Power Up; Time to push some electrons!

[Firestorm]

ahh. I see the pic now. Hang on.

[Firestorm]

Jack looks okay. Is that a teflon insulated coax? Probably okay too, if so.
Where is the grounding bus grounded? I can't spot it. And what's that other bare copper wire under the pots? From the pic, it almost looks like some of the lugs on the pots could be contacting the pot body (them PEC pots is big suckers) so make sure they aren't. Wiring on the output tube sockets looks kinda scary -- no accidental connections there?
I'll stare some more.

[gearhead]

Jack looks okay. Is that a teflon insulated coax? Probably okay too, if so.
Where is the grounding bus grounded? I can't spot it. And what's that other bare copper wire under the pots? From the pic, it almost looks like some of the lugs on the pots could be contacting the pot body (them PEC pots is big suckers) so make sure they aren't. Wiring on the output tube sockets looks kinda scary -- no accidental connections there?
I'll stare some more.

Yes, it is teflon coax.
Grounding buss is to the pots only. Had contemplated running it to a lug, but haven't, considering the sheer number of connections. The bottom copper wire was the original buss wire; held there by lugs from each pot -because- you can't solder to the pots. Ended up adding the central buss for easier access.

I checked pretty closely the pots after installation; there is not a lot of clearance and those PECs are huge. Used telfon tubing on the right side presence pot because I couldn't get around it. Will check further.

The output tube wiring was one of the last things I did, and it was in a rush; and does look scary. Afterwards, was tempted to rip it out and do again, but pressed forward. It's, by far, the least thing I'm proud of with this build. Am not used to flying components mid-air, and would definitely do it different if I had it all over to do with (wrong order of installation restricted solder iron placement and too much solder). Have checked the sockets for connectivity and tried continuity tests between most sockets to see if there was a short somewhere. Extensive visual inspection with maglight and magnifying glass. Will continue.

BTW, when I said low volume, couldn't hear anything till vol was at 9 oclock, and at full volume it was as if it was on, at most, 7 oclock. From initial sound on up, it was clipping (no clean).

Any ideas?

[Firestorm]

BTW, when I said low volume, couldn't hear anything till vol was at 9 oclock, and at full volume it was as if it was on, at most, 7 oclock. From initial sound on up, it was clipping (no clean).

I'm not understanding the volume description. 7 o'clock comes before 9 o'clock -- are you saying it got quieter? Where is the pot's zero volume setting?
It's significant that there's also no clean sound. Bad groiund somewhere or an open resistor?
I really would start by disconnecting V2 and jumpering V1a plate right to the PI input cap; then volume pot wiper to PI cap; etc. That way you'll know where to look.
BTW, what did you find in investigating the too-high V2 voltages noted by Allynmey?

[gearhead]

Sorry, not very clear.

At full volume on the knob (5 o clock) setting, it sounds only as loud as you would expect it to be if the knob was set at 7 oclock (just barely on)

WRT resistors, I couldn't seem to get a good voltage across the V2a 100k plate resistor. Most times it was 0, but since the B+4 and plate were usually 27V apart, that really puzzled me. Did measure the resistance across it when I shut down, and it was close-ish to 100k; I think I also did a continuity check to make sure it was not shorted. If it's open, couldn't imagine the plate could get the voltage it has.

It's like something is yanking around the V2a plate voltage (tried a different tube too) and sucking all the current too. That would either be from the V2a grid/cath or V3a grid (except it has that .1 uF blocking cap)?

[Allynmey]

Check your bias. It sounds as if the power tubes are running at or close to shut off. You should be pulling close to 40mA per tube. If it is, then you have probably grounded out a signal wire -or- your input shielded wire is bleeding signal to ground due to melting of the insulation. That's where I'd start.

Allynmey

[gearhead]

OOps, forgot to put bias measurement on my list. It was -33VDC at the node, but will check at the tubes.

What would the tubes be running at if the signal is grounded and they are running close to shutoff?

Did check the input (and output) jacks for continuity for signal and ground, but maybe I missed something (hmm, did I check -between- them?) Might just start at the top and work my way down. Did that for 2-3 hours before I even powered it up this weekend, but is worth a try.

How do all the other voltages look? Especially the PI and Power tube values.

Thanks!

[Firestorm]

If it is, then you have probably grounded out a signal wire -or- your input shielded wire is bleeding signal to ground due to melting of the insulation. That's where I'd start.

Allynmey

I think you had -27V on the output tube grids, so that shout be pretty far from cutoff. I'd definitely check for grounding along the signal path. Did you check for 1M resistance from the tip of a guitar cord to ground to rule out melted coax insulation?
The V2 business bothers me, too. Check resistance from V2k (at the pin) to some point on the grounding buss. 10K? Also check V2g to the grounding buss. 150K? If you're not using a good meter, it can sometimes give you phantom voltage readings.

[gearhead]

Tip to ground is 1M.

The V2 grid pin to buss was 150k-ish, and the V2 cath was 9.9k, although getting a reading off the buss was hard for the later. Pin to the far side of the resistor lead was easy, but no matter which meter I used (cheap or decent), it took some moving around to get it from the buss.

If the signal is grounding out, I haven't found it so far.

On the one hand, it's pretty frustrating. On the other, it is quite a puzzle; a challenge

[Firestorm]

On the one hand, it's pretty frustrating. On the other, it is quite a puzzle; a challenge

Well, that's why I seem to have become obsessively interested in this thread. In my heart I am still a repair guy more than a builder.
I hope you'll test each stage to the PI next se we can see where the signal is getting lost.

[Allynmey]

Looks like the output jacks are wired incorrectly. I sent a pic with a correction. I hope this helps.

[Firestorm]

Looks like the output jacks are wired incorrectly. I sent a pic with a correction. I hope this helps.

[Edited to minimize ranting ...]
Where did Allynmey get pics of your output jacks? You holding out on us? Consulting an actual expert instead of a seasoned dilletante (moi)?

Let us know how it goes. Next week I'm gonna build one of these myself so I can have as much fun as you

[gearhead]

FS, I highly value your inputs! Allyn sent me a PM and I just sent pics to him before I went to bed Wed. It's been like a blender set on puree this week at work (up at 0530, work, eat dinner, go to bed, rinse/repeat), so I haven't even been able to make the change he suggested.

If my signal was grounded at the output, how the heck did I get that horrific screech? Guess I'll find out.

The V2a plate voltage is at a level way higher than I was thinking it should be; is that typical at idle? Thought that ROM current for a 12AX7 is in the 1 mA range; 294-267=27v. Thru the 100k resistor, that's only about one fourth that (.27mA). The output change might change that, we'll see.

Thanks Allyn and Firestorm!

[Allynmey]

Gearhead, you have one wire across the hot and shorter clips of both jacks. The jack that has no plug in it could short to ground if you use jacks that are internally connected short>ground. It's best to bend down the shorters on cliffs if used on outputs.

[Firestorm]

The horrific screech is just feedback in power amp (output back to PI). You don't even need a preamp connected. Since you were reading 1M tip to ground, I don't think the coax is shorted.

The current in a typical 12AX7 is in the microamp range (set by the cathode resistor). With a 10K cathode resistor I think you're likely to see a drop sort of around 30 volts. I don't Allyn was suggesting V2 was off the map, just a little high.

I just saw Allynmey's latest note. If you used shorting jacks for both outputs, that would explain a lot.