AU-9900 static and hiss issue

[KeithD]

I'm in the process of restoring my AU-9900. It's gone well. Check on DBT then line voltage after each board. Did the F2568 regulated PS board first. Here is what was replaced.


All checked out OK, so moved on to F2567 replacing e-caps. Again, all ok. Same with F2570 equalizer board replacing e-caps and 2SC1313 and 2SA798 transistors.

Re-capped and replaced transistors on F2573, 2574, 2575, and 2576 together before testing. Here's the replacement list.



Passed DBT, but now I get a small background hiss at the speaker outputs and in headphones along with some intermittent and faint static/crackling. When I separate pre-amp and amp, the pre-amp is dead silent, so it appears the problem is coming from the Amp. However, I had not replaced anything in the amp at that point. Decided to got ahead and re-cap the driver board (F-2580). That went fine, but still have the same symptoms in the amp section. I'm not clear where to look for the issue, or where a good starting point is?

Also, when I listen through headphones or speakers, the background hiss is more noticeable with pre-out/main in connected then when not connected. Seems odd. I have not tried cleaning the pre-out/main in switch. The random static seems more like a bad transistor to me. Help!

 

[miromustang]

I had and I am still having the same issue as you KeithD on my AU-11000 which seems simmilar to your AU-9900 topologicaLly. I've replaced all components on the driver board literally and the issue remains. After a year visiting various reputable technicians I gave up and came to conclusion that nothing can be done to lower the level of noise/white noise. When looking at the bottom of driver board there are many bodge wires showing that Sansui guys had troubles with this model. If you look at 11000A model which came as a replacement later in production, boards are different and amp is totalLy silent.

 

[KeithD]

I had and I am still having the same issue as you KeithD on my AU-11000 which seems simmilar to your AU-9900 topologicaLly. I've replaced all components on the driver board literally and the issue remains. After a year visiting various reputable technicians I gave up and came to conclusion that nothing can be done to lower the level of noise/white noise. When looking at the bottom of driver board there are many bodge wires showing that Sansui guys had troubles with this model. If you look at 11000A model which came as a replacement later in production, boards are different and amp is totalLy silent.

Thanks for sharing your experience. Did you disconnect the driver board and see if the problem persisted? Just wondering if the output transistors could be eliminated as the source. I have to look at the schematic closer. I believe there are seven plug connectors to the driver board, and it would be good to know what is connecting to what. That's probably my next step.

I'm optimistic the problem can be solved. It is being caused by something. So, it's still a matter of investigating until it's identified. Just to be clear, did you replace all caps, transistors, resistors, trimmers, and diodes on the driver board? I think that would be everything. If so, and you heard no difference, the trouble must be elsewhere. I'm still wondering about the power supply. I replaced all of the transistors on F-2568 except TR07, since I did not have the replacement, and I did not read advice that it needs replacing. I did not replace any transistors on F-2567. I just don't know if this problem can be caused by an issue in the PS or not.

 

[miromustang]

Output transistors, in my opinion, should not be the cause for this issue.
I've replaced all the transistors driver and signal, capacitors and even all the resistors for which I thought might help the noise on the driver board. There might be something in the power supply but I did not bother with it too much - standard re-cap and uni-watt transistor replacement.
If you are going deeper into it and isolate the cause you will make a great favour to me and probbably somebody else too. I gave up on this Sansui a bit and went for AU20000 but I still have some hope for it. I am very reluctant to get rid of it.

 

[Hyperion]

For the signal path Bipolar Electrolytics I would have used UES range not UEP, and would also have increased their voltage rating to a minimum of 10V or 16V - I noticed you did this with other capacitors, but not these. The exception being C16 on F-2568 which correctly in my opinion is a UEP.

The AU-9900 has a signal path interrupted by many switches, which might be contributing the the noise. But you shouldn't have intermittent faint static/crackling, that certainly shouldn't be there. If you haven't done so yet, try replacing D05 and D06 (SV-02) with series pairs of 1N4148 diodes. Also F-2580 C03, C09, C13 and C15 could usefully be changed for C0G/NPO ceramics and could help with the background hiss noise. And another significant SQ improvement not related to the problems, change all the 4.7Ω base resistors (on F-2581's) for ½ watt metal films.

 

[KeithD]

For the signal path Bipolar Electrolytics I would have used UES range not UEP, and would also have increased their voltage rating to a minimum of 10V or 16V - I noticed you did this with other capacitors, but not these. The exception being C16 on F-2568 which correctly in my opinion is a UEP.

The AU-9900 has a signal path interrupted by many switches, which might be contributing the the noise. But you shouldn't have intermittent faint static/crackling, that certainly shouldn't be there. If you haven't done so yet, try replacing D05 and D06 (SV-02) with series pairs of 1N4148 diodes. Also F-2580 C03, C09, C13 and C15 could usefully be changed for C0G/NPO ceramics and could help with the background hiss noise. And another significant SQ improvement not related to the problems, change all the 4.7Ω base resistors (on F-2581's) for ½ watt metal films.

Thanks for those cap recommendations, John. I had followed the advice given in a thread I posted asking for same. @ivandezande recommended EP bipolar caps there, but will note yours for a future Mouser order.

