The Electricity Trust of South Australia v. Zellweger Uster Pty. Ltd.

[1986] APO 27

In the Matter of the Patents Act 1952

‑ and ‑

In the Matter of Patent Application 531592 in the Name of THE ELECTRICITY TRUST OF SOUTH AUSTRALIA

‑ and ‑

In the Matter of Opposition thereto under Section 59 by ZELLWEGER USTER PTY. LTD.

DECISION OF A SUPERVISING EXAMINER OF PATENTS:
         Patent application 531592 in the name of the Electricity Trust of South Australia (ETSA) was accepted on 5 July, 1983 and was opposed by Zellweger Uster Pty. Ltd. (ZELLWEGER).  The notice of opposition recites all the grounds in sub‑section 59(1) and was the subject of a hearing in Melbourne on 7 February, 1986 when the applicant was represented by Mr. R.K. Maddern, Patent Attorney, of R.K. Maddern & Associates, assisted by Mr. P.M. Foord (the inventor), and the opponent was represented by Mr. G.C. Pryor, Patent Attorney, of Davies & Collison.  At the conclusion of the hearing I reserved my decision.
The Invention
         The complete specification is entitled "Load Shedding Control Means". The invention is in the general field of what is usually referred to as "ripple control".  Ripple control is the art whereby an electricity supply authority controls the switching of loads connected to the power mains by superimposing a control signal on those mains in addition to the power supply voltage.  The mains power provided by electricity utilities is an alternating current (AC) supply.  Early ripple control systems used a control signal which was also an AC signal, but at frequencies higher than the mains frequency.  The control

signal, like the mains supply voltage, was generated by spinning machines, but the machine generating the control signals were spun at much higher speeds than were the machines generating the power supply voltage; resulting in the generation of a control signal which was a sine‑wave voltage at a frequency much higher than the frequency of the power supply.  In such a ripple control system, the loads were switched by receivers which detected the presence of the control frequency on the power mains.  To switch a load, the control frequency would be coupled into the mains for about 30 seconds.  The receiver at each load, on detecting the presence of the control signal on the power mains, would swtich the load from "on" to "off", or vice‑versa.  Such systems could be used, for example, to switch street‑lights on at dusk and off at dawn from one central control location.
         The specification of the present application commences by stating that:

"This invention relates to means of and method for transmitting and decoding information, for example for control of consumer loads in an alternating current electrical power distribution network, and is particularly applicable where there is a need to shed a load from the network under peak conditions for example.  However, the invention is not limited to such uses and can be used to control relays, for example, for the control of street lighting, shop lighting, traffic lights and other control functions for which ripple control has been, or could have been, used previously."

The specification then goes on to give a "brief summary of the invention", as follows:

"Briefly in this invention, an electrical information transmitting and decoding means for consumer load control comprises a transmitter which transmits a coded signal through the conductors of an alternating current electrical distribution network as a sequence of waveform distortions to a receiver which decodes the signal.  The receiver comprises two correlators connected in quadrature, and the transmitters and each correlator synthesise synchronous signals from the power frequency, all of which are locked thereto.  The D.C. voltage outputs of the correlators are squared and summed so that the receiver is independent of phase angle."

The specification then sets out a consistory statement in terms similar to claim 1, as follows:

"... the invention consists of means for transmitting and decoding information for example for control of a load in an alternating electric power distribution system having a power frequency which comprises a transmitter coupled to the system to derive power from the system and generate a signal the frequency of which is synthesised from and locked to the power frequency, and to superimpose said signal upon the power frequency wave‑form as a sequence of wave‑form distortions, a receiver also coupled to the system, said receiver having a detector circuit comprising a pair of correlators driven ninety degrees out of phase with each other by a control frequency which is synthesised from and locked to the power frequency and synchronous with said signal frequency, such that a signal input to the correlators appears as a D.C. voltage output thereof, a squaring circuit coupled to  the signal output of each correlator and a summing circuit coupled to the squaring circuits, so arranged as to square and sum the D.C. voltage outputs of said correlators."

