[marcelnl]

agree, if you measure a uniform voltage being created over both I/V resistors the output caps (or the soldering) are the most likely candidate...
 
Depending on how you connected those I'd work my way back from the output towards the I/V resistors, chcking everything with your DMM and resoldering all joints and checking for shorts.
 
TDA 1543 will come cheap, that is the good news, but you could swap the TDA 1543s between both channels to see whether that is the issue.
 
BTW the cost of replacement parts is not going to be that much, one way shipping should be bearable too....
 
Hope this helps
 
 

 

[MikoLayer]

yeah, I looked into USPS rates; if the total weight is below 4lbs and I could fit it into the small flat rate box, it is a tad under $15. What I am worried about is how long it takes to get to him and make it back here (well over a month for the whole transaction?) and whether he would not undo the mods I have applied to it during the repair job

 

[marcelnl]

I can only say, let him know you want to keep the mods whenever possible....
You did not listen to the thing for  months (well I hope you do not listen to mono in stereo) so another month is the least of your problems I reckon
 
But what about swapping around your TDA 1543s, that can be done quickly and will give you good info.
 
You could take the DAC out of its housing and put it in a plastic container, that would save weight...USPS is about the last carrier I;d advocate as everything I have received through them was a major waiting experience, there must be other couriers but I do live in Europe so your choices and service experiences may be different....

 

[MikoLayer]

Tried swapping the 4 rows of left and right TDA 1543 pairs, one at a time. No matter which combination I came up with, the result was the same regardless (only the left channel active). I guess the only way out now is to contact Kevin for repairs

 

[MikoLayer]

sent another message thru their ebay homepage and still no response.
 
I already told him I would send it back if he can't help me with troubleshooting it over email too  

 

[marcelnl]

yeah well customer support is there, yet it is not really core business making any money for Kevin I reckon ...
 
Helping you sort this out via email is going to be a tough one...so maybe you can find someone closer to home who can help you out?
Usually there are Audio DIY communities (DIY HiFi shop?) to be found in most larger towns, anyone with some experience in NOS DACs and a scope should be able to pinpoint the issue fairly quick I think.
 
Would the schematic of the DAC be of any help to you? you could request it from Kevin as I'm pretty sure he would share it (I only have 1 version and there are many small differences between various versions)
 
 

 

[MikoLayer]

yeah sounds like something I should ask next time he decides to drop me a line. Mine is v3.0, dated 2009.06.08. Is yours the same by any chance?

 

[marcelnl]

not sure, will have a look...think I have a 3.0 but have the diagram of the 2.5 will send it to you via PM once found
 

 

[MikoLayer]

thx!

 

[MikoLayer]

With the schematic now at hand, I have probed a whole bunch of combinations at the relevant points and found out there must be some discontinuity between the L10 ferrite bead and C36 output cap. When I bridge the either end of L10 with L OUT by holding the output cap by hand and bypassing the solder points built onto the PCB, I am rejoiced with beautiful sound in the right channel once again (finally!)
 
I haven't soldered anything on to L10 just yet, not knowing which of the two ends the C36 should be connected with; don't want to bypass the L10 altogether by accident, so it's only matter of figuring out the right one between the two. SO EXCITED
 
EDIT: by inspecting the trace on the underside of the PCB, I have concluded it should be the outter (further away from the center of the board, not in direct contact with C49) end of L10 that is to be joined with C36. 

 

[marcelnl]

good work!
 
Usually the PCB board traces and solder spots are flakey and I always try to avoid using them as they invariably create havoc at a later stage AND a PCB sounds awful when compared with hard wiring.
 
I did away with all of the output filter(just unsoldered the lot), so I connected the resistor between ground and the DAC output and a piece of silver runs runs straight from the I/V resitor into the output RCA connector. Not sure if your amp likes a filterless DAc, it may cause oscillations in some designs yet many amps are overengineered and have some form of filtering in the input stage...just experiment without and be prepare to put everything back in place if no real improvement is heard.
 
If you need a coupling cap you could run that straight from the resistor to the RCA output connector. I did not need an output coupling cap as my tube amp has an input coupling cap.
 
While you are at it, why not replace the coupling cap with something decent like a decent foil cap or better, as a bonus to an improved sound quality you need not worry about which end should be connected where as they are unipolar

 

[MikoLayer]

by couple cap you mean which? you mean to the TDA1543s like LosNir has done? Might need to order some caps then.
 
I revisited some earlier threads on HiFace and USB galvanic isolation, and settled to experiment with one of these: http://www.circuitsathome.com/measurements/usb-isolator What I need to clarify is whether the isolator would need to be powered for this application. My understanding from reading the replies in the article is that if the device to be connected draws more than 400ma the isolator should have its own PSU connected to the barrel-jacket type connector. Not sure how it would work with this dac (or others USB dacs for that matter). I know valab dac is self powered, but that's for the I/V section right? Is there anything else in the chain that would make it usb-powered?

 

[marcelnl]

nope was referrring to the output caps, but I just saw that is one of the mods you already did. But adding or replacing the buffer caps to the TDA 1543 is also very worth your while and will not break the bank. and after all....you have the thing ripped apart as it is
 
The Valab is self powered for all sections BUT the I/V part as that is done through the 330 Ohm resistor
I seem unable to find how much power it really needs, think they are referring to the power needs of the usb isolator itself or the device connected after it should that require power to operate. 
Your computer should be able to supply power to the isolator and the Valab is supplied separately from the USB part as te diagram is showing that VDD2 (+5V) is used to power the USB circuitry. What this thing does is providing a galvanic break of the connection between computer and DAC whils allowing data to flow think the valab does not need any power from USB to operate so that should be oK.
 
 
 

 

[repeet]

You could bypass the circuit board completely and solder your output cap directly to the dac chip. 
 

 

[MikoLayer]

where exactly in the circuit do those buffer caps go? I can't imagine loading the board up any more with something bulky, couldn't fit those mundorf output caps so had to drill some holes at the top of the enclosure; it's pretty cramped as is
 
I have done the DIR9001 mod without knowing that it had to do with SPDIF only. The thing is I feel usb 2 spdif converters like hiface cost too much, and would rather opt for USB. I have been following Leeperry's exploits in the usb isolators, will most likely order one soon myself