[cobaltmute]

Quote:

 
So the question is, where can this kind of hum come from? If it's not buzz then it's not coming from the PS circuitry. If it's just plain 60Hz hum then it has to be coming from the AC components somehow.
 
I guess you don't have a scope?

 
Unfortunately, no scope.  Never quite justified the price versus a project I was working on.
 
Quote:

My guess as well, if not source impedance maybe headphone impedance.  Perhaps not related to this amp, but I'd oscillation much like you describe (e.g. vol positions) w/my ß22, only w/250R DT880s, before I added Zobel network.   I'm going to strike this... It's only AFAIR, was a while ago & might not be entirely accurate.  Also that ß22 problem, unique to use w/my 2003 DT880s, was more of a squeal than a hum.   But it did come & go with vol positions like you describe.       Carry on

 
This is a good lead.  300Ω HD650 has the issue.  33Ω Cheap Senn earbuds exhibit a low level uniform hum through the whole volume range.
 
I think my next course of action will be:
 
1 - take the board out and clean it again and just do a check for any shoddy joints.  As jdkjake said, I'll check my joints on the e27 as well (although I think I'm going to make a board to replace it that has split grounds).
2 - ensure all my wiring is good, including grounds.
3 - if still having an issue, add a zobel.
 

 

[cobaltmute]

I got lazy:
 

 
and I have a dead quiet amp now.
 
0.015µF in series with about a 5Ω resistor.  No idea if those values are great, but it was the best I had in the boxes on my bench.

 

[jdkJake]

"It's not a habit, it's cool, I feel alive..."
 
I knew I should never have put those sockets into the amp to play with the gain. 

 
After much experimenting and endless resistor swaps, I have come almost full circle. I am back to the original 75K at R13 and R14 to bring the NFB setting back to the value Alex intended it to be. This setting brings back the sound I found so compelling in the first place. Don't get me wrong, by reducing the NFB, you can make the amp loose as a goose and as "exciting" as you want, but, to these ears anyway, the most balanced, accurate sound is at the values Alex originally specified (no surprise there I suppose). 
 
I did reduce the closed loop gain a bit. I am using 1.8K at R11 for what I calculate to be an overall gain of ~6.5. This gives a bit more play on the pot for my grados and puts the HD650 right into the sweet spot with plenty of room for obscene amounts of volume. Nice.
 
In any case, time to step back from the churn, button it up and just enjoy the amp. Really, I mean it this time... 

 

[jdkJake]

Quote:

...and I have a dead quiet amp now...

Are you going to leave it like that or try to track down the source of the problem?

 

[Cloud]

is this a ground loop breaker between PS ground and amp ground?
 
Quote:

I got lazy:
 
 
 
and I have a dead quiet amp now.
 
0.015µF in series with about a 5Ω resistor.  No idea if those values are great, but it was the best I had in the boxes on my bench.

 

[fishski13]

sachu,
beautiful build.  good to hear that the Hammond EI are free of noise. 

 

[Lil' Knight]

Is it a must to have the e12 in the amp?
 
Also, I have a spare Avel +/-30v 80a left from my B22 build. Is it ok to use it with the EHHA?

 

[jdkJake]

Quote:

is this a ground loop breaker between PS ground and amp ground?
 

 

 No, it is a Zobel network between signal output and ground.

 

[jdkJake]

 
Quote:

Is it a must to have the e12 in the amp?
 
Also, I have a spare Avel +/-30v 80a left from my B22 build. Is it ok to use it with the EHHA?

No, the e12s (plural, you would need two) are an optional component. They protect your headphones from excessive DC offset upon power-up as well if something should happen to the amp during use (unexpected circuit/part failure).
 
 
Is that transformer REALLY 80A? Even 800mA would be too much for this build. 

 

[cobaltmute]

Quote:

Are you going to leave it like that or try to track down the source of the problem?

 
I'll spend some time to try and track it down.  I'll still do the steps that I said, but if it doesn't work at least I know I have a solution.

 

[runeight]

If the zobel is from O/P to ground that particular solution shouldn't really work. The roll off frequency of the zobel should be above the audio band. If it isn't, then you loose HF response. If it takes out 60Hz or 120Hz then it is killing freqs in the audio spectrum. Something else must be taking place, even with the zobel in place, that is not related to the behavior of the zobel. That combination that you've chosen might even put a bump on the HF end of the spectrum.
 
I don't think that this addition can really be fixing the problem unless it is fixing a grounding problem or something similar.

 

[Lil' Knight]

Yeah, I still have a couple of E12s left from the last build. Just wondering if it's a must to have them.

My bad, the traffo is actually 80VA.
 
Quote:

 


No, the e12s (plural, you would need two) are an optional component. They protect your headphones from excessive DC offset upon power-up as well if something should happen to the amp during use (unexpected circuit/part failure).
 
 
Is that transformer REALLY 80A? Even 800mA would be too much for this build. 

 

[jdkJake]

Personal preference on the e12's. Then again, if you already have them, why not, right?

80VA is too much. Ideally, you are probably looking for something around 30VA, perhaps a bit higher depending upon the manufacturer. Unless you are going balanced or dual mono with a four board solution. Then it might be okay, still a bit high, but closer in the ball park. What is the part number?

 

[Lil' Knight]

How about a 50VA one? It's the Avel-Lindberg Y23 series that's recommended on the o22's web page. Since I'll order the boards from GlassJar, I might ask them if they still have a 50VA one on hands.

 

[sachu]

That works...but get the 25+25 secondary one instead of the 30+30 if your AC mains are steady at 120V. If its at 115VAC then you can go for the 30+30 transformer.