Truepath channel silent

[goneXC]

V5 and the mute LED work fine but VN10 only draws around 100mA and the right channel MOSFETs remain cold. Right channel offset is stuck at 2.63V no matter what I do with the R104 pot (Left channel is fine and adjustable). I cannot see any obvious faults - is this a Vboot issue?

[Scratchy]

Have you followed Audiophools test procedures here?
http://www.maedos.com/Guideline.html...Test_Procedure

[Jan]

V5 and the mute LED work fine but VN10 only draws around 100mA and the right channel MOSFETs remain cold. Right channel offset is stuck at 2.63V no matter what I do with the R104 pot (Left channel is fine and adjustable). I cannot see any obvious faults - is this a Vboot issue?

Could be, but if the boot voltage was bad, then the low FET could switch on anyway as it feeds directly from VN10, and then you would have a neative offset on the output. Do you have 10V over both VN10 47uF caps? Check if the diode on the bad side is OK and in the right direction, and that the gate drive resistor and diodes are OK. The 2.5V offset comes from the feedback path but normally nulls when you power up (have output). Checking the feeback network would therefore make sense.

[goneXC]

I tested the entire board for symmetry (one channel works so used it as a reference) and everything OK - in fact I was surprised at how exact the match was. Then checked the 3020 connections in the socket and #1 VN10 was dry (so why did it work at all?). Then got sticky mute problem - it WAS working and yes, it functions OK shorting #15 to V5 as per Audiophool though initial current is 33mA then drops to 24. Spent ages trying to sort it out but in the process V5 current has climbed to 160mA, then 180mA and now very briefly 900mA (turned it off very quickly) - nothing else attached. Suggests to me a progressive breakdown in the 3020. Comments please - this is beyond my experience!

[audiophool]

Hello goneXC,

Sorry to hear about your tough problem.

Undoubtedly although the soldering on #1 VN10 was dry, it was still conducting adequately.

900mA into the TA3020 would mean that the chip is definitely dead. Is it possible that during your work on V5, you induced a short somewhere and that the current is not going through the TA3020? Suggest you go through the V5 test to answer this question.

Problems in the output circuitry including the Vboot circuit can be notoriously difficult to trace, and the symptom you are describing is not one that we have seen before, although I think we can be pretty sure it is caused by

1. something wrong in the soldering/assembly/component in the output circuit (most likely), or
2. a defective gate drive circuit within the TA3030 itself (least likely).

Do you have an oscilloscope?

If you get fed up, you are welcome to send it to me. I would be interested in sorting out and documenting the diagnosis.

[goneXC]

Hello Audiophool; The V5 test before the 3020 is inserted is fine. Although I have been through the circuits several times I will do so again using a different approach to see if I can pick up an anomaly. It WAS working albeit with only one channel so I am more than a bit puzzled. I suppose in extremis I could just let it run at 900mA and poke around to see what gets hot - if the 3020 is cooked anyway...

I don't have an oscilloscope but can get access to one.

Thanks for the offer, I might well take you up on it if I cannot solve this.

[audiophool]

First you have to fix the 900mA problem. It's almost certainly caused by a dead TA3020. It's possible that replacing the TA3020 will also solve the silent channel problem as well.