Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › PCM2702 USB DAC Revision B
New Posts  All Forums:Forum Nav:

PCM2702 USB DAC Revision B - Page 8

post #106 of 670
10/15/05 at 6:11pm
Hi Alf,

Congrats on the latest layout, looking damn good.

Just a couple of things, nothing big at all.

I really do not like traces between caps and you have the 5V line running under the CL cap. Hard to see a way around it, shifting the CL via under IC1 up a bit and bring it in that side maybe? Dunno.

Now the 5V line is on the bottom layer is the idea of having ferrites on all 5V supplies worth thinking about again? Anyone have an opinion here? Wonder if it will improve channel seperation. Have lots of space under there and if people do not want to use them then can just be jumped out.

Can R2 & R3 be mounted flat instead of vertically? Hate vertical resistors.

About the only other point is that several of the tracks, especially on the bottom layer can be wider.

And like the idea of having different spacing for Cl & CR mentioned above.

There are lots of little tweaks of course to think about but it is finally almost there. Will print it out and have a good look over it.

Nice work, the end of the tunnel is near.

Botch...
Reply
post #107 of 670
10/15/05 at 9:34pm
Quote:
Originally Posted by Botch
I really do not like traces between caps and you have the 5V line running under the CL cap. Hard to see a way around it, shifting the CL via under IC1 up a bit and bring it in that side maybe? Dunno.
Can't do that. That puts the capacitor on the other side of the power input, so it is now working through a short inductor (the remainder of the trace to the pin) before it can de-couple the power. The trace must meet the capacitor before it reaches the pin.

I don't really see any issue with the trace under CL.


Quote:
Now the 5V line is on the bottom layer is the idea of having ferrites on all 5V supplies worth thinking about again?
Absolutely. It think this is most important.

Quote:
About the only other point is that several of the tracks, especially on the bottom layer can be wider.
Take care they are not made too wide. Avoid any temptation to create anything that resembles a power plane. Those tracks have a capacitive coupling to the ground plane, if that capacitance gets too big you have a distributed RF capacitor that can resonate.
Reply
post #108 of 670
10/16/05 at 7:00pm
Thread Starter 
I put together my notes regarding the parts used on the board. Please free to comment. In particular I would appreciate if you could make sure that all values/sizes/packages are correct and are available in your country. If not, please specify alternative parts. Also please suggest boutique parts the design would benefit from. I will try to accommodate all your suggestions.

Quote:
IC1 – PCM2702E. SSOP28 package.

IC2 – TPS6734. SO8 package.

IC3 – LM317/LM317A. SOT223 package.

IC4 – REG102-5. SO8 package.

IC5 – REG102-33. SO8 package.

IC6 – Optional. BUF634U. SO8 package.


USB – Type B receptacle. BERG 61729.

X1 – Crystal 12MHz CL=30pF. HC49 package.


L1, L5, L6, L7– Ferrite 600R. 1206 package. Murata BLM31AJ601SN1L

L2 – Radial. 18uH. >= 1A. High Q. 7.2mm diameter. 3.5mm pitch. Suggestions??????


D1 – Schottky diode. 1N5817.

D2, D3 – Standard recovery diode. 1N4004.


C1 – Ceramic. 47uF. 6.3V. 1210 package.

C2, C4 – Tantalum electrolytic. 33uF. 16V. Radial. 6.3mm diameter. 2.5mm pitch. Lowest ESR < 0.25R. SANYO OSCON SC/SA.

C3 – Ceramic. 1nF. 1206 package. AVX/EPCOS/MULTICOMP.

C5 – Aluminium electrolytic 820uF. 16V. 10mm diameter. 5mm pitch. Height <=15mm if used with 1455C enclosure. Panasonic FC.

C6 – Tantalum electrolytic 10uF. 16V. Radial. Up to 5mm diameter/pitch. AVX/MULTICOMP.

C7 – Tantalum electrolytic >= 1uF. Medium ESR (>= 1R, the more the better?). 16V. Radial. Up to 5mm diameter/pitch. AVX/MULTICOMP.

C8, C12 – Aluminium electrolytic 100nF (1uF????). Low ESR. 6.3V. Radial. 4.0mm diameter. 2mm pitch. SANYO OSCON SC/SA.

C9, C13 – Optional. Ceramic 10nF. 0805 package. AVX/EPCOS/MULTICOMP.

C10, C14 – Optional. Tantalum electrolytic >= 10uF (47uF). 6.3V. ESR >0.05R. Radial. 6mm diameter. 5mm pitch. AVX/MULTICOMP.

C11, C15 – Optional. Ceramic 470nF. 0805 package. AVX/EPCOS/MULTICOMP.

C16, C17 – Ceramic 33pF. 0805 package.

C18, C19, C20, C21, C22, C23 – Ceramic 100nF. 0805 package.

C24 – Aluminium electrolytic 47uF. Lowest ESR, ESL. 6.3V. Radial. 6mm diameter. 2.5mm pitch. SANYO OSCON SC/SA

CL, CR – Pro audio grade aluminium electrolytic 47uF. 6.3V. Radial. 6mm diameter. 2.5mm pitch. Black Gate Nx HIQ, ELNA Silmic II.


