Gramophone Pick-Ups

Radio Record, Volume III, Issue 45, 23 May 1930, Page 26

Gramophone Pick-Ups

A Guide to Convenience and Worth-While Results

By

CATHODE

BFORD going on to the matter of connecting the now-popular gramophone pick-up in circuit, it seems desirable to treat, at least briefly, the subject of pick-ups in general, and of their output. The reader will then he in a position to decide for himself the type of circuit best fitted to suit his own particular requirements. The amount of amplification neces: sary to give the best results from a pick-up naturally depends upon a number of factors. The most important of these are :- (1) The design of the pick-up-that is, its voltage output under normal circumstances. (2) The type of needle to be employed. (3) The grid-swing necessary to properly load the power-valve employed. These can be dealt with in turn. Of the numerous designs of pick-ups which have at various times been introduced, only two major types have survived. Of these, the single acting type illustrated in Fig. 1 usually has the greater output, but suffers from certain disadvantages. It will be seen that this pick-up derives its electrical output as a result of a vibrating reed varying the magnetic flux between twn pole pieces, both located at one side of the reed (the reed is, of course, caused to vibrate by its mechanical connection with the needle which follows the reeord groove). The disadvantages are Obvious. Firstly, serious amplitude distortion results from the fact that the flux varies as the square of the dis-

tance of the reed from the pole pieces; secondly, the major part of the flux necessarily passes through the reed, and is likely to cause saturation; thirdly, as the pull on the reed is in one direction only, a stiff and heavy reed and a powerful restoring force are necessary to ensure that the reed Will not fall on to the pole pieces. Advantages of Double-acting Movement, HE double-acting, or differential, movement is, for the reasons outlined above, fast ousting its rival. Here the flux in the armature when at rest is zero, as the diagonal fluxes from the pairs of opposite pole-pieces eancel out. When the armature is displaced, the greater part of the flux concentrates in one path, thus inducing an B.M.F. in the coil O (Fig. 2), but a certain -residue still flows in the reverse direction and tends to correct the amplitude distortion which the single-acting pick-up necessarily gives rise to as a result of the square-law principle mentioned previously. There is still a certain tendency for the armature to stick over on the pole-pieces, but this is so small that it is unnecessary te strengthen the reed to resist it. The necessary retorting force can be supplied by the rubber damping pads D. This ability to reduce the size and strength of the reed is of importance in another direction. Frequency distortion, or the tendency to give undue prominence to certain parts of the musical seale, resolves itself largely into a question of resonance in the reed. Any vibrating system involving mass

and a restoring force will resonate at one fundamental frequency (and possibly at harmonics of this frequency), which is determined by the magnitude of the mass and of the restoring force. If the pick-up is to have good characteristics this major or fundamental resonance should lie outside the range of fundamental frequencies used in speech and music. Frequency Adjustment. INCE the resonance cannot well be lowered to 80 cycles or less, it must be raised above 3500 cycles. Lessening the mass or increasing the restoring force both have the effect of increasing the resonant. frequency; but too great

a restoring force is undesirable, tending as it does to heavy record wear and restricted output. Thus the usual and proper process of design is to reduce the mass of the reed to the lowest possible amount and then to add just sufficient restoring force in the form of rubber damping to raise the fre quency to the desired figure and to maintain the reed centrally in the gap between the pole pieces. The rubber pads also, of course, absorb a certain amount of energy and thus reduce the effect of the inevitable minor resonances. As already pointed out, it is desirable, before essaying to design or decide upon a suitable amplifying circuit, to have some idea of the voltage output which may be expected from the pick-up to be used. The more sensitive pick-ups usually develop a maximum voltage slightly less than one volt, but those who desire more precise information may find Table 1, covering a number of commercial pick-ups operating from a fairy deep-cut test record, of some use.

Considering Voltage Output. ‘\ (THE reader, in studying the table, should not fall into the common error of measuring the quality of a pick-up by its maximum voltage output. Ags a matter of fact, some of the best of the pick-ups whose characteristics are given have only a very small output of any frequency. What is important is that the voltage output shall be substantially the same

at all frequencies in the audible range. A slight rise in output toward the extremities of the range is sometimes of advantage in counteracting the deficiencies in recording at these frequencies. The fact that the output is small is of no importance whatever, ~ since, except where an existing amplifier providing only a small gain must be used, any deficiency in this respect is easily made up by increasing the gain of the amplifier, The effect of the needle on the output must not be overlooked. A softtone needle having a relatively large amount of "whip" or spring" will veduce the output appreciably. As a guide to the relative outputs to be expected from various types of needle, in Table 2 are the results of some measurements with H.M.V. needles. Table 2, Relative output. Needle. (per cent. Tungstyle, extra loud ....+... 10 Tungstyle, loud ..... ccsceees M0 Tungstyle, soft ....e.- eecceee 80 Steel, extra loud ....... eooare 85 Steel, loud ...eccccceevcceess 80 Steel, half-tone ...ccscsesaess TH Steel, soft ........ rer eee 45 The Tungstyle needles appear to offer advantages, but their relatively high mass tends to lower the resonant frequency and Lring it within the audible range. The steel "half-tone"’ needle is a good compromise: its weight is small, yet there is not enough "give" to reduce the output seriously.

