    The step 1 preparatory work concerns no changes made to the printing material container identified in integer [1] of the claim. That integer only requires that it be adapted to be attached to a printing apparatus by being inserted in an insertion direction. That aspect of the container is not materially altered.

240    Step 2 involves minor physical alterations to the cartridge container in the form of the creation of an injection port or a slit in the gas membrane. These changes bear no relationship to the invention the subject of the integers of the claim.

241    Step 3 concerns changes made to the data contained in the memory. These concern ink level and cartridge status. Seiko contends that within the memory chip different “electrical states” (that is, relative presence or absence of electrons at each particular location for an EEPROM) or molecular states (in the case of a FRAM) are interpreted as a 0 or 1. The resetting and reprogramming constitute, Seiko submits, physical modifications to the memory as a “device” and the fact that the physical modifications are microscopic is irrelevant in the context of integrated circuits. It submits that the physical alterations made to the memory are highly substantial in the context of the patented article and essential to its functioning. When an original Epson cartridge is sold by Epson, the memory was in a certain physical state. The purchaser or user modified the memory by using the cartridge to print. Any implied licence to use the patent that article permitted such modifications. The cartridge was then discarded with the memory on a different physical state reflecting the ink level and the cartridge status. The step 3 process involves modifications to the memory by resetting the ink level to full and resetting the cartridge status to unused, in normal mode.

242    I accept that the changes described by Seiko are made within the memory, but the fact that data is stored in the chip by making minuscule physical changes to it is not to the point. The question is whether the original Epson category 1 cartridges have been materially altered insofar as they embody the invention as claimed. In this respect, Seiko’s submissions focus on the physical changes made to the original Epson cartridges per se rather than upon the manner in which those cartridges represent an embodiment of the invention.

243    Properly construed, the claim distinguishes between the memory chip and that which is stored upon it (whether it be data, software or otherwise). The combination of integers claimed is not for a monopoly referable to the content of the chip, but to the existence of the chip as part of the combination. Although I accept that a benefit arising from the claimed combination (reduced likelihood of shorting) is likely to be for the purpose of ensuring that the data content of the chip is not corrupted or destroyed in operation, the content of the chip is not part of the essential features of the claimed invention and does not form part of the monopoly in respect of which Seiko has rights by virtue of the patent. In that context it is not to the point that the resetting process involves microscopic physical changes to the memory chip. Those changes reflect only the information on the chip, which is not the subject of the claim.

244    Accordingly, I do not consider the step 3 alterations to be material.

245    I have already rejected Seiko’s contention that the research and development processes of step (4) are relevant; see [223] – [224] above.

246    This leads to the conclusion that the category 1 Calidad products are cartridges in respect of which the licence implied by a sale without restriction of the original Epson cartridges applies.

6.4    Category 2 – sold after April 2016

6.4.1    Summary of the modifications

247    Category 2 concerns Calidad 253 and 258 cartridges that have been sold by Calidad since April 2016. They were originally Epson 133 and Epson 138 cartridges respectively.

248    Step (1) - preparation and step (2) - refilling for category 2 products are the same as for category 1, with the addition of conducting a pressure test which is required to ensure that the pressure valve of the cartridge is within a prescribed range.

249    Step (3) - reset/reprogram involves rewriting the information on the memory chip when it is in test mode. This involves selection of a “rewrite” function on the equipment used, similar to the process described in relation to category 1. The evidence of Dr Hawkins, upon which he was not cross-examined, is that it would be very difficult to access the test mode and determine what instructions to issue to write the data in the test mode. Whereas in the normal mode, observation of how the printer interacts with the ink cartridges can provide significant insight into the interface protocol and the structure of the information being written and read, there are no analogous communications while it is in test mode, since such communications only take place during ink cartridge manufacturing in Epson’s facilities (Hawkins 56 (b)).

250    Seiko contends that in addition to data relating to ink level and cartridge status, in category 2 cartridges data relating to cartridge model, ink colour and date of manufacture is also reprogrammed. Calidad disputes that these additional steps are taken in relation to this category of cartridges.

251    The disagreement arises from [46], [47] of Mr Li’s affidavit of 22 September 2016. In it he states that in the case of Epson 133 and Epson 138 cartridges, the rewriting equipment used by Apex causes the data that is stored on the cartridge status and ink level addresses to be rewritten. He goes on to say that in the case of Epson cartridges that are equivalent to Epson 133 and 138 cartridges but which have different model numbers, in addition to cartridge status and ink level, the equipment enables data relating to cartridge model and ink colour to be rewritten. In [48]-[56] he then describes the process of rewriting data relating to these additional features. Seiko contends that in those later paragraphs, Mr Li in fact conveys that the additional steps are taken in relation to the original Epson cartridges 133 and 138.

