Agilent Technologies Deutschland GmbH v Waters Ltd

This is in appeal from a decision of Pumfrey J of 21^st December 2004 ([2004] EWHC 2992 (Ch)). The patentees, Agilent, sued Waters Limited for infringement of EP (UK) 0 309 596. In prior proceedings this court held that an earlier version (the “automatic device”) of the Waters device infringed the patent and that the patent was valid ([2002] EWCA Civ 612). Waters modified their device to that now accused of infringement (“the manual device”). Pumfrey J held that it did not infringe because it was not within the scope of the patent.

Agilent challenged that decision before us. In addition to resisting the appeal by supporting the Judge’s decision, Waters raised other defences in the nature of an estoppel. Mr Burkill QC made it plain that he only relied upon these if he did not succeed in supporting the Judge’s principal conclusion about the scope of the patent.

We accordingly decided to hear argument about that first. After we had done so we formed the view that the Judge was right and so informed the parties. Accordingly we did not hear full argument on the other defences, though in part both parties had touched upon them. It is not necessary to say any more about them now.

It is common ground that this case involves no question of general principle. The sole question is whether claim 1 of the patent on its proper reading covers the manual device, a point essentially of interest to the parties alone. So there is no point in re-setting out for general readership the clear description of HPLC given by the Judge in his paragraph 2 or of the patent in suit in its paragraphs 4-16. Likewise his description of the first, infringing device, in paragraphs 16-19 was uncontroversial and need not be repeated here.

I can now go to the alleged infringement, again gratefully borrowing the Judge’s description, omitting a few unnecessary words criticised by Mr Wyand QC for Agilent:

“20.

The pump and control system with which I am concerned in this action operates in what was referred to as “manual mode”. In manual mode, flow rate and stroke length are, in principle, independently selected by the operator. The machine is described in detail by Professor Smith in his report and in the Product and Process description served in the current action. The key feature is that the pump has four preselectable pump volumes. The flow rate is independently specified. If the machine cannot deliver the specified flow rate at the volume specified, then the operator is informed—but there is no automatic change of range as there was in the semi-automatic mode of the machine the subject to the first action. The result is that the machine no longer maintains any predetermined relationship between stroke length and flow rate other than that necessarily imposed by the maximum and minimum possible frequencies and the maximum and minimum possible stroke lengths. Any flow rate that is possible for a given stroke volume is permitted by the machine. There is no suggestion, as I understand it, that the machine does anything to suggest to the operator a change in stroke volume so as to use an optimum volume for a given flow rate. There is, however, a series of instructions for the use of the machine …. which do certainly contemplate that the user should take care to select manually the displacement which would have been fixed by the automatic machine for the specified flow rate.”

The key part of the claim is its characterising portion, broken down for convenience by the judge as follows:

“Control Means

a)

coupled to the drive means

b)

for adjusting the stroke length of the pistons between their top dead centre and their bottom dead centre respectively

c)

in response to the desired flow rate of the liquid delivered at the outlet of the pumping apparatus,

d)

with the stroke volume (i.e., the amount of liquid displaced during a pump cycle) being decreased when the flow rate is decreased and vice versa,

e)

such that pulsations in the flow of the liquid delivered to the output of the pumping apparatus are reduced.”

Mr Wyand QC submitted that these words covered the manual device as a matter of plain language. He pointed out the claim is to an apparatus and is not to a method. So when it says “for” it means “suitable for” i.e. “capable of.” The pre-characterising portion of the claim requires that the apparatus has a “selectable flow rate.” That must mean (as was agreed) a flow rate selectable by the user.

As far as (a) is concerned there must be a control means which is coupled to the drive means. Mr Wyand says it does not matter how – whether or not you are working to a predetermined relationship. I do not think Mr Burkill disagreed at this point.

Next (b). He submitted that this merely requires that the control means must be “capable of adjusting stroke length”. There is nothing here which requires automaticity. You simply ask, can the control means alter length?

As to (c) he submitted this is not confined to an automatic response to the selected (which means the same as “desired”) flow rate. It is enough if the user selects the flow rate and then selects what stroke length is needed to achieve that flow rate. Since that is what happens with the alleged infringement it falls within the claim.