-. = The method of designing such an amplifying circuit as will permit the pick-up to properly load the output valve or valyes will no doubt be fairly obvious. Assume that a single output valve of the 171 or 171A type is used in the output stage. Such a valve will, at maximum plate volts, have a grid bias of approximately 40 volts, , 8¢@ that the grid swing will have a peak value of 80 volts, or a R.M.S. voltage of 56 volts (the R.M.S. or root mean square voltage is 0.707 of the peak voltage). Now our table will tell us the maximum B.M..S. voltage which the particular pick-up in use may be expected to develop. We will assume that that in this ease it happens to be 0.5 volts. Then 56+-0.5=112, that is, an amplification of 112 must be provided, between the pick-up and the srid of the last valve. If a trans- ' former is used preceding the last valve, this in itself will provide an amplification corresponding roughly to its ratio, say, 3.5 to 1. The amplification still required will be 112+-3.5 =86. There are valves with an amplifi- ~ gation factor as high as this, and if ond of these were used to precede the transformer, the pick-up could be connected directly in its grid circuit. The writer would prefer, however, in such an instance, to use two stages of amplification preceding the power-valve, and to reduce the ample voltage which would then be available by means of a suitable volume-control. It is very comforting to have amplification iu hand to cope with such eventualities as a very lightly recorded disc. Adapting a Receiver for a Pick-up. T will be appreciated that there is always a certain doubt as to whether a single amplifying valve preceding the power-valve is adequate, especially where the power-valve is of really adequate size. There are circumstances. in which one is necessarily limited to an amplifier of this size, however (e.g., where a "detector and one alidio" receiver is to do service as amplifier), and in such cases it is well to remember that the amplifying valve should be one having a high amplification factor, while the pick-up should be chosen from those having a comparaticely high output. » Where a receiver of the "detector and two audio" type is available, a choice is ayailable between a two-valve

amplifier and a three-valve amplifier. Assuming, of course, that the amplifier is also to do service as a radio receiver, it will be clear that to provide a three-valve amplifier will necessitate employing the detector valve as one of the amplifying valves. This may conveniently be done with either a grid detector or a plate-bend detector by switching in the pick-up at a convenient point, at ‘the same time adjusting the bias applied’ to the grid so as to enable the valve to operate efficiently and without distortion as an amplifier. Figure 8 shows a suitable circuit for the introduction of the pick-up into the grid circuit of a grid detector. All that is necessary is a switch (single-pole double-throw or equivalent or a plug and socket arrangement, whereby the grid may be changed over from its usual connection (to grid condenser and leak) to the pick-up. The circuit is completed to filament through a bias battery, which may consist of a single dry cell unless the detector valve has a lower impedance than usual. It should be noted that in this circuit, unlike the case where it is a plate-bend detector which is to be converted, the switch is at high oscillating potential. Care must therefore be taken to avoid long wandering leads, while it is wise to choose a switch of small dimensions, a large mass of metal being undesirable. Any risk of impairing the performance of the set as a wireless receiver can be obviated by fitting the switch on a small block of ebonite, which can in turn be mounted on the baseboard in close proximity to the grid terminal. Convenience is sacrificed, but a panel switch is in this case likely to be a source of trouble, unless it can be wired with short leads. An anode-bend detector is more readily and conveniently modified, since the switch and pick-up may be introduced at a point of low oscillating potential. A suitable circuit is shown at fig. 4. Here, with the switch in the "radio" position, the lew-poten-tial end of the tuned circuit is returned to filament direct through the full bias voltage, while for gramophone reproduction it is completed through the pick-up, and a lower voltage tapping. The presence of the tuned circuit in series with the pick-up will not, of course, have any effect, as the current generated by the pick-up is