The preferred embodiment described in the specification utilizes a form of frequency shift keying (FSK) in which the control signal is switched from one to another of three frequencies.  The control signal is generated in a transmitter which includes a phase locked loop oscillator.  That phase locked loop is controlled by the frequency of the mains voltage and so generates a frequency which is determined by the frequency of the mains voltage.  In the preferred embodiment the phase locked loop generates a frequency of 256 times its controlling (i.e. the power mains) frequency.  The control frequencies for transmission down the power lines are generated by frequency division of the output of the phase locked loop.  Desirably, the control frequencies are in the range between 200 Hz and 400 Hz, and the specification cites the example of 237.037 Hz (237.037 = 50 x 256 ‑ 54).  The other two control frequencies are 224.56 Hz and 213.33 Hz.  The three control frequencies are used to signal an 11‑bit digital code.  The first bit of the code is a start bit and is signalled by keying the highest of the three control frequencies on for 0.8 seconds.  Each successive bit of the code is signalled by keying one of the two remaining control frequencies on for 0.8 seconds.  To signal that the next bit of the code is a "1" the higher of the two remaining frequencies is keyed on; to signal that the next bit of the code is an "O" the lower of the two remaining frequencies is keyed on.  By this means each successive bit of the code is always a different frequency.  The specification states that this enables improved bit separation and identification.
         The receiver end of the system also includes a phase‑locked loop, controlled by the power mains frequency.  By frequency division of the output of the phase locked loop the receiver can generate frequencies identical to the control frequencies generated in the transmitter.  Each receiver is programmed to search for a specific code, i.e. for a specific sequence of control frequencies.  Following reception of a start bit each receiver generates the sequence of control frequencies corresponding to the digital code assigned to that receiver.  This sequence of frequencies is fed into and controls a detector in the receiver.  The nature of the detector is such that it can detect a signal of a frequency equal to its control frequency.  This method  of detection has advantages over a conventional narrow‑band filter in that its frequency response does not depend on the values of any circuit components; accordingly the response of the detector is invariant even though the values of the circuit components vary with, for example, age and temperature.
         The detector includes two correlator circuits.  The control frequency from the power mains is one input to each correlator, another input to each correlator is derived from the frequency generated in the receiver.  The output of a correlator circuit

‑will always be zero unless the two input frequencies are the same; and

‑when the two input frequencies are the same, will be proportional to the cosine of the phase difference between the two input frequencies.

