Pars

I was planning on Roedersteins MK3s, and PRPs. I've always liked the BC (now Phoenix/Vishay) MRS and SFR series (0.4W and 0.6W), but only the 0.4W would fit a 0.2" spacing. I just replaced a pair of 100 ohm MRS on an I/V board with PRPs and could hear the difference (PRPs were better). So, being time to speak up, I am less than thrilled with this option. I know it tightens the layout up alot, but the tradeoffs are unacceptable to me. Tombstoning is not really acceptable either. Do you know of any really good 0805 resistors that aren't an arm and a leg pricewise? I know that some of the stepped atten. manufacturers think the resistors they are using are better than the leaded ones they used to use, but I haven't seen anyone identify what they are using.

Just noticed a problem on the schematic. Per post 72 in the headwize thread, only one of the two FB points for should be used. On the schematic here, both are being used. I would think using the point between the two 5k resistors would be best? [url=http://headwize.com/ubb/showpost.php?fnum=3&tid=6584&pid=59968&fpage=4]HeadWize: View Post [DIY Workshop

Looks good. I assume that is ground plane (top) that is not rendered? What are the resistor package sizes? I guess I assumed when Marc (luvdunhill) asked for the 0.200" and 0805 SMD, that these would be in addition to an RN60D type footprint... guess not. That's alot of money to Texas Components... My only other comment at this point would be that it is my preference for an input connector that it be a Molex KK 0.100 rather than the Phoenix blocks, but that is minor. I would assume that the mounting holes match a Dynahi board, as well as the output device mounting holes? I measured 0.45" c-to-c on the output devices on a Dynahi board I had. Mounting holes were 4 9/16" and 3.5" c-to-c. You're pretty fast with Eagle, aren't you I was trying to recreate your previous layout this weekend and only got about halfway done with it. And was glad I didn't have to determine component placement...

BTW, Charles Hansen has been noting some interesting observations on MOSFETs: DIYHiFi.org • View topic - push and pull me back into the limelight

Yeah, I hear you. Marc: Your analogy to power amps should probably be change to preamp as a closer fit in terms of noise performance. But yes, I would think with a shielded trafo (or perhaps use of a split bobbin instead of a toroid) that single box should be fine. My dynalo (V2 headamp board) has an unshielded Amveco in there that was supposed to be mounted on the board (not a PCB trafo though). Mine is just off the board and I don't hear noise thru it. Granted, I dont have any IEMs to check it with.

No. The original Gilmore dynamic (i.e., dynalo) used a +/- 16.4V PSU, which is probably what your Gilmore Reference uses. The Dynahi, etc. use +/- 30V PSUs. One other thing regarding the boards. The offset adjustment pots present on the Dynahi were not in the DynaFET schematic, nor on the board layouts that I have seen (though I could hav missed them). The pots were present on the google spreadsheet BOM however. The (2) 500 ohm resistors should be replaced with 620 ohm with a 10K pot in parallel with each.

Yeah, I hope you make it home safely and get checked out in the hospital with no problems. Take care!

It would be a good idea, I would think. Not sure who would do them though.

Also, here is an Eagle library archive including the dual package devices: 2SJ109, 2SK389 transistor-toshiba.lbr 2SA1349, 2SC3381 transitor-pnp.lbr, transistor-npn.lbr respectively Eagle_toshiba_trans.zip

Happy Birthday!

No, that's just for a production run.

I got some preliminary pricing from Imagineering for a board run at 25, 50 and 100 pieces, with an option to double it at about half price. I included electrical testing just to see how much it was, but probably would not recommend it. It was never done for the dynalo/dynahi board runs. 25 boards = $8.32 + $4 tooling = $12.32 50 boards = $5.43 + $2 tooling = $7.43 100 boards = $3.98 + $1 tooling = $4.98 plus shipping and delivery to each person.

Yes. 2 boards and I guess a sigma22 PSU. 80VA 2x30V trafo (or 30-0-30 or 60Vct). If you are going to do it in a single box, then trafo should be shielded if toroid; split bobbin or r-core should be fine as well. Additional sundry items such as case, volume pot, jacks, wiring, etc.

Uhhh, yes you're right Let me return to my confusion. So, you guys actually like diptrace? I took a look at it this morning, and , at least with the schematic editor.

They won't work for the duals (one device is flipped). They were laid out for singles. Singles in place of the duals have the flat of one device pointing 180 degrees from the flat of the other device (like the HS).

No that's what I'm saying. The layouts as already done would not work for the heatsinks because one device of each pair needs to be flipped.

Marc, The goal on these boards in terms of the front end and VAS stage is to Use singles (2SJ74/2SK170 and a BJT pair, currently perhaps 2SA970/2SC2240), and use the heatsinks per the diyaudio thread? A nice to have would be grouping the 2SJ/2SK together (back to back) and the same with the BJT quad? Not sure if that is doable. Referring back to the layout in post 192, the individual TO92s for each dual package device are not oriented opposite to one another, which was what I thought you needed to have for the heatsinks? Let me know if I am making incorrect assumptions on any of this.

Shielded transformer from SumR is cheaper than an Avel plus another case plus good connectors (though JAE and whichever ones were used in the krmathis build are good and not that expensive). Heat may be the trump card however, though I was planning a 2 ch. build, so single chassis = win.

IIRC, KG doesn't like active grounds. I'm not sure I've ever seen a 3 channel dynalo or dynahi. I'll let him weigh in and correct me if he cares to however. The ones I've seen are 2 channel for SE and 4 channel for balanced.

First stab at a BOM. I took the Google one from Headwize and started updating it with some info Marc provided. http://home.comcast.net/~youngc1/DynaFET BOM.xls

Hmm, I wonder. I wasn't sure if he might not even be a regular here or HF... board says Jacob Potter.

Yes, I was looking at that as yours is a very well done example of angle bracket. I just always wondered why most people did it that way instead of mounting the output BJTs straight up onto a rectangular section about the thickness of the board offset (3/8") or so? I guess the angle bracket does lend itself to the board mount for that side and all. It would be nice in the Dynafet project to make the boards so they could kinda interchange in a Dynahi as far as mounting, etc. so you could swap boards in the same amp chassis.

Not quite that far yet, but I'm sure they will. Boards are pretty much done other than putting the transistors in. I also need to decide how i will mount these to the heatsinks (angle bracket or standing up straight onto rectangular section aluminum). SumR transformer on order, plus case. BTW, did anyone happen to download j4cbo's Eagle schematics and/or board files on headwize? Looking at the board, it may be possible to do this in the freeware version of Eagle. Maybe. http://headwize.com/ubb/showpage.php?fnum=3&tid=6584&srch=dynafet;

Fuck you. Still contemplating post 161