[jcx]

easiest looking to me is to lift 4k feedback R, tack solder a shorting wire across the 470k to get a unity gain circuit around the OPA627 - see if DC V measures OK
 
the phase shift of the 1 K in the buffer input looks like its not a problem this time, the pole with the buffer input C ~50 MHz, is comfortablely above the OPA627 16 MHz GBW

 

[shake]

I cut the connection between the 1K resistor to the buffer.  Measuring from the resistor to ground, I get -11.4V.  And from ground to buffer input, I read -14V. Where am I derpin?
 
 
Quote:

easiest looking to me is to lift 4k feedback R, tack solder a shorting wire across the 470k to get a unity gain circuit around the OPA627 - see if DC V measures OK
 
the phase shift of the 1 K in the buffer input looks like its not a problem this time, the pole with the buffer input C ~50 MHz, is comfortablely above the OPA627 16 MHz GBW

?

 

[jcx]

check V on the supply pins at the op amp, buffer, op amp input, output
 
 reconnect the buff input to gnd, check the buffer input , output V - if you do it through the 1 K series R you can then "pull" the input node with another R clip-leaded from either rail or from your function generator

 

[shake]

Quote:

check V on the supply pins at the op amp, buffer, op amp input, output
 
 reconnect the buff input to gnd, check the buffer input , output V - if you do it through the 1 K series R you can then "pull" the input node with another R clip-leaded from either rail or from your function generator

I'll try that tonight, but I don't have a function generator.
I should also note that when I'm measuring the rails on my power supply, one side measure 14.8-14.9V and the other 14.6-14.7V.  Would that be a problem? I've checked and rechecked and rechecked it again and the wiring is correct, no shorts.

 

[shake]

I completely rewired the power to each buffer and op amp.  Each buffer and op amp are pulling 29.7V
I removed the bass boost and the offset dropped to 700mV in the right channel and about 30mV in the left channel (both from 13V).
Why would that have happened.?
And it plays sound now! Right channel only.  Left one is dead.

 

[Avro_Arrow]

Lets try building something a little simpler to start...
This uses the parts you already have.
 

 
You can add the extras later.

 

[shake]

Quote:

Lets try building something a little simpler to start...
This uses the parts you already have.
 

 
You can add the extras later.

Besides biasing and the bass boost, what would I add to that if it worked?  And what can I do to reduce dc offset even more?

I haven't had time to work on the amp recently, with work at the lab, work, and studying for the mcat, but I think one of the op amps I got was bad.  Digikey is sending a replacement now, kudos to them.

 

[G.Trenchev]

Add nothing.If you follow Avro_Arrow's schematic exactly,you will get less than 1 mv offset.

 

[shake]

with 1mv offset, adding bassboost should be safe though, correct?

 

[shake]

I rewired it the way avro suggested and the offset is still killing my headphones.  The left earpiece actually melted this time (using throw away earbuds).  The last culprit I can think of is the make a short on the buffer, but I can't see one.  The right channel has a 300mv offset, and I'm measuring up to 12V dc from the left channel's buffer output.
 
Should I just chalk this up to a $250 fail?

 

[Avro_Arrow]

That's an awful lot of money to spend just to have a "fail".
 
At this point we need to find out if your components are fried or working.
 
Use two fresh 9 volt batteries as your power supply.
If you don't have a couple of 9 volt battery connectors, just solder the
wires right to the battery. Don't worry about bypass caps or any other
components.
 
Wire the op amps like this to test them..."out" should be very close to zero.
 

 
Wire the buffers like this to test them.
Being open loop buffers, they will be some offset, but not much.
 

 

[shake]

I might just swap them out for BUF634s or the LME49600.  The lmh6321 is just too dang small.

 

[shake]

So I received the replacement opa627, got rid of the bass boost circuit, got rid of the buffers, and tidied up the circuit a bit more.
Damn, the opa627 sound fantastic  First thing I noticed was how I could feel every sound, not just the bass.  There's still a grounding issue that I need to sort out though. 

 

[Avro_Arrow]

Good to hear that you got it working...now just add BUF634...
I have OPA627 in my PPA.

 

[shake]

Would adding a buffer really benefit it?  It sounds fine as is, but I thought what I had sounded fine too... lol
At louder volumes (pot past 50%) I can hear humming prominently in the right channel.  What could be causing that?  When I get closer to and touch the volume knob, the noise increases.  With music playing, it goes away until I pause the music.
What's really weird is the noise gets louder even if I'm touching the pot that isn't connected to anything...
Should earth ground be connected at all?
 

 
The reason there are two knobs is cause I couldn't remove the bass boost pot, so I thought I might as well make it look fancy.
 
edit: Ok, it appears that the body of the alps pot is not connected to the ground pins, but still must be ground...