For example, the mathematical description of the amplitude of a control signal at the receiver input may be equal to
  A sin (w t + O_-)
where A is constant, w is the angular frequency of the control signal, and O_- is the phase difference between the control signal and the locally generated signal.  The locally generated signal would then be described as
  B sin wt
in the case where the control signal and the locally generated signal are of the same frequency; and B is a constant.  The output of the correlator would be
  C cos O_-;
i.e. non‑zero, but proportional to the cosine of any phase difference between the control frequency and the locally generated frequency.  To eliminate the affect of phase differences, a second correlator circuit is used, with the control signal as one input and the locally generated signal, phase shifted by 90o, as the other input; i.e. the correlator inputs are
  A sin (w t + O_-) and
  B sin (w t + ll)
  2
which can be re‑written as
  B cos w t
in which case the correlator output is
  C sin O_-.
         The output of each correlator is then squared in a squaring circuit; the outputs of the two squaring circuits are then summed in a summing circuit so that the output of the detector is
  (C cos O_-)2 + (C sin 0_-)2
  = C2 (cos20_-) + sin2O_-)
  = C2
(applying the trig. relationship
  cos2O_- + sin2O_- = 1).
         To summarize, the transmitter sends a sequence of pulses of different frequencies down the power lines.  When a sequence of pulses arrives at a receiver, the receiver generates a sequence of pulses specific to that receiver or to a group of receivers.  If the received and locally generated sequences are the same, the receiver performs a switching operation.  Although the frequencies generated by the transmitter and by the receiver are both locked to the power mains frequency, there will at the receiver be phase differences between the two sets of frequencies caused by the different phase distortion of the mains frequency on the one hand and of the control frequencies on the other hand as they travel along the power lines.  The effect of phase differences is overcome in the receiver by controlling two correlators, one by a local signal and the other by a local signal of the same frequency as the first local signal but in phase quadrature to the first.  The outputs of each correlator are squared, and the squares are added.
Evidence
         The main witnesses for the opponent are MR. FREDERIK JACOBUS BACKER and DR. DONALD ARCHIBALD HARVEY JOHNSON.  Mr. Backer is a professional electrical engineer and is a member of the Netherlands Institute of Professional Engineers.  Mr. Backer was general manager of the opponent company from 1965 (the year he came to Australia) to May 1973, and is and has been managing director since May 1973.  Prior to 1965 Mr. Backer was employed by Zellweger Uster Limited, the parent company of the opponent, at its factories at Uster in Switzerland from 1958 to 1964 in the capacities of design engineer and senior project engineer in the ripple control department of that company.  In Switzerland Mr. Backer was engaged in the design and development of ripple control transmitters and receivers and in the use of such equipment.  In 1959 he developed a frequency‑multiplex system to control and monitor distant ripple control transmitters over pilot lines.  In 1960 he led a team in the development of the first automatic load leveller, which allows a ripple control system selectively to shed and restore controllable load in response to variations in the demand of the electricity supply network.  On his appointment in 1960 to the position of senior project engineer Mr. Backer became responsible for the development of his employer's ripple control activities in Australia and New Zealand, and he visited those countries on several occasions prior to taking up permanent residence in Australia in 1965.
         Dr. Johnson, the opponent's other witness, is a senior lecturer in the Department of Electrical Engineering at the University of Melbourne.  He has held that position since 1975 and his responsibilities include teaching and research in data communications, facsimile transmission coding, speech compression/identification, and image processing.  Dr. Johnson has over 30 years experience as a professional communications engineer and as an academic with the British and New Zealand Post Offices, with STC, and in three universities.
         Mr. Backer has put into evidence a number of documents relating to the opponent's DECABIT, RETROBIT and MINBIT mains signalling apparatus.  Those documents include patent specifications, papers, and technical literature published by the opponent company.  From that evidence I am satisfied that before the priority date of the claims of the present application, there had been used in Australia and described in published documents in Australia mains signalling or ripple control apparatus which utilized:

‑a transmitter including a phase‑locked loop and other apparatus to generate a control frequency synthesized from and locked to the power supply frequency and used to superimpose a signal on the power mains, and

‑a receiver including a phase‑locked loop and other apparatus to generate frequencies synthesized from and locked to the power supply frequency,

‑the frequencies generated in the receiver being used to control the switching in a commutating filter.

One difference between this prior art and the invention as defined in claim 1 of the present application is that the prior art uses a commutating filter, whereas claim 1 uses the squaring and summing of the outputs of two correlators driven in phase quadrature.  According to documents put into evidence by the opponent, a commutating filter

"consists of a resistor R between input and output, and a number of capacitors which, one after the other, are sequentially connected across the output terminals by a rotating switch S.  If a frequency which is equal to the frequency of rotation of the switch is applied to the input terminals, then, by way of synchronism, the connection of one and the same capacitor consistently occurs during the very same phase angle of the sine wave and, consequently, after a while given by the time constant RC, every one of the capacitors is charged up to a voltage which is equal to the average of the instantaneous voltages of each particular phase interval of the sine wave.  Provided the number of capacitors is high enough and the load on the filter is low, the particular frequency under discussion passes the filter practically unattentuated.  (All that has just been said also applies to all frequencies which are multuples of the frequency of rotation of switch S, but as these are away from the fundamental by a factor of 2 and more, they can be suppressed easily after the filter by other means). ... It is obvious that the output voltage of the filter is closer to a sine wave the higher the number of capacitors.  ...  For the application under discussion, the type of filter just described is very well suited because ... no tuning and (no) highly accurate components are required.  ...  In an actual filter, switch S is, of course, not a mechanical device but fully solid state."

