Help with Sansui 9090db bias and dc offset.

[arfz28]

I have finally found the time too work on my friends sansui 9090db. Its been sitting for a couple of years and was stuck in the protection mode with a flashing red indicator. It was either I can try and fix it or it would sit on the basement self, he doesnt want to spend any money on the old sansui. Thanks to Lapete for the info on the R33 and R34 resistors on the power supply board. He said they could be open or out of tolerance. Now believe me I know hardly anything about fixing stereo equipment or electronics, about the only thing I can do is deox the pots and switches that is it, I can solder perty good at least get the job done.As with working with a multimeter not good at all dont understand most of the measurements. Anyway after reading the meters book I figured out how to check OHMS. I found that the R33 resistor measured 307 OHMS way to high because the resistor should be 180 OHMS, the R34 resistor measured nothing it was open.So I put in some new resistors 1/2 watt 180 Ohm, I think the factory ones are 1/4 watt 180 Ohm, but I read somewear on the forum to up them to the 1/2 watt ones so I did.Man this was the first time soldering a circut board I was very carefull and it turned out very nice I was scared to death of burning up the board but it was no problem.
So now time to see if the 9090db would go up in smoke or work. Crossed my fingers and hit the power button and the red light blinked for about 2 seconds and turned green wow!!!!! I got my cd player and hooked it up to the sansui and was using a set of junk speakers (old panasonic ones), and I have music couldnt beleive it the old sansui works. The only thing that doesnt seem to be perfect is the balance, the left channel is slightly more loud than the right, not buy much thow,if I take the balance control and move it about a quater of an inch to the right side it evens out, also the right power meter reads lower than the left when playing music. So I switched the speakers on the back of the receiver, still the left side was a shade louder than the right so I know the problem is in the amp and not the speakers. Dang, deoxed the volume and balance control again no luck. I tried the tape input with my cd player and aslo the AUX input still the same, even switched the CD plyers right and left cables from left to right and right to left nothing but the same problem.Every control and switch I cleaned the best I could even the dolby switch which is a bite to get too. Any ideas beside using the balance control to even things out?
I know also that there is something to do with DC offset and bias that can be checked but I dont know how to do it could this be the problem maby?How hard is it to bias and check the DC offset any help would be great.

 

[glen65]

Start here first and see if any of this info helps.
http://65.219.61.150/forums/showthread.php?t=5634

 

[arfz28]

Thanks Glen I read that post about 3 days ago and that is were I found out about the dc offset and bias. Not much on exactly how to do the 9090db, but I do know I could smoke the amp if not carefull, that is why I will not touch the pots on the power supply board without exacly knowing what to do.I had know idea that you had to bias a solid state amp I didnt think you had too, I have a scott 299a tube amp and I know how to bias it thanks to Crag Ostby who walked me threw it. I thought biasing was only for tube equipment but I am wrong sounds like the solid state needs it too.I dont know if the bias or dc offset wiill fix the balance problem or not its just a guess.

 

[glen65]

One thing I would want to do is to see if the preamp signal is
the same comming from both channels first before going too far. This could save you from going on a wild goose chase.

 

[Ecaroh]

I think your problem may be one of those pots (I know you already cleaned them). I think that's more likely than bias/offset adjustment. More likely the volume pot than the balance pot, unfortunately. The balance pot should be easy to check by disconnecting both channels from the pot, i.e., bypassing it. Taking the volume pot out of the circuit would be a nastier task.

Steve.

 

[EchoWars]

Bias and/or offset won't do what you describe.

You got dirty switches. Maybe a dirty relay. Remote possibility of some real component failure. The odds are on the first.

 

[CDFixer]

On several X0X0DB units that I have worked on, I have found some resistors on the driver board (the one that plugs into the power supply board) that were bad. All of these are on the bottom edge of the board, there are a couple of 150 or 180 ohm (depending on the driver board vintage), and four 4.7 ohm and a couple of 10 ohm resistors. I have found these have increased in value and things seem to work but the sound is either distorted or uneven.

Also be sure an clean the dolby switch this switch caused all kinds of problems with these units.

