Seeking Pure Reproduction

Radio Record, Volume I, Issue 27, 20 January 1928, Page 12

Seeking Pure Reproduction

A STUDY IN COMPROMISE

We are all looking for the perfect receiver, except perhaps the few who think they actually possess it. But such a piece of apparatus docs not yet exist. here are certainly many good receiyers, some of which might be classed as "nearly perfect,’ judged by our present standard. Yet a broadcast receiver is really after all a study i compromise. In the first place, the receiver can only at the best reproduce what is sent out by the transmitting station, which primarily decides just what range of frequencies shall be put "over the air." As a rule the broadcast station is capable of sending out a wider range of frequencies than the average receiver and loudspeaker can reproduce. ‘The highest speech frequencies are about 6400 cycles per second, and the highest note of the piano about 4100 cycles, If We atrange that our apparatus shall cut off frequencies above 6400, speech will be as perfectly reproduced as possible, but if we are content to cut off above the highest piano frequencies at 5000, then the final s in speech will be mutilated or lost, but musical notes up to 5000 frequencies will be received. But at these high frequencies the ear hegins to be less sensitive, and the tones are weak. It is generally conceded that a cut-off at 5000 will not interfere with the intelligibility of speech. A similar compromise has to be arrived at on the low musical notes, speech not entering into the question at this end of the scale. ‘The lowest note of the piano is about 27 ercles per second, and the low pipes of organs may give a 32 or a i6-cycle tone. / modern broadcast transmitter will actually transmit a 380-cycle frequency, but few sets will reproduce it. Generally speaking, it is the low frequencies of the musical scale that at the moment present the most ditficult problem. If the detector valve is followed by an audio transformer with a primary winding of inefficient impedance, the lower musical tones will be under-amplified, and the cxtremely lew notes will not be amplified at all, It is then an impossibility for any device in the second stage to replace the lost frequencies, so it is essential to ensure

that the ontput of the first audio valve shall be as near perfect as possible. The following valve must Le able to easily handle the volume passed along to it, and if heavy volume _ is aimed at this must be a = superpower valve, or, alternatively, a pair of such yalyes operating one stage of push-pull, amplification, one of the -best sneans of obtaining distortionless output with transformer coupling. In any case there is a tendency for the lower musical frequencies to be amplificd less than the bigher. The use of reaction tends to emphasise the lower frequencies, ‘ It is important to make sure that the detector stage of the set is functioning properly. ‘ind the correct values for the grid leak and _ condenser, and let them both be of the fixed yariely by good makers. A fixed grid leak ensures freedom from any trouble likely to arise ,from variable contacts. A good yalve socket is an important item, as much noise, and sometimes distortion, can be caused by a socket giving a poor fit for the ~yalve pins. , An instrument of great value to the serious constructor is a milliamseter. Connected in the plate circuit of the last valye, it indicates the plate current in milliamps, possibly 20, if a super-power valve. If no distortion is present, the meter reading will be quite steady, but if there is distortion, extra current will be drawn in the plate circuit, and the meter needle will "kick" with an amplitude of, perhaps, two or three milliamps. Such a meter is also useful for many other testing purposes. As has already been pointd out in this column, tone can be to a great extent regulated by careful adjustment of the values of fixed condensers oi the audio side, particularly on the. first transformer, if so coupled. When the output of the receiver is heayy it is impossible to obtain good reproduction aud satisfactory volume by attempting to pass through the loudspeaker windings the whole output, (Continued Next Page.)