252    It may be that Mr Li could have been clearer in drawing a distinction between the different classes of Epson cartridges in this section, but in reading paragraphs [46]-[56] as a whole it appears clear enough that in [48]-[56] Mr Li is referring to the processes undertaken to cartridges that are equivalent to Epson 133 and 138 cartridges but which have different model numbers. Thus, at [50] when he says “[i]n addition to the data relating to ink level and cartridge status…”, he is referring to the steps taken in relation to this category of cartridge, not the Epson 133 and 138 cartridges.

253    It is true that in cross-examination Mr Li accepted that “reprogramming” is a term that he uses in relation to altering the data concerning date of manufacture, model number and ink colour, and answered affirmatively to the proposition that reprogramming occurs “in relation to Epson 133 and 138 and their equivalents”. But the words “and their equivalents” tended to roll up both the 133 and 138 cartridges together with cartridges that are equivalent, but which have different model numbers. I do not consider that the cross-examination served to establish that [48]-[56] of the affidavit should be read down to have the meaning for which Seiko contends. Accordingly, I find that for category 2 cartridges, the rewriting that takes place concerns cartridge status and ink level.

254    Step (4) - test mode research and development involves a significantly greater degree of research and development than is required compared to normal mode research and development. As with the cartridges that have their memory chips rewritten in normal mode, it is first necessary to obtain samples of the Epson printers in which the Calidad cartridges are intended to be used and to obtain used and unused samples of the original Epson cartridges that are to be replaced. They must be inspected in order to understand the path that the ink follows inside the cartridge and to understand where the sensor is. In order to determine how the memory chips of the Epson cartridges operate it is necessary to understand the electrical circuitry by removing a black seal on top of the memory chip, removing protective layers to reveal circuitry around each transistor, photographing each block of circuitry, including each transistor (and there may be hundreds), and conducting electrical testing. It is also necessary to understand the interface protocol by which the printer and chip communicate, which involves use of an oscilloscope, to determine the data stored on the chip by collecting the communication between the printer in the chip and determine what the data means and the memory addresses at which information needs to be read or written.

255    In addition to these steps, which are in common to the steps taken in normal mode research and development, it is necessary to determine how to activate the test mode and determine the applicable interface protocol for the test mode. This involves the use by Apex of specialist third-party technology. Details as to how to convert the chip into test mode is a business secret which is confidential and apparently not known by Apex. Evidence was not adduced as to precisely what happens. It is apparent that custom programmed electronic equipment is also necessary to reset the ink level and cartridge status and “reprogram” the data concerning date of manufacture, model number and ink colour. The process used is maintained as confidential. I infer that it is costly, time consuming and requires considerable specialised skill and experience.

6.4.2    Consideration of category 2 cartridges

256    In my view, the modifications described above do not have a material bearing on the manner in which the original Epson cartridges described embody the invention as claimed.

257    The work described in relation to steps (1) – (3) is not materially different to the steps summarised in relation to category 1 cartridges, and the conclusions expressed above in relation to those steps applies equally for category 2 cartridges.

258    The primary difference between category 1 Calidad cartridges and category 2 Calidad cartridges lies in the fact that the resetting or reprogramming of the memory chip is undertaken in test mode. As noted above, it is very difficult to access the test mode in order to determine what instructions to issue to write the data in it. A greater degree of research and development is required in order to be able to rewrite the contents of the memory chip. However, the fact that there is an additional degree of difficulty in preparing to make the alterations to the memory is not legally relevant. The analysis for category 2 chips is not different to that of category 1 cartridges.

259    This leads to the conclusion that the category 2 Calidad products are cartridges in respect of which the licence implied by a sale without restriction of the original Epson cartridges applies.

6.5    Category 3 – sold after April 2016

6.5.1    Summary of the modifications

260    Category 3 products are Calidad 253 and Calidad 258 cartridges that were modified from cartridges other than Epson 133 and Epson 138 cartridges (that is, other than category 2 Calidad 253 and 258 cartridges). This is because whilst the Calidad 253 and 258 cartridges are marketed as alternatives respectively to the Epson 133 and 138 cartridges, in certain circumstances, each is produced by modifying different Epson cartridges. For instance, the Calidad 253 cartridge in some instances are made using Epsom 125, 98, or ICLM50 cartridges, or from cartridges acquired from overseas that have a corresponding model number to Epson 133.

261    Step (1) - preparation and step (2) - refilling are the same as steps (1) and (2) in category 1 cartridges.

262    Step (3) - reset/reprogram in test mode for category 3 cartridges involves modifications made in test mode to; model number, ink colour, ink level, cartridge status and date of manufacture. It involves, in one rewriting process, resetting this data. The process takes a few seconds.

