477,392

PATENT SPECIFICATION

Application Date: May 22, 1936. No. 14449/36.

Complete Specification Left: April 17, 1937.

Complete Specification Accepted: Dec. 22, 1937.

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

PROVISIONAL SPECIFICATION

Improvements in or relating to Alternating Current Measuring Instruments

We, ALAN DOWER BLUMLEIN, of 32, Audley Road, Ealing, London, W.5, and ERIC LAWRENCE CASLING WHITE, of 32, The Rise, Hillingdon, Middlesex, both British Subjects, do hereby declare the nature of this invention to be as follows:-

This invention relates to instruments for measuring the value of small alternating currents of the type commonly known as valve voltmeters.

Previously known direct reading instruments of this type generally used either a leaky grid rectifier or a diode rectifier; but in the case of a leaky grid rectifier the arrangement suffered from the disadvantage that it could not be used to cover a very large range of input voltages and, in the case of the diode rectifier, although the range was good for relatively high applied voltages, the value of the rectified current was very small especially if a high value of load resistance were used and a very sensitive measuring instrument was required.

It has also been proposed, therefore, to use a diode rectifier followed by direct current amplifying stage; but, as normally used, this arrangement suffers from the disadvantage that the input impedance of the amplifying stage is low, particularly for alternating currents of high frequency, and it becomes impossible, therefore, to use the high volume of load resistance in the diode circuit which is required, particularly if *** value readings are to be obtained. Furthermore, owing to the curvature of the characteristic of the amplifying valve the range of inputs for which measurements can conveniently be made ** small and if an attempt is made to improve linearity by choosing a valve with a more suitable characteristic (i.e. one which has a wider grid base) there is the further disadvantage that there will be a large standing current which *** to be compensated for in the measuring instrument.

It is, therefore, an object of this invention to provide an improved instrument, and more particularly, one in which true peak readings can be obtained and which shall be sensitive and have a calibration which is reasonably linear over a wide range of alternating current inputs without an undue amount of standing current.

According to the invention the improved thermionic voltmeter comprises a diode rectifier followed by a direct current thermionic amplifier in which provision is made for feeding in the negative sense to the input of the said amplifying valve a potential derived from the current in the output circuit thereof for the purpose of increasing the input impedance of the amplifying valve.

In the preferred form of the invention, the negative feed-back is obtained by causing the potentials on the cathode of the amplifying valve to vary with the potentials on the grid thereof, as for example, by inserting the load impedance of the amplifying valve between cathode and earth, the input potentials being applied between grid and earth.

Such an arrangement will have a high input impedance at all frequencies and it is, therefore, possible to use a high value of load resistance in the diode rectifier circuit and hence true peak readings can be obtained, and the instrument will have improved sensitivity. In addition, it will respond to a greater range of input potentials and the calibration will not be greatly affected by changes in the operating potentials of the anode circuit.

In one arrangement according to the invention, a diode rectifier has connected in series between its anode and its cathode circuit the secondary winding of an input transformer and a high load resistance shunted by a condenser which imparts a long time constant to the diode, the value of the resistance and condenser being, for example, 5 Megohms and one micro-farad respectively. The end of the resistance nearest to the cathode of the diode is directly connected to the grid of a triode amplifying valve and the other end of this resistance is connected to earth through a suitable source of biassing potential. The anode of the amplifying valve is connected directly to the positive end of a source of high potential and in its cathode to earth circuit a resistance of say 50000 ohms and a meter are connected in series. The provision of the resistance in the cathode to earth circuit of the amplifying valve causes the potential of the cathode of the valve to vary in accordance with variations in grid potentials for the purposes above mentioned.

In operation when there is no alternating current input to the transformer, the amplifying valve grid is baissed negatively with respect to its cathode and very little current flows through the cathode circuit and hence through the meter. When an alternating current is induced in the secondary winding of the transformer, rectified currents appear across the load resistance and the voltage drop is such as to make the grid of the amplifying valve more positive, thus increasing the current through the meter.

In a modification of this arrangement the end of the load resistance which is normally connected to earth, is instead connected to a point on the resistance between the cathode of the amplifying valve and earth. Since the potential of this point will to some extent follow the change in potential of the cathode the effective value of the load resistance will be raised and if for example, the actual value of this load resistance is 1 megohm by connecting its end as described to a point one fifth of the way down from the cathode the effective value of the load resistance will be approximately 5 megohms. This modification enables a very advantageous value of load resistance to be created for obtaining true peak readings while keeping the actual value of this resistance, which is the connection between the grid and cathode of this amplifier valve, a reasonable size.