I believe the background hiss is more significant than it was pre-restoration; however, it is very faint, so I can't tell for sure. It is slightly stronger with pre-out/main in connected than disconnected. Yes, there are a lot of switches, but working all of them does not present a noticeable difference in the sound in headphones or speakers. I believe if you set the filter/tone switch to "defeat" the signal does not go through those boards, is that correct? I will also say the intermittent static decreases with time. I listened to about 30 minutes of music last night, and by that mark I was not hearing static. To give you an idea of the static level, with music playing at and volume at -33 and level set at -20 db, I don't notice the static. It does make the DC offset jumpy. It can swing about 30 mV when there is static.

I also wondered if I had a dirty pre-out/main-in switch, but disconnecting the connector to the driver board with pins 1-4 (signal from pre-amp) made no difference.

I will change those diodes. I do have 1N4148 on hand. I only changed the diodes recommended in the other thread I mentioned, which are the Zeners on F-2568. I'll take a look, but I doubt I've got those small capacitors or resistors on hand.

 

[ivandezande]

The UES caps have a short life compared to the UEP caps. I’m also not into “audio” caps.

 

[KeithD]

If you haven't done so yet, try replacing D05 and D06 (SV-02) with series pairs of 1N4148 diodes.

Well, that was a major advance forward. After replacing those diodes, with tone/filter in defeat position, there is no static. When I switch to tone/filter, I get a bit of static, but much reduced. I think my next step is to check the KSA992 replacements I put in for 2SC798. Maybe one of them is bad. I did check each one to find pairs with matched hFe and Vf in my stock of 100. But after assembling and soldering them in, I put some heat shrink over each pair and shrunk it down with soldering iron heat. Maybe one got too hot and did some damage? I already pulled the pairs on the midrange board, but no change, so next I'll check the ones on the treble/bass board.

Thanks so much John for pointing to those diodes!

 

[LAUST9000]

Well, that was a major advance forward. After replacing those diodes, with tone/filter in defeat position, there is no static. When I switch to tone/filter, I get a bit of static, but much reduced. I think my next step is to check the KSA992 replacements I put in for 2SC798. Maybe one of them is bad. I did check each one to find pairs with matched hFe and Vf in my stock of 100. But after assembling and soldering them in, I put some heat shrink over each pair and shrunk it down with soldering iron heat. Maybe one got too hot and did some damage? I already pulled the pairs on the midrange board, but no change, so next I'll check the ones on the treble/bass board.

Thanks so much John for pointing to those diodes!

Hi Keith

Found this thread searching answers and troubleshooting regarding some hissing noise from my AU-9900.
The unit has been completely recapped - but no other parts have been changed.
Was wondering if you ever resolved the hissing issue?
Regards
Laust

 

[KeithD]

Hi Keith

Found this thread searching answers and troubleshooting regarding some hissing noise from my AU-9900.
The unit has been completely recapped - but no other parts have been changed.
Was wondering if you ever resolved the hissing issue?
Regards
Laust

A good question. The restoration of my AU-9900 is mostly documented in this thread. In the end, the remaining static and hiss was coming from the tone control boards where I had replaced 2SA798 ICs with 2xKSA992s. The solder pads for the ICs are very small and close together. I damaged some in the desolder/resolder process, and I could cause the static just by touching the replacements with a chop stick. In the end, I purchased tone control boards on e-bay from a scrapped AU-9900A. Once I put those in the static/hiss from the tone control path was gone. Were I to do this all over again, I would not have changed the 2SA798 ICs. They were not causing me problems before replacement. 2SA798 is an IC that is known to potentially go bad over time, but if they are working OK, might be best to leave them alone.

As you saw in post #8, the major source of static was those diodes on the driver board. I also replaced the "uniwatt" transistors on the driver board, which I detailed in my other thread linked above.

The solder pads on the boards in this unit are a bit delicate, so you've got to be careful not to lift them when you desolder components.

 

[LAUST9000]

A good question. The restoration of my AU-9900 is mostly documented in this thread. In the end, the remaining static and hiss was coming from the tone control boards where I had replaced 2SA798 ICs with 2xKSA992s. The solder pads for the ICs are very small and close together. I damaged some in the desolder/resolder process, and I could cause the static just by touching the replacements with a chop stick. In the end, I purchased tone control boards on e-bay from a scrapped AU-9900A. Once I put those in the static/hiss from the tone control path was gone. Were I to do this all over again, I would not have changed the 2SA798 ICs. They were not causing me problems before replacement. 2SA798 is an IC that is known to potentially go bad over time, but if they are working OK, might be best to leave them alone.

As you saw in post #8, the major source of static was those diodes on the driver board. I also replaced the "uniwatt" transistors on the driver board, which I detailed in my other thread linked above.

The solder pads on the boards in this unit are a bit delicate, so you've got to be careful not to lift them when you desolder components.

Thanks a lot.
I got some reading to do. Good thing I enjoy the subject.
I will try another set of speakers soon. At the moment I’m using Technics SB-E100’s and they are fairly sensitive at 95db.
The Sansui just make them sing - and hiss a bit