Other evidence has been put in by the opponent's witnesses to prove that commutating filters on the one hand and the technique of squaring, then summing the squares of, quadrature modulated signals on the other hand, are both
     ‑   common general knowledge, and
     ‑   obvious electrical equivalents of each other.
The relevant evidence includes evidence that various papers were widely distributed to electricity supply authorities or read before a learned society overseas but before an audience including Australian engineers, opinion evidence from the expert witnesses, and an extract from a text book on communications engineering.
         The final piece of evidence I need consider in any detail is Swiss patent specification 259157.  The specification describes a "device for the frequency analysis of electrical oscillation processes".  The device is designed for frequency analysis of an input signal.  The input signal is modulated in each of two modulators by two local signals each of the same known frequency but in phase quadrature.  The outputs of the modulators are squared and the squares are summed, so that an output will be produced when the (known) frequency of the local oscillator is set equal to the (unknown) frequency of the input signal.
Law
         The opponent has raised issues concerning:

‑the extent of the "common general knowledge" of the "addressee" of the specification, and

‑whether the addressee would understand a number of passages in the specification.

The law relating to these issues is conveniently collected together and summarized in the decision of the Court of Appeal in Valensi v. British Radio Corporation Ltd., (1973) RPC 337 at pages 374 to 377, which I quote extensively below.

"A number of authorities were cited to us but we propose to refer only to those which are relevant to two questions of law which are indeed connected with each other.  These are:

(1)Who is the notional addressee of the specification? and

(2)To what extent can the addressee be called upon to exercise his own skill and knowledge in following any details in the specification or in correcting errors in it?

In Plimpton v. Malcomson ... Lord Jessel M.R., said that the specification was addressed to people who were not ignorant of the subject matter but not to people of superior intelligence who would be capable of invention.  The Master of the Rolls was considering a simple type of invention (an improved roller skate), and he regarded as the proper addressee an intelligent but not exceptionally intelligent workman.

In Edison & Swan Electric Light Co. v. Holland ... the Court of Appeal held that the objection of insufficiency failed when the directions in the specification were sufficient to enable a person having reasonable knowledge and skill to make the invention, though it might be that some trial and experiment was required ...

In Vidal Dyes Syndicate v. Levinstein ... Fletcher Moulton, L.J. quoted ... the following passage from the judgment of Lord Westbury L.C. in Simpson v. Halliday ...

"When it is said that an error in a specification which any workman of ordinary skill and experience would perceive and correct, will not vitiate a patent, it must be understood of errors which appear on the face of the specification, or the drawings it refers to, or which would be at once discovered and corrected in following out the instructions given for any process or manufacture; and the reason is, because such errors cannot possibly mislead.  But that proposition is not a correct statement of the law, if applied to errors which are discoverable only by experiment and further inquiry.  Neither is the proposition true of any erroneous statement in a specification amounting to a false suggestion, even though the error would be at once observed by a workman possessed of ordinary knowledge of the subject.  For example, if a specific‑

action describes several processes, or several combinations of machinery, and affirms that each will produce a certain result, which is the object of the patent, and some one of the processes or combinations is wholly ineffectual and useless, the patent will be bad, although the mistake committeed by the patentee may be such as would be at once observed by any ordinary workman."

Fletcher Moulton L.J. said at page 280:

"It is quite immaterial to say that skilled chemists would, after experimental research, find the requisite proportions and temperatures."

In Gold Ore Treatment Co. of Western Australia v. Golden Horseshoe Estates ... a Privy Council case about a process for extracting precious metals from ores, Lord Dunedin, delivering the judgment of the Board, said ...

"The learned judge has described the class of people to whom a specification such as this must be taken to be addressed in language so concise and accurate that their Lordships do not hesitate to adopt it.  He says 'The specification is therefore addressed to those persons engaged in gold‑mining in Western Australia who would be concerned with the extraction of gold from its ore, and who would have a knowledge of the existing cyanide process and a sufficient knowledge of the chemistry to understand and work the process.  It is not addressed to the working miner at the one hand, or to the expert chemist at the other, but to the mine manager or his metallurgist or assayer'."

In No‑Fume Ltd. v. Pitchford ... the Court of Appeal held that a patent for a form of smoker's ash receptacle was valid notwithstanding that exact dimensions were not given in the specification.  Romer L.J. said:

"The test to be applied for the purpose of ascertaining whether a man skilled in the art can readily correct the mistakes or readily supply the omissions, has been stated to be this. Can he rectify the mistakes and supply the omissions without the exercise of any inventive faculty?  If he can, then the description of the specification is sufficient.  If he cannot, the patent will be void for insufficiency."

