Sansui A-1100 Driver stage issue and distorted output

Bassblaster

Super Member
I decided to pull out the Sansui A-1100 then been sitting on my shelf for over a year.

This model uses the STK3102II for the driver. whats going on it the right channel works perfectly no issues at all. Left channel is intermittently VERY distorted or doesn't work at all. I sat down with the scope and found the left channel is oscillating between 60 and 85ish kHz. i tested the input pins of the STK driver and the oscillation is on the left input so the issue has to be before the driver. when i tap the F-4773 control board (page 6 in the service manual) with a screw driver the signal cuts out, gets softer and louder, most likely a bad solder joint somewhere.

SERVICE MANUAL - https://www.hifiengine.com/manual_library/sansui/a-1100.shtml
STK3102II DATESHEET - http://pdf.datasheetcatalog.com/datasheet/sanyo/STK3122.pdf

1kHz tone - pin 15 on the STK
IMG_0292.JPG
1kHz tone - pin 1 on STK
IMG_0290.JPG
no signal, pin 1 on STK
IMG_0289.JPG
another no signal, pin 1 on STK
IMG_0294.JPG
 
its with an 8 ohm load and no load attached, it made no difference
it was running on a separate outlet not a powerbar.
all sliders and switches were cleaned. cool thing is the functions and tape switched are controlled by an IC so if the buttons were dirty it wouldn't hurt the signal path.
my wall voltage is a bit high at 126V, but im failing to see how that would cause the issue at hand here.

this thing isnt anything special. it would be nice to have as a beater test amp.
 
Last edited:
If you think you have pinned the issue down to the power amp stage including the STK - then I would be looking at C2, C5, C6, C10 & C11. And if you are going to replace them, replace with C0G/NP0 ceramic type.

Don't be too dismissive of the intermittant fault you discovered, it could be important - I suggest trying to find that first.

Good luck.
 
Last edited:
i'll pull the F-4773 control board and look for anything wrong there. The 5 caps you listed, i found C10L and C11L but i cant find C2, 5 and 6. where are they at?
This thing is a royal SOB to work on because there's not room in it at all.
 

Attachments

  • IMG_0304[1].JPG
    IMG_0304[1].JPG
    120.7 KB · Views: 25
It looks like the ones you can't find, kC2, kC5 & kC6 are on the F-4762 Driver Amp PCB. ;)
 
Yep, go for the low value ones first, kC2 last in other words.

I am guessing they are all ceramic capacitors? (kC2 may not be) if they are, remember to replace with C0G/NP0 ceramic type.

My thinking is that one or more of them has gone o/c or resistive, thus removing the anti-oscillation protection. It's your best bet to get the oscillation stopped - otherwise it may be the STK.
 
Yep, go for the low value ones first, kC2 last in other words.

I am guessing they are all ceramic capacitors? (kC2 may not be) if they are, remember to replace with C0G/NP0 ceramic type.

My thinking is that one or more of them has gone o/c or resistive, thus removing the anti-oscillation protection. It's your best bet to get the oscillation stopped - otherwise it may be the STK.
i may aswell just recap that entire board. not very many caps on it. it even has some bi-polar caps on it. i have a thing of 470pF C0G caps but nothing else.
 
okay lots to report now.
after re-soldering the entire F-4773 control board AND the F-4774 board and the intermittent sounds is mostly gone. when i tapped the F-4773 the right channel became distorted but changing the input to something else then back solved it.
the oscillation appears to be better, but i didn't fully test because i was sidetracked. the one time i checked i only saw a bunch of noise.

the LEFT channel just got more interesting, with a scope the wave is perfect where the emitter resistors meet WITH NO LOAD ATTACHED, the oscillation was NOT affected by the load.
IMG_0306.JPG
IMG_0308[1].JPG

NOW WHEN I ATTACH AN 8 OHM LOAD THIS HAPPENS
IMG_0307.JPG
IMG_0309.JPG
 
You have an open emitter resistor, or other resistor, or a blown driver/OP transistor.
Should i test the inputs and outputs of the STK driver with the load attached? and i already pulled all 4 outputs and they all test good in a transistor tester. Maybe its a rare case where it breaks down at voltage.
 
I was thinking that without a load attached the outputs are seeing infinite resistance which would preclude an oscillation issue. Also, that trace makes me suspect as hyperion stated that you have lost half of the push-pull bjt output which points to a failed emitter, opt etc.
 
the changing inputs thing is funny because its not a problem with the button as the inputs are IC controlled, and its not arching and pressing buttons really fast because it has a little beeper that beeps on EVERY button press.

As others have stated, how did you gain access to internals of pots and switches to properly clean them?
Spraying 'stuff' through the tiny holes or into their general direction is not cleaning them.

The pots, push buttons and switches need to be activated or cycled multiple times during the actual cleaning stage as the action of the contacts swiping each other is what removes oxidization - if and when you can actually access and spray the contact surfaces with deoxy.
I had to completely remove the front panel. and yes i know ive cleaned tons of sliders and switches.
 
i haven't checked driver output WITH a load, stand by im going to do that right now
LEFT OUTPUT of the STK WITH LOAD.
IMG_0314[1].JPG
IMG_0315[1].JPG
Well, looks like its either the STK pack or something on its PCB. would this be a bad time to say the STK pack gets REEEEALLY hot in maybe 5 seconds.
 
Last edited:
Meh, starting to wonder if its even worth to fix this, like its cool because it just SCREAMS the 80's. but the STK pack alone is worth more than the entire amplifier is.
 
I think ill give it some new caps since those are cheap and if that fixes the issue then yay! if not then for all this amp is worth i can gamble with an eBay pack. at least the STK3102II isnt as common as say the 0050 so its less likely to be counterfeit. if that doesnt fix my issue then ill just part it out and most likely sell the other 3 pieces. this thing has a real meaty DC fan, heatsink and transformer.
 
Bassblaster, it isn't the caps.

Check any ceramic emitter resistors, and if they are fine, the odds are still on the STK. Your oscilloscope shots prove that.

I like your little DSO138 kit BTW. :)
 
Well, what do you guys think about this piece of BPC? Anything else i should test or scope?
EDITTTTT: I FORGOT the scope pics with the oscillation on post 1 was on pin ONE which in the INPUT pin. so the STK might be fine! we all kinda forgot about that...
 
Last edited:
Well it magically works now. it was 3AM and i resoldered the entire driver board and swapped the outputs and it started working perfectly!

Now since there's no way in hell i was going to get the emitter resistors back in under the heatsink, i just mounted them on the bottom. since this is just a piece of BPC and its not worth anything i didn't really care how it looks
IMG_0361[1].JPG
IMG_0358[1].JPG
Plenty of room
IMG_0359[1].JPG
IMG_0362[1].JPG
 
Back
Top Bottom