Seeking help with VibroChamp build

[gary h]

Hi all,

This is my first time here. I'm in SF CA. I have built a few tube amps before and I have a basic idea of tube construction, and I am stuck on a project and need some help. I am building a replica Fender vibro champ using the Mojotone parts kit. Stop me if you've heard this one before but, something is not right. For a while I was getting anemic output with wild oscillation if the intensity was anywhere outside half way. I am not sure how that figures into the core problem. I have done the basic troubleshooting things: triple checked parts, orientation, poked for bad joints, tested resistor and pot values. Mojotone included both a schematic and wiring diagram, and with one exception they are completely consistent. I'd like to say my work is the same. I am attaching a schematic of my voltage readings, which are generally high, but probably within reason. Please comment if I am too naive about that.

The DC voltages went down a bit once I started testing with a 100mv 1k sine wave on the input. I lowered the AC using a variac until the filament voltages stabilized at 6.3 volts; this ended up being 116 VAC.

The bad news is that the output transformer is toast. It went slowly, after having it running on and off for several hours. It finally farted (literally) when I increased the input signal to about 500mV. I have never known a transformer to go bad so this is a first for me. Looking online, it seems that they rarely go bad, and when they do it is usually due to excessive current and/or voltage. I have had either an 8 Ohm speaker or 8 Ohm 20W resistor connected to the amp output throughout, so I am wondering where I could be drawing excess current on the primary side.

I put my DMM inline with the AC line and the amp draws 0.45 amps with the volume at 3 and the 1k test tone on input. This seems a little high to me, like maybe double what it should be? I figure that all the tubes together should draw around 33 watts at max, filaments and plate dissipation combined, the lamp and some leakage current should add up to 0.25-.3 Amps (x 120 =33W). Maybe you could correct me if I am missing something. Perhaps there is a better way of measuring current draw.

All looks fine through the preamp section, I have an o-scope and the signal responds as it should for a 12AX7 gain stage. I originally suspected the tremolo section could be the problem so I tested the amp with the second 12ax7 removed, and the footpedal shorted. Perhaps that is not advised, please let me know.

The bottom line is I need to install a new output transformer but I don't want to put it in and have it burn up again.

Any advice on how to proceed with locating the problem is greatly appreciated. I have triple checked all the components are correctly placed and oriented, poked with the chopstick for bad joints, and retested all the resistor values. I did not test the caps before building, but I did check the PS filter cap, which has the correct capacitance and super low ESR values. (I originally had the first PS filter stage connected to the extra 40uF cap. I used PSUDII which showed very little voltage difference and half the ripple. I assume the best reason to keep it original is...to keep it orignal.

Any thoughts, comments, criticism, wisdom, humor, or new curse words I should know are welcome. And thanks for any or all.

Gary

[The Dude]

I can't see your document (it requires login). My first thought is that the OT primary leads need to be reversed. If wired incorrectly, you will get positive instead of negative feedback and the amp goes into oscillation. If you encounter the problem when you get your new transformer, try reversing the primary leads. On the high voltage problem, how high was filament voltage without lowering the variac? It's not uncommon, depending on power transformer input AC ratings, for voltages to run slightly high.

[Mick Bailey]

I also wonder if you have some ultrasonic oscillation going on as well as audio oscillation. This usually results in low output, excessive current consumption and sometimes damaging primary voltages. Often though the power tubes will show excess idle current, or red-plate. What checks did you make on the OT to establish that it's defective? When scoping, have you been looking above the audio range - if not you'll probably not capture HF oscillation.