As above mentioned, the diode rectifier is arranged to have a long time constant, and, consequently, the peak values of the potentials are registered accurately, and owing to the appreciable time taken for the charge on the condenser in the cathode circuit of the diode to leak away, the peak reading remains apparent for a sufficient time to enable the meter reading to be observed.

In order to reduce the reading of the meter to zero when there is no signal to be measured, a suitable source of current may be used in known manner to back-off the current flowing in the meter and additional biassing means may be used to bias the diode anode negatively with respect to its cathode so that the instrument may be used to measure the value of large input currents.

Dated this 21st day of May, 1936.

F. W. CACKETT,

Chartered Patent Agent.

COMPLETE SPECIFICATION

Improvements in or relating to Alternating Current Measuring Instruments

We, ALAN DOWER BLUMLEIN, of 32, Audley Road, Ealing, London, W.5, and ERIC LAWRENCE CASLING WHITE, of 32, The Rise, Hillingdon, Middlesex, both British subjects, do hereby declare the nature of this invention and in what manner the same is to be performed, to be particularly described and ascertained in and by the following statement:-

This invention relates to instruments for measuring alternating potentials of the type commonly known as valve voltmeters.

Previously known direct reading instruments of this type generally used either a leaky grid rectifier or a diode rectifier; but in the case of a leaky grid rectifier the arrangement suffered from the disadvantage that it could not be used to cover a very large range of input voltages and, in the case of the diode rectifier, although the range was good for relatively high applied voltages, the value of the rectified current was very small especially if a high value of load resistance were used, and a very sensitive measuring instrument was required.

It has also been proposed, therefore, to use a diode rectifier followed by a direct current amplifying stage; but as normally used, this arrangement suffers from the disadvantage that the characteristic of the amplifying valve is not straight over a sufficiently wide range. Alternatively, if an attempt is made to avoid this drawback by choosing a valve with a more suitable characteristic (i.e. one which has a longer grid base), there is the further disadvantage that there will be a large standing current which has to be compensated for in the measuring instrument.

It is the chief object of the invention to provide an improved valve voltmeter having sensitivity approximately independent of the amplifier valve constants.

It is a further object of this invention to provide an improved valve voltmeter which shall have a substantially linear response over a wide range of alternating potential inputs without an undue amount of standing current.

According to the invention the improved thermionic voltmeter comprises a diode rectifier for rectifying an alternating voltage to be measured, a direct connected amplifier to which the rectified potential from said rectifier is arranged to be fed, means for feeding in the negative sense to the input of the amplifier valve potentials derived from the output circuit thereof, and a direct current meter disposed in the anode- cathode circuit of the amplifier valve for measuring the anode current of said valve, the feedback potentials being of such amount and applied in such manner that the response of the meter to the said alternating voltage is substantially independent of the amplifier valve characteristics.

In the preferred form of the invention, the negative feed-back is obtained by causing the potentials on the cathode of the amplifying valve which may be a triode to vary in phase with the potentials on the grid thereof by inserting a resistance between cathode and earth which is large compared with the reciprocal of the mutual conductance g, of the amplifying valve.

The rectified input potentials are applied between grid and the lower end of the cathode resistance. Under these conditions the change of anode current of the amplifier valve due to a rectified voltage applied to the input terminals is substantially independent of the amplified valve characteristics. The input impedance of the amplifier grid circuit is particularly high due to the negative feedback and the diode load resistance can therefore be made very high in order to obtain substantially true peak rectification; whereby the rectified potential derived from the alternating potential, or is at least accurately proportional to said amplitude.

Further, the high resistance of the diode circuit results in a high impedance presented to the source of alternating voltage to be measured by the input terminals of the instrument; this is a desirable feature in that it is generally desirable to take as little current as possible from the said source, particularly if the said source has a high internal impedance.

Furthermore, the advantage is secured that the arrangement responds linearly to a considerable range of input potentials, and the voltage sensitivity of the arrangement will be very little affected by change in the operating potentials of the anode circuit, or by substitution of another amplifying valve of reasonably similar characteristics.

The meter which serves to indicate the anode current of the amplifier is preferably placed in the cathode lead of the amplifier, and may itself constitute part of the feedback resistance in the cathode lead.

The method of carrying the invention into practice will be readily understood from the following description with reference to the accompanying drawings of which the two figures are each circuit diagrams of an embodiment of the invention.

