VTA ST 120 blowing fuses - solved

My amp has started blowing fuses after installing rectifier tubes, gone through a several fuses and a couple GZ34 rectifiers. With only the 12AU7 tubes, all the voltages from all from the secondary windings of the power transformer match would should be from reading the schematic. After adding the rectifier weather by itself or any combination of left/right KT-88 power tubes installed I'll see the rectifier tube flash and a blown fuse after anywhere from a few seconds to a few minutes. This last time 5 or so minutes with all tubes installed, thought it was fixed because I replaced the 12AU7 tubes from my other amp. Actually was finishing up biasing the tubes at that time when it blew. Only thing I can think of is something in the power supply circuitry doesn't warm up gracefully and breaks down shorting the power supply? Is it possible a failed section in the quad-cap, ESL or SCM? Now looking for suggestions for things that I could check without removing any components unless I have to. Now I'm coming to the experts for advice of what I can check, next troubleshooting step?

Any help appreciated, Dave

If the GZ34 flashes when no KT88 is mounted you have 2 things left:
the GZ34 and the power cap ( and maybe the B+cap's on the pre-board. This last thing may be eliminated by disconnecting the B+ that feeds the board, if the fuse still blows
the area is reduced to GZ34 and power cap)

Try a NEW gz34 first. If it has flashed once probabilities is that it will flash again.

Do you have the means to test tubes?
What prompted you to replace the rectifier tubes in the first place?
How old are the old tubes, and what is their source?
And, what is the source of the 'new' tubes?

It is generally a bad idea to install a rectifier tube in a system with no load on them. Less so with Bob's designs as his caps can operate at a higher voltage than OEM Dynaco products. But still never a good idea. And a flashover is NEVER a good symptom.

I would suggest you put everything on hold until you can verify the bona-fides of the new tubes under actual test. Leave them in the tester for at least five minutes, watching like a hawk. Then test for shorts and gas. If they pass completely, then start with the power tubes.

Best of luck with it. At this point, I am suspecting that you got some defective rectifier tubes or, the main quad-cap has problems. I would not want to advise you one way or another until you at least can eliminate one or another issue first.

BUT: if you have a good VOM and know how to use it under these conditions, you have one rude-and-crude test for the quad-cap - which will require that it be disconnected.

With the amp unplugged and turned off:

a) Disconnect all four sections.
b) Discharge the cap. I have a clip jumper that I use for the purpose.
b) In OHMS, check each section to ground (the cap shell). The VOM should climb very slowly to infinity - this may take several minutes. But eventually it should reach into the mega-ohms, at least. Anything else, that section has failed.
c) Check from section to section. Same as above. Note that you should discharge, again, between each test.
d) If your VOM has a Capacitance Test setting, do the test. But, note that if an electrolytic cap tests very high, that is often the sign of an internal short. So, anything more than +50% from rating should be viewed with suspicion pending an ESR test.

Good luck with it!

Thanks for the advice, I'll try disconnecting the B+ with my last new tube and a different set of power tubes. Unfortunately I have no tube tester, however I do have a spare ST120 in use that I can pull known good tubes from. Bought the second amp, well because they just sound terrific and to try rehash some 30 year electronics knowledge, theory and eventually used for bi-amping. As they say, know enough to be dangerous. lol Anyway had the spare on my bench with 8ohm dummy loads and wanted to see if I could see a signal on my scope and flash, things went down hill from there.

Tube testers and cap checkers are wonderful tools... but the majority of them perform their tests at low voltage. If your amp is operating well enough to give you five minutes of biasing time before the fuse blows--I am guessing that testers may give you false negatives. Testing a quad cap at 9V is a whole other matter than testing it at 525V.

I will echo others... do not power up your amp with an installed rectifier tube unless you have at least one set of output tubes installed.

Powering up with a "dim bulb" may give you some insights into what is going on with your amp--especially the power supply.

Since your amp will power on, and since you have a spare VT-120... Use a clamp-on ammeter ($10 at Harbor Freight) to compare the line current for the two. Once with no tubes, then with 12AU7 only, then with all tubes but the rectifier, finally all tubes. A comparison of the two amps under these conditions may be informative.

Maybe you could try debugging your quad cap at high voltage. If necessary, find a guitar shop near you ask them to sell you a couple 47uF, 450VDC caps. Put these in series and add 1/2W, 220kOhm shunts across each cap. Use this contraption to sequentially replace each individual segment of your quad cap. If the amp starts working you have found the bad quad cap segment.

FYI: I avoid using resistor loads on the bench. Blowing a fuse after 5 minutes of playing beautiful music is a different scenario--in my experience--than blowing a fuse after 5 minutes of (unknown) oscillations due to a feedback path problem. Not having a speaker(s) connected has tripped me up on occasion.

...j

Thanks for all the advice troubleshooting, learned alot in the process and built a dim bulb tester. Was able to verify the power supply cap values all good and no shorts on the transformer. During my testing I was moving the 6.3VAC filament wires out of the way to reach the points for taking a measurement then one pulled loose. Went back with a little beefier soldering iron and secured up all the power supply connections. The amp is now back in service and gained some valuable troubleshooting knowledge in the process.