ICON DATACOM LTD v. SOLARBRITE ELECTRONICS LTD

HCA985/2005

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 985 OF 2005

---------------------

------------------------

J U D G M E N T

-------------------------

1.  The plaintiff and the defendant are both companies in the electronics industry. One of the plaintiff’s products was a NetPos-2000. This was a ‘point of sale’ machine commonly seen in banks and retail outlets into which a credit card was put and information then appeared on an illuminated window on the machine.

2.  This case concerns the merchantability of a small component inside the machine which was supplied to the plaintiff by the defendant.  The component in question was an “EL panel”, short for electronic luminescent panel.  An EL panel is a rectangle of plastic through which an electronic current is passed which causes it to light up and thereby illuminate the window on the POS machine.  The cost of each EL panel was under HK$4.

Background to the contract

3.  The parties first met in 2002.  

4.  At that time the representative of the defendant who had discussions with the plaintiff’s representative was a Mr Kenneth Kong.  He spoke and had meetings with Mr Augustine Fung, the general manager of the plaintiff, Ms Yip Wan Ching, a marketing executive of the plaintiff and Mr Tam Pui Tat, the engineering manager of the plaintiff.  All three of these plaintiff’s witnesses made statements and gave evidence at the trial.

5.  For the defence however Mr Kenneth Kong neither made a statement nor gave evidence.  Neither did his successor Mr Jeff Lee (who took over in 2004) and neither did Mr Thomas Tse nor Mr Victor Chan who were referred to as the bosses of Mr Kong and Mr Lee.  The defendant’s main witness was Mr Kwong Chi Wai, the marketing director who was senior to all the above named but who had had no personal dealings with the contract at the material time.

6.  The content of the discussions which took place in 2002 are of importance.  I shall refer to them in more detail later.  I merely note at this stage that only the plaintiff’s witnesses can give first hand evidence of what transpired and what was said at that early stage of contractual relations.

Events leading up to the April 2004 order

7.  After the initial discussions the defendant, in October 2002, supplied two sets of sample ELs to the plaintiff.  One set was of 3 panels, the second was of 120 panels.  The samples were intended to comply with a diagram prepared by Mr Kong which also referred to a model number EL108CF.  The diagram itself merely showed the dimensions of the panel.  Further requirements had been discussed, according to the plaintiff, at the meetings.  Mr Kong had also been given an LCD module together with a plastic mounting to work to.

8.  Both the set of 3 samples and the set of 120 samples were fully tested over a two-month period.  The plaintiff was satisfied with them and, in October 2003, an order for 1,200 panels was placed and in January 2004 a further 2,000 panels.  All these were the same EL108CF and all worked properly.

9.  In March 2004, the plaintiff placed another order for 5,100 panels.  The POS terminals into which these panels were to be put were intended for a Taiwanese customer.  The order had been received from the plaintiff’s Taiwan agent “Chia Hsin”.  The ultimate customer in Taiwan was the Tai Shin Bank.

10.  It is this batch of 5,100 which is the subject matter of this claim.  They were sent to the plaintiff in two batches of 1,100 on 6 April 2004 and 4,000 on 27 April 2004.  They were then sent to the plaintiff’s manufacturing sub-contractor, Surface Mount Technology Ltd (“SMT”) for assembly into the finished POS terminals.  After the first twenty panels had been installed SMT noticed some problems in five of them.  Those five were returned to Mr Kong who sent them off for testing.  After receiving the ‘defect report’ from the manufacturer he acknowledged them to be problematic.  However as a result, it seems, of certain assurances made by Mr Kong the production of terminals continued.  The manufacturer had stated that the “mistake” would be avoided in future.

11.  The entire Taiwanese consignment, 3,100 terminals, was factory tested and all were approved.  They left the factory, therefore, in good working order having been subjected to a 8-hour “burn in” test.  The evidence further stated that Chia Hsin, in Taiwan, conducted further tests, as a matter of routine, before delivery to the customer.  This test was a 65‑hour test.  Again no faults were detected and accordingly the terminals were distributed to the various branches of the Tai Shin bank in Taiwan.

The problems with the 5,100 order

12.  Within weeks, in May 2004 and increasingly in June 2004, problems with terminals were reported back.  There were two types of problem; firstly, “snowflaking” which was the appearance of black vertical lines at random places on the screen which distorted the words and figures thereon and secondly, a failure of the panel to light up so that the screen was either dim or had no backlight at all.  In round terms, the failure rate of the panels in the whole batch exceeded 60% (later tests indicated).

