Is my OT fried?

[Gainzilla]

Hi all,

I'm working on a cathode biased push/pull el84 build and ran into an issue last night. "Finished" the wiring, double checked that all my voltages looked good with the tubes out, then checked again with the tubes in (connected to a speaker also). More on the voltage later...

The issue is that I've got no sound at the OT. I looked over the wiring and found a couple small wiring issues while debugging, but still nothing. I started tracing the signal with a sound generator and an audio test probe, looking for where the sound drops out. It literally stops at pins 7 on both el84s, and I get a hum. I hear tone at pin 2, though. I tried de-soldering the primaries from the tubes, wondering if a dead OT would prevent me from getting a tone there. Still no signal.

I re-checked the OT connections, and took a center tap reading to make sure it was getting B+. All Good there.

The thing that leads me to think it might be the OT (a hammond 125E) is that while wiring up the power supply I accidentally wired the CT incorrectly... Can't remember how exactly but all filament voltages looked good, but I blew a fuse when I turned off standby. Current wiring is definitely correct.

So about the voltage... There's something weird (to me anyway). PT is a hammond 269EX, rectified by a pair of UF1007 diodes. Filer caps are 40uf, 450v. I'm getting nearly 450v of B+, which is much higher than what I was expecting. Not a bad thing, just wasn't expecting that based on my transformer specs and the voltage I was getting before (this used to be an AX84 P1).

I'll get exact readings of all B+, Preamp/Power Tubes, Filament voltages tonight after I get a new battery for my MM.

So. Before I go ordering a new PT, and based on what I know so far, what do you think?

Thanks in advance!

[g1]

Post all voltages for one of the EL84's.

[Gainzilla]

Will do, later tonight.

Thanks!

[Mick Bailey]

Something certainly wrong with your B+ for a 390v CT transformer. Unless your MM is acting up because of a low battery.

Check the DC resistance from each side of your OT primary to CT. Even a shorted transformer will generally produce some low-volume distorted sound, but if the windings are open it's dead.

[Gainzilla]

Thanks for the info, Mick. You could be right. It definitely was acting weird right before it died. I'll get a new battery tonight and a fresh set of readings, as well as a reading on the primaries. Crossing my fingers for a working OT, although then I still need to figure out what else is going on.

[Gainzilla]

Ok guys, here are my findings. Anything jump out at you?

OT primary 1: 83.6ohms
OT primary 2: 77.9ohms
OT 1+2: 161.5ohms

B+: 295v
B2: 295v
B3: 292v
B4: 266v

EL84 2
Pin 2: 0v
Pin 3: 41v
Pins 4/5: 6.4vac
Pin 7: 295v
Pin 9: 295v

EL84 1
Pin 2: 0v
Pin 3: 41v
Pins 4/5: 6.4vac
Pin 7: 295v
Pin 9: 295v

12ax7 3
Pin 1: 265v
Pin 2: 0v
Pin 3: 5v
Pins 4/5/9: 6.4vac
Pin 6: 269v
Pin 7: 0v
Pin 8: 5v

12ax7 2
Pin 1: 151v
Pin 2: 0v
Pin 3: 2v
Pins 4/5/9: 6.4vac
Pin 6: 137v
Pin 7: 0v
Pin 8: 1v

12ax7 1
Pin 1: 161v
Pin 2: 0v
Pin 3: 2v
Pins 4/5/9: 6.4vac
Pin 6: 160v
Pin 7: 0v
Pin 8: 2v

[pdf64]

What is the schematic?
The EL84 look to be in cut off; assuming cathode bias, check the cathode circuit, eg cathode resistor value.

The cathode voltages on V3 looks strange, doesn't line up with any circuit I can think of; suggest you check that over.
Pete

[Mick Bailey]

Your OT looks OK unless you have a shorted turn, which won't show up with a MM. But I think your problem lies with the PI & output section.

[J M Fahey]

EL84 and PI cathode voltages look VERY high , enough to turn those tubes of.
Plate and screen voltages are essentially unloaded V+ value, further confirming that tubes are cut off.

If they all received fixed bias I would guess "too high negative voltage bias, driving them into cutoff" but since they are all (presumably, you STILL haven´t posted any schematic ) auto biased through cathode resistors to ground, all I can think is that either cathode resistors are open (not impossible but unlikely) or (most likely) they are not grounded , period.
Those cathode 5V or 41V you measure are the cathode voltage with a 1M resistor to ground: your multimeter internal resistance.

Trace back DC voltages to ground, first at cathodes (you already did that) .then the "grounded" end of cathode resistirs, then any wiring in the path, then the "ground" point they lead to ... which I guess is not ground at all.
Or you simply missed the grounding wires.

[Gainzilla]

Wow, that was truly impressive sir! You nailed it. No connection from the cathode resistor/bypass cap to ground. :facepalm: I must've checked that and missed it 5 times.

[edit] Here is the schem:


With the following variations:
- mosfet cathode follower
- mosfet loop driver
- .01 uf PI input and NFB caps
- 220k grid leak resistors
- 82k/100k PI plate resistors
- Cathode biased with 120r//47uf (which I just noticed is a 50v. Is that ok?)... Actually, currently, un-cathode biased as they are not connected to ground. lol

Anything look weird just based on the bullet points above? I'll add the ground tonight and see if I can get it to make some sound.

Thanks!

[g1]

PI voltages are still a problem, cathode voltage seems very low.

[Gainzilla]

Do you mean the 5v? What range should I be expecting?

[g1]

You've got roughly 20K of cathode resistor, if each side of PI was running at 1mA current, you would have 40V at the cathodes. I'd be expecting at least that. What node of the supply is the PI running off and what voltage is at that node?

[nickb]

PI voltages are still a problem, cathode voltage seems very low.

Yes ...and the voltages at the PI grids seems wrong too. Did you accidentally tie the grid resistors to ground instead of to the cathode resistance tap?

[Gainzilla]

I may have. This is a pretty ambitious build for me, having only built single ended amps and one hundred watt Marshall clone. This time I'm going a little off script.

While troubleshooting, I simplified the wiring a bit... Bypassed the FX loop, and presence circuit, but saw other designs where there is a path to ground there. Certain AC15 models for example (But then other models only ground the tail and use the "NFB" cap as an input for the second channel). Initially I wired the presence circuit as such: The union of the Tail Resistor and NFB/Input Cap received the NFB signal, then sent that signal to the Presence pot, where it had a path to ground. This circuit was a little new for me and I chose to remove it until I got the power amp working.

So anyway, yes. The NFB/2nd input cap and Tail resistor are currently tied together and grounded. What is the best practice here? only ground the tail resistor if not applying input/NFB? But if applying NFB let the path to ground be at the presence pot?

Cheers!