AU-555A adjustment

[biguej]

Hello,

In 2013 I bought a Sansui AU-555A.
Since then it had been working fine, at least for my expectations, except for the phono but I did not use it.

A few months ago, after reading, I don't remember where, that after 50 years an amplifier with original electrolytic capacitors (EC) should be considered a fire hazard, I decided as a first step to replace these capacitors on the phono (F-1267) board which had electrolyte (presumably) stains. And it actually fixed it.

More recently, one channel completely died so I decided it was time to replace all the remaining EC. I found a very detailed post on this forum from Leestereo about restoration and upgrade of an AU-555A but I chose to stay as close to the original as possible, at least for the moment.

The main (F-1266) board had also a lot of stains, so after de-soldering all the EC, I tried to clean the board with alcohol (EtOH/MeOH), cotton buds and a toothbrush. But finally I dipped the board in alcohol just up to the board surface for 24 hours in order to dissolve the dirt. The bath was completely yellow after that bath so I finally rinsed and dried the board.
By the way, 4 resistances that I measured too low when mounted, but at the correct value when de-soldered, were also replaced.

Then I soldered the new EC, double-checked everything and reassembled the system.

I bought a 8.2 Ohm/50 W resistance (manual recommends 8 or 16 Ohm/50 W) from a local store to load the system while doing the adjustments. The seller told me to be careful and not use it for too long because it is supposedly inductive and could re-inject a voltage inside the amp. I did not really get that, but I got that it gets freaking hot when pushing the volume knob quite high.

Then I followed the instructions, feeding a +/- 205 mV (manual recommends > 200 mV) 1 kHz sinusoid in the left Aux input and measuring the output wave on the left A channel with the load in place. For both right and left channels, I obtain a nice looking sinusoid. If I play with the corresponding adjustable resistance, I can make appear and disappear clipping but only on the upper part of the sinusoid, never at the bottom. The same applies at low or high volume. The same also applies if I bypass the pre-amp board (F-1272) by feeding directly to the main amp. after disconnecting the bridges.

Since it is the first time I am doing that, I am not sure if the clipping behavior is correct or not. At least I reckon I cannot achieve what is recommended by the manual, namely adjusting the resistance in order to get a symmetrical clipping at both the lower and upper part of the sinusoid. I don't know if adjustable resistances age well, so in case, I also bought replacement ones that I could use.

On the schematics that I have, some voltages are indicated, so I plan to start measuring these to provide more clues. However, I was wondering if I need to perform these measurements while the amp is loaded and if so, at what volume etc.

If my description rings some bells, I would really appreciate recommendations and comments.
Thank you very much for taking time to read this.

 

[Watthour]

With a cap-coupled output and quasi-complimentary pair operating from a single power supply, balance is more difficult to control. It may help to read through this thread and consider some of the suggested modifications:

https://audiokarma.org/forums/index...ade-of-a-sansui-au-555a.816754/#post-11495731

 

[biguej]

Hello,

I had to leave this project for a moment but I am back on it.

I also had to replace the 4 variable resistors and finally I was able to adjust the clipping by feeding a +/-205 mV sine amplified at full volume, I just needed a 10x probe (see picture 1 for right channel and picture 2 for left channel, voltages must be multiplied by 10).

Then I adjusted the main-amp current as described in the manual, at 20 mA.

Finally I assembled everything and connected the amplifier to the speakers and started a test. Unfortunately I did not get any sound from the left speaker, unless I switched to mono mode. I tried with another cable, playing with the tape mon switch but it did not help... So I am back on the bench.

