Sansui G-9000 Doubts with the power supply ripple.

[CBarboni]

Hello!!!
I am restoring a Sansui G-9000 that had a serious heating problem on one channel.
I did a thorough restoration based on other restorations published in AK.
After many twists and turns on the right channel, I cannot get the BIAS current below 20mV. (manual say 16mV)
And I could see that when the voltage ripple entering the F-2806 board on the external leg of resistors R23 and R24 is high (1.2 or more mV), the BIAS goes up to approximately 100mV.
But if the ripple stays at 600mV, the BIAS stays at 20mV.
That led me to think that there was a problem with the rectification (diodes or capacitors) on the power supply boards.

I measured the voltage at the transformer output and the waveform is clean and smooth.
After the SS-5 and SS-5R diodes it is not as nice as I imagined. And after the large capacitors (which are new Nichicom) there is a ripple of 600mV.

It seems too much ripple for this power supply.
Can anyone give me some opinion? Or have you seen this? What is the correct ripple for this equipment?

I am attaching some photos that show everything I said above.
Thank you in advance.

 

[Hyperion]

Between 20 and 50mV is normal (high current PSU supply), ripple with no input signal.

Obviously you still have a bias issue, you need to get that fixed. Other than that, the signal will modulate the PSU voltages, depending to some extent on the status of the main PSU filter capacitors. This is not a concern unless you are unable to produce the full specified output power or it causes distortion (or both).

 

[CBarboni]

Thanks @Hyperion.
The main filter capacitors are new.

Yes, there is a problem with the BIAS setting.
The way it works is as follows.
When turned on, the BIAS starts to rise to around 75 or 100 mV.
5 minutes later, the BIAS drops to around 20 mV.
After about 10 minutes, it rises again to around 50 mV or 75.
The problem is that at 50 mV everything starts to heat up. Previously, the BIAS reached 250 mV and the heat was perfect for cooking a fried egg.
I prefer the BIAS to be around 20 mV to ensure that the heat is not too much.

I am still looking for problems but I have already reviewed everything. I am still accepting suggestions.
Thanks

 

[Mike_S]

Is there a circuit that adjust bias based on temp? My amp has one, if that is out of spec you will not be able to set bias.
Most manuals do say to wait at least 10min before adjusting bias.

My thought is something is out of spec, or not installed correctly. Maybe check all the diodes around the bias circuit.

 

[CBarboni]

Is there a circuit that adjust bias based on temp? My amp has one, if that is out of spec you will not be able to set bias.
Most manuals do say to wait at least 10min before adjusting bias.

My thought is something is out of spec, or not installed correctly. Maybe check all the diodes around the bias circuit.

Thanks for the suggestion.
Yes, there is a part of the circuit that adjusts the BIAS with a small potentiometer and there are also 2 STV-3H diodes to adjust based on the amount of heat.
I'm thinking of changing those diodes because it's the only thing I didn't change.
Today I did a test measuring the temperature on the heat sink where those diodes are and I noticed that there is a limit near 40 degrees Celsius (104F). The BIAS had gone up to 50mV but it doesn't go down less than that. That happened approximately 10 or 12 minutes after powering on.
The difference with the other channel is that on the left channel I can lower the BIAS up to 5mV with the potentiometer at full. But on the right channel it only goes down to 20mV.
Tomorrow I'm going on vacation for a week and I won't be able to do tests but I will read all the suggestions that come in.
Thanks