compworld
Active Member
Both 18 and 19 have 66.1 dc voltage!
Need Help Sansui AU-20000 protect mode
- Thread starter compworld
- Start date
Both 18 and 19 have 66.1 dc voltage!
richardpp
Well-Known Member
Very carefully check the rail voltages, +68 Vdc at point 25 and -68 Vdc at point 26 on each of the Emitter Resistor Boards (F-2497).
Power down then check for shorts between collector and emitter of the output transistors with a DMM. A simple check would be to measure the resistance between the orange wire (emitter) and the red wire (collector) for each transistor on each heatsink assembly. Measure both ways. You need to probe inside the heatsink to reach three of the orange wires, the other three are on the board’s edge. Where you can’t reach the red wire, probe the transistor case instead which is the collector.
A reading of 0 to low ohms indicates a short and the transistor is no good. An infinity reading indicates the CE junction of the transistor is good, not necessarily that the transistor itself is good. Ideally they should be tested on the bench, but this might give you a quick indication of why the amp is in protection.
Edit: Make sure main filter caps have discharged to a safe level before doing resistance checks.
Power down then check for shorts between collector and emitter of the output transistors with a DMM. A simple check would be to measure the resistance between the orange wire (emitter) and the red wire (collector) for each transistor on each heatsink assembly. Measure both ways. You need to probe inside the heatsink to reach three of the orange wires, the other three are on the board’s edge. Where you can’t reach the red wire, probe the transistor case instead which is the collector.
A reading of 0 to low ohms indicates a short and the transistor is no good. An infinity reading indicates the CE junction of the transistor is good, not necessarily that the transistor itself is good. Ideally they should be tested on the bench, but this might give you a quick indication of why the amp is in protection.
Edit: Make sure main filter caps have discharged to a safe level before doing resistance checks.
Last edited:
compworld
Active Member
Ironically got 68v plus and minus and and was checking the resistance across collector and omitter and 3 or 4 were ol and 6 were somewhere around 1.2M and 2 started lower but kept up after a minute. Here comes the good part started unit back up and it came out of protect! shut down and turned on back in protect! unplugged unit for a few minutes and fired it up and came out of protect and was playing music for a while. I would like to get to the bottom of the issue.
richardpp
Well-Known Member
… and I’m sure you will, they are encouraging results, but tread very carefully.
Be aware that the +/- 68 voltage supply lines have no current limiting protection, a shorted supply can do a lot of damage. Please have a look at this at this post by AK’er Trnsfmr R.I.P., ‘a must fix for safe operation’.
I don’t think there’s any getting around the fact that the heatsink assemblies will need to be removed at some stage.
Be aware that the +/- 68 voltage supply lines have no current limiting protection, a shorted supply can do a lot of damage. Please have a look at this at this post by AK’er Trnsfmr R.I.P., ‘a must fix for safe operation’.
I don’t think there’s any getting around the fact that the heatsink assemblies will need to be removed at some stage.
Take a look at my restoration:
https://audiokarma.org/forums/index.php?threads/early-au-20000-restoration-project-oh-boy.732316/
Install fuses in the B+ and B- leads to protect the outputs.
If yours is an early build you should remotely mount the large power resistors since the heat will damage the PCB.
I suspect that since this is an intermittent problem and you'll need to rewire the pass throughs on the power amp section of the amp.
https://audiokarma.org/forums/index.php?threads/early-au-20000-restoration-project-oh-boy.732316/
Install fuses in the B+ and B- leads to protect the outputs.
If yours is an early build you should remotely mount the large power resistors since the heat will damage the PCB.
I suspect that since this is an intermittent problem and you'll need to rewire the pass throughs on the power amp section of the amp.
compworld
Active Member
It's not hard, it just takes a degree of commitment.
compworld
Active Member
Tested the 12 transistors on one of the outputs and got 2 2sb554 that are showing up at diode on transistor tester so maybe that a good portion of the problem, going to pin the board while it’s out.
compworld
Active Member
Sorry about the delay but got all the transistors replaced and she is still stuck in speaker protect. Got 68mv on one channel and 89mv on the other on pin 18 and 19 of the protect board. The output boards are all pinned and the thermistors are 198 ohm and 298 Ohm. Kinda stuck any help appreciated, also the tr01 on the output boards are not present on either board.
richardpp
Well-Known Member
Verify that the power supply is wired correctly for the mains voltage in your area, and that you are getting the required supply voltages to the relevant boards:
+/- 68Vdc supply, both regulated and unregulated, to each F2507 Driver Board. In case of missing supply, inspect the traces on F2511 and F2512 for damage.
+/- 85Vdc to F2508 Regulated Power Supply Board
+24Vdc at the speaker relay coil on the Protector Board F2511
Be aware of exposed primary voltages on the protector board, please take care! Otherwise, good work! 68mV/89mV is a big improvement on 66.1 Vdc. I’m curious to know what you used to replace the Toshibas output transistors. Also what type did you use to replace the 2SC1364 transistors on the Protector Board F2511?
