Help! So close! Left channel failed after five minutes; [RESOLVED]

[Necplur]

Crack worked for five minutes on first power up!  (sounded GREAT!)  But then the left channel got some noise (and you could hear things when I gently tapped the tube), then stopped working all together.  Before the failure, everything was great:  Resistance checks, voltage checks, both LED's, tube glow, etc.  I'm new to soldering, so this thing is a hot mess.  That said, nothing looks too egregious...

The Good:

• Tube glow from the big tube (6080?), check
• Tube glow from the smaller tube (12Au7?), check (both front and back)
• LED at A8 lights
• ** Resistance Checks Look Good **
• Terminal  3.  0Ω
• Terminal  6:  0Ω
• Terminal  7:  2.9kΩ
• Terminal  8:  0Ω
• Terminal  9:  2.9kΩ
• Terminal 10:  0Ω
• Terminal 12:  0Ω
• Terminal 14:  0Ω
• Terminal 20:  0Ω
• Terminal 21:  0Ω
• Terminal B3:  2.9kΩ
• Terminal B6:  2.9kΩ
• RCA Red:  93.7kΩ
• RCA Blk:  101.6kΩ
• Ground tab (the screw bolt and nut next to Letter E?):  0Ω

The BAD:

• The LED at A3 does NOT light
• ** The Voltage checks have some problems **
• • Terminal  1:  78V
  • Terminal  2:  168V
  • Terminal  3:  0V (though it seemed like it might have 1V but then i would check like ten more times and it would be 0)
  • Terminal  4:  163-165V
  • Terminal  5:  149V (this seems high, compared to the 50-100V in the manual)
  • Terminal  6:  0V
  • Terminal  7:  102V
  • Terminal  8:  0V
  • Terminal  9:  150V (this seems high, compared to the 50-100V in the manual)
  • Terminal 10:  0V
  
  Lemme know what you guys think!

[Paul Birkeland]

The solder joint at A4/A5 not being well soldered would give these exact issues.

[Necplur]

No luck with that.  I removed the wire at A4/A5.  Snipped off the end (I just couldn't get it clean enough.)  Re-trimmed, routed through A5 to A4 and soldered A5 & A4.  Same readings.

Redo B7/B8?  Or something else?  I didn't damage the small tube, did I?

[Paul Birkeland]

Can you post some build photos?

[Necplur]

Oh, boy.  This is gonna hurt.  This is embarrassing (especially the burnt wire), but here it goes.  I want this fixed!  Man, those LED's were tough to maneuver.  Thanks for the help, btw.  And feel free to be blunt (I'll take solace in egg nog and knowing this is my first project.)

[Paul Birkeland]

If terminal 2 and terminal 4 don't show exactly the same voltage, then I would suspect that there's an issue with the wire that connects the two together.  Otherwise I don't see any reason in your build photos for that half of the 12AU7 to not draw any current.

Do also note that it's not your LED.  A damaged LED will fail open and allow more like 12V or so to appear at A3, but since you have 0V that's not the case.

[Necplur]

The voltages really are the same.  They do fluctuate, though.  So, I plug it in and it will go between 168.0V and 168.8V.  I'll unplug it and plug it back in, and it might fluctuate between 169.0V and something close (and when I do this second iteration, the fluctuation is the same range on Terminal 2 & 4.)

Are the 10W 3K resisters supposed to be so hot, being on only long enough to take voltage measurements?  (I'm a dummy, I shoulda known.. being in front of the vent!)

I re-did A4/A5 again.  I also replaced the wire between Terminal 1 and A6 (the burn was bothering me.)

Still no luck.

I included more pictures, in case any of them shed any light.  Also,

[Paul Birkeland]

Do both halves of the 12AU7 glow?

Yes, the 3K/10W resistors get hot.  One of those resistors right now will be getting hotter than it's supposed to because you have an operational issue with your amplifier.

[Necplur]

Yes.  Front and rear dot on the top glow.

[Necplur]

Here is a pic (sorry I didn't include it before, the forum was erroring out on posting.)

[Paul Birkeland]

Can I see the octal socket wiring as well?

[Necplur]

When uploading the attachments, the forum is erroring out.  Sorry about the constantly edited reply.

[Paul Birkeland]

The soldering at 5L looks a bit suspect, but that shouldn't cause this issue.

What I would do is get a wooden chopstick and run the amp flipped over, then gently poke around the amp to see if you can put a little pressure on a specific connection to get that A3 LED to illuminate. 

[Necplur]

OK, I have an update(!!!):  I press down (absolute last connection I try!) on the A3 LED, between A3 and the LED, and...

• the LED lights
• Terminal 5:  75V
• Terminal 9: 104V

So, that makes it seem things look good!

I've rather molested that LED getting it in there.  Hopefully I can do this.  I will try to re-solder the A4 side of it (or do you suggest something else?)

[Paul Birkeland]

If the LED goes out again I would turn the amp off and measure the DC resistance between terminal 3 and A3.