13.  Subsequent events can be stated shortly.  The bank took a quick and unilateral decision to reject the entire consignment.  They were returned via the agent, Chia Hsin, without much further discussion or negotiation.  Moreover the agent stipulated that if they were repaired or reworked they could not be marketed through them into Taiwan.

14.  The plaintiff gathered them in and arranged for new EL panels, supplied by Solarbrite, to be inserted.  The reworked terminals proved to work properly thereafter and they were resold to a customer in the PRC for a much reduced overall price.

15.  Thus, all the facts, in a nutshell, produce a somewhat puzzling outcome.  The early consignments, in 2003, the subject consignment in April 2004 and the replacement consignment in July 2004 all passed both the factory testing and the agent’s testing and were deemed to be of merchantable quality and suitable to be distributed into the market place.  However, within weeks serious problems materialized with the middle consignment whereas no such problems ever arose with the other two consignments.  They were all made in the same factory.

Legal basis of plaintiff’s case

16.  Following discussions at trial, in spite of the absence of any reference to the Sale of Goods Ordinance, Cap. 26 in the pleadings, I accept the defence submission that the plaintiff’s case is one of breach of implied terms as to merchantability and fitness for purpose as provided by section 16 of Cap. 26.  No evidential basis has been established for a case based on a contract for sale by sample which would invoke section 17 of Cap. 26.

17.  Section 16(2) and (3) provides as follows :

“(2) Where the seller sells goods in the course of a business, there is an implied condition that the goods supplied under the contract are of merchantable quality, except that there is no such condition—

(a) as regards defects specifically drawn to the buyer's attention before the contract is made; or

(b) if the buyer examines the goods before the contract is made, as regards defects which that examination ought to reveal; or

(c) if the contract is a contract for sale by sample, as regards defects which would have been apparent on a reasonable examination of the sample.

(3) Where the seller sells goods in the course of a business and the buyer, expressly or by implication, makes known to the seller any particular purpose for which the goods are being bought, there is an implied condition that the goods supplied under the contract are reasonably fit for that purpose, whether or not that is a purpose for which such goods are commonly supplied, except where the circumstances show that the buyer does not rely, or that it is unreasonable for him to rely, on the seller's skill or judgment.”

18.  “Merchantability” is defined by section 2(5) as follows :

“(5) Goods of any kind are of merchantable quality within the meaning of this Ordinance if they are—

(a) as fit for the purpose or purposes for which goods of that kind are commonly bought;

(b) of such standard of appearance and finish;

(c) as free from defects (including minor defects);

(d) as safe; and

(e) as durable,

as it is reasonable to expect having regard to any description applied to them, the price (if relevant) and all the other relevant circumstances; and any reference in this Ordinance to unmerchantable goods shall be construed accordingly.”

19.  The key issue is therefore the EL panels’ fitness for the purpose for which they were intended.  It is therefore now important to consider the evidence relating to :

(a) the original discussions between the parties to establish what the defendant agreed, either expressly or by implication, to provide and their knowledge concerning the purpose for which the EL panels were intended and

(b) an analysis of what they actually provided.

Question (b) will involve a consideration of the expert evidence.

Early discussions

20.  In their written statements Mr Augustine Fung and Mr Tam Pui Tat stated as follows on this issue :

Mr Fung :

“Between September and October 2002, Mr. Kenneth Kong, the Senior Sales Engineer of Solarbrite Electronics Limited had meetings with me and my staff, Ms. Eva Yip and Mr. Tam Pui Tat on 2 to 3 occaisons at the office of Icon to produce Solarbrite’s Electroluminiscent Lamp Backlight Panel for Icon’s business. …

During the sessions of meetings, I explained Icon’s business framework, market profile, information of POS terminal application, end-user needs and new products to Mr. Kong. My colleague, Mr. Tam, and I made a request to Mr. Kong that we wanted the most common colour of EL Panel namely bluish green. Miss Yip discussed with Mr. Kong the proposed pricing and order placing arrangement of the Solarbrite’s EL Panel by Icon. Mr. Tam explained the technical requirements of the EL Panel which was necessary to fit Icon’s POS Terminal identified NetPos-2000 model.

