Problem w/ST-70...looking for suggestions

[joeriz]

My ST-70 recently developed a problem in that the right channel has less high frequency extension (and perhaps less volume overall...but if so, not by much) than the left. I have traced the problem to the inside of the amp by switching the interconnects coming from the preamp, switching the speaker leads, and switching all tubes. The problem follows the speaker leads.

I re-built my ST-70 about 10 years ago with the Welborne Labs kit so it's not like it's a 40 year old in need of a complete rebuild. However, I supposed something inside has gone wrong. I'm just looking for suggestions as to where might be the best place to start looking. The Welborne mod makes it somewhat difficult to swap out parts so I'd like to keep that to a minimum although I realize that may not be possible.

I dunno....coupling caps, perhaps?

Thanks,
Joe

[mesherm]

If the problem follows the speaker leads maybe the problem is with the speakers not the amp. Run one speaker in the left side with an 8 ohm resistor on the right side instead of speaker. Then connect the other speaker to the left side and see if the speakers sound different. Do the same except with the right side. If it is in a speaker then open the speaker up and see if a tweeter or a crossover cap has gone bad.

[joeriz]

Thanks mesherm but I'm not following you. If I swap the speaker leads on the back of my amp and then my right speaker gains highs while the left loses highs, I would think that would indicate that the problem is in the amp, no?

Also, I neglected to mention that I also swapped speakers and the problem stayed in the right channel.

[mesherm]

Just a misunderstanding. I assumed when you said the problem follows the speaker leads that you ment that the right speaker (for example) would sound muffled even when its leads were switched to the left side of the amp thereby "following" the leads. But you have clarified by saying that a speaker swap doesn't swap the problem.
OK so you have tried tube swapping and it didn't follow the tubes.
You have tried interconnect swapping so its not at the source.
So if its inside the amp its either in an output trannie ,the driver board, the input terminals or the speaker terminals (feedback wires).
Start with easy to check stuff first, such as RCA and speaker binding post connections. Something loose or not grounded, a solder joint gone bad in the feedback wire.
If that checks then the solder joints on the driver board are next to check.
If nothing is obvious then I suggest feeding a 1000 hz tone through both channels in mono mode and checking at various points with your DMM set on AC volts. Most new meters will read something at 1k hz and if you check the same point in the right and left circuit you should get the same reading. You don't really care if it is absolutely accurate, you are comparing left with right. Sooner or later as you work your way through you should be able to isolate where the problem is. The fact that it is affecting the highs does seem to suggest its a capacitor problem. Check any electrolytic bypass caps first.

[joeriz]

Thanks for the tips.

So I would just feed a 100Hz tone from a test CD into the amp with my preamp set to mono? From there I can test any point in the circuit for a reading? Do I test in reference to ground?

Sorry, I've never had to do this level of sleuthing before.

Thanks again.

[mesherm]

Well you have said that the problem seems to be with the higher frequencies thats why suggested using a 1000 hz or maybe even higher.
I don't know whether 100 hz would be high enough to notice a difference.
Look at the schematic and pick out some points along the signal path from input to output maybe marking them on the driver board in advance.
Then, yes, connect your DMM black lead securley to circuit ground and probe with the red with the meter on AC volts. The voltage should increase as you get closer to the output tubes but it should be very close with mono signal right to left. Those are the differences you are looking for. Starting from the inputs you will have maybe 100 to 500 millivolts of signal ending with maybe 40 volts or so at the driver tube grids. The readings right to left should be fairly close as you start and then as you pass the problem area you should see a difference. All this assumes the high freq rolloff is measureable. You might have to increase the frequency to 5000 or even 10000 hz. Even if the meter isnt that accurate at those frequencies, you are looking for differences in left to right. An O-scope would be perfect for this but I am assuming that you don't have one of them beasts.
You might even before this do the same thing but with the DMM on DC volts. Checking the DC voltage with inputs shorted or mute, at each of the driver board tubes plates and cathodes. That might also find a possible source of the problem if its caused by a bad resistor. Chances are though since the higher freqs are affected that is a capacitor or something in the signal path. I'm not familiar with the Welborne modified circuit so I can't suggest possible good test points.
If you have a CD test disc you might run a frequency sweep for 20 to 20k hz and see if you can tell at what frequency the rolloff starts to occur.

[joeriz]

Thanks! I appreciate all of the help.