01-12-2016, 09:13 PM
Even back in the valve era, 50+ years ago, it was the “conventional wisdom” that a triode pair with series feedback used as a disc input amplifier required DC heating in order to avoid undue hum levels, given that the input stage cathode, to which the feedback was returned, was not at signal ground. The inconvenience of having to use a metal rectifier for the heater supply probably mitigated against widespread use of this approach. An unlikely example was the Jason J2-10 Mk III stereo amplifier of 1959. This used an ECC83 with DC heating and series feedback for its input stages. A more sophisticated example was the Revox G36 tape recorder of the mid-1960s; this used DC-heated half-ECC83s for the input and head amplifiers, although I think that their respective feedback pairs were made up by AC-heated half-ECC81s.
So it would appear that the 1s37 has simply followed established good practice in using DC heating. The feedback pair would probably suffer from a relatively high hum level if AC heated, and the situation would likely be worse for the grounded grid head amplifier, presumably provided for very low output cartridges, perhaps including some moving coil types.
An aspect of series feedback RIAA-equalized amplifiers is that they have an inherent HF error, with the actual curves having another HF pole after which they flatten out, the position of this being dependent on the open-loop gain relative to the amount of feedback. For the most part this seems to have been treated as negligible and ignored, even in the H.P. Walker design, where the departure from the RIAA curve is visible in the published curve. Sometimes it was corrected by an appropriate passive RC circuit following the RIAA-equalized stage, as in the Quad 33. Apart from the primary objective of obtaining an immaculate RIAA curve, possibly there was some advantage in continuing the 20 dB/8ve downslope well into the supersonic range, in that any noise, clicks, etc., in this range coming off the disc would be attenuated before they got very far into the amplifier.
Cheers,
Steve
So it would appear that the 1s37 has simply followed established good practice in using DC heating. The feedback pair would probably suffer from a relatively high hum level if AC heated, and the situation would likely be worse for the grounded grid head amplifier, presumably provided for very low output cartridges, perhaps including some moving coil types.
An aspect of series feedback RIAA-equalized amplifiers is that they have an inherent HF error, with the actual curves having another HF pole after which they flatten out, the position of this being dependent on the open-loop gain relative to the amount of feedback. For the most part this seems to have been treated as negligible and ignored, even in the H.P. Walker design, where the departure from the RIAA curve is visible in the published curve. Sometimes it was corrected by an appropriate passive RC circuit following the RIAA-equalized stage, as in the Quad 33. Apart from the primary objective of obtaining an immaculate RIAA curve, possibly there was some advantage in continuing the 20 dB/8ve downslope well into the supersonic range, in that any noise, clicks, etc., in this range coming off the disc would be attenuated before they got very far into the amplifier.
Cheers,
Steve