On 2 October 2002, I approved the mechanical and/or dimensional specifications of Solarbrite’s EL Panel which Mr. Kong had provided me according to the requirements that Mr. Tam and I explained to Mr. Kong at the meetings.  I made it clear to Mr. Kong that Solarbrite’s EL Panel had to meet Icon’s requirement because the NetPos-2000 was designed specifically to meet the needs of Icon’s end-users.  Shortly afterwards, Mr. Kong delivered 3 Solarbrite’s EL Panels made according to the approved mechanical and/or dimensional specifications.  Mr. Tam conducted preliminary tests on the 1^st sample.”

Mr Tam :

“During the meetings, Mr. Fung and I explained to Mr. Kong that Icon’s business plan was to use the EL Panel in the NetPos-2000 terminal (‘NetPos‑2000’). The NetPos-2000 is ICON’s 3^rd generation product that is an enhanced version based on the previous 10 years of R&D achievement. The NetPos-2000 uses the Sipex’s EL driver IC SP4422A to drive the EL Panel. The NetPos-2000 is designed to meet the high standard of industry requirement for credit card and debit card transactions. Over 60,000 NetPos-2000 are now installed in retail shops, restaurants, gasoline stations and shopping malls throughout Taiwan. There is now produced and shown to me marked Exhibit “TPR-1”, a copy of the specification sheet of SP4422A.

I informed Mr. Kong of the mechanical dimensions, colour, mounting method, drive characteristics and the purpose of Solarbrite’s EL Panel in the NetPos-2000.  I provided a LCD module together with a plastic mounting frame to Mr. Kong so that he knew what technical requirements Icon was expecting.

On 2 October 2002, Mr. Kong sent by facsimile a copy of the mechanical and/or dimensional drawing of the EL Panel based on Icon’s requirements.  We relied on Mr. Kong’s knowledge and advice of the quality of the EL Panel.”

21.  In his oral testimony Mr Fung further stated that he had “made it clear” to Mr Kong that the EL panels were to be fitted into the NetPos system for use in banks and should have a 5-year life span.  He also remembered telling Mr Kong that the ‘SIPEX’ driver would be used in the system.  He specifically remembered Mr Kong mentioning an alternative driver but Mr Fung rejected the alternative in favour of the SIPEX with which the EL panels would have to be compatible.

22.  Mr Tam’s oral testimony was that in addition to the drawing of the panel later sent by Mr Kong and approved by the plaintiff he told Mr Kong at the early meetings the size, colour, mounting method, the use of the SP4422 SIPEX driver, that the end user was a bank and that a 5-year life span was expected.

23.  As already mentioned there was no evidence from the defence on this issue.

24.  Mr Abraham Chan, counsel for Solarbrite submitted that little weight should be attached to the oral testimony because it did not appear in the same detail in the written statement, there was no documentation to support much of it and the conversations took place eight years earlier.

25.  I am satisfied however that the plaintiff’s evidence does reflect the truth of what took place.  The witnesses more detailed oral evidence is, I am satisfied, an elaboration of remarks such as “I informed Mr Kong of the … drive characteristics and the purpose of …” and “Mr Tam explained the technical requirements of the EL panel which was necessary to fit …” which are very likely to have been said between these two parties to this contract.  The recollection that Mr Kong mentioned an alternative driver supports this finding.  Mr Kong was more than a mere salesman.

26.  Thus Solarbrite had specific and technical knowledge of the purpose for which the EL panel was to be fit.

What were the characteristics of the panels from the batch of 5,100 which caused the Taiwan bank to return 3,100 terminals?

27.  There is no doubt that a significant proportion of the terminals developed problems after a short time which caused the end-user, the bank, to return 3,100 of them.

28.  The plaintiff must satisfy the court that the problems were, on the balance of probabilities, attributable to the EL panels.

29.  Other possible explanations have been advanced by the defendant.  Explanations such as exposure to moisture, negligent handling, non-compatibility with other components or other environmental reasons.

30.  On the technical side it has also advanced the argument that some variations in the manufacture of EL panels is inevitable and the production of “electronic noise” is also inevitable. Accordingly, the plaintiff’s design of its terminal should have been more robust so as to accommodate such variations and should have included increased “shielding” which would have minimized the problems associated with the inevitable “electronic noise”.

31.  Before addressing these issues however it is necessary to review the expert evidence to try and answer the question ‑ what was wrong, if anything, with the EL panels in the 5,100 batch?

