First off, I want to thank Russ and Brian for taking their time to make such a great product and for being so quick to ship. I’ve been eagerly awaiting getting a buffalo dac for about half a year now and it’s exciting finally having everything coming together now.

Unfortunately I’m not terribly experienced with this kind of stuff, so I’ve jumped into something a bit over my head and have had to get some help from a few friends to figure things out. Last night I finally got some sound to come out and its sounding leaps better than the computers sound card (no surprise there) but there seems to be a bit of distortion in the midrange (around the frequency of where an acoustic guitar is). The odd thing is I only get this distortion when using xlr out to my speakers (buffalo directly to power monitors). I tried wiring the outputs of the buffalo to a 3.5mm jack and then connected that to an adapter cable back to xlr and to the same speakers and the distortion is no longer there. I figured I must’ve wired xlr right but after double checking I believe I have it right. The odd thing is the distortion sounds identical to when I had accidentally wired the toslink module into spdif on the buffalo instead of d1. I hope this is understandable, I’m at a complete loss for words when it comes to using the right tech terminology here. If you have any ideas as to whats wrong I’d love some insight. Also the manual mentions something about breaking a few points on the pcb if youre sending this signal to a powered source, but I’m up in the air on this because I’m uncertain if I’ll be using headphones or a preamp. Will this create a substantial difference or is it something I can leave for now? Thanks in advance

Ethan

Also posting pictures can help a lot, especially of the jack wiring.

Make sure you wired like diagram on this page:
http://en.wikipedia.org/wiki/XLR_connector

If your getting music, the lock LED should be on. My guess is backwards LED(s).

Glad to help.

Cheers!
Russ

Where are you connecting VA GND? Can you post a pic?

Edited by user Tuesday, May 26, 2009 2:33:18 PM(UTC)  | Reason: Not specified

Yes I cut the traces to enable the output resistors, 4 of them, and I measure 22R across the gaps. Ground is connected to the VD (gnd) terminal. I "reasoned" that would be a good ground point to give even distribution across the board. I rechecked the gnd jumpers on the back side and they seem good. I have 5V on the + terminal of VD, and I have +- 15.00 going into both (left & right channel) VA terminals. I tried the board with and without the volumite, but the distortion is the same. A 1kHz sine wave producing 1.6V on the right channel (peak to peak, on both the + and - output terminals) measures 1.24V on the left channel. I can hear a superimposed buzz on a 400Hz sine wave. In playing Spanish Harlem (Rebecca Pidgeon) the string bass sounds like a rubber band being played on the left channel, but the right is perfect.

Any points on the board where I could compare left and right signals?

RossG

Did you measure continuity across the jumpers?

You might want to check the DC bias at the inputs and outputs of the OPA1632. Also check for DC on the outputs. Check both accross and to GND. Take not that in some spots DC bias to GND is normal, but not at the final output.

Are you running PCM or SPDIF?

If SPDIF try pulling the on-board microcontroller.

Please send some close up pics of the top of the board.

Cheers!
Russ

Thanks for the reply. I'll look into what you suggest and get back to you tonight--with a pic or two.

RossG

Here is a shot of my entire chassis. It's messy still, haven't cleaned up from when this had the original Buffalo board. I'll switch it to just two power supplies soon. Note, the large green wire is common ground which is a 'star' configuration originating at the upper right PS.

RossG

There are two opa1632 per side. It looks like you measured only the one on output. How does the other one look?

Those measurements appear normal.

The only thing that jumps out at me is the Volumite connection.

You should be using all four wires out from the BUF32S and omitting the use of the VREG on the Volumite.

Also can you please tell me the IC parts that are closest to the DAC. They are the two opamps that are oriented 90degs different than the opa1632s.

Cheers!
Russ

Lots of questions to answer....but first, welcome back Brian... Hope the move went well.

"are you feeding VD with +12v?" No, I measure +5.00 on my meter across VD. The 12 V supply (pictured above) is for the mux and toslink.

"are you using the buf32S firmware?" Yes, it's in the volumite controller.

"tell me the IC parts that are closest to the DAC"
IC 13: L49720,
IC 3: L49710
IC 2: L49710
IC 14: L49720

I wired the volumite as perscribed.

Quiescent voltages across IC 9 & IC 12 seem normal, i.e. they are the same for each chip
1: 1.656V; 2: 1.653V; 3: +15.00; 4: 1.657V; 5: 1.657V 6: -15.00; 7: 14.91V; 8: 1.657V

What I've discovered is that a sine wave will fold over with the volumite pot at about 3/4 rotation. I'm sure this is the distortion that I'm hearing. Pins 4 & 5 of IC 9 & IC 12 start to fold over (reaches saturation?) at just over .7V (peak to peak).

And when I'm playing music, it's the peaks (such as when the string bass is plucked on Spanish Harlem) that I can hear the distortion. I have to turn the volumite control to about the half-way point for the distortion to clear, but the music is hardly audible. BTW, I using the Sympatico for my power amps--blanced inputs naturally.

When I disconnect all the wires going to the volumite and put in the original BUF32s firmware that came with the board (pluged into IC8), the Buffalo is at full volume (as I would expect) and fully distorted.

Should IC 9 & 12 be reaching saturation at .7V P-P output? With the original Buffalo I was able to crank the volumite up all the way and not get any distortion.

Is it possible that I have the wrong firmware chip?

RossG