luvdunhill

A Zobel network shouldn't be necessary with the Dynafet due to the feedback design. If you want this sort of protection, and output choke might be a better choice. For example, say you're using a Piezo tweeter in your speaker design. In this case, the load doesn't tend to be inductive at high frequencies, but capacitive. In this case, the Zobel network is a poor choice and the output choke would better protect your speakers. The problem is that the values in the Zobel need to be chosen carefully with respect to the design for optimum performance. With this said, the only person at this point who can really comment on this is Kevin, so I'll freely admit this is just speculation on my part.

first hit on google seems to have it for $0.19 a foot.

I agree, looks fantastic! What about Justin's suggestion of the input filter? Here's one that I used, and I've also swapped the location of R34-R35 to be after the caps. I'm partial to polystyrene caps here (e.g. Newark 96K9640), but micas are another possibility. While we're on the subject of caps, I've had very good luck using Teflon (PTFE) caps in feedback loops. Would it be possible to add a second pad to support this? This would not be a standard option, but more for me to tweak with The Voltronics stuff at Mouser is good, but very expensive. Another choice would be Vishay/BC sold at Newark. Both places, just search for "PTFE trimmer". All that really needs to be added is one pad... what that be possible? Finally, what about adding the Wima FKP2 bypasses? I still think this is a good idea wherever possible.

I was referring to the input transformers

so, is there a ground plane on both sides of the board? I know amb does this with his boards a lot... I'll look over it some more tomorrow... edit: also, I would recommend cutting off the ground plane around the output devices, or is that what the dashed line is for? I'm confused I suppose

hey, do you have any rough costs for case cost? any estimates from your beta22 builds would probably be right on, plus the differential cost in the cases. Also, the transformers come in a PCB version.. it would be cool if those got integrated into the I/O boards. If you can get OEM cost, I could probably get some other people interested as well to help the numbers.

sure, just PM me anytime Back to your regularly scheduled program!

I don't have any PSU boards, but I do in fact have some boards if you'd like, just like the ones above. I'll sell them for $15 for a stereo pair and I'll pay shipping and PayPal fees. If you don't do anything with them, it won't hurt my feelings. Just think of me as a DIY enabler Also, you can get an F5 "kit" from tech-diy that I can order for you and shove in the envelope. See here: http://tech-diy.com/Store/CloneKits.htm Of course, the real cost is the case and the power supply....

you can practically count the number of components on 2 hands for this one. Maybe I can get some secret camera footage of swt61 doing a F5 for your viewing pleasure.

F5

I'd appreciate it not being open to the outside world. I got all kinds of spam from the Pearl deal and I'm trying to keep this small enough to be advantageous from a cost standpoint but not unwieldy. Plus, I'm counting on the group to help weed out :asshat:s, as this could be a serious drag.

oh, I forgot about -=her=- ... thanks for the reminder.

yeah, good luck with that

1x Copper Flat Bar .500" x 3" x 12" - Class II 1x 3003 Perforated Aluminum Sheet .063" x 24" x 24" 2x 6061-T6 Aluminum Square Bar .375" x .375" x 48" UPS shipping Total: $60 seems to be an extremely good deal, we'll see what it looks like when it gets here.

With the GBP in the crapper, I think now would be a good time to try some Sowther transformers for conversion

SMD resistors can get very exotic. Susumu has some 5ppm, 0.1% parts as does Alpha: http://www.susumu.co.jp/english/pdf/products-j01-04.pdf [url=http://www.alpha-elec.co.jp/e_machine.html]Alpha Electronics Corp.

I like Susumu and the KOA RK73A. I believe both DACT and Goldpoint use KOA resistors.

would tombstoning these resistors be acceptable?

heh. There are a lot of other (reasonable) 0.200" options out there, like the Phoenix 0.6W MRS (I think?) series and the blue Vishay series. You can also tombstone larger resistors. If this isn't all that interesting to people, now's the time to speak up. Jacob: Can we remove the ground plane from around the output pad and up, so there isn't any chance of the L-bracket scratching through the solder resist layer? For the large output resistors, I like the Panasonic ERX series, which are available in 1W-3W. http://www.panasonic.com/industrial/components/pdf/AOA0000CE18.pdf What are you thinking for the caps? Something like Nichicon UHE 50V, perhaps? http://products.nichicon.co.jp/en/pdf/XJA043/e-he.pdf I'd also add some film bypasses, something like Wima FKP2. There is a 10nF/63 DCWV that is PCM 5mm 5.5mmx7.2mm that might work well

I think it would. At least that was the case when I played with it on a design I've been working on. I found a good write up somewhere in the documentation for the Leech amp I'll find and post. I pretty much followed the concepts like a recipe and it worked perfectly. Also accounts for phase issues, though in a slightly less elegant way.

a LPF would also require recalculating the global compensation cap, right?

Basically, the input JFETs need to be just like the Vas duals, except the 4-pin side needa to be placed back to back (i.e. one rotated 90 degrees) and the pins 5mm away from each other. Like this: -0-0-0- 0-0-0-0 -------- 0-0-0-0 -0-0-0- where the distance between the second row of text and the fourth row is 5mm.

the input JFETs are not correct. They should be layed out just like the 7-zip packages.

Pete: We all appreciate your enthusiasm, but you do realize that the lead time for the parts (then resulting matching) will be the long pole in tent here? We will have plenty of time for the boards, trust me