[gary h]

Hi and thanks for your quick replies. Sorry about the doc, please try viewing again, Dropbox didn't think I was me, I believe it is now "shared". Answers to your questions:
I did not detect HFO on my scope, but that could just be because I wasn't looking for them. That will have to wait until I get a new OT.
The voltage on the filaments was 6.7 volts, which I realize is acceptable. Since I have the variac, and the posted DC voltages were significantly lower, especially at XC, I thought it couldn't hurt.
The OT emitted a suspicious smell, (no smoke) which led me to check continuity of the primary windings first before probing anywhere else. The primary leads indeed were open circuit. And after removing it from the amp, there is no continuity between primary windings.
The high frequency oscillation premise sounds like my next order of business. What range of frequency are we thinking? Will HFO likely occur without the load of the OT? Is it safe to run the amp without the OT in place? If not, can I substitute a resistive load to mimic?
As far as reversed polarity, I imagine I can simply scope grid of the 6V6 and +/green secondary of the (new) OT and determine that they are in phase, correct? I seem to remember having both on the scope before the trouble, and I like to think I would have noticed if they were out of phase. Something for the list.

Thanks for the help, I really appreciate it.

[Dave H]

As far as reversed polarity, I imagine I can simply scope grid of the 6V6 and +/green secondary of the (new) OT and determine that they are in phase, correct?

For the feedback to be negative I think the grid of the 6V6 and speaker output (green wire) should be out of phase.
You can check by disconnecting the feedback. If the output increases the feedback is negative.

[Mick Bailey]

It's safe to run the amp without the OT, but there's then no plate voltage so the output stage is disabled. The voltages show the 6V6 is idling at 16.9W, which isn't unusual and doesn't explain why the OT would fail. Scope the grid of the 6V6 to see if there's any higher frequency upwards of10khz - just keep stepping up on your scope and see what's there.

Lead dress can contribute to oscillation, also twisting the OT primary leads can improve matters. If you do find signs of parasitic oscillation, Fender installed a 330pf cap from grid to cathode on the 6V6 socket on some amps as a fix for poor lead dress. Other improvements can be to install a 1.5K grid stopper as close to the socket as possible. You could try either of these to see if there's any difference.

[pdf64]

Without an OT, the output valve screen grids will be at HT voltage whilst the anodes are floating, close to 0V. I think that will result in excessive screen grid current.

[gary h]

Thanks for the advice. One thing that puzzles me, I thought that negative feedback lowers output not increases...a slight decrease in amplification in order to provide stability. Please correct me on this if I am wrong. I have a new OT on order, and will test the circuit for short amounts of time with the scope to address all feedback and oscillation concerns. I am also posting a picture of my lead dress for constructive criticism.

[Helmholtz]

I thought that negative feedback lowers output not increases...a slight decrease in amplification in order to provide stability. Please correct me on this if I am wrong.

Yes, NFB lowers stage gain, typically by a factor of 2 to 10 with a tube power amp.
That's why gain increases when you disconnect the NFB, as Dave H indicated in post #5.

NFB doesn't lower available output power.

Power amp oscillation may cause an OT primary short but hardly an open primary.
It takes more (current) to melt the wire, like maybe a shorted power tube.
But it could just have been a poorly manufactured OT, exhibiting a bad internal solder joint.

[gary h]

Just to follow up with those who helped advise me on my Vibrochamp build, the problem has been solved...at least for the time being. As was mentioned by Helmholtz and possibly others, I shifted my attention to the output transformer, which indeed was shorting to the case. I installed and tested a Hammond replacement and the amp is scoping beautifully, including the tremolo section. I opened up the original OT and noticed a little burnt spot at one of the external laminated windings right where the paper touches the metal case. I changed nothing else within the circuit except to disconnect the feedback as advised by others. Reconnecting it does lower the output slightly as it should.

Thanks again for the help, I wish I had been able to detect this sooner, but I would not have guessed a new XO would be bad, and not just faulty. Any other advice on how to test the amp to ensure all sections are working not only correctly (which they appear and sound to be) but within ideal ranges. For instance, what is the most practical way of testing the 6V6 for proper current draw without interrupting the cathode or plate wiring? Thanks again,

Happy holidays.

[The Dude]

To calculate bias current, measure the voltage drop across the existing cathode resistor and use ohms law to figure current.