Can one bypass a PS choke for the purpose of troubleshooting?

[Chuck H]

Some of that is meant to cancel in the power amp because it's push/pull. So if you haven't tested with both tubes yet...

Sans that:

Since the filter caps were replaced I might first double and triple check to see that all the grounds are good and all wiring is correct.

[Randall]

I can't do much more until I get more fuses. I used my last 500mA HT fuse. I did carefully verify the grounds, but I can't really lift the caps to measure as they sit. This board is so difficult to get under it just begs for more trouble, so I spliced them in topside (which I never like to do). I'm afraid whatever heat it would take to desolder them might disconnect a leg from underneath (this happened with the first turret I tried to remove a leg from.) I've ordered more and will bite the bullet and go the long road by carefully pulling the board and installing them through the board, thereby giving me a chance to inspect the underside.

As I remember, the ripple is approx 30vac, sure seems like one of those caps is not right somehow. This is on both nodes with all tubes in with the smooth switch in vintage mode, which it has consistently been. Perhaps it was stressed with all the noise the bad rectifier made during this long troubleshooting process?

And Chuck, yes I will re-check all connections, but the grounds do look good, I looked closely at that.

[Randall]

I replaced the filter caps by lifting the board and carefully checking all grounds and connections. I installed the new caps after first measuring them. Put the board back together and still the same problem, the power supply is swinging all over the place with all tubes installed. This is on both nodes with no diff between bias or mode switch settings. I have tried separate pairs of outputs at a time in different sockets to try to rule out a tube. However, I find that with only preamp tubes, but no outputs the PS is steady. I also see with outputs installed the cathode bias voltage across R12 is swinging like Frank Sinatra. Which leads me to wonder how the bias bypass cap C1 could be involved. Maybe charging and discharging, causing a big bias modulation? Do I have anything with this? I'm grasping here.

[g1]

In post #13, it sounded like you solved the fuse blowing problem with a new rectifier, but still had a ripple problem.
But now you mentioned running out of fuses. Is there still a HT fuse blowing issue? Or was it the mains fuse blowing before, or both?
The bias modulation sounds like motor-boating, or tremolo? Is it audible?

[The Dude]

I am wondering about the coupling caps (C19 & 20) off of the PI. Can you lift one end of them and see what happens when you fire up?

[Chuck H]

Two great suggestions. Bias shifting on the grids due to a leaky cap or instability causing a LF modulation were the first two things I though of. Thing is, there's no NFB on that amp and no bias supply. These are two things I would have checked, if they were there?!? It a real brain twist. Still no bead on it.

Where exactly does the ripple seem to be coming from and what is the actual frequency??? That info could help A LOT. Reading the progress, are we thinking in the right way to call this ripple? Is it fluctuating in a symmetrical manner?

[Jazz P Bass]

If the cathode cap on the bias resistor is suspect, lift it.

[Randall]

I used my last HT fuse while it was fired up with the fluctuating V's. I wouldn't say there is still a fuse blowing problem, I assumed it failed because the B+ was dipping up and down. Perhaps ripple is not the correct description. I measure between about 4 - 12 HZ, so a LF oscillation seems a better one. It is audible, I can hear it beat along with the fluctuation on my Fluke, and it does sound like a tremolo. Removing the PI tube has no effect on this fluctuation. The B+ is fluctuating quite a bit at a low frequency, I can't tell exactly how much as it's hard to measure such a moving target. I have to say, this one is getting the better of me.

[The Dude]

I'm sure you got this, but removing the PI isn't the same as lifting the coupling caps. They're still attached to the PI plate supply. If they are indeed leaky or intermittent, you'll still see a problem with bias on the output tubes.

[Chuck H]

That amp does have a tremolo new fangled SS trem circuit. Have you tried disconnecting it? I just looked at the schem for that thing. EEEK!

Try measuring the voltage at the power tube grids. If there's any DC present you should suspect the trem circuit which should be the only thing (besides a leaky PI coupling cap) applying DC to those grids. It's possible that a previous failure took out that circuit and it's malfunctioning OR it was malfunctioning and caused the previous failure.

EDIT: Sorry I couldn't find an image to upload, but here's a link to the schem. http://bmamps.com/Schematics/vox/Vox...2005)%20SM.pdf

[pdf64]

Is the 30Vac ripple an rms value from a meter, or p-p from a scope?
Agree that a voltage survey of the output tubes should be the next step.

Just to note that it isn't a good idea to operate a cathode biased amp that uses a shared bias resistor without a full compliment of functional tubes.
Perhaps make an exception for a few seconds for test purposes.
Ideally they should be reasonably well matched on static plate current.

[Randall]

OH, for Pete's sake!!! Tremolo! It is the tremolo, it was on. I completely missed that. It's a cathode bias type tremolo, it was swinging the current draw, and thus the supply voltages. There is no off switch on the amp, only a speed and depth knob. Turn those down, and the amp is working fine, and likely has been working for the last week, before I chased that red herring. How blind I can be sometimes. Oh well, this lesson learned cost me four filter caps, and a couple of hours of my time (conservatively). But, in the end I'd say it was worth it to have had the help and goodwill from the excellent techs on this forum. I value that very much, and want to thank each and everyone who has generously offered their expertise in this and my past conundrums. Thanks guys!

[Chuck H]

Hmmmm... If that trem is operating at 12Hz I might suspect something is wrong with it anyway!?! The way it's arranged in the schem I could see a power amp failure buggering it.

Glad it's working out so far. And especially glad it ONLY cost you four caps instead of some much more expensive iron

[Randall]

Chuck, the trem seems OK as far as I can tell. The approx 12 Hz was measured with the Hz setting on my fluke, so maybe I'm missing a point there, but the amp seems to functioning well, so off it goes!

[g1]

I used my last HT fuse while it was fired up with the fluctuating V's. I wouldn't say there is still a fuse blowing problem, I assumed it failed because the B+ was dipping up and down.

Houston...