@zackthedog, thanks so much for all of your help on this. Here are my replies below. You raise some good points.
What would really be helpful in diagnosing the wayward voltages is if you could measure the voltage readings at the actual tube socket pins, not at the power capacitors. Specifically, it would be useful to know the voltage measured from V1 pin 3 to ground (should be 2.2vdc), and from V2 pin 3 or 8 to ground (should be 13vdc). "More or less" isn't good enough here, these voltages need to be *very* close to what the chart indicates, within .2 volts or less.
I've already done that and posted the results above, but have also measured the caps as they are upstream. Since those voltages are over, it still strikes me as only logical that overvoltage on downstream components is caused by this. Help me understand why I might be wrong about this. As for change since I fixed the OPT ground issue, my initial read of the V1 P1 voltage was 130, and after I fixed the ground on the OPT, I re-measured and it's 126 vs spec of 88, so still too high. That's 38 VDC over spec which is pretty close to the over-voltage coming off the cap board. The difference isn't equal because there's another resistor between that power takeoff point (G half circle) and V1P1.
As for "more or less," here's why I say so. I have an old, cheap analog VM with a needle. If I measure voltages over 250V, I have to use the 1000 V scale, look at the meter remind myself how many volts per hash mark, and then multiply by four. Anything I read below 250 V is more precise. The lower voltages (<50) are very precisely measured, at least to the capability of my VM. So for example, V1P3 measures 2.5v vs spec of 2.2. I've got to believe that that's close enough for an amp of that vintage. Also, I was initially reading voltages by clipping the ground lead of the VM to the ground rail and then using the other lead as a probe. Given the voltages, it's a nerve wracking business. Later I decided to go alligator clips on both sides, which will produce a more precise reading as the VM doesn't read instantly, and I can calmly wait until the needle stops moving without worrying about my hand slipping.
Also, please check the voltage from each test point for the output tubes to ground (should be about 50vdc).
Done. See the table below. I've checked all the voltages noted. I've only written about the ones that are out-of-spec, since stuff that's working properly isn't worth talking about, hehe. V3 and V4, which get their power from the OPT, which is in spec, are also in spec. And the transformer is in spec too.
I suspect what is happening is that both 12AU7s are worn and not operating properly. You're dropping too much voltage across the 22K resistor because the second stage of V1 has 130 volts on its grid instead of 88, which is causing it to draw way too much current. I would strongly urge you, before changing any resistors, to check these voltages, which will tell us a good deal about the tubes themselves--and also to beg, borrow, buy or steal a fresh pair of 12AU7s to test in the amp. At the very least, try swapping out the pair from the other amp.
After the OPT ground lead fix, V1P6 increased to 240V vs 208 prior and 280V spec, so it's now 14% below, within Heathkit's 20% range. As for the age/quality of the tubes, that's worth looking into. While trying to figure out my distortion problem, I did swap the tubes, but frankly don't remember whether I measured before and after and if there was any difference, so I'll swap again and see what I see. As you correctly point out, I have no idea about the wear state of the tubes, though they came in boxes with hand-written measurements on them and the assurance of the seller that they had been checked. Of course that might mean nothing. But I can say, that the amp sounds very sweet and nice, so that argues against a bad tube, though it's clearly not 100% definitive.
The power transformers are underspec'd and will normally run hot, but checking the voltage from the output tube test points to ground will also tell us if they are drawing too much current and possibly putting a strain on the power transformer. You'd be best off to remove the cages during operation, as Pio1980 mentions.
OPT Tube voltages are all within 2% of spec, so I'm good there.
The Mullard 12AU7s are indeed very desirable but unless you have tested them, they may be worn out and not worth anything. ;-). 5881s are nice, but the Gold Lion KT66 will produce a more open and rich sound. They're the only good currently-manufactured KT66's IMO and are very nice. I use them in my Williamsons with excellent results.
As for the 5R4's, flaking getter (the silver stuff) is not good and these too should probably be replaced. If there is white powdery residue inside the tube, it is shot and should not be used.
What else? Do you think it'd be risky to even try the 5R4 with the flaking getter? It has now fallen to the bottom of the tube.
And if anyone reading this is in the Boston area and has a tube tester, I'd be eternally grateful if you'd be willing to help me test some tubes.
Thanks,
Kim
