Power Output

Radio Record, Volume III, Issue 33, 28 February 1930, Page 35

Power Output

(Concluded from: last week.) N coming to the question of determining the magnitude of the load imposed on the output valve by a speaker, we are faced with the difficulty that manufacturers of commercial speakers are singularly reticent regarding this important information, while measurement of the reactance and resistance of a speaker is a task of some little difficulty, complicated by

the fact that both the reactance and the effective resistance vary with frequency. It so happens that the writer has chanced upon curves relating to the Amplion "Lion" speaker, and no doubt these are fairly typical of speak-~ ers aS a class. They are reproduced in figures 8 and 4. The impedance of the speaker may be determined by' taking the figures for both reactancé and effective resistance at a particular frequency and adding them vectorially (i.e. adding their squares and extracting the square root of the result). The increasing impedance of this type of speaker (a reed-driven cone) at the higher frequencies leads one to wonder what frequency should be chosen for "matching" purposes, It will usually be found that the only "matching" which can be done may be accomplished aurally, since the effect of lowering the impedance of the power valve in comparison with the speaker is to produce a preponderance of low tones, whilst the opposite process bnirgs up the high tones. ‘The conditions for maximum valve output are usually satisfied somewhere in the

lower part of the audible range when a pleasing balance is obtained. ; In view of the increasing impedance of the reed-driven speaker at the higher frequencies, one might anticipate that the reduced current occasioned thereby would spell a lacking in high tones. The resonant frequency of the reed is itself fairly high, however, with the result that, instead of its being necessary to bolster up the high tones, it is usually advisable to use a power valve of fairly low impedance. in order to produce a comparatively large current at low frequencies and a satisfactory response in this part of the scale. It is interesting to note the effect of the pentode on the reed-driven speaker. Here we are robbed: of the possibility of keeping the current at

low frequencies appreciably highed than at the shriller portion of the scale (by reason of the high impedance of the pentode as compared with any nor-« mal speaker). Thus, the current being appreciably constant throughout the audible frequencies, the speaker reson« ance takes its full effect and we have the characteristic shrillness of a pene tode-driven reed speaker, A step-down transformer is of some assistance in eliminating this. The moving coil speaker requires separate consideration. Here the ree quirements for successful operation ig that the coil current shall be substane tially constant throughout the audible register, This requirement can readily be satisfied with a triode power-valve if the moving coil is not made too large, But the pentode is also very much at home under these conditions, and exe cellent results can be had from.a singl@

pentcde feeding a moving coil speake» having somewhat more turns on the coil than is usual. As regards the im: pedance of a coil-driven speaker, thig is a matter demanding a somewhat lengthy explanation. The average home-constructed instrument, however, with no step-down transformer and a coil of about 1000 turns, has an impedance of about 4000 ohms at 50 eycles, and at 4000 cycles and rather less than that at intermediate frequencies.