Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › SOHA II troubleshooting
New Posts  All Forums:Forum Nav:

SOHA II troubleshooting - Page 2

post #16 of 28
11/25/09 at 1:08pm
The difference just means that the tubes are not perfectly internally matched, but those voltages are not out of line.

If your power is perfect then there must be something wrong, obviously, on the board. I know that you've done this already, but I think a very thorough look at all of the wiring and solder joints would be the next step. Use a magnifying glass if you have one. The first guess is that there is some intermittent connection somewhere that may be temperature or vibration dependent.

If you don't find anything from this step then we'll have to try some other troubleshooting that may be a bit more challenging.
Reply
post #17 of 28
11/26/09 at 2:58am
Quote:
Originally Posted by Johnthurston View Post
When I get down into the "tail current for the input stage", I find the difference between TB+L and TPL on one channel is only about 15V, while on the other channel it is 20.4V. The target is 19V. This measurement moves between the channels when I swap the tubes between channels.

What is the significance of this voltage, and (how) does this relate the periodic cut-out I've been experiencing?

John Thurston
Juneau, Alaska
First the tube at 15V CCS drop is no good replace the tube. It is probably causing the tube to spit out some low hz AC which could trip the e12. The CCS needs 19V's to maintain constant current.

Also I actually found that my computer was causing the dropouts. Foobar would stutter a little when playing a CD and then the e12 would trip, very odd. But when I play from the harddrive never a problem. I also had a DIY headphone cable that was shorting at the plug that would cause the e12 to drop out, fixed that and all is well.
Reply
post #18 of 28
11/28/09 at 6:45pm
Thread Starter 
Quote:
Originally Posted by runeight View Post
The difference just means that the tubes are not perfectly internally matched, but those voltages are not out of line.

If your power is perfect then there must be something wrong, obviously, on the board. I know that you've done this already, but I think a very thorough look at all of the wiring and solder joints would be the next step. Use a magnifying glass if you have one. The first guess is that there is some intermittent connection somewhere that may be temperature or vibration dependent
I've pulled the board out of the chassis and gone over it with my loupe. It is sometimes hard to get a good view from the component-side, but I find no stray solder bridges or cracks. The worst thing I can find are several places where the solder does not fully fill the hole. From the back-side, I can identify no cold joints and (with one exception) they all appear to have nicely domed solder which covers the whole pad. So in the cases where the solder doesn't appear to fill the hole from the top-side, the back-side still looks like a good joint.

The exception was the ground-lead. It was added last so it was very difficult to reach between the heatsinks. I've re-flowed that joint from the back-side and added a bit more solder to fully cover the pad and flow into the (relatively) heavy, stranded ground wire.

The behavior of the amp has not changed Suggestions of where to go from here would be appreciated .
Reply
post #19 of 28
11/28/09 at 6:54pm
Thread Starter 
Quote:
Originally Posted by regal View Post
First the tube at 15V CCS drop is no good replace the tube. It is probably causing the tube to spit out some low hz AC which could trip the e12. The CCS needs 19V's to maintain constant current.
This should be relatively easy to check. Being a newbie at this, I don't have a drawer full of tubes from which to grab a replacement, but I can get some ordered up.

In them meantime, is it possible to test this hypothesis by running the amp on a single channel? Put another way; if the tube is to blame, should I expect the amp running with only the suspect tube to exhibit the cut-out problem, or will removing the other tube change everything too much to mean anything.

Quote:
Originally Posted by regal View Post
Also I actually found that my computer was causing the dropouts.
This was my first suspicion many months ago. I have discounted it because I see the same behavior regardless of the signal source. I've used my squeeze-box, two different computers, two different CD players, and one pre-amp. I don't think it is caused by the source.
Reply
post #20 of 28
11/28/09 at 7:18pm
Hate to drag you through this but what happens if you disconnect the source and let the amp sit. Does it drop out? What happens if you bounce it around with no source? Does it drop out? What happens if you rotate the volume pot fast?

Unfortunately, running with one tube won't be successsful because of the way the circuit works.
Reply
post #21 of 28
11/28/09 at 8:24pm
Thread Starter 
Quote:
Originally Posted by runeight View Post
Hate to drag you through this but what happens if you disconnect the source and let the amp sit. Does it drop out? What happens if you bounce it around with no source? Does it drop out? What happens if you rotate the volume pot fast?
If you are willing to contribute your time, I am certainly willing to have you drag me through the weeds. I hadn't thought to try running the amp with no signal input....