....

In Henriksen v. Tallon ... Lord Reid said:

"The patentee is representing to the Crown in seeking the patent and telling the skilled addressee after its publication that if the skilled addressee follows his directions he will produce an instrument that is useful at least in the sense that it will work.  He is entitled within fairly wide limits to leave it to the addressee to choose appropriate material from a class which he specifies if he makes it plain that the choice is left to the addressee."

We think that the effect of these cases as a whole is to show that the hypothetical addressee is not a person of exceptional skill and knowledge, that he is not to be expected to exercise any invention nor any prolonged research, inquiry or experiment.  He must, however, be prepared to display a reasonable degree of skill and common knowledge of the art in making trials and to correct obvious errors in the specification if a means of correcting then can readily be found.

We agree with the judge that where the art to which the invention relates is so complicated and difficult as that of colour television it would be inappropriate to suppose the addressee to be a workman on the shop floor.  The judge took the view that in 1939 anyone considering manufacturing apparatus within Valensi's patent would refer the specification to one of the four or so existing design research teams in this field, and that such a team should be taken as the notional addressee.  Such a team would include men such as Dr. White, the very highly qualified gentleman called as an expert witness for the plaintiffs, or Mr. Bedford, the very highly qualified gentleman called for the defendants.  In our view this is too wide a departure from the principles laid down in the cases referred to.  Dr. White and Mr. Bedford were already, in 1939, men of highly inventive capabilities, and it is somewhat artificial to suppose them studying the invention and trying to make it work while studiously avoiding any inventive steps of their own.  In 1939, when research into colour television had been going on for a number of years, it is obvious that between the leading experts and any manual workers who were constructing experimental sets there must have been skilled technicians from whom a team could be constituted which would possess the common knowledge of those in the relevant arts but would not have the capacity to make elaborate additions to or modifications of what was disclosed in a specification.  It is to a team of this category of people that we think the specification should be taken to be addressed.

Further, we are of the opinion that it is not only inventive steps that cannot be required of the addressee.  While the addressee must be taken as a person with a will to make the instructions work, he is not to be called upon to make a prolonged study of matters which present some initial difficulty : and, in particular, if there are actual errors in the specification ‑ if the apparatus really will not work without departing from what is described ‑ then, unless the existence of the error and the way to correct it can quickly be discovered by an addressee of the degree of skill and knowledge which we envisage, the description is insufficient."

I think it highly material to the facts of the present case to compare the erroneous reasoning of the trial judge in Valensi v. British Radio Corporation, 1972 RPC 331 at 451, where the judge stated:

"Sufficiency is, of course, a question of fact, and in my judgment one must be realistic in testing it.  In fact, the question whether a company should embark upon the manufacture of a new television system involving novel transmitters and receivers is one which would certainly in the first instance come before the chairman and board of directors.  Few, if any of them, would be likely to understand Velansi's specification at its date, and inevitably the decision to manufacture if made would involve prior reference to the company's research department if it had one.  Ex hypothesi such a research department in 1938, unless it happened to incorporate one of the four teams referred to above, would have to seek outside assistance, and the search for such assistance would inevitably lead to one of the teams in question and to Dr. White or his opposite number in one of the other teams.  I am, therefore, satisfied that Dr. White or someone else with similar qualifications is properly to be regarded as the notional addressee on questions involving the circuits and electronic side of this invention.  If, as he says was so, he would have been able by using techniques known at the date of the specification to fill in the details of the general and diagrammatic drawings of the specification so as to obtain workable results, then in my judgment that is enough ..."

