Sansui AU-7500 relay not clicking after partial recap/transistor replacement

[cobascis]

I have been following @Leestereo's excellent thread documenting the full service of a Sansui AU-7500.

I replaced all the large chassis mounted caps in the power supply first. Tested the unit and relay popped on, speakers idled. I then recapped the power supply PCB, tested, all is well. (When I test, I am using a 60w DBT, if it dims, then I plug into wall voltage).

This is where I got a bit too hasty and messed up:

I recapped the main driver board with the exception of C811 and C812, as I didn't have those values on hand. I then replaced the 4 transistors (making sure to check the alignment). I should've tested at this point, but I had a few more caps lying around so I decided to replace C903 on the protector board with the exact same value electrolytic, and I replaced C601,C602, C625, C626 on the phono/tone board, as well as TR 601, TR602.

That is all I have done to the amp. I tested with a DBT, no short. Then I booted it up. No click from the relay. If I run through Aux or Phono and turn the volume up a bunch I can hear very tinny high frequency signal leaking through... but the amp isn't switching the speakers on. I just doubled checked all cap alignment. I replaced C801 and C802 on the driver board with bipolar electrolytics.. but I doubt that would be the cause.

I was very careful while doing this and am really puzzled. (Other than my foolish mistake of not testing before moving on to other boards) Thoughts? Could a mistake on the driver board stop the relay from working? Or should I replace the cap I swapped on the relay /protection board back with the original? I tried two correct value caps there and no change.

 

[cobascis]

I feel like the only reasonable explanation is flipped transistor installation, but I checked to make sure the 992s where ECB, and they match the board and Leestereo's thread. Just spent another hour staring at this thing and can't figure it out. Everything appears to be correct. Wouldn't a flipped transistor cause a short?

I'd imagine if the relay isn't going, I should be focused after the preamp section of the amp, right? Driver Board/ Power Supply / Protection Board

 

[sregor]

Start with the simple. Check fuse F02 (1 amp).
Recommend you put a sleeve of tape on your probes so only the tip is bare, or use clip leads and connect with power off.
On board F-2041 (relay) board check points 9Y and 9Z For DC - these are the outputs before the relay. Takes about +/- 0.5 volts (r more) to inhibit relay should be less than 50 mV on a properly looking amp. If DC something has gone wrong in the amps.
If everything good up to here, start trouble shooting the Protection circuit. Make sure you have supply voltage on 9T and 9S. The way the circuit works is if too much DC on an output, there will be enough voltage to turn tr901 on, which turns tr902 off. If Tr901 is off, there should be voltage on base of tr902 and the relay should energize.
One comment - the transistor on the F-1215 board tr903 is a "ripple filter" and there is at least one missing connection in the schematic - if you have wired it the same as originally found, it should be okay. (hard for me to be sure as I'm not looking at one...)

 

[sregor]

I feel like the only reasonable explanation is flipped transistor installation, but I checked to make sure the 992s where ECB, and they match the board and Leestereo's thread. Just spent another hour staring at this thing and can't figure it out. Everything appears to be correct. Wouldn't a flipped transistor cause a short?

I'd imagine if the relay isn't going, I should be focused after the preamp section of the amp, right? Driver Board/ Power Supply / Protection Board

FWIW I always verify base of transistors with meter - collector is in center, and will only conduct in one direction to base...not all transistors are the same, and is a simple sanity check.
Depending on the transistor, a flipped transistor may not show up as a short.

one more comment - the "ground" for the protect circuit is floating (but limited by diodes) and most accurate measurements of voltage (except for the output from the amp) is using pin 9U for ground reference.

 

[cobascis]

Start with the simple. Check fuse F02 (1 amp).
Recommend you put a sleeve of tape on your probes so only the tip is bare, or use clip leads and connect with power off.
On board F-2041 (relay) board check points 9Y and 9Z For DC - these are the outputs before the relay. Takes about +/- 0.5 volts (r more) to inhibit relay should be less than 50 mV on a properly looking amp. If DC something has gone wrong in the amps.
If everything good up to here, start trouble shooting the Protection circuit. Make sure you have supply voltage on 9T and 9S. The way the circuit works is if too much DC on an output, there will be enough voltage to turn tr901 on, which turns tr902 off. If Tr901 is off, there should be voltage on base of tr902 and the relay should energize.
One comment - the transistor on the F-1215 board tr903 is a "ripple filter" and there is at least one missing connection in the schematic - if you have wired it the same as originally found, it should be okay. (hard for me to be sure as I'm not looking at one...)

Hugely helpful post, thank you Steve. All 4 1a Fuses are good. As is the main 3a fuse.

Here's what I got (measuring these from the solder point on the underside of the board that is labeled on top)
9Y: 0.198 V DC
9Z: -0.322 V DC

So, both way over 50 mV. Maybe enough to trigger relay?

I'm having trouble finding a 9T and 9S. Which board are they on?

At 9V on the relay board I have -12.45 V DC, and at 9U I have -12.5 V DC.

Good tip on checking the transistors, I verified these are ECB (as I installed them).

 

[sregor]

9V and 9U are the transformer leads, and should have AC between them, but the - DC bothers me. Did you check the fuse? The transformer is floating, so DC can be there, but shouldn't be that high...It's a little tricky as it does float, and can go negative. Tg\here are diodes which prevent it from going positive with resect to ground.
For DC measurements in the protect circuit, use 9U for the ground reference, not regular ground.
The 198 mV and -322 are high offsets, but shouldn't stop relay from turning on.