of audible frequencies, and these will be unaffected by the tuned circuit. Only 9 very. simple alteration is necessary where the grid voltage on a plate-bend detector is controlled by a potentiometer. As can be seen from Fig. 5, all that is necessary in this case is to provide a switch by mearis of which the pick-up may be short-cir-ecuited when not in use. When potentiometer settings for the alternative methods of reproduction have been determined, it is convenient to mark appropriate positions of the control knob. Omitting the Detector Stage. [tz may be desired, in cases where a pick-up having a considerable output is employed, to avoid pressing the detector into service. Naturally, where the radio receiver to be employed as amplifier is of the detector and one audio type this is impossible, but where the receiver has two audio stages, these are usually adequate to give satisfactory results from the more sensitive pick-ups without having recourse to the detector for further amplification. The pick-up is inserted at ‘Some appropriate point in the grid circuit of the first audio valve. In the ease of a stage of resistance-

eapacity amplification, a suitable arrangement would be that shown on Fig. 6, while for a _ transformer-coupled stage, the circuit of Fig. 7 wouid be suitable; there is no need to alter the bias of an audio stage when the pickup is introduced as it is already adjusted for distortionless amplification. It has been assumed, so far, that it is desired to leave the pick-up permanently connected in circuit. Where it is desirable to keep it available for other purposes (e.g.. testing receivers) while being able to use it at will, a plug and jack arrangement is the most eonvenient. The pick-up is connected to the plug, and the jack permanently wired into circuit. Generally speaking, the reader will experience little difficulty in adapting the circuits already given to plug and jack operation. For example, a transformer-coupled first audio might have the connections from transformer to plate and B+ made through a closed double-circuit jack (the transformer being connected to the two outer springs). The insertion of the plug would then place the pick-up in series with the transformer primary and break the connections to plate and B+.

A rather more difficult application is that shown in Fig. 8, where a closed double-circuit jack is again used, this time to connect a pick-up in the gridcircuit of a plate bend rectifier, at the same time automatically adjusting the bias to a proper value for amplification. Here, the lower spring (the shorter of the two connecting with the plug) is connected to the low potential side of the tuned circuit. The next spring up (a short one) is connected to a point on the bias battery suitable for plate bend rectification. The third spring up (another short one) is left unconnected, and the top spring might also be left unconnected were it not for the necessity of providing some form of volume control. As it is, it will be seen that it is connected to one terminal of a variable resistance which operates as a volume control, the other terminal being connected to the lower spring; lastly, and let it not be forgotten, the frame of the jack is connected to a low value of bias on the battery or other source of negative grid voltage. Volume Control. "HE mention of volume control brings us rather aptly to a discussion of

the two usual methods of accomplishing this very necessary function. It is well-nigh impossible to do without* some form of yolume control. In the circuit of Fig. 8 a variable resistance of high value was shunted directly neross the pick-up; the method is illustrated more clearly in Fig 9a. This arrangement, however, is open to certain objections. Where the pick-up in use is one inclined to shrillness-that is, one having a resonance peak at the higher frequencies-it is unobjectionable. In fact, it may be advantageous so long as there is not too much reduction of yolume necessary. In the case of a more normal pick-up where the high-frequency response is already rather feeble, a shunt resistance may well have the effect of wiping out the higher frequencies altogether. The effect of shunts of various resistances. is shown graphically in Fig. 10. Remembering that we wish to preserve frequencies up to 5000 or 6000, the desirability of avoiding low shunt resistances is not hard to see. Yor this reason, then, a potentiometer control of volume is to be preferred. This, method of control is illustrated in Fig. 9h. There is still a shunt re-

sistance, of course, but it can be of a high value-say 100,000 or 200,000 ohms.-so as not to affect the response eurve. : The circuits given have, of course, been shown in the form of battery-

operated receiver circuits. It is not anticipated that readers will have any difficulty in applying them to a.¢c, receivers as the major connections are the same. It is proposed, however, in certain future a.c. designs, to incorporate special provision for the inclusion of a pick-up, and these designs will doubtless be of assistance to readers wishing to modify an existing a.c. receiver for gramophone reproduction.

Volts (r.m.s.) 250 1000 2000 4000 Make. eycles cycles cycles cyces emplion O. 0. 0.5 T.H. 0.85 0.4 0.55 0. 3 Spot : 0.85 0.4 0.25 0.05 Brown No. 3 0.15 0.2 0.85 0.15 Brown No. 2 1.4 0.75 61.6 0.05 Burndept .. 01 0.05 = 0.05 0.1 Celestion ... 0.4 0.3 0.35 0.05 G.D.C, coseee O83 0.25 U.85 0.05

Table 1.

Igranic 0.55 0.25 0.6 0.05 Igranic (tone- aTm model) 0.8 0.4 0.85 0.15 Loewe 0.15 0.05 0.05 0.05 Philips 1,5 1.0 0.05 Varley 23 0.35 0.05 0.05 Webster 15 0.75 0.6 0.5