In the arrangement of Fig. 1 according to the drawing, is shown a normal full wave rectifier arrangement comprising a combination 1 of two diode tubes with their anodes connected to the ends of the secondary winding of an input transformer 2 and their cathodes connected together and to the high load resistance 3 shunted by a condenser 4 the value of the resistance 3 and condenser 4 being for example, five megohms and one micro-farad respectively, and having a time constant much greater than the period of the lowest frequency to be applied to the rectifier. The end of the resistance 3 nearest to the cathode of the diode 1 is consecutively connected through a resistance 14 to the grid of a triode amplifying valve 5 and the other end of the resistance 3 is connected to the junction of the potentiometer resistances 11 and 6 forming part of the anode circuit of valve 5. The anode of the valve 5 is connected to the upper (positive) end of the resistance 11, which is connected directly to the positive terminal of a source of high potential (not shown) the negative terminal of which is grounded. In its cathode to earth circuit a resistance 7 of say 5000 ohms and a milliammeter 8 are connected in series.

The centre tap in the secondary winding of transformer 2 is connected to a sliding contact 10 of the potentiometer resistance 11, whereby a low positive potential is applied to the anodes of the diodes in rectifier combination 1, to bias them so that there is a potential on the grid of tube 5 which is positive with respect to ground. The potential applied to the anode of tube 5 may be of the order of about 250 volts. A resistance 13 is connected as shown between one side of the meter 8 and a sliding contact on resistance 6 for the purpose of annulling in well known manner the initial steady current flowing in meter 8. Suitable values for resistances 11 and 6 in a particular case were 9000 and 200 ohms respectively, and for resistance 13 2000 ohms.

The provision of the resistance 7 in the cathode to earth circuits of the valve 5 causes the potential of the cathode of the valve to vary very nearly in accordance with variations in grid potentials so that the current increase through the valve for a given increase of grid voltage is less than it otherwise would be. The effective input impedance of the valve 5 is thus increased, and its response is rendered linear and substantially independent of the valve characteristic. The small non-inductive resistance 14, of the order or, for example 100 ohms, is preferably provided in series with the grid of valve 5 for the purpose of preventing self-oscillation of the valve.

In operation when there is no alternating current input to the transformer 2, the grid of valve 5 is biassed negatively with respect to its cathode and a small current flows through the cathode circuit which is compensated in the meter. When an alternating voltage is induced in the secondary winding of the transformer 2, rectified voltages appear across the load resistance 3 and the voltage drop is such as to make the grid of the valve 5 more positive, thus increasing the current through the meter.

The operation of the negative feedback D.C. amplifier may be understood by the following analysis with reference to Fig. 1:

Let R be the load resistance in the cathode lead which is common to the anode and grid circuit. In Fig. 1, R comprises resistance 7 and the meter 8.

Suppose the resultant signal applied between grid and cathode will be

EQUAT. HERE where EQUAT. HERE is the change of voltage across R.

Now EQUAT. HERE where EQUAT. HERE is the change of anode current due to the signal

But EQUAT. HERE where µ is the "amplification factor" of the valve and R* is the anode slope resistance of the valve..

EQUAT. HERE

 

i.e. EQUAT. HERE _ _ _ _ _ _ (1)

Therefore, if µR is large compared with Ra, we have

EQUAT. HERE approximately _ _ _ _ (2)

EQUAT. HERE

approximately for normal values of µ _ (3)

It will thus be seen that the change of anode current resulting from a signal voltage EQUAT. HERE is approximately independent of the valve and controlled only by the load resistance R.

A current meter inserted in the anode-cathode circuit of the valve will therefore indicate linearly the voltage applied to the input

terminals AB, and the calibration of the meter will not be affected by changing the valve provided that the said valve is operated on a normal part of its characteristic such that the above condition EQUAT. HERE is satisfied.

This condition may also be expressed as EQUAT. HERE where g is the mutual conductance of the valve, since g=µ/Ra.

For example, if g=5 mA/volt then 1/g=200 ohms, and a suitable value of R would be 1000 to 5000 ohms or more according to the sensitivity required.

It will be seen from equation (3) that the "output" potential developed across R is EQUAT. HERE approximately.

That is, the voltage gain of the amplifier is approximately unity, and in fact somewhat less than unity. The amplifier is, of course, a current amplifier rather than a voltage amplifier and serves to provide substantial variations of current in response to, and accurately linearly proportional to variations of voltage applied by a source which is itself not capable of supplying current.

It will therefore be appreciated that the anode current of the amplifier varies with the input voltage EQUAT. HERE in such a way as to produce across R a voltage variation nearly equal to EQUAT. HERE. Thus, if a steady rectified potential of 1 volt corresponding to an alternating potential of peak value 1 volt applied to the rectifiers) is applied to the input terminals AB, a potential change of approximately 1 volt will appear across R, corresponding to a change of current of 0.2 mA if R is 5000 ohms.