Mainiy About Construction

(CONTINUED FROM PREVIOUS FAGE)) SEEKING PURE REPRODUCTION

{Continued from Page 22.) which consists of separate direct-cur-rent and alternating-current componients. These two combined will be too heavy for the fine winding of the Speaker, so the correct practice is to by-pass the direct cutrent, so that it goes through the plate circuit only, and not through the speaker, which only receives the alternating current which conveys the impulses that actuate the diaphragm. This by-passing of the direct current is the function of the choke-filter that has frequently been referred to. ‘There are several methods of connecting up the choke, and these will shortly be dealt with and their respective merits discussed. It will be seen that although the choke, by reason of its great number of turns over an iron core, offers great resistance to the alternating current, it is important that its resistance to direct current must be kept as low as is practicable in order not to unduly reduce the plate voltage. So far, everything being properly artanged, a fairly good reproduction of the broadcast may be assured, but there is the loudspeaker to deal with, and this is quite frequently the worst offender in the circuit. If a horn type is in use, then the middle and higher frequencies may be quite pleasingly reproduced, but below a certain frequency, depending upon the particular dimensions of the speaker, the trouble begins. Below this point, which may be 200 cycles, the speaker cannot. reproduce faithfully any musical tones, and if it left them alone, matters would be fairly tolerable. But the speaker endeavours to reproduce the sounds that are below its compass:or range, and can only present them in distorted form, either as "blasting" or as discordant blur. Some of the lower sounds it is able to reproduce, an octave higher as harmonics, and_ this sonietimes gives an impression that the lower notes are being reproduced. It is possible to govern the range of a speaker within certain limits. A fixed condenser of suitable value across theleads will make the tone more inclined to mellowness than without it, but the value must not "be overdone, otherwise | "woolly"? tone is produced. A fixed | condenser may be placed in series in one of the leads in order to determine to some extent what depth of frequencies shall be reproduced at the lower end of the scale. This condenser, which in any case should not be less than say .005 or .01 mfd., according to the speaker, if increased, will allow "more of the low frequencies to be reproduced, and by careful experiment can be regulated to cut off where the frequencies begin to distort. This arrangement is most suitable when a horn speaker is used in conjunction with a cone, The horn is then left to take care of the high frequencies only, and the cone the lower. Once the correct value of the condenser has been determined, it can be permanently fixed inside the base of the horn, not ‘ across the two leads, but in series with one of them. In most cases quite an amount of detrimental "mush" will thus be cut ont. A good cone speaker will reproduce most of the deep organ notes in ordinary use, and reproduce them without distortion if volume is not overdone, and provided the amplifiers are ‘delivering the gogds," there will be little trouble at the lower end of the scale. Few cones will give much strength of tone above the middle frequencies, although odd makes stand out in this respect, but on very high notes they are all weak, This is where the highpitched horn mentioned above steps in and preserves the balance in a pleasing and satisfying manner, when a speaker of each type is used.

RESISTANCE COUPLING. It is necessary in amplifiers to have an amplification characteristic that is within certain limits independent of frequency, and the ordinary valve comPlies with this condition. But in the case of the methods -employed in coupling the valves, a compromise is arrived at between the possible and the ideal, Pure resistance coupling, if it could be carried out upon the ideal

lines that introduce too. many practical difficulties, would be near to perfection, byt. in. actual practice it is not .convenient to tise resistance alone, so the varying but not the direct potential of the plate resistance is conveyed to the grid of the next valve through a condenser of sufficient capacity. This type of amplifier depends sclely upon the amplification of each valve, and to overcome this disadvantage valves with a specially high amplification factor are now made for resistance-capacity coupling. A great advantage of this method of coupling is that on the lowest organ notes, there is not the great falling-off in amplification that is present in highratio transformer coupling, and the amplification at twenty cycles may not fall as low as half that in the middle of the audible range. The use of high value grid leaks or larger coupling condensers is necessary to effect this, but in practice the amplification of such very low notes would cause trouble, and distortion would take place through overloading of the valves, besides trouble caused by the slow reduction of excessive charges by dissipation through the high value grid-leaks. With frequencies up to about 3000 the degree of amplification can be regarded as independent of frequency, or at an even rate through that range, above which it falls off, Whilst a high Voltage is necessary on the plate for best wperation of resistRuce coupling, the drop across the high resistance keeps the actual voltage on the valve at a much lower figure than is given across the battery terminals, and consequently the best working voltage is frequently not supplied. | Consequently, whilst even amplification is secured over a. wide range of frequencies, there is a tendency for valves to be overloaded, with the result that distortion is present in what is generally claimed to be a distortionless amplifier, IMPEDANCE AMPLIFIERS,