When I disconnect the signal and let it sit at a low volume (10%) it didn't drop out.

When I rapidly rotated the pot up to full, it didn't immediately drop out, but after about 15 seconds it did so. It remained out for a few seconds, cut back in, dropped immediately out for several more seconds before cutting back in. I have been able to reproduce these symptoms several times.

I then slowly (over 15 seconds) rotated the pot up to 75% and let it sit there. After about two minutes, it cut out for several seconds.

What does this tell us?

Please recall that per an earlier discussion on this amp, I'm running R4E at 3k rather than the stock 2k.
Reply
post #22 of 28
11/28/09 at 8:49pm
Please drop it from about 1" height while powered up with no input and the vol pot at zero. Does it drop out from physical stress? I don't think so, but let's be sure. Slightly twist the enclosure. Does it drop out? Push on the tops of the tubes. Does it drop out?

When you do the vol pot experiments put a voltmeter at TB+. See what happens. Then do the same for TB+L and TB+R while jacking the vol pot around. Let's see if any of these move around a lot.
Reply
post #23 of 28
11/28/09 at 10:25pm
Thread Starter 
Quote:
Originally Posted by runeight View Post
Please drop it from about 1" height while powered up with no input and the vol pot at zero. Does it drop out from physical stress? I don't think so, but let's be sure. Slightly twist the enclosure. Does it drop out? Push on the tops of the tubes. Does it drop out?
I am unable to correlate drop outs with twisting, pushing, dropping or knocking.

Quote:
Originally Posted by runeight View Post
When you do the vol pot experiments put a voltmeter at TB+. See what happens. Then do the same for TB+L and TB+R while jacking the vol pot around. Let's see if any of these move around a lot.
I don't detect any change on TB+ with the cut-out event.

The behavior on TB+L/R is a little different. The values on them do not change as I rotate the pot, but if I crank the volume up and leave it there, the voltage (on either) starts at about 61.3 and climbs very slowly until the cut-out occurs. When this happens, the reading is 62.2 +/- .1 Both TB+L and TB+R behave the same way.
Reply
post #24 of 28
11/28/09 at 10:36pm
Do they stay there? Or do they drop down back to 61V after some time?
Reply
post #25 of 28
11/28/09 at 11:21pm
Thread Starter 
Quote:
Originally Posted by runeight View Post
Do they stay there? Or do they drop down back to 61V after some time?
They drop back
Reply
post #26 of 28
11/29/09 at 12:21am
Would you kindly repeat the experiment measuring TPL and TPR. Also, what tubes are you using?
Reply
post #27 of 28
11/29/09 at 5:55am
Thread Starter 
Quote:
Originally Posted by runeight View Post
Would you kindly repeat the experiment measuring TPL and TPR. Also, what tubes are you using?
I'm using 5963 tubes.

I'm going to re-do this experiment in the morning when I'm more alert, but my preliminary results are...
When I monitor TPL and TB+L with a pair of DMMs, TB+L climbs slowly as previously reported. The cut-out occurs but TB+L does not immediately start falling back. I'm going to have to reproduce this in the morning because this is different from what I have recorded from my previous experiment. TPL hovers around 40.1 but climbs to 41 +/- .1 immediately preceding the cut-out.

So there are the preliminary numbers. I'll try to reproduce the results and compare L/R tomorrow.
Reply
post #28 of 28
11/29/09 at 11:46pm
Thread Starter 
And today, the amp is behaving itself better so the cut-outs are much farther apart, but here is what I found :P

Code:
TPL+/-TB+L+/-
40.0.161.3.1
These measurements just don't vary much. They drift around by a tenth of a volt, but its very slow. Even during cut outs, I'd consider those numbers to be stable.

Code:
TPRTB+R
45.061.2Stable about 15 mins elapsed
45.1-45.461.4-61.9Climbing slowly about 10 mins elapsed
45.6-45.862.2-62.3Cutting in and out
45.6-45.762.0Cutting in and out

45.562.0Stable again for 5 mins
These measurements were taken with a pair of DMMs connected to either TPR/TB+R or TPL/TB+L The cutouts are usually too brief to permit getting a reading from both the left and the right channels during the same event.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › SOHA II troubleshooting