Troubleshooting help needed for brand new VTA ST-120 - solved !

Still my question : why not return it to the seller ( or send it to bob ) for proper repair ??

That is a possible solution, but I’d rather try fixing it here first. Does anybody have any ideas?

This seems like a perfect logic problem. A person who really understands these amps should be able to look at the facts and surmise what the problem is.

alanhuth wrote:That is a possible solution, but I’d rather try fixing it here first.  Does anybody have any ideas?  

This seems like a perfect logic problem.  A person who really understands these amps should be able to look at the facts and surmise what the problem is.  

You reported this 11 days ago. In that time frame the amp could have been transported
fixed and returned by bob latino. ( or the seller) .
Are you located on an isolated island ? Or very keen on guesswork ?

.

Does that mean you don't think it's worth posting this and asking if anyone has any ideas?

I would think that if there was a mis-wire or something like that, a systems-thinking person might be able to predict that based on the measurements. I'm guessing only a very limited set of possibilities could yield this result.

.

Here is what changed to make the amp work.  Sorry I have no idea why the images are so big:

The ground wire was just placed there to take the picture. it was soldered properly

alanhuth wrote:Thanks for the helpful suggestions, corndog.  My hearing is not great, but I can tell the difference between a very loud speaker you can hear outside the building and one which you need to put your head right next to the speaker to hear.  This is not subtle.

Is it generally understood to be a good idea to re-solder everything?  I was afraid that it might make matters worse - convert a good solder joint to a bad one.  Is that not a thing?

I ordered some LM334s.  We’ll see if that helps.  

Does anybody have a schematic or voltage chart for the OCTAL BOARD?  Bob doesn’t have one, and Roy is temporarily unavailable.  

Did you receive/install the LM334? I had this problem on my ST120 w/AB and 9-pin board.

It had been months since this occurred for me. I had a tough time getting my AB board installed. That's working fine now. Then after about 2-3 hours of running the amp, I heard a crackle and volume was lower on the LHS. I swapped the outer driver tubes, tried different driver tubes, swapped speaker channels and the problem followed. I opened the amp up, looking for bad joints but couldn't find any. I asked Bob and he said it could be the LM334. I got one a couple of months ago and installed it last week (summer fun doesn't include soldering). Problem fixed!!! Try it and be gentle with the heat with these.

I also used a tiny screwdriver to tighten some of the tube pin sockets and a dental brush w/ some DeOxit5 to clean them out.

If you can't touch up a joint, use some good solder braid, clean it with 99% isopropyl and do it over again. I put heat sinks on the IC's leads to be safe. I got my soldering re-education from YouTube.

Where is a good place to order LM334s? I have a spare quad of 6550s. Three spare driver tubes, and a spare rectifier. Sounds like spare LM334s would be good to have on hand. My VTA-120 sounds great, but I have red plated one 6550 and cooked a a 10 ohm resistor. Would prefer to able to repair it quickly if something else fails. I do enjoy the sound, and enjoy the engagement that doing my own repairs provides.

Bob sent me an eBay link to a seller in PA. It was about $6 and change for the one I got. I just saw them on amazon today as well.
In the past I've gotten components from Mouser but if Bob was ok with the Ebay seller I was too.

Yes the LM334s were replaced. No difference. One person with a lot of Dynaco experience said he thinks that it could be shorting inside the output transformer cap. Another idea I heard was that the reason the RHS is low is because something screwed up in the transformer may have caused the NFB/16ohm to overload the feedback on the board, causing it to shut down. Now that’s what I call Sherlock-level sleuthing. No idea how to test that theory though.

Did you get to the bottom of this? I'm happy to help. BTW: you're absolutely doing the right thing by 1) keeping the AMP. 2) Fixing on your own... Transformer post is what's relevant...

In the Midwest we call this a snipe hunt.

Your Primary transformer DC measurements sound OK.

IF YOU LEFT SECONDARY GROUND WIRE CONNECTED: then
Both secondary measurements to chassis sound too high. The low measurement is more correct, but I'd expect 10ths of ohms.

Measure speaker ground post to chassis: (should be effectively 0 ohms)
Measure speaker ground post to each tap: Should be tenths of ohms doubling with each tap

If ground to chassis is broken, that will break negative feedback and that channel will be louder.

As I recall, there will be 2 black wires on speaker grounding post; one to secondary, one to chassis. I doubt the latter on the loud channel.