With a view to improving the re-sistance-coupling system, a low-frequency choke coil was substituted for the high resistance in the plate circuit, By this means a much higher inductance js obtained than is possibie on the primary winding of a good transformer, and the great drop in high-tension voltage of resistance coupling is avoided. But this System, still retaining the grid leak, is liable to the choking effect caused by a heavy signal charge being unable ; to get away rapidly. Jn order to get rid of this effect, the dual-inpedance circuit has been devised, in which a low-fre-quency choke of very high impedance (200 to 250 lenries) takes the place of the leak between grid and filament. This is the principle of the ‘T'ruphonic’’ amplifier. Commercially, the two impedances are wound upon the two outside legs of a shell type of core, the plate tmpedance being of considerably lower value than that connected to the grid. Three stages of dual-impedance amplification may. be used with success, but a good transformer is frequently used instead for the last stage, and increased voluine is thereby obtained. For great volume the type of tube for the three stages would be 201A, 112, and 171. An cutput filter or output transformer should also be employed, and valves must be given full rated filament voltages, ample plate voltage, and correct grid bias, A_ good method of adapting this method of amplification in the Brown-ing-Drake receiver is to incorporate one Stage in place of the first audio transformer, retaining the A.F.8 in the second stage of the audio amplifier. Experimenters can test the idea by connecting up the primary of a spare audio transformer as the plate impedance and the secondary of another as the grid impedance, This amplifier allows of a greater degree of amplification of the lower frequencies than does any other system, and, moreover, by varying the values of the coupling condensers, the amount of such amplification can be adjusted to suit the loudspeaker and the taste of the listener.

f SELECTIVE TUNING. * As a receiver is made more selective either by improvement of coils or tuning gear, or by the addition of a wave-trap, it becomes more necessary to pay careful attention to the tuning. On a nearby station many listeners make a practice of reducing volume by de-tuning the aerial circuit, and such a method has been advocated at times by British technical journals. Whilst with an unselective receiver little harm may be done by adopting this method, when selectivity is secured, both or all dials should be very carefully tuned to the exact maximum position, otherwise distortion 1s liable to be introduced. Volume must then be cut down by other means, the best being firstly the cutting-down of reaction to nil, and then the lowering of R.F. filament voltage. Detector voltage may also be reduced, but dimming the audio filaments should never be resorted to, as that will cause tone to suffer. A variable high resistance across the speaker is not a good volume control for most conditions, as if there is overloading of valves, operating the resistance only reduces volume without removing the overloading, which is simply made more bearable by its lessened intensity. Want of fine tuning on the local station may easily give an impression of distorted transmission, especially when one tunes in to a distant station that must be tuned finely to be brought in clearly, and finds reception from that station free from distortion at equal volume to the local. The Wellington station, 2YA, is a sharply-tuned station, and this assists listeners to obtain high selectivity the more readilv, and at the same time careful tuning is necessary in order to do full justice to both station and receiver. But there should he no attempt to extort from the receiver more volume than it is designed to give, and a ten-cuinea ontfit should not be expected to equal one costing several times that amount. Dealers are not very keen on the idea of listeners tinkering with factory-made sets, and they are not to he blamed for this attitude. If you wish to experiment, buy a set of components’ or ‘make any you feel capable of constructing, hook them up on a board, make alterations to your heart’s content, and with care and patience you will soon improve the initial hook-up Into a set that will give results superior _to the average cut-price factorymade article, and at same time you will learn a great deal about the why and wherefore of radio. :