T 1500/20 23-05-2024

Main request, inventive step, Article 56 EPC

1. For the reasons set out below, the Board came to the conclusion that the subject-matter of claim 1 of the main request was obvious to the person skilled in the art starting from the disclosure of document D1.

2. Like the present application, document D1 concerns the operation of a square wave power supply (inverter) that provides power to critical loads such as PFC loads when there is a fault in the commercial power supply (line power). Like the present application, document D1 identifies that when there is a fault in the commercial supply, and the system transitions to inverter operation, PFC loads may experience a problematic drop in voltage (see D1, column 11, lines 15 to 26). To prevent this, as set out in D1, column 11, lines 26 to 40, the system controller 50 begins operation of the inverter 56, following transition from line power, using an inverter pulse width which is "at or near the maximum inverter pulse width" and "wider than that normally required to provide the desired steady state output voltage level under normal operating conditions". The inverter pulse width is then gradually reduced until the RMS output voltage level of the inverter is at the desired RMS voltage level, e.g. 120 or 230 volts. As regards the detailed operation of the inverter, document D1 explains in column 10, lines 35 to 37, column 13, lines 33 and 34 and column 14, lines 22 to 24 that there must be a dead space between the switching of the inverter switching devices 70, 71. Such a "dead space" is a well-known requirement for any inverter to avoid both of the switches being on at the same time and thus short-circuiting the source. It is more usually referred to as "dead-time".

3. Document D1 does not disclose to lock the phase of the square wave power supply to that of the commercial power supply, however the Board concurs with the examining division that locking the output of a UPS (uninterruptible power supply) to the frequency and phase of the commercial power supply was part of the common general knowledge of the person skilled in the art. Indeed, the appellant acknowledged this to be the case both in the grounds for appeal (page 9, first paragraph) and in the oral proceedings before the board.

4. Furthermore, the Board concurs with the finding in section 1.6 of the contested decision that it would have been obvious for the skilled person, starting from the disclosure of document D1, to consider locking the phase of the square wave power supply to that of the commercial power source in order to achieve the known benefits of doing so. This could be to avoid disruptions and glitches on transition from the commercial power supply, see document D2, column 2, lines 17 to 21.

5. Furthermore, the Board considers that it would have been self-evident to the skilled person that the commercial power supply may fail at any phase angle of its AC cycle. Hence, when attempting to implement the system of document D1 with the square wave inverter output locked to the phase of the commercial power supply, the skilled person would have had to make a decision whether the inverter should begin its operation as soon as possible after the failure, possibly emitting only a partial pulse in the first half-cycle before the "dead space" between switching is reached, or whether the inverter operation should be delayed until the next half-cycle in which it is able to emit a pulse "at or near the maximum inverter pulse width" before the "dead space" is reached. Document D1 discloses in column 2, lines 16 to 19 that some loads may fail when power is interrupted for as little as 2 ms, and discloses in lines 34 to 39 that a break in the input voltage signal provided to a PFC load can cause instabilities and oscillations in the PFC power supply. The Board concurs with the the examining division that these indications would lead the person skilled in the art implementing the system of D1 with an output locked to the phase of the commercial power supply towards emitting a partial pulse in the first half-cycle before the "dead space" between switching is reached, rather than waiting to the end of the first half cycle in order to be able to emit a pulse at or near the maximum inverter pulse width. Hence, the Board concurs with the examining division that the reasons set out in paragraphs 1.6, 1.7 and 1.9.2 of the contested decision would lead the skilled person in an obvious way to the first of these options and hence to the subject-matter of claim 1 of the main request.

6. The appellant submitted that the "dead zone" according to the present application was a parameter that was only used in the first half-cycle after a transient condition and was different to the "dead space" of D1. The Board did not find this convincing. It is clear from the explanations on page 14, lines 4 to 6 of the application as filed that the square wave must be "closed" (i.e. finished) before the dead zone time t54 to t55. In the Board's view, the skilled person would understand this to be the case every time the polarity of the inverter changes, not just in the first half cycle.

7. The appellant also submitted that in addition to improving synchronization of the square wave output voltage to the commercial power source (as discussed above), the claimed invention also led to a reduction in the complexity of the corresponding circuit. However the appellant did not provide any support for this assertion and no reduction in complexity is evident to the Board.

8. Also, the appellant argued that it was important in D1 that the pulse width be immediately increased to a pulse width at or near a maximum pulse width and that this taught away from the use of a partial pulse in the first half cycle. The Board did not find this convincing. Whilst D1 indeed discloses in general to use the maximum pulse width after failure of the commercial power supply, the Board considers that when attempting to synchronise the square wave power supply to the commercial power supply upon failure, the skilled person would immediately realise that it would not always be possible to use that maximum pulse width in the first half cycle after the transition.

