Need advice - Power supply upgrade problem [resolved]

[borism]

Well, I decided to go another upgrade step and install Triad c7-x chokes plus replace the 100 uf electrolytes with a 33 uf film capacitor.

As far as I can tell everything went smoothly. I checked resistances and all voltages afterwards and all were within 5% of target values. The only curiosity was when I measured the Triad chokes before installing them. They both measured at ~240 ohm.

The problem arose when I connected the amplifier to the speakers. Upon start up there was a soft thump followed by a low frequency warble (something like D-D-D-D....). It was loud enough to cause me concern. Afterwards, the amplifier played very nicely without any issue. The next day the same warble repeated and one of the 2A3s blew.

I replaced both tubes with the Sovteks, repeated the measurements which were all excellent. I thought maybe the warble tone was due to a tube going bad. However, the same noise repeated itself. So I turned off the amplifier and am looking for suggestions before dismantling the "upgrade".

Of note, I mounted the Triads on a stand-off and small wooden platfrorms. I did connect each choke housing on one of the mounting screws to the ground tabs on the output transformer.

Here are some pictures before and after:

[Paul Joppa]

Sounds like a feedback through the power supply during the interval where the 2A3 is conducting but the 12AT7 is still warming up. Or it might be some other cause I haven't thought of. This is the sort of thing that happens when you change the circuit. (We had a similar issue with the Seductor during development, and modified the circuit to eliminate it.)

It's possible that simply changing the final capacitor back to 100uF might resolve it.  Temporarily shorting the 10-henry choke with the discarded 270 ohm resistor would at least test this diagnostic guess. A delayed high-voltage start up might be needed to solve it. Probably there are other solutions that have not yet occurred to me.

If you have an oscilloscope, watching the turn-on transient at the 2A3 grid, 2A3 plate, power supply high-voltage rail, and driver plate might give better diagnostic data. Or if you have a SPICE-type simulation software and are familiar with its use, that might give some insight.

[borism]

Thank you Paul for your quick reply. I don't have an oscilloscope or am familiar with the SPICE-type simulation software. I can basically follow a cookbook. So, I decided to go back to the tried and working power supply configuration.

It was a nice experiment though.

[borism]

Well, I just had to find out. So, I reinstalled the Triad chokes instead of the 270 ohm capacitors in the power supply. However, this time I left the 100 uf /450 V electrolytic capacitors in place.

As Paul suggested this did the trick. There is no strange noise upon power-up. So, in hindsight the 33 uF film capacitor was to small.

If 33 is not enough and 100 uF definitely works, any ideas how small a film capacitor might work without any issues on power-up?

[Paul Joppa]

You could try using the 33uF as a bypass, keeping the 100uF for a total 133uF. The ESR of the electrolytic, which is much higher than that of the film cap, will force the film to dominate at higher (i.e. audio) frequencies.

[borism]

Thank you Paul. Any suggestion how I could install the 33 uF bypass on the original Stereomour? There is definitely no place on the board with the 100 uF electrolytic in place.

[Paul Joppa]

Thank you Paul. Any suggestion how I could install the 33 uF bypass on the original Stereomour? There is definitely no place on the board with the 100 uF electrolytic in place.

There are spare terminals on the PSU board for just this purpose, marked HV+, for the positive end. Ground is available at terminals 6 and 10. Outer foil to ground, if you can determine which end that is, but it's not terribly critical for this application.

[borism]

Thank you Paul! That is very helpful. I was hoping it was something that simple.

[borism]

I put in the 33uF "Audiophiler" (as per Doc) capacitors using Paul's instructions to bypass the 100uF electrolytic in the power supply and success - everything works. However, I can't hear any significant difference. So, the effect must be very subtle or possibly the film capacitors need break in time. There is certainly no degradation.