The Addressee of the Specificaton
         I am satisfied that the notional skilled addressees of the present specifications are not researchers such as Dr. Johnson and Mr. Backer; although if a company was contemplating the manufacture of ripple control systems in Australia then it would probably refer research and development to workers of a similar level of expertise.  Power supply engineers with electricity supply authorities may be closer, in terms of skills and knowledge, to the notional skilled addressee, and the opponent has adduced some evidence tending to show that some of the prior art has been disclosed to engineers such as these.  That evidence shows that some of the prior art was published at an overseas conference attended by some Australian power supply engineers.  That evidence also shows that documents describing some of the prior art were distributed to electricity supply authorities in Australia.  There is not, however, any evidence tending to show that any of the engineers who attended the overseas conference actually remembers anything of the prior art published to them; nor is there any evidence tending to show that any of the documents distributed to electricity supply authorities were ever read by any power supply engineer, let alone remembered by any such engineer.  Given this state of the evidence I cannot find that a knowledge of this prior art is "such as every worker in the art may be expected to have as part of his technical equipment", (Automatic Coil Winder v. Taylor Electrical Instruments, 61 RPC 41 at 43). Similarly I cannot find that, to a power supply engineer, commutating filters on the one hand and the technique of squaring, then summing the squares of, quadrature modulated signals on the other hand are obvious electrical equivalents of each other; nor can I find that the contents of Swiss patent specification 259157 are part of the common general knowledge of a power supply engineer.
         In any event, even if I could find that these various items of prior art were part of the common general knowledge of a power supply engineer, I do not think that the opponent's case would be advanced very far as I do not think that such an engineer is the notional skilled addressee.  On an application of the principles exhaustively considered in Valensi v. British Radio Corporation I think that the skilled addressee should be taken to be a production engineer in charge of the production of ripple control equipment; and no evidence has been adduced to show the extent of the common general knowledge of such an engineer.
Anticipation
         Claim 1 is as follows:

Electrical load control means for transmitting and decoding information in an alternating current electric power distribution system having a power frequency, comprising:

a transmitter coupled to the system to derive power from the system and generating a signal the frequency of which is synthesised from and locked to the power frequency, and to superimpose said signal upon the power frequency waveform as a sequence of waveform distortions,

a receiver also coupled to the system, said receiver having a detector circuit comprising a pair of correlators driven ninety degrees out of phase with each other by a control frequency which is synthesised from and locked to the power frequency and synchronous with said signal frequency, such that a signal input to the correlators appears as a D.C. voltage output thereof, a squaring circuit coupled to the signal output of each correlator and a summing circuit coupled to the squaring circuits, so arranged as to square and sum the D.C. voltage outputs of said correlators.

The invention as defined in claim 1 differs from the closest prior art ripple controllers at least in the mode of detection employed, i.e. the prior art uses commutating filters whereas the claim defines correlation, squaring and summing.  That prior art accordingly does not prior publish claim 1; nor does claim 1 lack novelty.  In the light of my finding as to common general knowledge, I cannot find that claim 1 is obvious.  Nor is the invention as defined in claim 1 anticipated in any way by the contents of Swiss patent specification 259157.  That specification discloses apparatus for the frequency analysis of a signal of unknown phase relative to a local oscillator, which could be incorporated into a receiver as defined in claim 1, but the apparatus does not anticipate claim 1.
Section 40
         The opponent has raised a number of section 40 objectons to the complete specification.  I think it fair to say that most of those objections to the description arise because the expert witnesses for the opponent would, in describing the operation of various parts of the circuitry, choose terminology which differs from that used by the applicant.  Even if the terminology chosen by the applicant could be described as erroneous, I think that the "errors" are such that "any workman of ordinary skill and experience would perceive and correct".  (See Simpson v. Halliday, above).
         There are, however, two objections to the description which cannot be so described.  These are set out in paragraphs 16 and 18 of Mr. Backer's declaration, as follows:

"16.In the paragraph bridging pages 10 and 11 of the specification there is a description of the load current wave‑

form as illustrated in Figure 5 and a statement to the effect that the "off" period is essential for maintaining operation of the circuit.  It is not clear to me why this feature is essential and moreover the feature does not appear to be included in the statements of invention which begin at page 4, line 26 and page 6, line 1 of the specification nor in the claims.

17......

18.In the second paragraph of page 12 of the specification, there is a discussion of a technique for increasing the energy that can be injected into the oscillator circuit by means of a small low voltage boost transformer 42 as shown in Figure 4.  The text includes a reference to four thyristors 19 which need to be "selectively fired" to ensure the boost transformer voltage always assists in oscillation.  There is no basis however upon which one can determine the manner in which the thyristors are selectively fired, and accordingly the specification appears to be deficient in describing how the circuit of figure 4 would operate."

