Help with S.E.X

[jeffrey]

Well, I managed to fininsh the s.e.x. However, I've still got a few problems. There's a definite hum that comes out of the amplifier (as in I can hear the physical amp humming, not coming out of headphones or speakers), and a hum that I hear through both channels through my headphones only, and I do not hear it through my speakers. It's considerably worse through one of the tubes (If I switch that tube to the other channel, the hum follows it), but I do hear a little bit coming through the other channel as well. The humming does not change with the volume, it stays constant no matter what.

For reference, the two headphones that I used it with are the sr80i (32ohm impedance) and k702 (62ohm impedance), and I used it with 90db 8ohm speakers. 

Could the humming through the headphones be what the 2 120ohm resistors are there for? If so, where do they go? If not, where should I be looking?

Is the humming that I hear coming out of the amp normal? If not, what could be causing that?

[Paul Birkeland]

If the amp itself is humming, that's generally an indication of a load issue.


How were the voltage measurements? 

Did you double check that all the caps are in correctly?  A backwards cap can make the amp hum quite a bit until the cap blows.

[jeffrey]

The caps are definitely oriented correctly. I've triple checked them against the pictures in the manual.

I found that somehow I'd not soldered one of the wires coming out of the output transformer. I'd forgotten to solder one of the wires out of the output transformers, from B3 to terminal 10 of the output transformer. I've resoldered it, but I think I might have shorted something since. I broke one of the probes on my cheap multimeter (I think it's a sign that I should really go and buy a decent DMM ) so I wasn't able to get all of the measurements, but they're pretty far off of what they should be.

(I measured 140V from the wall)
All measured from terminal 23:
1: 300V
2: 0
3: 0
4: 600
5: 295V -> 300V (Between)
6,19: 600V,600V
7,17: 0V,0V
8,18: 0V,0V

15,25: 0V,0V

A2,B2: 600V,604V (These fluctuated a bit, so they're approximate)
A5,B5: 522V,521V

Once I get back, I'll fill in the rest of the measurements (Along with resistance checks as well)

EDIT: The tubes no longer light up

[jeffrey]

First, I should say that it is working now. I found a short with two of the resistors on the 'A' tube which I fixed.
Having soldered the black wire from the output transformer, the sound coming out of the amplifier is quieter -- I have to put my head pretty close to the transformers to hear anything. I still have the problem with the humming in both channels (One of them is considerably worse, probably due to the tube).

There is actually a slight hum in the left speakers (the same channel that has a very noticeable hum when using headphones), it's just very hard to hear.

The resistance checks are pretty much spot on. The voltages look to be okay
Here they are, from Terminal 23 to:
1: 227V
2: 0V
3: 0V
4: 457V
5: 227V
6: 433V
7: 0V
8: 0V
9: .2V
10: 454V
11: 410V
12: 0V
13: 0V
14: 0V
15: 20.8V
16: 434V
17: 0V
18: 0V
19: .019V
20: 455V
21: 413V
22: 0V
23: 0V
24: 0V
25: 18.2V
C1: -3.5V
C2: 3.21V
C3: 0V
C4: -3.6V
C5: 3.4V
A1: .02V
A2: 406V
A3: 20V
A4: 0V
A5: 80V
A6: 2.6V
A7: -3.6V
A8: 3.1V
B1: .01V
B2: 414V
B3: 18V
B4: 0V
B5: 76V
B6: 2.6V
B7: -3.5V
B8: 3.1V

[Paul Joppa]

Does your meter have a 200mV AC scale? If so, you can measure the AC hum voltage at the speaker terminals; then repeat that with the tubes exchanged. This will help determine if there is a wiring error or a tube problem.

Another thing to check is the orientation and soldering of the heater power diodes (the fat Schottky ones behind the power transformer on the right). If one is backwards, or if there is a solder failure, that can cause some DC in the winding which introduces mechanical hum in the transformer. The diodes have heavy leads which conduct heat very well, so it requires  more heat to get them hot enough for a good solder joint - this make it an easy mistake that is also difficult to diagnose.

[Jim R.]

And you might want to replace the batteries in your meter -- flakey readings are often an issue with the bateries being too old and weak.

-- Jim

[jeffrey]

The diodes are correctly oriented (The two right diodes have the striped end facing to the right, and the two on the left have the stripe to the left. This is from the perspective shown in the picture in the manual). I checked the continuity between each of the diodes and the terminals that they are connected to, and they all show 0 ohms between them.

Sadly I don't have a meter that shows AC mv's. But if it were a wiring error, wouldn't the buzzing/humming stick to one channel no matter which tube was in that channel?

Edit: The humming is approximately 120hz. It's about two octaves down from the 'A' above middle C (440hz)

[Paul Joppa]

I was thinking of a noisy heater supply, which might couple from heater to cathode differently in different tubes, but still have its source in a faulty heater supply.

Measuring the hum is for the purpose of determining whether it's unusual (a building error somewhere) or "normal" at least for one of the tubes, in which case a replacement tube is in order.

In the original design, the 120 ohm resistors would reduce the sensitivity of low-impedance phones. It's a standard, though kind of out of date currently. The resistor has the effect of messing with the bass quality for some low impedance phones, and the most appropriate switching headphone jack seems to be not in production anywhere these days, so we dropped it. But that means sensitive low impedance phones, which are no longer always designed to work well with the traditional standard but are optimized for iPods and similar gear, will expose the noise floor much more than before. Hence the question about whether this is an assembly error or not.

[jeffrey]

Okay, I just received two new 6dn7's. The really loud humming in one channel is gone, so I think it might really have been the tube.

The only thing I hear now is a really quiet humming coming from both channels equally in my headphones. At this point I'm willing to put in the resistors to see what happens. Where should they go?