R2 – Metal film resistor 750R 1/4W. Axial.

R3 – Metal film resistor 120R 1/4W. Axial.

R4 – Metal film resistor 1.5K. 1206 package.

R5, R6 – Metal film resistor 22R. 1206 package.

R7 – Metal film resistor 1M. 1206 package.

R8, R9 – Metal film resistor 330K. 1206 package.
Reply
post #109 of 670
10/16/05 at 7:21pm
it looks like IC5 can't be found on digikey or mouser
Reply
post #110 of 670
10/16/05 at 7:23pm
Quote:
Originally Posted by Alf
I put together my notes regarding the parts used on the board. Please free to comment. In particular I would appreciate if you could make sure that all values/sizes/packages are correct and are available in your country. If not, please specify alternative parts. Also please suggest boutique parts the design would benefit from. I will try to accommodate all your suggestions.
All of these parts are impossible to get here in Portugal, so everything must be bought online.

regarding C5, you say:
C5 – Aluminium electrolytic 820uF. 16V. 10mm diameter. 5mm pitch. Height <=15mm if used with 1455C enclosure. Panasonic FC

is there a noticeable benefit if we use a bigger cap, say 1000uf or 2000 uf in a non portable version?
also does the mint board fit in the case using this particular cap?

The best thing would be a kit( the same deal guzzler offered) with all the parts necessary to complete a board excluding the IC's,( I have extra pcm's, both regs and the buf634, from the first usb dac, when I bought a few of each) and excluding also rca jack's etc.....

Just another question:

comparing to the original USB dac by guzzler and 00940, what improvements in sound should be expected?
Reply
post #111 of 670
10/16/05 at 7:49pm
why MINT? I thought this DAC should be in the same league with at least PIMETA, right? wouldn't it be wasteful to put it with MINT?
Reply
post #112 of 670
10/16/05 at 8:15pm
Quote:
Originally Posted by diablo9
why MINT? I thought this DAC should be in the same league with at least PIMETA, right? wouldn't it be wasteful to put it with MINT?
It is certaintly not be wasteful because with a mint you can still put it in the smallest hammond case and have a really portable great sounding dac/amp solution. of course with a pimeta you get better amplification but you will need an enclosure at least 50% bigger.

I will build a few of these, one portable with a mint and a home dac only version powered by a tread if possible, so the portable version must be the smallest possible!!
Reply
post #113 of 670
10/16/05 at 8:17pm
Quote:
Originally Posted by MASantos
because with a mint you can still put it in the smallest hammond case and have a really portable great sounding dac/amp solution. of course with a pimeta you get better amplification but you will need an enclosure at least 50% bigger.

Personally I think a mint is the way to go,such as the SMD mint which is being developed in another thread.

I will build a few of these one portable with a mint and a home version powered by a tread if possible, so the portable must be the smallest possible!!
looks like a great idea, you convinced me. for the SMD MINT, could you give a link?
Reply
post #114 of 670
10/16/05 at 8:24pm
Quote:
Originally Posted by diablo9
looks like a great idea, you convinced me. for the SMD MINT, could you give a link?
Sure no problem:

http://www6.head-fi.org/forums/showthread.php?t=136596

It's really small and allows the use of single opamps

PS: I never remember how to do those neat links!!!
Reply
post #115 of 670
10/16/05 at 8:34pm
thanks for info. I wasn't aware that thread is already going onto PCB ordering stage. definitely joining you guys!
Reply
post #116 of 670
10/17/05 at 8:58am
Thread Starter 
Quote:
Originally Posted by Botch
Remove C4. As we have C5 I can not see a need for it. Or does anyone have a problem with this?
Quote:
Originally Posted by MASantos
is there a noticeable benefit if we use a bigger cap, say 1000uf or 2000 uf in a non portable version?
I have been thinking about this lately. I tend to believe that C5 might do more harm than good. TPS6734 is a low noise chip. Bundled with a very low ESR capacitor C4 it is capable of delivering <15mV ripple. C5’s value is too high to have any effect on this. On the other hand I am not really sure how TPS6734 will behave with an additional output capacitor.

At this point I am thinking about removing C5 and keeping C4. LM317 should not need it anyway. C4 should be able to serve both chips well.
Reply
post #117 of 670
10/17/05 at 9:01am
Thread Starter 
Quote:
Originally Posted by diablo9
it looks like IC5 can't be found on digikey or mouser
Digikey part number: REG102UA-3.3-ND
Reply
post #118 of 670
10/18/05 at 6:01am

ac power

Alf can the dac be powered by a tread? This way the board gets cleaner power? should there be any audible diference?
Reply
post #119 of 670
10/18/05 at 6:35am

Alf and others, great thread!

Just a question on the USB.
Is the design prepared for soldering just the wires of a USB cable to the PCB instead of using a USB B-type plug?
Reply
post #120 of 670
10/18/05 at 5:26pm
You should be able to solder the cable directly the board. I think the only concern is the stress applied to the leads if it is pulled, which entails finding a cable relief adapter for the case. Its probably more elegant to just have the USB B adapter on the PCB board.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › PCM2702 USB DAC Revision B