Expert evidence

32.  The court was assisted by two expert witnesses. Dr Chi Kwong Li, associate professor of the Hong Kong Polytechnic University, for the plaintiff and Professor Kwok Hoi Sing of the Hong Kong University of Science and Technology for the defendant.  Both had impressive C.V.s, both endeavoured to assist the court and in many respects agreed with each other.

33.  Dr Li had carried out extensive tests on two sets of samples.  Six panels were randomly selected from the batch said to be defective and six from the batch which was problem free.  They were subjected to various tests to see what differences there were in the two sets of samples.  The testing commenced in August 2008 and it was finally concluded in December 2008.  Dr Li found the two sets of six samples each to have markedly different physical properties.  His conclusions were that (1) the “defective” EL backlight caused the LCD display to be corrupted, (2) the degrading and failure of the EL backlight causes glitches on the signal lines of the LCD interface, and (3) the “defective” batch aged more rapidly and had a reduced life when subjected to an accelerated life test. 

34.  Dr Li described the differences in the data obtained from the tests, in some cases, as “huge”.  He opined that the defective batch was not suitable to be put in the market place.

35.  Professor Kwok was aware of the test results and was shown a demonstration of the performance of both the ‘good’ panels and the ‘bad’ panels.  He stated that the difference in performance was due to the variations in the electrical characteristics between the two batches.  He stated that the system designer should take variations in performance into account and produce a “robust” design.  The question for the court is whether the variations in performance are too great to be accounted for and thus the ‘bad’ panels were outside the performance parameters of even a more robust design.

36.  Professor Kwok added that the EL panels were a “passive” component and provided they emitted light, the suppliers’ job was done.  It was agreed between the experts that both batches were capable of emitting light. The difference was that the “bad” batch required a lot more electricity to maintain the light and as a result the durability and reliability were diminished. 

37.  The expert evidence was illuminating.  Ultimately the difference between them came down to their respective opinions about whether or not the ‘bad’ batch could have been used commercially or not.  This, of course, is a question for the court.

38.  I was satisfied, firstly, that the panels as tested in 2008 were in the same or similar condition as they had been in 2004.  The “good” batch had been good in 2004 and remained good when tested in 2008.  Their luminescence and durability were both commercial.  By the same token the ‘bad’ panels did not perform adequately in 2008 and I am satisfied the results were a reasonably reliable indication of how they had performed in 2004.

39.  I am also satisfied that their failings, or their poor performance, were not matters that the plaintiff’s design should have catered for.  All other batches both before and after the ‘bad’ batch performed well but the bad batch became dim or expired and caused ‘snowflaking’ very early on in their lifespan. I reject the defence submission that a terminal should have been designed that would accommodate such wide differences as were detected in the ‘bad’ batch.

40.  Applying these findings to the overall facts I am satisfied that the defendant supplied the plaintiff with panels that breached section 16 of Cap. 26 in that they were not reasonably fit for the purpose which was known to the defendant.

41.  Given that all other batches worked well, given that the returned terminals worked well after the EL panels were replaced and given that samples for the ‘bad’ batch, after testing in 2008 were shown to have significantly different properties from the sample from the ‘good’ batch, I am satisfied that, on the whole of the evidence, it is more likely than not that :

(a) the different properties in the ‘bad’ batch were the cause of the problems;

(b) the defendant had produced thousands of panels which were compatible with the terminal as designed by the plaintiff but the ‘bad’ batch’s differences made them incompatible; and

(c) the defendant knew the purpose for which they were intended and the plaintiff relied on the defendant to produce a product that would perform accordingly in the market place.

42.  On the issue of causation, Mr Chan submitted that the “most likely” cause of the NetPos display problems was the inadequate design of the LCD interface.  Given the above findings the plaintiff’s design of the entire terminal should not be criticized.  Mr Chan continues that the court should not exclude “other possibilities” such as poor storage of this particular batch, or poor handling by the end users or quality problems with other NetPos components.  On the evidence these are speculative and cannot, in my judgment, disturb the key factors that the significantly different properties in the bad batch in 2004 are very likely to have been the real cause.

