Hello all, and particularly JRC,
I have done a LOT of work on the S-5000 this weekend, and it is going to take quite some time to write all of if up. However, while rebuilding the bias supply, I think I may have stumbled across a reason why JRC has such high plate voltage. After replacing all the caps and diodes in the bias supply, the bias voltage wound up being quite a bit higher than before. As a result, the tubes were biased pretty cold, and were only pulling around 7 watts each. However, during this time I noticed that my plate voltage climbed way up, almost to where JRC was seeing his.
I guess it seems pretty straightforward, drop the plate current and the tubes load down the HV+ less. I've never seen it quite so dramatic as this, but I think it may be a good starting point.
JRC - Can you please do the following and post results?
1) Measure the ACTUAL resistance of your 12 ohm 1W cathode resistors.
2) Measure the voltage drop across them
3) Measure your plate voltage again since we will be using it for math
Calculate your plate dissipation (per pair) of output tubes, and then the wattage. If you're not familiar with those calculations, go here and plug in the numbers:
http://www.webervst.com/tubes1/calcbias.htm
From this we can figure out just how many watts per side you are pulling and your idle current. I have a sneaky suspicion that while your plate voltage is quite high, your dissipation is going to be pretty low. This would explain why your tubes aren't glowing cherry red.
One of the things I did this weekend that I need time to write up is I replaced those 12 ohm cathode resistors with 4 x 10 ohm 1W 1% precision resistors. Now each tube has its own cathode resistor, and because they are 10 ohms each, we can very quickly determine the idle current in each tube rather than each pair, and because it's 10 ohm, no calculator is needed!
More to come.
JRC, please post what you find out. I think we're close here!
Cheers,
-forkinthesocket