The passage of the specification referred to in paragraph 16 of Mr. Backer's declaration appears in the part of the specification headed "Detailed Description of the Preferred Embodiment", and is part of the description of an embodiment of the transmitter.  That embodiment includes a transformer for coupling the output of a transmitter into the 11 Kv bus in a distribution sub‑
station.  The passage is as follows:

"The transformer load consists of the air cored inductance 21 in series with the capacitor 20, and switched by the back to back thyristors 19 as described above.  However, the load current does not flow continuously and as illustrated in Fig. 5 has an off period between each pulse.  This off period is achieved by making the resonant frequency of the LC combination 21/20 higher than 1.15 times the control frequency ... firing the thyristors.  The actual resonant frequency is not critical and hence it is not necessary to use expensive critically tuned components.  The off period is essential for the successful operation of the circuit and assists in maintaining simplicity of design.  The gap inbetween the pulses causes the LC circuit 21/20 to receive energy from the power system and hence oscillation is maintained ..."

I think when the above quoted passage is taken in context it is clearly a description of one preferred embodiment of the transmitter, and in that embodiment it is necessary to set the resonant frequency of an inductance/ capacitance combination higher than 1.15 times the required control frequency in order to achieve gaps between the half‑wave pulses of that control frequency.  Accordingly it is not necessary to include these features in all the claims to ensure fair basis, but it would be necessary to include them in any claim directed to this preferred embodiment.  Claim 7 is such a claim, and is as follows:

"Electrical load control means according to any preceding claim wherein said transmitter comprises an inductance in series with a capacitance, a pair of thyristors connected back‑to‑back, a transmitter control unit connected to control elements of the thyristors arranged to control the frequency at which said thyristors fire to be lower than the resonant frequency of the inductance/capacitance, and a transformer coupling the transmitter to said power distribution system.

Although this claim, as required by the corresponding passage of the description, defines the firing frequency of the thyristors as being lower than the resonant frequency of the LC combination, the required numeric relationship between those frequencies is not defined.  To that extent claim 7 is not fairly based on the matter described in the specification.
         Paragraph 18 quoted above is also a criticism of a description of a preferred embodiment of the transmitter.  The relevant part of that description is as follows:

"The level of oscillation obtained in the basic circuit shown in Fig 3 is difficult to predict in practice as it is influenced by the amount of resistance in the circuit (within the inductance and transformer windings).  It is advantageous to boost this oscillation to control its magnitude, to allow more resistance in the circuit (that is less copper in the windings) and to make more effective use of the capacitors, being the most costly transmitter components.  Extra energy can be injected into the oscillator circuit by means of the small low voltage boost transformer 42 shown in Fig 4.  The four thyristors 19 need to be selectively fired to ensure that the boost transformer voltage always assists the oscillation.  Control of oscillation magnitude is then obtained by an adjustable damping resistor 43 across the resonant circuit."

The criticism of this part of the description is that "There is no basis however upon which one can determine the manner in which the thyristors are selectively fired ...".  This is true, but on the other hand I am unable to find that the "skilled addressee" of the specification would not be able to so choose the timing of the energization of the thyristor control electrodes as to "ensure that the boost transformer voltage always assists the oscillation"; I am unable, in the present state of the evidence concerning the level of knowledge and ability of the "skilled addressee", to find that such a person could not "supply the omissions without the exercise of any inventive faculty". (See No‑Fume Ltd. v. Pitchford, quoted above).  Accordingly, this objection to the sufficiency of the specification fails.
         I note there is an error in claim 1.  Lines 12 to 14 of that claim, which refer to the receiver, define "a control frequency which is synthesised from and locked to the power frequency and synchronous with said signal frequency ...".  In contrast, according to the description, the control frequency and signal frequency need not be in phase, i.e. need not be synchronous, and so claim 1 is not fairly based.
         The opposition has been unsuccessful on all grounds raised in the evidence and at the hearing.  I accordingly award costs in favour of the applicant.  However I have found that claim 1 includes an error and that claim 7 lacks fair basis, therefore  I direct the applicant, within 60 days of the date of this decision, to propose amendments accordingly.

(A.J. EVANS)