Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Comments for my PCM2702 DAC layout?
New Posts  All Forums:Forum Nav:

Comments for my PCM2702 DAC layout? - Page 5

post #61 of 613
Quote:
Originally Posted by Citizen86 View Post
haha glad you cleared that up. I'm looking forward to this one, hope you guys make a kit that ships internationally, that would be great for people like me where it's difficult and expensive to order from more than one place
Yep - that's the plan and definitely international. I haven't calculated it, but my guess is that international customers make up about 30% of Beezar's business.
post #62 of 613
Quote:
Originally Posted by pinkfloyd4ever View Post
tom you know we're putting cover sheets on the tps reports now right. mmk great i'll just send you another copy of that memo
Where's my stapler!?
post #63 of 613
Oh man.. Is there any good substitutes to the Pana FM/FC, Nichicon PW? Maybe 0.5-1mm shorter? The 12.5mm cap is gonna make for a veeery tight fit. =x
post #64 of 613
Quote:
Originally Posted by JamesL View Post
Oh man.. Is there any good substitutes to the Pana FM/FC, Nichicon PW? Maybe 0.5-1mm shorter? The 12.5mm cap is gonna make for a veeery tight fit. =x
You can save a millimeter if you go down to a 560uf 6.3V Pana FM, but that's about it. I'm sure I don't need to suggest this to you, but if your clearances are go/no-go based on a 1mm difference, you may be designing things too close. Sometimes there's that much variance in the capacitor's rubber pad or the plastic wrap.
post #65 of 613
you could maybe save that much by mounting it laying down on top of the dac chip ?
post #66 of 613
Hehe, I like to play it close.. I didn't even look at the specsheet when I got this case, but everything is working out perfectly.
Turns out I had an extra half millimeter available in the case.
post #67 of 613
After listening to the BantamDAC for awhile, I have to say that cetoole was 100% right about the Wima MKS film caps. They are just tons better than Black Gates as output couplers. I can't tell whether it's the Wima's or the Bantam, but it sounds better than the VitQ Alien I built, too.

EDIT:
Speaking of which, that VitQ Alien bit the dust. Cetoole and I have talked about this for awhile: I've blown 3 or 4 Aliens in the last couple of months while fiddling with different output coupling caps.

In the Alien and Bantam design (and others) the DAC chip is directly connected to the input of an amp, with only a small capacitor protecting it (the coupling cap). There are instances where plugging and unplugging the DAC exposes it to shorting from the plug, spikes, etc. We conjecture that this is what's blowing out mine and other's Alien DACs.

I think it's prudent to always connect the USB power source first and wait awhile before connecting it to an amp. This ensures that the coupling caps are fully charged. If you always have your DAC connected to your PC and it's always on first, there seems to be little danger. I have a couple of Aliens that have lasted a year under a constant connection. However, start plugging/unplugging and fooling around with powering up the amp, DAC and connectors at different times, and it seems there's a good chance you can fry the DAC chip.
post #68 of 613
Any thoughts as to the benefit (if any) of putting small resistors in series with the outputs to help protect the DAC chip from shorts? 100 ohm or so?
post #69 of 613
Any space on there for a delay circuit? :P

( we believe in you Colin! )

Tom, Thanks for those cautionary words. It may be what killed the single channel on a chip I had to replace on an alien recently. It had a switch on it, so it's not unlikely to be linked.
post #70 of 613
Well, it's just conjecture so far - and no indication that the Bantam would be susceptible. However, the precaution seems prudent and not very imposing.
post #71 of 613
Real good info, thanks tomb!! Something I'll definitely remember when using the DAC. And still looking forward to the Bantam!!
post #72 of 613
Quote:
Originally Posted by Pars View Post
Any thoughts as to the benefit (if any) of putting small resistors in series with the outputs to help protect the DAC chip from shorts? 100 ohm or so?
As it turns out, Colin had sent me an e-mail just before you made this post. I'll test it out when the boards come in. It should be easy enough to solder a couple of 100 ohm resistors in the output leads.
post #73 of 613
Based on comments from the prototypers, this is the final layout:


Bottom:


The changes were relatively minor, but due to the very small size of the board, some of the larger components were relocated. In summary:

1. The selected USB Type "B" jack had a conflict with C14.
The prototypes that were built solved this by cutting away some of the metal casing of the USB jack.

The new layout basically swaps the positions of the USB jack with C7, the large electrolytic. This places the USB jack close to the PCM chip, but all measurements indicate more than enough clearance.

2. An LED was requested.
While an LED is not necessarily functional to the DAC, it was still felt that the visual power indication from connecting to a USB cable was a valuable feature to have. A standard 3mm LED was added and an LED resistor (R9).

3. It was noted during prototyping that using the TPS regulators required tantalums between output and ground instead of electrolytics (C11, C13).
The new layout uses dual SMD and through-hole pads at these positions to preserve the dual compatibility with the REG101/2, if those become available again in the future. However, the TPS chips perform well and the use of the SMD ceramics and no electrolytics seems to work quite well - more convenient, too! I have remarked before that there seems no reason to persist with the REG101/2's and haven't changed that opinion.

4. An extra output ground wire position was requested for convenience and performance advantages in braiding and grounding output leads.

Two "OG" positions are now on the board.

Also, we are recommending the Wima MKS caps in all cases as the default build. Frequency response with loop-back to an M-Audio Transit is only down 0.5dB at 20Hz, and equals or exceeds the Transit's response at 20KHz. I suspect the Transit has a lower input impedance than the typical 50K pot input of our headphone amplifiers - I have not noticed any loss of bass, period - after repeated listening. On the contrary, the Wima's bass response is thumping and slamming with outstanding detail - it beats my Black-Gate-output DACs to pieces. All of a sudden, I can hear this metal sheen with the BG's that I never noticed before.

The Wima's do take a while to break-in. There is a mid-bass "thickness" that is not unpleasant in the beginning, but it disappeared completely in about a week. Humble Homemade HiFi recently added Wima's to his list - it's an interesting read for their use in the Bantam and also the Millett Hybrid/MAX/MiniMAX:
Humble Homemade Hifi - Cap Review

He reviews the MKP10's that we use on the MAX, but the MKS's for the Bantam are very similar. (MKP10's are unworkable for the Bantam, due to the size at the ratings needed.)
post #74 of 613
Nice work! I haven't tried it, but do you think maybe an LED on the /SSPND pin is more useful than a generic power LED? It should give more information about system status.
post #75 of 613
Quote:
Originally Posted by error401 View Post
Nice work! I haven't tried it, but do you think maybe an LED on the /SSPND pin is more useful than a generic power LED? It should give more information about system status.
Good suggestion, but -

the BantamDAC boards have arrived(! ) and will be available for purchase this weekend.
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Comments for my PCM2702 DAC layout?