 

[arfz28]

Thanks for the help guys, I went ahead and cleaned all the contols and pots again maybe a tad bit better balace now not perect thow its mostly noticable at very low volume and when you turn the sansui up it seams to even out a lot better. I dont think any more cleaning is going to get me anywere, its really livable thow not really to noticable once the sansui is turned up to lisening levels.
I went ahead and read that post by echowars on DC offset measurements I took the readings and it wasnt too good in the left channel the one that is a tad bit more loud than the right. I came up with these measurements.
LEFT CHANNEL 210MV
RIGHT CHANNEL 34MV
Dang if that left channel had about the same reading as the right it woudnt be too bad but 210mv is way to high from what I have read in echowars post.
Now if I could only figure out how to lower the left side. I see some little pots on the driver board at the top, there is 3 of them one blue one and two white ones.Im not going to mess with them unless somebody can tell me which ones are for the dc offset, most likely the white ones because their are two of them wonder what the blue one is far? I will try and get a service manual for the sansui 9090db maybe there is something in there on how to set the dc offset.Well so far I learned a few things and am having fun hopfully I can figure this out.

 

[LBPete]

I came up with these measurements.
LEFT CHANNEL 210MV
RIGHT CHANNEL 34MV
Dang if that left channel had about the same reading as the right it woudnt be too bad but 210mv is way to high from what I have read in echowars post.
Now if I could only figure out how to lower the left side. I see some little pots on the driver board at the top, there is 3 of them one blue one and two white ones.Im not going to mess with them unless somebody can tell me which ones are for the dc offset.

arfz,
There are four little pots on the top of the driver board. They are arranged in left and right pairs. The two inner most pots are for DC offset, the left one is for the left channel and the right channel. They are very touchy to adjust. A little movement can give you a lot of voltage change. With a little patients, you can get it to spec. The service manual says 0+/_ 10mV.

Good luck.
- Pete

 

[Ecaroh]

arfz,
There are four little pots on the top of the driver board. They are arranged in left and right pairs. The two inner most pots are for DC offset, the left one is for the left channel and the [other one is for the] right channel. They are very touchy to adjust. A little movement can give you a lot of voltage change.

Arfz --

Please heed what Pete says about touchiness. If you really want to do this -- when you start, just twist the screw the tiniest possible amount, we're talking the smallest fraction of one degree that you can do. Then see what happens with the meter.

If the meter sails way off, then slowly slowly slowly produces a reasonable reading -- think about stopping right there. Or if the reading is worse than it was, you could try correcting for that.

The problem is: the pot is not only touchy, it is also old and it may not act right. In other words, a smooth turn may not produce a smooth, linear change in resistance. It may act up badly. (This is complicated by the fact that if you have some leaky caps, then you get the effect of a very very very slow return to the correct meter reading.)

The risk of all that is: the bias is also affected. And it is possible to do real damage with bias -- like ruining an output transistor.

Steve.

 

[arfz28]

LAPETE SAYS,There are four little pots on the top of the driver board. They are arranged in left and right pairs. The two inner most pots are for DC offset, the left one is for the left channel and the right channel. They are very touchy to adjust. A little movement can give you a lot of voltage change. With a little patients, you can get it to spec. The service manual says 0+/_ 10mV.
Well I went down stairs too check how many pots were on top of the driver board I swore I only seen three, two white ones and a blue one. AND checking again three OF them is all there is, two white pots marked VR01 and VR02, one blue marked VR03, this sucks someone has worked on this unit before, where the other blue pot is saposed to be in the vr4 slot they have soldered in a resistor, I have no idea what this is about. The resistor they soldered in looks like the size of a 1/4 watt resistor, it is color coded with 2 red strips and two gold ones. I ought to take a picture of this driver board so you can see it. Anyway what are the blue pots for Bias then? MAN why would anybody put in a resister instead of replaceing it with the right piece so a person could make an adjustment down the road this way its imposible. WELL at least the two middle ones are there for dc adjustment.

 

[eradoncic]

They are very touchy to adjust. A little movement can give you a lot of voltage change. With a little patients, you can get it to spec. The service manual says 0+/_ 10mV.