The arrangement comprising resistances 6 and 13 is well-known to neutralise the effect of the initial current of valve 5 on the meter 8 by passing an equal and opposite current through the said meter, controlled by the slider 12. Change in the valve emission of the H.T. voltage may necessitate re-adjustment of 12 in order to restore the initial reading of the meter to zero. Providing this is done the voltage calibration of the meter will remain substantially unaltered by changes either of the valve 5 or the H.T. voltage, but only so long of course as the valve is working under such conditions that EQUAT. HERE. For example, it may be necessary to re-adjust the slider 10 which determined the initial grid bias.

Fig. 2 shows another arrangement according to the invention, where diode 11 is used to measure the peak positive potential across resistance 17. The diode has a conventional resistance 3 and capacity 4 load in it cathode circuit, and the potential developed across condenser 4 is applied to the grid of the triode 5, which operates the meter 8.

The arrangement shown is intended to work under conditions where there is a considerable d.c. potential across resistance 17 as well as a.c. potentials, so that even when the resistance 3 is taken to a positive point on the potentiometer 18, 19, 21 and 22, the diode 11 can still be conducting on any of the peak positives to be measured. The meter 8 is put into the cathode circuit of the valve 5 and the resistances 18, 19 form the remainder of the cathode circuit of this valve of which the anode is also connected directly to the positive high tension supply. The upper end of resistance 22 is also connected directly to the high tension supply and by varying this resistance any desired zero reading (not representing zero current in this case) can be obtained on meter 8. The resistance 3 is taken to a tapping between resistances 18 and 19, the resistance 19 supplying negative bias for valve 5. Since the tapping point to which resistance 3 is connected follows to a certain extent the potential of the cathode of valve 5, which in turn follows the potential of the grid of valve 5, the effective value of resistance 3 is much greater than its normal value as regards the operation of the rectifier 11. For example, the tappings may be so chosen that resistance 3 only has across it about 1/5th of the potential that exists between cathode of valve 11 and earth, in which case its effective value will be five times as great as if it were connected directly to earth. This allows good peak rectification to be obtained and a long time constant for resistance 3 and condenser 4 without excessively high values of either resistance of condenser. The cathode load of the valve 5, which comprises the meter and the potentiometer resistance 19, is made large compared with the inverse of the slope of the valve, so that the cathode potential of 5 follows substantially the grid potential and thus the voltage calibration of the meter is practically independent of the characteristic of valve 5. Of course it must be arranged that the valve 5 has sufficiently long grid base to pass the maximum current required for the highest peak indication without causing the grid to become positive with reference to the cathode, which would immediately spoil the effective high impedance load to the diode 11.

In the above described arrangements, if desired, the meter 8 may be inserted between the anode of valve 5 and the positive high tension supply. Also the anode of rectifiers 1 or 11 may be connected to the grid of tube 5 instead of the cathodes as shown. In the case of Fig. 2 this change will make the arrangement read peak negative potentials.

Having now particularly described and ascertained the nature of my said invention and in what manner the same is to be performed, I declare that what I claim is:-

  1. A thermionic valve voltmeter comprising a diode rectifier for rectifying an alternating voltage to be measured, a direct connected amplifier to which the rectified potential from said rectifier is arranged to be fed, means for feeding in the negative sense to the input of the amplifier valve potentials derived from the output circuit thereof, and a direct current meter disposed in the anode-cathode circuit of the amplifier valve for measuring the anode current of said valve, the feedback potentials being of such amount and applied in such manner that the response of the meter to the said alternating voltage is substantially independent of the amplifier valve characteristics.
  2. A thermionic voltmeter according to claim 1, in which the negative feedback is obtained by causing the cathode potential of the amplifier valve to vary in phase with the grid potential.
  3. A thermionic voltmeter according to claim 2, in which the negative feedback is obtained by means of a load resistance in the cathode lead of the amplifier valve, the said resistance being large compared with the reciprocal of the mutual conductance of the said valve, and the rectified potentials being applied between the grid of the valve and the lower end of the said load.
  4. A thermionic voltmeter according to claim 3, in which the said load resistance is constituted partly by the current meter which serves to indicate the anode current of said valve.
  5. A thermionic voltmeter according to any of the preceding claims, in which the diode rectifier has a load resistance which is returned to a point in the cathode circuit of the amplifier whereby the rectified potential appearing across said diode load resistance is a small fraction of the total rectified potential applied to the input terminals of the amplifier.
  6. A thermionic voltmeter substantially as described with reference to Fig. 1 or Fig. 2 of the accompanying drawing.

Dated this 16th day of April, 1937.

F. W. CACKETT,

Chartered Patent Agent.

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

Leamington Spa: Printed for His Majesty’s Stationery Office, by the Courier Press. – 1938.