A few questions:
Is the main muting relay on the protector board closing?
Is the unit on DBT?
Have you attempted to set the bias?
Did you check the 0.47 Ohm emitter resistors were within 10% tolerance, not shorted or open … bit late for this, sorry.
... and a couple of suggestions:
Take your thermistor resistance measurements directly from pins 24/25 and 25/26 on F2511 respectively to rule out wiring issues.
There is a single VD1212 on each of the driver boards and a pair on F2508 Regulated Power Supply Board. these can be changed out for two 1N4148 diodes connected in series. Otherwise I would refrain from changing out any other parts unnecessarily until you find the fault.
Edit: ... "inspect traces on F2511" to "inspect traces on F2511 and F2512"
... "exposed mains voltages" to "exposed primary voltages"
+/- 68Vdc supply, both regulated and unregulated, to each F2507 Driver Board. In case of missing supply, inspect the traces on F2511 and F2512 for damage.
+/- 85Vdc to F2508 Regulated Power Supply Board
+24Vdc at the speaker relay coil on the Protector Board F2511
Be aware of exposed primary voltages on the protector board, please take care! Otherwise, good work! 68mV/89mV is a big improvement on 66.1 Vdc. I’m curious to know what you used to replace the Toshibas output transistors. Also what type did you use to replace the 2SC1364 transistors on the Protector Board F2511?
A few questions:
Is the main muting relay on the protector board closing?
Is the unit on DBT?
Have you attempted to set the bias?
Did you check the 0.47 Ohm emitter resistors were within 10% tolerance, not shorted or open … bit late for this, sorry.
... and a couple of suggestions:
Take your thermistor resistance measurements directly from pins 24/25 and 25/26 on F2511 respectively to rule out wiring issues.
There is a single VD1212 on each of the driver boards and a pair on F2508 Regulated Power Supply Board. these can be changed out for two 1N4148 diodes connected in series. Otherwise I would refrain from changing out any other parts unnecessarily until you find the fault.
Edit: ... "inspect traces on F2511" to "inspect traces on F2511 and F2512"
... "exposed mains voltages" to "exposed primary voltages"
Last edited:
compworld
Active Member
I've been on dbt to make sure nothing smokes and i get first relay to click and nothing on the speaker protect. at some point in its lifespan r36 got hot on driver board and looked burned out bad but tested good but replaced it anyways. There was a short on f-2497 to heatsink. Removed them again and rechecked all 12 and they are good but wondering if there was damage to driver board since r36 looked like a firecracker. Got all plus and minus 68vdc on the outputs and there is 24 vdc on the relay. It still stays in protect even with driver boards removed.
compworld
Active Member
2508 board voltages -81.9, 80.8, x, ground, 32.3, ground, -32, 69.7, ground, -69.7 and I tested 24 and 25 wire and get nothing 0L and 26 and 27 289 ohm
compworld
Active Member
stuck a 300 ohm resistor 24 and 25 connect and she came right out of protect! If that thermistor is bad what can be done to fix cause I’m pretty sure they are not attainable?
richardpp
Well-Known Member
These measurements look ok. Double check the wiring between F2497 and F2511. If you have 198 Ohms at the thermistor you should also have 198 Ohms at Pins 24/25.
One of the Murata PTFM series of thermistors could well do the job.
compworld
Active Member
Oddly i was getting reading at the thermistor but that has now disappeared. Is the thermistor actually that effective to begin with?
richardpp
Well-Known Member
To begin with, it is effective only in the sense that it is providing a reference voltage under normal operating conditions when the resistance value should be relatively stable.
When the thermistor senses a rise in temperature above a set point the resistance increases and causes the protection circuit to trip.
On the AU-20000, protection circuit trips when the thermistor reaches a resistance value of ~ 2.9KOhms. Of course there are other contributing factors as to when the protection circuit will activate in an overheat situation.
You might find post #9 and #11 interesting thanks to @EchoWars and @Hyperion:
https://audiokarma.org/forums/index.php?threads/pth487a-thermistor-substitute.231983/
When the thermistor senses a rise in temperature above a set point the resistance increases and causes the protection circuit to trip.
On the AU-20000, protection circuit trips when the thermistor reaches a resistance value of ~ 2.9KOhms. Of course there are other contributing factors as to when the protection circuit will activate in an overheat situation.
You might find post #9 and #11 interesting thanks to @EchoWars and @Hyperion:
https://audiokarma.org/forums/index.php?threads/pth487a-thermistor-substitute.231983/
Last edited:
compworld
Active Member
Thanks everyone we actually found the issue with the thermistor. the trace was messed up i guess we fixed it and now have it reading again and she fired right up and i was able to get the offset close to 0 and the bias around 35mv and ran it for 6 hours straight and she sounded great! Thanks all and onto the next project.