263    The reason for additionally changing the data in relation to cartridge status and ink level (which does not take place in category 2 cartridges) is the same as set out in relation to category 1 above. It is necessary to rewrite the data stored in the memory address relating to cartridge model because where an identical cartridge is identified by Epson using a different model number (for instance an overseas equivalent) the model memory address must be rewritten to represent the model number that is used by Epson for that cartridge in the intended region for sale (here, Australia) as otherwise the local printers will not print. Mr Li explains that it is necessary to rewrite the data related to the date of manufacture stored in the memory address because the series of bits that record this information are interpreted by the printer and displayed on the printer screen of the computer as a prompt when the ink in the cartridge has expired or will soon expire. Step (4) - test mode research and development process for category 3 cartridges is the same as for category 2, as set out above.

6.5.2    Consideration of category 3 cartridges

264    Steps 1 and 2 for the category 3 cartridges are the same as for category 1. However, step 3 is more complicated, and involves additional modifications to the contents of the memory chip. However, for the reasons given in relation to category 2 cartridges, the analysis and outcome is the same.

265    This leads to the conclusion that the category 3 Calidad products are cartridges in respect of which the licence implied by a sale without restriction of the original Epson cartridges applies.

6.6    Category A – sold after April 2016

6.6.1    Summary of the modifications

266    Category A is said to represent category 2 or 3 cartridges that do not have the gas membrane cut during the modification process. It is not in dispute that this represents 95 per cent of cartridges. The Calidad parties submit that this separate category is necessary and appropriate in the analysis. Seiko contends that it is a minor difference that does not warrant such treatment.

6.6.2    Consideration of category A cartridges

267    It follows from the analysis and conclusions expressed in relation to categories 2 and 3 that the category A cartridges have also not been the subject of material change such that the conclusion expressed in relation to those categories would be any different.

268    This leads to the conclusion that the category A Calidad products are cartridges in respect of which the licence implied by a sale without restriction of the original Epson cartridges applies.

6.7    Category 4 – sold prior to April 2016

6.7.1    Summary of the modifications

269    Category 4 cartridges are Calidad 260H cartridges that have not been sold since April 2016 and that were modified from cartridges other than Epson T200XL such as cartridges acquired from outside Australia that have a corresponding number to the Epson T200XL and the Epson T200 (category 1 includes Calidad 260H cartridges that have been marketed since April 2016).

270    Step (1) - preparation and step (2) refilling processes for category 4 cartridges are the same as for category 2 cartridges.

271    Step (3) - chip replacement process is quite different to the “reset/reprogram” stage (step (3)) for the category 2 cartridges. Mr Shen gives evidence that instead of reprogramming or resetting the integrated circuit chip in these cartridges, Ninestar employees remove the integrated circuit boards from the printer using a cutting tool. The circuit boards are then sent to AMC which arranges for the manufacture of a compatible chip that contains: bits in the ink level memory address representing that the ink volume capacity was full; bits in the cartridge status memory address which represent that the cartridge was “unused”; bits in the date of manufacture memory address; bits in the model memory address representing the relevant cartridge model and ink colour suitable for its proposed use. AMC arranges for a third party to place the new memory chip on integrated circuit boards. When a printed circuit board is replaced, it is not on the same cartridge from which it was removed, but rather on any original Epson category 4 cartridge available.

272    Analysis of the evidence in relation to this category of cartridges indicates that the steps taken were to; remove the integrated circuit board from the cartridge using a cutting tool; manufacture generic memory chips using a third-party commissioned by AMC; test the chips at wafer level; program the generic memory chips; dice the wafers into individual die (each die being one chip) and good dies (those chips which have passed testing) picked and placed into waffle packs; replace original chips with generic memory chips on the integrated circuit board; and reattach the integrated circuit board to the cartridge.

273    Step (4) - compatible chip research and development processes for category 4 may be summarised as follows. It is first necessary to obtain samples of the Epson printers in which the Calidad cartridges are intended to be used and to obtain used and unused samples of the original Epson cartridges that are to be replaced. They must be inspected in order to understand the path that the ink follows inside the cartridge and to understand where the sensor is located. In order to determine how the memory chips of the Epson cartridges operate it is necessary to understand the electrical circuitry by removing a black seal on top of the memory chip, removing protective layers to reveal circuitry around each transistor, photographing each block of circuitry, including each transistor (and there may be hundreds), conducting electrical testing and constructing and simulating the circuitry in Electronic Design Automation software and testing its performance. It is also necessary to understand the interface protocol by which the printer and chip communicate, which involves use of an oscilloscope, to determine the data stored on the chip by collecting the communication between the printer in the chip and determine what the data means and the memory addresses at which information needs to be read or written.