[Misterrogers]

So I seem to have a gremlin. It periodically unlatches, and at times I just have the right channel, sometimes both. Got some hunting to do. I'll tell you though - when it's working, it sounds really, really nice. Very detailed and clear, with great extension top to bottom. If I can get the circuit stable, it's my best sounding CTH so far (initial impressions, etc. etc. ) I'm checking for more shorts, but any thoughts as to what I should be checking? Are there specific test points that would shed light on why it's de-latching at times?

 

[sachu]

What are you reading on pins 2, 3 and 6 of ICE opamp?

 

[Misterrogers]

~12v, ~12v and ~24v. I've also noticed that when I turn the volume all the way down it'll unlatch. Turn it up and it'll re-latch, usually with only the right channel.

 

[cfcubed]

You are in advanced territory Mr Rodgers  Perhaps obvious but If you've a spare standard ICP regulator (7824) around would try swapping that in for the Belleson Super Regulator.
 
In hindsight it might have been good to get the "standard" build going w/7824 then try using the super reg.   Even tho the current problems may be unrelated to the reg (e.g. the usual full inspection & CTH debug steps & readings in my sig apply).
 
Edit:  One thing to consider is your coupling caps, both for the channel in/out/loss (solder joints) & nuisance e12 trips.   I found that no matter what I could not use fancy-pants V-cap TFTFs as C4s in CTH w/o e12 trips.

 

[sachu]

I honestly would get rid of the E12 if it were for a personal amplifier. The risk you run of course is busting your drivers if the amp goes kaput. 

 

[cfcubed]

Quote:

I honestly would get rid of the E12 if it were for a personal amplifier. The risk you run of course is busting your drivers if the amp goes kaput. 

Would not be my recommendation, if CTH is constructed properly & e12 is tripping it is just doing its job   If one wanted to limit e12's behavior to only catastrophe avoidance, nearly disabling it, they could raise the "sense" resistors (R1E, R2E) to some high value like 100k and the "speed" resistor (R4E) to say 3.3K.
This would leave some degree of protection against build errors, shorts in case, etc.  And retain a bit of power on delay and power off DC protection.

 

[Misterrogers]

Ok. 'Downgraded' to the standard ICP and C4 caps - the good news is I'm still missing my left channel. Gotta step away, but this gives me hope that the circuit can handle one or both of my upgrades - once I sort out my channel issue. I'll start measuring when I can sit again, but I'd appreciate any direction you'll can provide.

 

[Misterrogers]

Voltages are good up to tube socket pin 1 and 6. 1 is ~94v and 6 is 77v. Setup says that either the tube opamps are bad, or I have a wrong component in. I assume the tube opamps are ICL and ICR? What voltages should I read at what pins to vet those? Thanks again for any/all help.

 

[cfcubed]

Since you have one working channel & one not this should not be so bad to debug.  Just take great care not to short things tho, using fine needle probes (I use pins taped to mine only exposing the pin tip).
 
You are correct about the ICR/L and can use the schematic here:
http://www.cavalliaudio.com/diy/cth/main.php?page=schematics/ipschematic
along with the opamp datasheet to discover the opamp pin functions.    You can see that the tube in conjunction w/ICR/L's influence on cathode bias is what brings the raw TB+/100v (94v in your case) down to about the 80v you should be reading on pins 1 & 6.    SO your 94v reading (your left channel) is like what we get when no tube is inserted, IOW if you pull the tube you should get about 94V->100V for both pin 1 & 6 WRT ground.
Thus in addition to your suspicion of ICL as the culprit (e.g. swap ICR & ICL as a test?) you should re-confirm the tube socket soldering & the tube itself as good. 

 

[pabbi1]

Will 6aq8 work here, or is the mu too high?
 
Getting close on mine... maybe tomorrow.

 

[Mullet]

The 6AQ8 might work. If the pin outs are compatible and the plate voltage can accept around 80v then it might work. What mu does the 6AQ8 have?

 

[cfcubed]

6aq8's 57 mu is way too high, would probably have same probs 12at7 has - search thread.
See CTH tube info link in my sig. And with your DIY history hard to believe you don't have a 12au7 or 6922 class tube laying around

 

[pabbi1]

Honestly, I had sent all my excess tubes to friends (spring cleaning(s)), or with amp projects that found new homes, and only happened to have one of these around by accident. In searching the list, and searching that thread, I only saw a passing mention of 6n23p, and nothing on 6h30, though I hear whispers the latter is quite good (and which I have several of).
 
The 6n1p I have found to be noisy in other amps (Bijou), that will settle down, but nowhere as good as the 6n23, so... just wondering.
 
 

 

[sachu]

I have a very nice RCA 6CG7 clear top. Will send it your way. 

 

[pabbi1]

OK, e12 works fine, LED goes red to green, SG->og is 12.04v, OG to L & R out is 12.03v, TB+ is 97v, and...
 
Pin 1 to Pin 6 is .6v - across 2 tubes... just making sure, but heater switch "ON' is 6.3v with the default resistor values, using H1 & H2?
 
In the meantime, here is a money shot: