Fixing an Aune T1 (amp/dac) TUBE DAC Board

[cowboybob]

https://www.onsemi.com/pub/Collateral/NCP1117-D.PDF
low drop out regulator 5.0v
see page 14 for markings

Excellent, pfofit! Thank you. Thought that's what it was (in a general sense), but I just couldn't make out the part number(s)

bittgata, see above post (#39).

This is my interpretation of the circuit we're investigating (It is only a very basic circuit example, not the actual circuit):

The "bad" DAC appears to have a fault in the V+ side. Don't know why at this time.

 

[bittgata]

I didn't have a solder wick and I can't find my solder plunger but I managed to improvise. The plate voltage seems to be unaffected, it changes a bit but still around -0.4V. Also the light never comes on with the Vin desoldered.
Thanks for the info pfofit, I don't know long I would I have had to search for it.

 

[cowboybob]

Ok.

Sounds unlikely, but it does appear that the bridge (1/2 of it, anyway) circuit has failed. We'll start with the output of the bridge:

Check the voltages from ground to the points below:

Both under "AC" should be ≈16VAC, + ≈ +22VDC and - ≈ -22VDC.

 

[bittgata]

On the plus pin I get 23.5VDC and on the minus -23VDC.

 

[cowboybob]

On the plus pin I get 23.5VDC and on the minus -23VDC.

Ok. So the bridge appears to be working as it should. Something in the circuit after the bridge is causing the problem.

We're going to attempt to isolate what that "something" is. We'll test the KIS-3R33S PS's load(s).

TEST 1. Momentarily re-connect Vin to the trace.

Disconnect the Vout pins of the KIS-3R33S PS from the PCB traces (as you did with the Vin pin):

Re-power the PCB and note the voltage on tube pin 1.

 

[bittgata]

In the picture below Test 1 the arrow is pointing at the ground, I think that's a mistake but if it's not a mistake I have no idea you mean to show me.
Also Regarding Test 2 you say " If, pin 1's voltage drops to previously noted very low levels", the voltage on pin1 was always low with or without the vin connected so it can't drop to the very low level because it's already there. Do you mean if temporarily reconnecting the Vin while the dac is powered on doesn't do anything to change the voltage on pin1 then proceed with test2 or did I get that wrong?

 

[Les Jones]

I agree that the arrow is poining to the wrong pin. The Vin pin is the one to the left of the arrow. You also say in post #42 that disconnecting it does not change the plate voltage
I think CBB may have miss read post #44 as I did when I first read it. Where you said "On the plus pin I get 23.5VDC" I initially read it as "on pin 1 I get 23.5VDC"
You will have to trace the track on the board from the positive output of the bridge rectifier until it gets to the plate pin on the valve. Show all of the components connected to this track.

Les.

 

[cowboybob]

bittgata, please excuse that error.

I rewrote Post #45 to just test the loads of the KIS-3R33S PS and how their absence affects (if at all) the bridge PS circuit. Please re-do the test in Post #45.

Les is right, although, at the moment, I'm working backwards with traces and components we can see.

The traces for the ±22VDC supply are very hard to follow. I think it's only a double sided board but can't really say for sure that there's not a third layer between the two.

 

[bittgata]

I managed to disconnect the both Vout connections and then I temporarily reconnected the Vin. The plate voltage still has the same very low value.
PS:
I've mentioned this before but I imagine you have other things to do besides reading the entire thread every time I post something.
Though there is a large difference between the very low plate voltage of the "bad" DAC and the ~22V of the "good" DAC the Vout is also lower ~16V on the "bad" DAC vs ~22V on the "good" DAC. I also checked the Vin voltage after I desoldered the Vout and it did not change, still at around 16V.

 

[cowboybob]

OK. It's looking like the problem is in the primary, ±VDC supply circuit.

I've been attempting to "reverse engineer" the primary power supply (pps) and it's slow going, especially without the actual board in my grimy hands. Many of the traces disappear under components as well as pass between the top and bottom through a "PCB via". This is not going to be easy and might very well end in failure.

I'd like to avoid just a willy-nilly parts replacement approach, but that may become necessary.

But first, how comfortable are you with exposing (by removing the dark green cover film) a PCB trace? And then severing that trace? The idea is to isolate the pps, going backwards from the loads (as best as I can identify them) until the "bad" components can be discovered. Part of this process is to restore the trace if that particular change proves nothing.

 

[bittgata]

First thing first, I am not at all uncomfortable severing a trace but there is a high chance I will screw that up somehow so I would rather not. Regarding the PCB film, probably the blue coating in this case, I did not even know that could be removed but would it help if I just removed the film without severing any traces? How and can this be done with components connected to the board? Would this have long term effects/ would I have to apply another coating?

Without doing any of the above I've been thinking the reverse engineering might take a lot more time so it might be worth to wait for the ESR meter and start testing every component that I can, might be most or just some of the 2 and 3 pin components, before doing anything else.

Could you direct me to the areas or some of the components that look like they might have anything to do with the ±VDC supply circuit? I will test outside of this if I don’t find anything but it’s better to have a place to start.

PS: Is it likely or possible that board itself has developed a defect?

 

[cowboybob]

Sorry for the delay.

First thing first, I am not at all uncomfortable severing a trace but there is a high chance I will screw that up somehow so I would rather not. Regarding the PCB film, probably the blue coating in this case, I did not even know that could be removed but would it help if I just removed the film without severing any traces? How and can this be done with components connected to the board? Would this have long term effects/ would I have to apply another coating?

I understand. But I have exposed (the blue comes off rather easily) and severed many a trace. If you choose to do so, I can guide you. We would only sever traces between components (to isolate them, or additional down stream circuits) in easily accessed areas of the PCB. There are no long term effects I am aware of or have experienced. And replacing the coating (after removing any flux) is recommended after you've reconnected the trace edges.

... so it might be worth to wait for the ESR meter and start testing every component that I can, might be most or just some of the 2 and 3 pin components, before doing anything else. ...

Works for me. Although, understand that the ESR meter is only used for testing capacitors, not many of which, anymore, have more than 2 pins.

I'd start with the two electrolytics (silver top with the embossed cross, note polarization - gray stripe should be minus\Gnd) and the two polyesters (C33 & C34, red rectangles - NOT polarized).

At the moment, I very curious about the cap above with the red X (the smoothing cap for the +22VDC side of the PS.

Don't forget that the negative VDC side of the PS seems to be working just fine.

 

[bittgata]

I've ordered one of these. Not sure If it's gonna give me a reading for the 3 pin components in circuit but if it does I can just compare them to the good dac so it shouldn't even matter if it's accurate or not.
I ordered it at the same time with a few things I already received so hopefully I'll have it soon.

 

[Nigel Goodwin]

I've ordered one of these. Not sure If it's gonna give me a reading for the 3 pin components in circuit but if it does I can just compare them to the good dac so it shouldn't even matter if it's accurate or not.
I ordered it at the same time with a few things I already received so hopefully I'll have it soon.

I don't really see as an ESR meter is going to help you? - electrolytic's go high ESR due to poor design, misuse, high temperatures, and old age - none of which apply here.

What you need to do is post the schematic - if you can't get hold of it, then draw it out. Without that everyone is just making totally blind guesses, valve technology is crude and simple, and dead easy to repair - but you can't do it with a blindfold on wearing boxing gloves (which is pretty well what's happening here).

 

[cowboybob]

True, Nigel.

But the problem seems to be in the (also simple) ± VDC PS, for which there are, at the least, some identifiable traces and components that shouldn't be too difficult to remove/replace (if, of course, that's where the problem lies).

I think it's a worthwhile exercise for the OP, even though it may very well get stopped dead in its tracks if anything else, other than the PS's is/are the problem. Back engineering the board would be, essentially, impossible and not worth the effort. Especially at a distance.

 

[bittgata]

After one and a half years I’m back. I had some house related problems to deal with after which I haven’t had time and now I have time.

Long story short, I know very little about electronics, I have two boards of the same type mentioned throughout the thread, one fully functional and the other defective and I’ve just tested and probably, I think, determined cowboybob was correct in saying it’s a problem in the VDC PS so despite the time gap most, if not all, needed info is probably in this post. If anyone still wishes to help me that would be awesome. The tests/measuring results are below.


On the good board
Taking polarity into account I get about 21.5V on all the caps(the two big caps, C33, C34 and C38) as soon as the board has power.

On the bad board
I get about 22.5V on the big –cap and the C34 cap.
On the big +cap the electrical polarity is reversed (compared to what it should be acording the markings, the other cap and both caps on the good board) and it only slowly rises after power up, after 30s it’s still at -0.3V.
On the C33 cap I get 23V, however, unlike the other caps, the voltage drops gradually and very slowly after power off, whilst on the good board it’s at 1.2V two seconds after power off.
C38 does the same thing as C33.

The cylindrical part between the two poly caps and the big electrolytic cap is marked L3 so I assume it’s an inductor, I have no idead what it’s doing there and beyond measuring it's resistance, which is about 0.3ohms on both boards, I how or even if I should test it.


I also got a very cheap ESR meter. I’ve used it for some tests with the capacitors still on the board. For whatever reason I got 1.3ohm resistance on the leads but I got at least one ESR value for all the caps with the same value as the lead’s resistance but it did vary a lot between tests so I won’t mention it but I will mention the capacitance and vloss because even if they’re off by a lot they were fairly/somewhat constant between tests for the same part.
The big caps are 3300 microfarads, 25V, the poly caps rating is obviously not the one below but there are clear differences between the boards, maybe this will paint a clear picture for someone else because I thought the +electrolytic cap was bad but after I saw these results I have no idea.

_________________bad board____________________________________good board
big -cap_________ 3458 microfarads, vloss3.5%__________________ 3493 microfarads, vloss3.9%
big +cap_________3376 micro farads, vloss3.5%__________________3612 microfarads, vloss3.8%
c34______________3470 microfarads, vloss3.5%___________________3481 microfarads, vloss3.9%
c33______________7609 microfarads, vloss 0.7% _________________ 3612 micro f, vloss 3.8%
c38______________8003 microfarads, vloss 0.7%__________________3556 microfarads, vloss 3.2%

Useful info mentioned in other posts: the PSU is 15-0-15V AC and the voltages on the bridge (below C33 and C34 in the first picture) vary a little between the two boards, 21.5V positive and -21.5V negative on the good board and 23.3V positive and -22.9V negative on the bad board. cowboybob told me this is fine.

All suggestions are welcome,
Thanks!

 

[rjenkinsgb]

That sounds like the bridge rectifier has failed and instead of directly supplying each cap with + / - to 0V, one has no proper power feed and the circuitry connecting to the (working) opposite polarity one is causing the reverse bias.

This is the principal of that style of dual-polarity PSU:
https://qph.fs.quoracdn.net/main-qimg-225584e3f63ba8b4033f354824101c08

Check continuity from the bridge + & - outputs to one terminal each of the main caps. The other terminal of each should be the 0V or ground and the centre tap of the AC input.

If continuity is OK, replace the rectifier.

 

[bittgata]

Continuity is not ok. I'll add two pictures with continuity in the area on both boards, green is continuity, red is no continuity. The small C38 cap on the other side has continuity with the poly cap on the plus side of the circuit on both boards but I could not show it because it's on the other side.
Could it be a bad trace or is it something else?

 

[cowboybob]

... If continuity is OK, replace the rectifier.

I agree.

Did you find your solder plunger? Removing a 4 pin device ( the bridge) can be a bear and takes a lot of patience, especially if you don't have a soldering device that will span all four connections at once.. A "rocking horse" action (along the long axis of the bridge) is sometimes needed. And be very aware of surrounding components to avoid ancillary damage.

<EDIT> Just saw your new post (added while I was hen pecking my response...). Now I am flummoxed...

 

[cowboybob]

Using a known, good circuit ground (from elsewhere on the PCB), what are the voltages present, on the good board, for the four connections in the below picture?

Then, if you would, please post the voltages for the same points on the "bad" board.