Pars

I've never seen pics of the one(s) that KG built, and I don't recall him saying what he used for a PSU in them. He's the only one I know of who has built a DynaFET. On another note, have board pics been posted yet?

I'm not sure I can remember an order where I didn't forget something? Oh, maybe the ones that I went over for several days before placing

That is very good advice. Make sure the amp etc. is turned off before moving/connecting/disconnecting ICs and other pieces of gear.

Happy Birthday!

In the board pics on amb's site, it looked like the bottom ground plane extended under the sinks. I couldn't tell for sure whether the holes were actually tied to the ground plane or not, and being at work, can't measure one.

I'm not vvs, but I'll take a guess. The heatsinks are probably tied to board ground (looks like it on the sigma22). Normally you would use a ground loop breaker, which is a low value resistor and x-rated cap (AC line) in parallel, from the board ground to the case (i.e., wall ground). The heatsink touching could potentially short this ground loop breaker out, rendering it useless.

It seemed to me that ep3 showed us she might not be as erased as they think? I liked the 3rd ep personally; probably even better than the 2nd ep.

If you slap a Burson in it, it will be like putting a "bend over" sticker on, cuz at their price, you just did Colin gave a very good answer, BTW.

Happy Birthday Nate! Now, go shovel that driveway! Just kidding...

I figured you meant 2U, just wanted to make sure no one started ordering 1U cases

A 3" case is 2U, not 1U (U = 1.75"). The 1U cases from modushop are 40mm; 2U are 80mm. The 2.5" heatsinks would make for a tight fit. And I concur on the pins... not knowing exactly what they were, I tried to desolder some in an MD10 I was fixing. I didn't try real hard, but I could not get them to budge. Threaded as the stock ones are is far preferable.

Maybe I missed it, but looking at his initial build pics on headwize, he used onboard heatsinks on the sigma22s?

I would think it would be difficult (or at least not particularly clean) trying to use off-board heatsinks on a sigma22, as the output devices are on both ends of the board. Maybe I'm thinking inside the box too much?

In the original schematic, it is a 1K feedback resistor with a 100 ohm to ground, so should be a gain of 11 ((Ra / Rb) + 1). Per amb's advice on headwize, in order to change gain it is probably preferable to increase Rb (the 100 ohm) instead of lowering 1K to prevent loading the output. Something like 221 or 330 ohms in series with the 100 ohm and shorted by a DPDT switch (4.1 or 3.3 respectively) might work. Not sure how to adjust the compensation cap in parallel with the 1K. Intuitively I would think its value would need to decrease?

The original schematic shows a 1K feedback resistor with a 10pf compensation cap. I'm not sure how to calculate gain in this, so I don't even know what the stock gain is. If it is like the dynalo/dynahi, it is probably around 11?

I was asked by Icarium to mention the possibility of incorporating selectable gain into the design. I don't need it, but wanted to bring it up as a possibility. My thought was perhaps two positions? I know there was discussion regarding this in the Dynahi build thread on HF a couple of years ago, and it is certainly doable. Would probably require a different compensation cap, etc. Any interest in this?

I wouldn't bet on it. These are almost certainly current (I) out DACs. All of the I-out DACs that I have seen have different quiescent currents that you need to deal with to put the I/V input at virtual ground with a very low input impedance (I-out DACs really want to drive a dead short or 0 ohm impedance load). The Buffalo can operate in either current out or voltage out fashion (programmable? I haven't read the datasheet). The chances of the SF output stage being configured just right for the Buffalo are slim or none.

There would have to be a mechanism to specify options... if these are just off the shelf then my guess is they are not shielded, screened, etc.

Here's the glassjar spreadsheet for the sigma22. sigma22 BOM.xls.zip

The Talema transformers are (IIRC) the same as the Amveco sold here at Digikey (although they don't have one that is big enough here). If you are putting these in a single box build, you would want to get the transformers shielded, and possibly potted in a can. Not sure if electrostatic screening is required or not? So in answer to your question, can Talema provide units like this for you? As far as I know, the DynaFET would take a +/-30V PSU. If that PSU is a sigma22, then 2x30V is the correct transformer for that. If the PSU were a Gilmore Dynahi PSU, you would want to reduce the transformer secondaries to 2x28V; maybe lower. 2x30V will run too hot on one of these.

Really good review: Album review: Neko Case's 'Middle Cyclone' | Turn It Up - A guided tour through the worlds of pop, rock and rap

You might point him to the previous thread where you indicated it didn't work as soon as you got it home. I would be very doubtful that he wasn't aware of the issue however .

In my opinion (but I haven't actually checked the figures), getting the Sigmas as kits from Jeff at glassjar is probably cheaper (or as cheap) as buying everything as a group. Looking at his BOM spreadsheet where you can configure what you want, a sigma with board, but no transformer would be ~$66 plus shipping per kit. I can send or post the spreadsheet if anyone wants to see it.

Happy Birthday!

I'm with Grawk that the first thing I would be doing is getting my money back from the seller. You had problems with it immediately upon getting it home (which is when I would have contacted the seller). I'm also not sure it would be that easy for your repair person to isolate it to the D20400A? At any rate, these do come up occasionally: http://www.diyaudio.com/forums/search.php?s=&action=showresults&searchid=1164282261&sortby=lastpost&sortorder=descending