9. Hence, the main request is not allowable for lack of inventive step in the sense of Article 56 EPC.

10. In view of this finding, the other issues concerning the main request which were identified in the communication pursuant to Article 15(1) RPBA do not need to be addressed in the decision.

Auxiliary request I - inventive step, Article 56 EPC

11. In paragraph 2 of the reasons for the contested decision, the examining division noted that the applicant had indicated that the auxiliary request (corresponding to present auxiliary request I) merely represented a rewording of the claims of the main request in view of possible objections under Article 123(2) EPC and found that the substance of the independent claims was therefore the same as has been discussed in view of the main request.

12. For the present auxiliary request I, the appellant did not challenge this finding. In their written submissions the appellant merely referred to their respective submissions on novelty and inventive step for the main request (see sections II.2 and II.3 of the statement setting out the grounds of appeal and of the submission of 22 April 2024) and in the oral proceedings they did not make any submissions on auxiliary request I.

13. The Board's findings on inventive step for the main request therefore apply mutatis mutandis to auxiliary request I. Hence, auxiliary request I is also not allowable due to a lack of inventive step in the sense of Article 56 EPC.

Auxiliary request II - inventive step, Article 56 EPC

14. The subject-matter of claim 1 of auxiliary request II differs from that of auxiliary request I in that the square wave power supply calculates pulse widths which correspond to an upper specification limit and in the first half-cycle after transition, a pulse of that width is output if that can be done before the dead zone is reached.

15. The examining division held that the technical differences introduced into claim 1 of auxiliary request II were merely the result of considerations relating to the loads. According to the examining division, auxiliary request II merely imposed an additional maximum voltage constraint that the person skilled in the art would be in a position to implement, on the basis of his common general knowledge.

16. The appellant argued in respect of auxiliary request II that neither of the documents D1 and D2 disclosed the tailoring of an unlocked pulse width in the first half-cycle after a transient condition to the instantaneous needs of loads coupled to the square wave power supply to enable improved synchronization of an output voltage of a square wave power supply to a voltage waveform reference synchronized to a voltage waveform of a commercial power source while an effective output voltage was maintained at the required level. According to the appellant, the pulse width on transition was calculated so as to correspond to an upper specification limit in order to avoid a pulse width, and hence voltage, that would be too high for a smooth transition. The pulse width calculated according to claim 1 was a more moderate pulse width than the full pulse width that was taught by document D1. This allowed for a fine-tuning according to an upper specification limit.

17. The Board concurs with the examining division's findings. It has to be considered a matter of routine design when implementing any power supply to adapt the output to the constraints of the loads being supplied. If the loads have some upper (voltage) specification limit, the Board considers that it would be obvious for the skilled person implementing D1 to calculate and set the maximum pulse width according to that upper specification limit, in order to avoid providing the load with a voltage that is greater than that allowed by its specification.

18. Considering the appellant's arguments, it has to be recalled that D1 does not in fact disclose to emit exclusively the maximum pulse width (and hence voltage level) on transition. On the contrary, at column 7, lines 47 to 50 D1 explains that the term "maximum output voltage level" used therein also includes voltage levels which are approximately maximum. Furthermore, in the embodiment (column 11, lines 26 to 33), D1 discloses to use an inverter pulse width which is "at or near the maximum inverter pulse width" on transition and suggests a pulse width of "at least approximately 7/8ths" of the maximum. Thus, the disclosure of D1 allows for some freedom in the setting of the pulse width on transition. If confronted with a load having a (voltage) specification limit, it would be a straight forward matter for the skilled person to use this freedom to set a pulse width which ensures the specification limit is not exceeded.

19. For these reasons the Board finds that auxiliary request II does not meet the requirement for inventive step in the sense of Article 56 EPC.

Auxiliary request III - inventive step, Article 56 EPC

20. According to auxiliary request III, the square wave power supply not only calculates but also sets pulse widths corresponding to the upper specification limit, and the upper specification limit is at about 110% of the commercial power source.

21. The appellant submitted that with these features the output was optimized to a specific level to meet the needs of average loads coupled to the square wave power supply, whereby the complexity of the method was reduced in terms of both computation and implementation.

22. These arguments failed to convince the board that the claimed combination of features was not obvious. Document D1 already discloses setting the initial pulse width after transition, so to do so at a certain level calculated to take account of a specification would be obvious for the reasons given above for auxiliary request II. Furthermore, an upper specification limit of about 110% of the commercial power source, is an entirely normal specification limit for a load that is to be connected to the commercial power source.

23. For these reasons the Board finds that auxiliary request III does not meet the requirement for inventive step in the sense of Article 56 EPC.

Conclusion

24. In the absence of a request which meets the requirements of the convention, the appeal had to be dismissed.