David McGown wrote:
Stuart,
Thanks. I intend to add LED bias to mine, cathode resistor bias w/bypass was expedient. The bias voltage is 12.2V, so need two parallel strings of 7 red LEDs each to handle the current for each tube. Need to cut up a perfboard to wire them up. I want to use 3mm LEDs since that is the most compact, the 5mm LEDs do not quite fit 0.1" spacing in a string.
My volume pot is an Alps RK40 100K stereo conductive plastic ("Black Beauty") that was last in a preamp 15 years ago. I cleaned and lubed it and it is smooth!. Balance between sections is very close, I did some random resistance checks between channels and they measured very close to one another. From dB perspective, the amount of difference is practically nil. I was contemplating using a stepped attenuator, but if the Alps sounds good in the big rig, then I will keep it in. It definitely sounds great in my nearfield system.
I consider this as a trial run for a DHT preamp. I have Coleman regulators that I can setup for 300Bs or other DHTs. Current source load the plates using Gary Pimm CCSs. Cap couple using the best caps Charlie recommends (Dueland, of course), maybe Slagle autoformers for volume controls. May consider transformer output for a low/no gain preamp. Only issue with 300B is the biasing, I have to think about the best way to tackle that. I will be a good retirement project in a year or so.
David
David,
Interesting you've brought this up. Re-thinking preamps here. The gain circuits are fine. I'll either end up with an Aikido/6SN7 version, or the 12B4 in a new chassis. The one in use now is an old PAS chassis with a slab of Delrin in the bottom, new faceplate and jacks on Delrin. Very little room to add anything, including Deulund or Miflex caps.
The first issue is the low input impedance of either a P&G pot, (10k when paralleled), or an LDR attenuator (~7k). The second issue is that of the load to be driven, and hence the size of the output coupling caps. In the case of a tube amp, the input load is rarely under 100k, so even a 0.47uF cap is sufficient, but using SS crossovers (25k for the CX2310) or the miniDSP (10k), or both results in a load requiring a huge coupling cap.
So, my thought is servo buffering both ends. On the input side, a simple 6SN7 should do fine in the Broskie No Gain No Pain circuit for which I have Guy's servo boards. The output side should probably be something like an ECC99 in the servo follower, either Charlie's version of the Rankin/Berning stage, or another of Guy's Broskie servo boards.. The input stage would be completely DC coupled, the output stage would need at most a 0.1uF cap at the input, and be DC coupled at the output.
At 0.1uF, even I can afford the Duelunds! No worries about driving low impedance loads and difficult cables.
I like simplicity, but despise big and expensive coupling caps. Either I buffer the preamp input, or DC servo the output of every source that I build. Ditto for the output stage.
Stuart