IF YOU DID DISCONNECT the secondary ground: then
Possibly defective transformer. There should not be a path to ground for any second wire to chassis

I found a photo. The transformer GND is via a nut and screw right next to the output ground post. Check the torque and solder on those.

Is the LM334 a current source for the long tailed differential driver, same as the monoblocks?  I don't know the ST-120 specifically.  I'd say that would be a prime suspect, particularly since you've tried tube swaps.  Other suspects include the triode/ultralinear switch and the negative feedback loop (but if the NFB loop were wide open you would certainly be getting gross distortion on the louder channel).  Also check that all the output transformer leads go to the same places, left and right.

If you're getting a 'scope that's great.  Really all you need, other than a signal source.  Not only can you check for signal level differences but also changes in the waveform.  In a pinch you could use the cal pulse from the 'scope.  But if you don't have a steady sine oscillator they can be bought cheaply or built simply.  You can use a test CD as one person suggested.  Or an audio test app for a smart phone (I have "Audio Tool").  You'd then just need a stereo sub-mini to two RCA's cable.  When using a stereo source make sure the two levels match.

You don't really need a table of voltages or voltage gains since you have a stereo amp and can "comparison shop" between channels.  Drive the left and right in parallel and put the scope in "AC coupling".  If the signal source immediately nosedives, stop.  One of your inputs is shorted (a tube amp has very high input impedance, typically in the 100K ohm range).  Otherwise, check the output level of your first stage, left and right.  The DC blocking on the scope's  AC coupling setting should be enough for even plate voltages but if you don't trust it, clip a capacitor to the scope probe and use that, otherwise remove the clip and use the bare probe lead (may take a few seconds to float down to where you can see it if it has a large DC voltage on it).  If you get all the way to the driver (check both the inverted and non-inverted outputs) and find the discrepancy is there, check the cathodes of the dual triodes.  If there's a difference, left and right, it almost certainly is your current source amp.  Otherwise, you may have a bad output transformer.  With a schematic and pictorial with the pinouts labeled (which you would have gotten with a kit.  For your assembled kit you may have to beg Bob for those) all this should take you no more than 15 minutes.  Otherwise, you may have to surmise where the stage outputs are by consulting a tube manual, ARRL Handbook, etc.  May take you a bit longer then, but the circuits are  not that complicated so even that shouldn't take too long.  Therefore, I recommend it before shotgunning your op amps (which I don't believe were socketed).  Unsoldering even an 8 pin DIP is enough of a PITA where I would go with simple signal tracing first.

It goes without saying, but I'll say it anyway: use extreme caution while probing inside the amplifier chassis.  You should be able to probe with one hand holding the scope probe and the other out of the way or in your pocket.

Also, if you still don't have your scope yet and are nervous about probing for voltages in a live amp, one thing you could do is check all the corresponding resistor values.  When I build a kit I check each resistor with a meter--color codes are often ambiguous to my eye.  You don't say how it came to be that a "brand new" amp has this problem.  Was it factory built?  It should be under warranty.
But a kit built by an amateur could easily have a wrong resistor somewhere in one channel.

Good luck.
 
p.s. if measuring resistors in circuit they may not measure as marked, but they should be the same, left and right.

I'd not go anywhere near the long tail section until you understand the transformer differences.

Transformer differences are much more difficult to evaluate.  To go there, I'd consider unsoldering them, driving the plate windings with a signal generator, and looking at each winding and tap with the scope.  But major PITA, much bigger than unsoldering the op amp (which apparently has already been tried anyway).  As I maintained before, basic signal tracing can be accomplished in minutes without taking anything apart.

The physics behind a transformer may have some complexities, but IMHO, debugging them is dead simple. He has a measurement that indicates an enormous channel discrepancy and one that is incorrect for a transformer if the measurement has been properly performed. And it's a measurement that would explain the behaviour. On the other hand, All of the basics have been performed in terms of swaps, front to back. And the current source has been replaced. The transformer measurement should be explained or repeated. Since he used the chassis as the reference, and the secondary is grounded to the chassis, and the ground from Chassis to Transformer is weak by design (nut and screw), it's a simple and fast check without so much as plugging the amp into the wall.

Turns out that there was a short between the secondary and the transformer cap, which had been taken off for painting and then not re-assembled properly.  There were also a couple of less consequential problems that may have affected the measurements, but that was the main one.  

Thanks to all you Sherlocks for weighing in.  It's nice to know that we have some smart cookies here.