My ST70 that I assembled from a kit back in 1966 and have been using happily ever since recently developed a problem holding a steady bias voltage on the output tubes so I began looking into my options for repairing it. I have come up with the following three options:
1) Minimal repair. Replace the selinium diode, bias supply capacitors and maybe the output coupling capacitors on the driver board.
2) Go a bit farther. Do the above plus make another stab at replacing the main power supply capacitor (I tried this several years ago but the CE Manufacturing quad capacitor that I bought was a waste of money as two of the sections failed open after a few months so I went back to the original Dynaco quad cap which still works fine but is getting a bit long in the tooth).
3) Go nuts. Do all of the above plus replace the driver board with one of these VTA gizmos that is supposed to make the amp sound better (I think it sounds fine now, but how would I know since I am hardly an audiophile and have been listening to it as is for 46 years now).
So, looking into option 1) I measured the resistance across the output coupling caps and found infinity on all 4 of them. Of course, infinity on my Fluke may not equal infinity at 400 volts and the consensus seems to be that the old black cats need replacing so I guess I should do so. I then measured the resistance of the "matched" 270k bias resistors and found that they were pretty far from being matched yielding values from 268k on the low side to 323k in the high side. Digging further, I found that most of the other resistors on the driver board were also well out of tolerance including the "matched" 47K resistors in the phase inverter. YUCK! Therefore, I resigned myself to having to replace everything on the driver board and maybe even the board itself.
This brings me to my first question. What does "matched" mean. Is 1% good enough and, if not, what is and how many 1% resistors do I need to buy off of Mouser in order to be sure of getting a good match?
Looking into option 2) I found a nice 80,40,30,20 quad cap at Dynakit but, being still a bit leary of a new quad cap, I also looked at the SDS power supply board being sold by Triode USA. That way I figure if some cap goes bad I can just replace the failed cap without having to shell out $40 or God knows how much a few years from now for a whole new quad cap if they are even still available. Plus "they" say that the much greater capacitance on the SDS board also makes the amp sound better. The main problem I see with the SDS board is that they are using two 250 WVDC capacitors in series.to get 500 WVDC. Now I measured the voltage at the first two sections of the quad cap (D and C on the schematic) with all the tubes pulled except the GZ34 and got 485 volts, so 500 WVDC seems to be pushing it quite a bit. Plus these capacitors are +/-20% so the voltage on any one of the two in series could range from 200 volts to 300 volts for a 500 volt input. Now I know that the balancing resistors are supposed to take care of this, but they are using rather high resistances at 330Kohms (RC = 129 seconds for the 390 uf caps) so one of these caps could be overloaded for quite awhile after turn on.
This brings up my second set of questions. Am I being too leary of the quad cap? Have others had problems with early failures of these newly manufactured quad caps, or did I just get extremely unlucky on my first attempt? If I do go with the SDS board I will be using a different set of capacitors for it. I found some Nichicon LS series 220uf 350 WVDC caps to use for sections 2,3 and 4 (C, B and A on the ST70 schematic). Are these good capacitors for this application? I couldn't find any ESR data on them. I also found some 100uf 400 WVDC Panasonic ED series capacitors for use on the first section attached to the GZ34. These are good low ESR caps. Is 50uf too much here? The specs for the GZ34 say 60uf max, but I have seen some posts that say I should keep this down to 40-45uf. How much capacitance do I need to make the amp sound "better". Is 50, 110, 110, 110 enough? There must be some point of diminishing returns. For that matter the 40, 80, 30, 20 that I'd get from the quad cap may be plenty. I just don't know.
Now to option 3). If I do go with the VTA driver board I'd like to know a little more about it. The schematic and parts list would be helpful, but they seem to be a closely guarded secret. My questions have to do with mating it to the power supply. Does it use both of the B+ voltages from the original driver board (A and B on the Dynaco schematic) or just one of them? I see that there are some filter caps on the driver board. What are their values and what B+ voltages are they hooked up to? Do they appear in parallel with both sections 3 and 4 of the main power supply cap, just one of them or none of them? If I go with the SDS power supply board also, what parts on it can I avoid having to buy? I see that the bias supply on the SDS board is redundant. Are any of the main filter caps redundant also? This goes back to my question of how much capacitance do I need in the main power supply. Do I still need the 6.8K resistor between sections B and C and the 22K resistor between sections A and B of the quad cap or does the VTA driver board require some other values?
Sorry for such a long post. I thank you for your patience.