Damages

43.  The issue of quantum of damages is also by no means straightforward.  Bearing in mind the claim for $2.84 million arises from goods originally costing less than $20,000 it is easy to understand why the defendant’s initial reaction to the claim was that it was “out of all proportion”.  Particularly when they had replaced all the returned panels.  When considering the question of what damages flowed from the breach which were reasonably foreseeable some factual issues should be revisited :

(1) All the panels referred to in this case passed all the factory testing and the agent’s quality control procedures in Taiwan.  Given that a ‘bad’ batch found its way into the market place that quality control is shown to be wanting.  That is not the defendant’s fault.  It seems that the tests done at the factory and by the Taiwanese agent could not have been adequate tests for gauging future durability and reliability, only present efficiency.

(2) The bank’s decision to return all 3,100 terminals was both quick and unilateral. It seems little attempt was made to effect repairs on site.  Moreover, Chia Hsin’s contribution seems merely to have taken the bank’s side and rejected the goods whilst adding a stipulation that they would not re-market them anywhere in Taiwan.  The defendant had no control over these decisions.

(3) Only a proportion of the batch of 5,100 panels exhibited problems.  The entire consignment of 3,100 terminals were returned but they were not all bad.  In round terms something over 50% of the panels were problematic.

(4) Mitigation of loss

The plaintiff was right to mitigate their loss by repairing (quite easily) the returned terminals and reselling them.  However, it is arguable that they undersold them.  Of the 3,100 returned terminals the plaintiff sold on 3,000 to one customer in China but only 1,000 of them were sold as Pos terminals (at a reasonable price of HK$1,450 each).  The other 2,000 were sold as PIN pads at less than HK200 each.  The only reason given for this very cheap disposal was that the plaintiff wanted to sell them all at once to one customer.

The evidence revealed that Pos terminals remained in the market place for three to four years after this incident.  A different backlight using an LED improved them.  The extra cost of installing an LED display rather than an EL panel was minimal.  In short, reasonable mitigation could have been 3,000 x $1,450 = $4.35 million which would have eliminated the entire loss claimed for the loss of the purchase order with the Tai Shin Bank.

(5) Methodology of calculating the plaintiff loss of business

Mr Chan submits that in any event, regardless of the arithmetic involved in calculating the mitigation of loss, (discussed above) the plaintiff has adopted the wrong method of assessing the loss.  It should be a loss of profit calculation. Clearly, I do not need to address this submission if I find that, in any event, a proper mitigation of loss would have reduced the loss of sales income to nil.

Approach to quantum

44.  The plaintiff claims under six headings :

(a) Rework costs by Chia Hsin	$226,831
(b) Transportation of returned NetPos	14,407
(c) Consigned material costs for reworking	112,960
(d) SMT’s charges for replacement of EL panels	243,517
(e) Loss of Tai Shin Bank P.O.	4,001,400
(Less mitigation income)	1,825,000
	2,176,400
(f) Dead stock	69,030
Total	$2,843,147

Item (e)

45.  In my judgment selling 2000 NetPos system for under $200 as PIN pads (i.e. simple PIN readers) when they were worth about $1,500 each was not responsible.  I disallow item (e) because a reasonable effort should have and could have been made to achieve 100% mitigation.  It is noted that the 1,000 that were sold to the PRC were sold at a higher price than some that had gone to Taiwan.  Moreover I agree with the criticism of the methodology employed in that no attempt has been made to quantify the net loss of profit, if any.

Items (b), (c) and (d)

46.  I consider these to be reasonably foreseeable consequences of supplying defective EL panels.

Items (a) and (f)

47.  The invoices supporting item (a) are described as ‘Maintenance Service Charges’ and ‘Technical Support Charges’.  In the absence if any other evidence explaining why it is a reasonably foreseeable loss it should not be allowed. 

48.  Item (f) relates to the purchase of surplus (different) modules which could not be sent to Taiwan after the main order had been cancelled.  They were re-sold the following year at a reduced price.  The purchase of these different modules did no more than provide the plaintiff with good stock which was never ‘dead’ as claimed but merely sleeping.  In the context of this action I consider it to be too remote to be met by the defendant.  Moreover, no explanation was given as to why it was resold at a reduced price.

49.  Items (b), (c) and (d) total $370,885.

50.  In rejecting items (a), (e) and (f) I have taken into account the general matters outlined above.

Conclusion

51.  I give judgment to the plaintiff in the sum of $370,885 together with interest from 27 May 2005 to be agreed.

52.  As to costs, I direct the parties to file brief written submissions informing the court of all matters which may be relevant to the issue of costs, on or before 22 February 2011.  Thereafter I shall hand down a final costs order.