Hoffman Plexi6v6 Troubleshooting Help

[MiniVanMan]

Looking for some help troubleshooting a Hoffman Plexi 6v6.  This thing ran great for about 2 years then started eating tubes.  From what I can tell V5 is red-plating for some reason.  I replaced both power tubes and was able to set the bias.  It ran for a few hours, then V5 crapped itself again.  Replaced the tubes again (FU Russia for picking fights and driving up the price of tubes.) and now I'm getting nothing. 

V6 vibrates hard when an input signal is supplied (guitar string is plucked).

I've checked every voltage on the schematic and it all looks okay.  It's running a tad hot at around 410VDC at pins 4 and a bit over 400VDC on pins 3 of V4/V5.

I swapped V3, but haven't swapped V1 or V2.   

I'm kind of lost at this point.  I've checked resistance on most of the resistors, but haven't started pulling caps yet.

I'm going to go ahead and swap V1 and V2 while I wait for some potential advice.   



https://el34world.com/Hoffman/files/Hoffman_Plexi_6V6-V2.pdf

[stratomaster]

Messing with the preamp will get you nowhere with a bias issue. Stop feeding this thing tubes. You either have a problem in the bias supply or an issue with the coupling caps from the PI. 

Slow down. Leave the power tubes out, and take some voltage measurements.  What does pin 5 of the 6V6 sockets read?

[Latole]

Follown Stratomaster advice . He know what he is talking about ; Power tube bias issue.
Leave 6V6s unplug until bias is fixed.

[MiniVanMan]

Messing with the preamp will get you nowhere with a bias issue. Stop feeding this thing tubes. You either have a problem in the bias supply or an issue with the coupling caps from the PI. 

Slow down. Leave the power tubes out, and take some voltage measurements.  What does pin 5 of the 6V6 sockets read?

At this moment Pin 5 on both sockets is -39.5 VDC.  This is with the bias set to 23 mV.  If I bring the bias pot all the way down I get over (under) -50 VDC. 

[stratomaster]

Messing with the preamp will get you nowhere with a bias issue. Stop feeding this thing tubes. You either have a problem in the bias supply or an issue with the coupling caps from the PI. 

Slow down. Leave the power tubes out, and take some voltage measurements.  What does pin 5 of the 6V6 sockets read?

At this moment Pin 5 on both sockets is -39.5 VDC.  This is with the bias set to 23 mV.  If I bring the bias pot all the way down I get over (under) -50 VDC.

With the tubes out you shouldn't be able to read any mV across the bias sense resistor. To confirm: you have no 6V6s in the amp currently.

[MiniVanMan]

Messing with the preamp will get you nowhere with a bias issue. Stop feeding this thing tubes. You either have a problem in the bias supply or an issue with the coupling caps from the PI. 

Slow down. Leave the power tubes out, and take some voltage measurements.  What does pin 5 of the 6V6 sockets read?

At this moment Pin 5 on both sockets is -39.5 VDC.  This is with the bias set to 23 mV.  If I bring the bias pot all the way down I get over (under) -50 VDC.

With the tubes out you shouldn't be able to read any mV across the bias sense resistor. To confirm: you have no 6V6s in the amp currently.

I was just stating the mV measurement as a reference.  I redid the bias a bit earlier this evening and with the bias pot the most negative I was at the -50 VDC on Pin 5. So, when I pulled the tubes it wasn't the most negative I could go.  I can get that measurement if you want.  I just didn't mess with the bias pot.

[stratomaster]

Ok.  Did you build the amp, and will you be performing the troubleshooting yourself?

Given that pin 5 was within reasonable ranges and matching that narrows the possibilities to:

Two consecutive sets of tubes with a short in one of them that were coincidentally in the same socket. (Unlikely, but not out of the realm of possibility).

A failing coupling cap from the PI that leaks slowly and/or only under load and/or heat.

If you're performing the troubleshooting I'd recommend lifting the coupling cap that feeds V5 (lift the leg on the negative voltage side) and temporarily adding a 1M resistor to ground. Measure the voltage across this added resistor.  It should be very near 0.  Let the amp run for a few mins and monitor that voltage.  If no change, apply heat with a heat gun or blow dryer and monitor. 

If no change, then reattach the cap and reinstall the tubes.  If you have a lightbulb limiter hook it up and power the amp on while monitoring the voltage at pin 5.  Preferably on both tubes, but if you only have one meter put the probes on pin 5 of both tubes.  The difference should ideally be 0v.  If at any point it deviates by 3v or more shut the amp off. This will confirm bias runaway due to a leaking cap. The sign (positive or negative deviation) will tell you which side was beginning to runaway if you keep track of which probe was attached to which tube, but I'd replace both with at least 400v rated replacements, 630v preferable.

[MiniVanMan]

Thanks a ton.

Yes, I built the amp.  It was my first.  I am doing the troubleshooting.  I figure the price of a few tubes is worth the experience learning how these things work well beyond just reading a schematic and hookup diagrams.  The journey from ignorance to competence is never cheap.

I live in a world of consecutive bad parts being installed.  Usually though, some parameter changes between the bad parts.  In this case, nothing changed which is why I started to steer away from the power tubes.

I was looking at those caps off of the PI but don't currently have a good way, or knowledge, how to measure them under load.  Your advice was very helpful.  Plus I love DIY'ing tools.  I'll build a Lightbulb Limiter over the weekend.

I'll take a look at those caps and change them to 630V.  Pretty sure they're 450V, but a chance they're only 400.

[Willabe]

.... but I'd replace both with at least 400v rated replacements, 630v preferable.

Why 630v for the PI caps?

[stratomaster]

Because he's already replacing them and the cost difference (and even size difference in many cases) is negligible.

250v is marginal, 400v would be marginal  the case of a shorted PI plate resistor (an admittedly unlikely failure mode). 630v is the next standard voltage rating up and would withstand a short.

[Lectroid]

Follown Stratomaster advice . He know what he is talking about ; Power tube bias issue.
Leave 6V6s unplug until bias is fixed.

++

[Brownie]

I always save two of the boring brand name Orange Drops for PI coupling caps. I tried SOZO blue ones in a Quad Reverb, and you couldn't chase the bias down. It wandered all over the place til I put in Orange Drops

[MiniVanMan]

Just an update. 

I'm still working on this.  I've definitely identified the OT (Output Transformer) as causing a problem.  Replacing the OT and I'm no longer getting that bright glowing out of the power tubes. 

What led me to the OT was during testing one of the speaker cabinets that was being used wasn't producing any sound.  Inspection found one of the terminals had vibrated loose.  I'm not sure how often the amp was attempted with that cabinet hooked up.  Also found a bad speaker cable that was being used.  Since the user of the amp is a male, teenage'd offspring, there's minimal information to be had at this point.   

Repairing the poor connection in the cabinet and using a new speaker cable had zero effect on the red plating.  But I replaced the OT anyway.   

The old OT measures just fine with no power applied (no shorts).  I'm suspecting that when power is applied one of the windings is arcing or something along those lines. 

For now the red plating is gone.   

Since the cabinet was a glaring problem, I focused in that area and haven't put the 630V caps in yet.  I have them.  But I'd really like to know if it was just the OT or if the coupling caps are a problem as well. 

It's a good thing power tubes are so cheap. 

Again, the journey from ignorance to competence is often paved with stupidity and disposable income.