Here is what I tried so far:

1) right channel

+/- 205 mV sine fed from a function generator to aux input --- pre-amp ---> +/- 2 V good looking sine measured at the pre-amp output
+/- 2 V sine fed from a function generator to main-amp input --- main-amp ---> stable asymmetrical clipped signal (see picture 3, voltage x10) - adjusting R812 can only make things worst

2) left channel

+/- 205 mV sine fed from a function generator to aux input --- pre-amp ---> +/- 2 V good looking sine measured at the pre-amp output
+/- 2 V sine fed from a function generator to main-amp input --- main-amp ---> asymmetrical clipped signal with slight fluctuations on the top part (see picture 4, voltage x10)

I have the following questions:
- Since I can measure an amplified signal even on the left channel, why can't I hear anything from the left speaker?
- Is it normal that the main amp, amplifies asymmetricaly?
- I was suggested to replace the bias transistor, no matter what for the 555A. Are they responsible for the bad amplification?

Thank you.

 

[ecluser]

When we do the "symetrical clipping" adjustment, we must have an input voltage just sufficient to drive the amplifier at the edge of clipping. Hard clipping as you do is useless. Personally I am using triangular signal. The clipping is much easier to observe with a triangular signal. If you think about it, the top and lower parts of a sine wave signal are already flat.

Unfortunately I did not get any sound from the left speaker, unless I switched to mono mode.

The problem is before the mode switch. Clean again the input selector and tape monitor switch.

 

[biguej]

Hello,

Thank you for the advice.

I readjusted the clipping with a triangular signal (1.5 Vpp @ 1 kHz) directly fed in the main amp.
I obtained something similar for both channels (picture 5 = right; picture 6 = left) even if for the left channel there is still a flickering on the top part of the signal.

I also cleaned the selector and the tape monitor switch without any success. I tried with headphones and in fact I can hear a very very faint signal on the left channel.

I also measured the reference voltages indicated in the manual:
-> 1,2 V: measured 1,2 V (L) and 0,6 V (R);
-> 9 V: measured 8,3 V (L) and 8,2 V (R);
-> 27 V: measured 23,6 V(L) and 19,5 V (R);
-> 55 V: 50,8 V;
-> 56 V: 51,3 V.

I also wanted to check the clipping when introducing the signal in the Aux input. I had to change the input signal for a triangular 200 mVpp @ 1 kHz. The clipping is not symmetric anymore and the signal shape is affected (picture 7 = L; picture 8 = R). Is that normal ? Should I favor an adjustement when the signal goes through the whole chain?

I am still confused by the fact that the signal seems to be very similarly amplified on both channel whereas nothing can be heard on the left. Is it because I only tried with 1 kHz?

 

[ecluser]

The shape is not the same on the aux input because that triangular signal contains a lot of harmonics and the frequency response of the preamplifier section is not as flat as the power amplifier section. If you play with the tone controls you will see that it changes the shape of the output signal. When you have the possibility to connect the input signal directly to the power amplifier, it is preferable for this adjustment.

For the weak signal at the left output, even on headphones, but not on the scope, I think that the output capacitor is bad, or there is a bad solder on one side of that capacitor. The input impedance of the oscilloscope is so high that the signal can pass to the scope. Connect your headphones and the scope at the speaker outputs. The left signal on the scope should be very low by comparison to the right output signal.

 

[biguej]

Hello,
Thank you for your suggestions.

I did not know about dry solder joints. So I de-soldered and pumped the old Sn on all the capacitors I had replaced and also on some wires, before soldering back everything as cleanly as possible.
Facing no improvements, I took out the tape monitoring switch. I realized it was possible to dismantle it so I did it, cleaned the contact with fine grain sand paper and added a bit of oil before a final re-assembly. I took some photos that could be helpful for someone else.
No improvements either. Then I swapped both coupling capacitor and there is no change: the left channel is still the one that is not working and the right is ok.

So the issue is from somewhere else.
I will continue investigate the left channel for possible bad solder joints.
Is it possible that one of the transistors is faulty or the symptoms would be different?

 

[biguej]

Out of ideas I swapped my aux cable and realized the cable was bad
I tried with another cable and indeed both channels are working.
I had already tried this good cable before so I must have solved the issue with one or several of the things I tried before (switch, solder joints etc).
Well thank you very much for the help!