Hi

I've been piecing together plans for a (cheap) amplifier that will ultimately end up in my office at work. It's going to be a push pull built out of an ST-70 chassis, output around 2w for small speakers/possibly headphones. So far it looks like I'll only be using half the chassis, so I thought it might be a cool idea to try to fit the DAC in and make it complete. I'm looking at the Opus voltage out DAC right now.

I made it through the whole development thread (Whew!). Looks like it's been a fun ride. I had a few questions that I couldn't find on the data sheets if you have a moment.

1) Is there any DC component to the differential outs? I'd like to avoid cap coupling if I can. I could even compensate in the cathodes as long as it is a constant DC.

2) Being as the DAC will be the only input (no phono at work :P) I was thinking of utilizing the built-in attenuator instead of putting in a passive between the DAC and driver section. I have a basic2 stamp that has been collecting dust since high school... I should be able to whip up a program for the DAC volume control (and some fancy-pants LED stuff or IR remote while I'm at it hehe). The attenuation must be controlled in software mode as I understand it. Does it completely ignore the hardware switches when in software mode? Will I need to account for setting all the registers on startup in my program?

3) One of those annoying semantics things that always bites me if I don't ask... Is the listed 2v rms (5v supply) output measured on each output with respect to ground, or with respect to each other? I need to make sure I don't go overboard with gain.


Thanks for any help!
Chris

Really? I don't know much about I/V conversion, but I do know that a current output DAC likes to see the lowest possible impedance. I'm guessing the compliance is well within 2v if there is a passive on board though. The only thing I could think of for tubes is to have a low value resistor as an I/V and then send that to the grid.. which is admittedly not an ideal solution. It would offer me finer control over the gain structure though.

I saw some circuits based on the 6em7 and reports that it sounds pretty good, so that's what I'd like to try. The chassis I have will already have a power supply, 4 octal sockets, and OPT's. So I can build the amp for the cost of the 6em7's and wire. The nice thing is with the 6em7's, the amp takes up very little space and the whole "driver section" of the ST-70 chassis is left open for a DAC. The gain structure looks like ~45-50 in the driver stage. Output probably needs a swing of around 45 volts for full power, so I'll need around 1v peak-peak from the DAC.

To get that from the COD will require something like a 130R resistor across the DAC outs. I unfortunately don't know much about DACs, but it seems like it wouldn't be the ideal solution. I don't know. The Wolfson chip does that work for me.. and likely in a better way.

Trick would be getting around the DC component.. as long as it is constant I can simply up the cathode resistor to compensate.. but will it stay constant with the on-chip attenuator? Ugh. Of course.. cap coupling would eliminate all problems... but what fun is that? :P

Edit: Heh, I replied before seeing your edits... Yea, I was going to do a semi-differential amp. The driver and output stage will be direct coupled. That will allow me to elevate the output tube to use the 400v B+ directly on the plate and also provide a longer tail for better symmetry.

Edited by user Wednesday, March 26, 2008 7:36:01 PM(UTC)  | Reason: Not specified