Mr Wyand further submitted that “in response to” distinguishes some prior art – for instance where you controlled the flow rate simply by controlling pump stroke. Here the claim covered the case where you chose the flow rate and the control means allows adjustment of the stroke length in response to that flow rate, whether that was done manually or by an internal look-up table (automatically).

Patent claims are to be construed as read by the notional skilled man in context. So Mr Wyand went to the body of the specification. The passage on which he relied (col. 10_2-27) reads:

‘After this start up procedure, the pump starts with its normal operation. The inlet valve 4 is opened by the solenoid 5 under the control of the inlet controller 46 and the piston 10 moves down from the top dead centre thereby sucking solvent into the first pump chamber. According to the present mode of operation, the stroke length, i.e. the distance the piston travels between the top dead centre and its bottom dead centre is dependent on the flow rate which a user has selected at the user interface 42. From the information about the desired flow rate passed from the user interface 42 to the system controller 41, the system controller 41 computes the corresponding stroke length using a predetermined mathematical relationship between flow rate and stroke length (or stroke volume which is proportional to the stroke length). An example of such a predetermined relationship between flow rate and stroke volume is explained below with reference to figure 4. Although in the present mode of operation there is a predetermined relationship between the flow rate and the stroke volume the control circuitry of the present invention yet permits to put the coupling between flow rate and stroke volume out of action so that a free selection of the stroke length or volume becomes possible.[my italics].

Up to the last sentence the description is of automatic operation: the user selects the desired flow rate, the controller (which is the “control means” of the claim) “computes” (i.e. uses an internal look-up table) the stroke length. Although nothing is said explicitly here about frequency, it is clear from the description in fig. 4 which is related to fig.5 that this is also under control – the look-up table sets both. The Judge explains this at para. 14.

Mr Wyand concentrates on the italicised bit of the last sentence. He submits that it is saying that a pump where there is no coupling between the (chosen) rate and length (which corresponds to volume) is within the contemplation of the patentee as being within his invention and so must be within the claim. You can have free selection (by the user) of both stroke rate and length. He submits that it this must be so, for the patentee is hardly likely to have gone out of his way to point out a way round his claims.

I am unable to accept these submissions. Firstly, and principally, I cannot accept the reasoning I have set out in paras. 10 and 11 above. It is the “control means” which is to adjust the stroke length “in response to the desired flow rate” not the operator. The operator is not part of the control means of the claim. I see no escape from this as a matter of language.

Nor am I impressed by the argument based on the passage in col.10. For what it says is that the circuitry of the invention enables “decoupling” – that is not saying the invention includes that which cannot be coupled. Putting it another way, the passage does not envisage a device in which the operator must independently enter stroke length and flow rate. To adopt Mr Burkill’s analogy, an automatic gearbox which can have a manual mode is not the same as a gearbox with only a manual mode. Actually the passage is against Mr Wyand rather then for him, for there is nothing in the circuitry of the alleged infringement which enables decoupling.

The Judge expressed his view thus:

“24.

In such an arrangement [i.e. that described in the last sentence of the col.10 passage] it seems to me clear that the machine is required to fix the pumping frequency having regard to the flow rate and stroke volume specified by the user. Accordingly, the response called for by feature (c) of the claim is not present. Such a mode of operation, if the only mode of operation of the machine would not infringe the claim. I read the words ‘free selection’ as being deliberately chosen in antithesis to the adjustment called for by the claim.”

Mr Wyand criticises that, saying that it ignores the fact that it is the control circuitry of the present invention which is being talked about, not that of the specific embodiment. I cannot see that the Judge was making any error. Whether it is one or the other does not matter – what is being contemplated is only a device which can be decoupled, not one which cannot be coupled.

I would add this, however. If the patentee was really describing and claiming a system in which the user chose flow rate and stroke length or volume, leaving the device to set the frequency, I cannot imagine that he would have done what has been done here. Of course it is always easy to say after the event that different words might have been used – but the difference here is too great to be explained by a mere choice of words – a different conception is involved.

I would dismiss this appeal.

I agree.

I also agree.