[Punnisher]

Very cool design.

Excuse my ignorance, but how hard would it be to hook up a single ended source and/or headphone to this amplifier?

Also, any news on boards becoming available for purchase?

 

[grawk]

You'd have to use some external method to balance them. This circuit is differential only.

I like the cinemag CMLI 15/15B transformers for that.

 

[Steve Eddy]

Quote:

Originally Posted by grawk /img/forum/go_quote.gif You'd have to use some external method to balance them. This circuit is differential only. I like the cinemag CMLI 15/15B transformers for that.

Except the secondaries on the 15/15B aren't balanced.

To get a balanced output with which to feed the balanced input of this amp, you'd need to use a pair of them per channel with the bottoms of the secondaries tied together (along with the shields/cans) as a quasi center tap.

You could use something like a CMOL-2x600T2 (which has both balanced primaries and secondaries) and have just one transformer per channel, but if memory serves, the price of a single CMOL-2x5600T2 is more than a pair of 15/15B's.

se

 

[jcx]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif Except the secondaries on the 15/15B aren't balanced. ...

that's OK - the amp's input is differential with balanced input impedances (AC and DC to the Vcom active gnd) and its own DC bias path so a floating source like the xfmr secondary would be just fine

just maybe not the ultimate possible feedthru/common mode rejection that could be had from a CT balanced sec with core, shield and CT gnd


in fact the active Vcom servo lets the amp directly take SE input as well - just connect the neg inputs to source signal gnd

there is a little weirdness with the Vcom servoing the output of the op amp while there is a follower with Vgs offset that changes with signal level between the op amp output and the feedback R giving a distorted common mode V at the amp output - but at a fairly small amplitude if the output Q are biased Class A

 

[Steve Eddy]

Quote:

Originally Posted by jcx /img/forum/go_quote.gif that's OK - the amp's input is differential with balanced input impedances (AC and DC to the Vcom active gnd) and its own DC bias path so a floating source like the xfmr secondary would be just fine just maybe not the ultimate possible feedthru/common mode rejection that could be had from a CT balanced sec

Mmmm. But with the asymmetry, I don't think it would quite float.

Actually, what you could do is just flip it around. Drive the secondary from your unbalanced source and drive the input of the amp from the balanced primary.

Win/win.

se

 

[Steve Eddy]

Damn it, jcx! Hold still, will ya?

That's what, your third edit?

se

 

[jcx]

well I'm not competing for post count

I try to label meaning changing edits but just adding info that shouldn't invalidate anyone's comments to a post while the discussion is on the same page cuts down on clutter

 

[Steve Eddy]

Quote:

Originally Posted by jcx /img/forum/go_quote.gif well I'm not competing for post count I try to label meaning changing edits but just adding info that shouldn't invalidate anyone's comments to a post while the discussion is on the same page cuts down on clutter

I know. I was just being facetious.

I read one thing in the EMail notification, a second when I read it here, and a third after I'd posted my reply.

se

 

[luvdunhill]

grawk:

it would also depend on the input impedance of the attenuator and how much it varies.