 

[cobascis]

9V and 9U are the transformer leads, and should have AC between them, but the - DC bothers me. Did you check the fuse? The transformer is floating, so DC can be there, but shouldn't be that high...It's a little tricky as it does float, and can go negative. Tg\here are diodes which prevent it from going positive with resect to ground.
For DC measurements in the protect circuit, use 9U for the ground reference, not regular ground.
The 198 mV and -322 are high offsets, but shouldn't stop relay from turning on.

yup, fuses are good (continuity checked all). Those numbers were with a chassis ground. I will recheck those values with 9U as a ground.

To be clear, 9Y and 9Z are little dots on the PCB that have no pin/test point. I am looking directly under the board for the soldered lead to test. (There are four wires soldered approximately under those dots)

 

[cobascis]

Putting the black probe on 9U, and the red probe on 9Y and then 9Z, I'm getting way more DC now. around 12V DC for both.

Also, 9V and 9U have 21.6 V AC between them, and 9V only has .01V DC when using 9U as a ground. Totally different reading than with the chassis ground.

 

[sregor]

Putting the black probe on 9U, and the red probe on 9Y and then 9Z, I'm getting way more DC now. around 12V DC for both.

Use 9U for measuring voltages around the transistors. The protect circuit floats, so voltages are relative to 9U. 9Y and 9Z are from the amplifier, and should be measured relative to ground. The protect circuit uses diodes D901-D904 to clamp the emitter of tr901 when the output goes positive, and to clamp the base to -o.2 volts when the output goes negative, and if too far either way, will turn on tr901 which shuts the relay off. Voltages around tr901 (relative to U9) should give you a direction to look for the problem.

 

[cobascis]

Use 9U for measuring voltages around the transistors. The protect circuit floats, so voltages are relative to 9U. 9Y and 9Z are from the amplifier, and should be measured relative to ground. The protect circuit uses diodes D901-D904 to clamp the emitter of tr901 when the output goes positive, and to clamp the base to -o.2 volts when the output goes negative, and if too far either way, will turn on tr901 which shuts the relay off. Voltages around tr901 (relative to U9) should give you a direction to look for the problem.

Black probe on 9U:
TR901:
Base: 13.3V DC
TR902:
Base: 18.91V DC

 

[sregor]

Black probe on 9U:
TR901:
Base: 13.3V DC
TR902:
Base: 18.91V DC

R906 open (10 ohm)? Emitter voltages? Transistors in right? ....

 

[cobascis]

R906 open (10 ohm)? Emitter voltages? Transistors in right? ....

Black probe on 9U:
TR901:
Base: 13.3V DC
Emitter: 19.81 V DC
TR902:
Base: 18.91V DC
Emitter: 19.75V DC

R906 reads 10.3 ohms in circuit. I didn't touch these transistors, also, they are stock. Only swapped the 4 on the main driver board, and two on the phono pre board. And c903 is the only thing I touched on the relay board at all. (replaced with exact same value, triple checked the polarity).

 

[sregor]

Looks like something open somewhere. One side of r906 goes to 9U. The other end goes to the emitters of tr901 and 902. 19 volts across it should get very hot! I'm signing off, so see what tomorrow brings....

 

[cobascis]

Looks like something open somewhere. One side of r906 goes to 9U. The other end goes to the emitters of tr901 and 902. 19 volts across it should get very hot! I'm signing off, so see what tomorrow brings....

So perhaps I should start checking for open resistors on the boards that I worked on? No idea how I messed this up, just checked all the cap orientation for the 4th time... everything is correct.

 

[sregor]

Check the voltages on R906. One end is connected to 9U - if not reading zero, then trace it back to the connection. At this point I would be looking for open connections - bad solder joints or even a crack in the runs on the PCB. It doesn;t seem to be a bad component problem, but a bad connection or cracked case - again it is only conjecture on my part....

 

[cobascis]

I found what may be the (or at least a) problem.

I checked continuity between capacitor legs on the driver board. C805 and C815 both were continuous from positive to negative terminal. The original board had tons of flux between joints, and I believe it was maybe being conductive between things. There were no visible solder bridges, but likely more solder than needed as I didn't wick out the old stuff. I desoldered them, scrapped between connection points, cleaned board with ethanol.

When the caps were both out, I checked for continuty between the holes in the PCB. Nothing. Reinstalled and turned it on. No relay. I then checked continuity again, and they both had continuity still. I pulled the caps again and they didn't have continuity out of the circuit.

Should I consider these caps fried?

 

[cobascis]

Check the voltages on R906. One end is connected to 9U - if not reading zero, then trace it back to the connection. At this point I would be looking for open connections - bad solder joints or even a crack in the runs on the PCB. It doesn;t seem to be a bad component problem, but a bad connection or cracked case - again it is only conjecture on my part....

Maybe disregard my previous post. Perhaps the caps just still had some voltage in them. I just tested again and they beep quickly and stop, which I read is normal for testing continuity as the cap charges and discharges.

R906 has about 19.9V DC on both sides of it when using 9U as ground reference. 8.5V DC with chassis ground reference.

 

[sregor]

Maybe disregard my previous post. Perhaps the caps just still had some voltage in them. I just tested again and they beep quickly and stop, which I read is normal for testing continuity as the cap charges and discharges.

R906 has about 19.9V DC on both sides of it when using 9U as ground reference. 8.5V DC with chassis ground reference.

So 9U trace goes directly to r906 - find the open....or jumper the trace.

 

[cobascis]

So 9U trace goes directly to r906 - find the open....or jumper the trace.

sorry, could you explain what you mean by that?

jumper what to what and what would the goal be, and what exactly do you mean by find the Open?

 

[sregor]

The red trace should all be the same voltage - somewhere there is a break - or bad connection at 9U point...