Hi guys,
I'm not familiar with sooo old Sansui stufs, but I'm pretty experienced with post 1987 models. Based on my expirience AND several japanaise reparation sites the "standard" and easy performing refreshment of old gear can be sumarized as follows:
1. Change all trim-pots (DC and bias) with multiturn ones. Do it one by one, setting them "half way" minimum to maximum, using simple DMM
2. Change all on board caps with some modern decent ones, e.g Panasonic FC or similar (there should not be too much there)
3. Check the speaker relay

Adjust bias around 40mA, and DC as suggested. The vlues up to +/- 30mV is acceptable, and this can be acheived even better using multiturns trimpots.
If you can not obtain a service manual, you don't need a tech for this, as you can't go wrong if you're working step by step (e.g: change one DC trim pot, turn on the amp, adjust that channel and so on).

 

[EchoWars]

VR1 and VR2 adjust the offset, VR3 and VR4 set the bias.

Looks like someone replaced the VR4 pot with a 22 ohm resistor, which means you have a driver board with a thermal diode for temp compensation/bias adjust (the low value gave it away). Why this was done, I dunno. As eradoncic suggests, replacing the pots with multiturn pots is a good thing (but you'd need to verify for certain that you have a thermal diode driver or transistor driver before you decided on a value to use...the transistor version has a 1K pot, and the diode version has a 100 ohm pot).

The offset is pretty easy to adjust, but with the stock pot is will be very touchy. Once you try to set it you will understand the value of a 25-turn trimpot.

Setting the bias is also a good thing, but Sansui's F'd-up procedure for bias on the 9090db doesn't help much and is a screwy way to get the job done. I'd recommend that someone who has some real experiance with this amp tackle it if you are considering doing more than battling the offset.

 

[dsk]

EW, I did not retain the email you sent to me describing all the stuff you did to my 9090db. If you have it, post it in this thread as an example of all the stuff that really should be done. As I recall, you went way beyond what was necessary to fix the fault it had, and this made me feel very good about the receiver, and about the repair cost. Arfz28, my recommendation: if your friend doesn't want to spend anything to fix the Sansui, it must mean he does not want it. Ask him to give it to you. Send it to EW, and tell your friend he can buy it back from you for the repair and shipping costs if he wants it back after he sees it and hears it post-EW.

 

[EchoWars]

Eh....I can't find it.

If arfz28 expresses interest, I'd be happy to go into detail on the rebuild, but I get the impression that he'd simply like it up and running with the correct adjustment pots.

 

[rek50]

This isn't the E-Mail, but maybe it's the basic idea. "The inside has been cleaned as best I can (it's a big improvement), and all the pots and switches were cleaned. The new Burr-Brown opamp is in (see pic), as well as the zener supply for it. In the pic, the board is still wet from the DeOxit spray. The wiring looks flaky with the board flipped over to take the pic...when reversed the wired are all relaxed and look a lot better (but the mod is still function over form, so neatness is secondary In the pic, to the right of the opamp you can see the BlackGate caps used for the input to the preamp. There are also a couple of BlackGate caps at the input to the phono section, as well as many many many new Panasonic FC capacitors in the power supplies, tone controls, protection circuits....etc etc. About the only part of the amp that doesn't have new caps is the tuner and the Dolby section (which does you no good since no one is broadcasting in Dolby...too bad)."

 

[LBPete]

As eradoncic suggests, replacing the pots with multiturn pots is a good thing (but you'd need to verify for certain that you have a thermal diode driver or transistor driver before you decided on a value to use...the transistor version has a 1K pot, and the diode version has a 100 ohm pot).

For what it's worth, I checked a couple of driver boards, the early ones with the thermal transistors have white colored bias pots, the later ones with the thermal diodes have blue colored bias pots.

Setting the bias is also a good thing, but Sansui's F'd-up procedure for bias on the 9090db doesn't help much and is a screwy way to get the job done.

How did Sansui screw up the bias adjust procedure? I've been working on an 8080DB and can't get the bias current on either channel up to the 30mA spec. One side won't go above about 18mA and the other side won't go above 24mA. The service manual has you connect the ammeter across a fuse terminal, with the fuse removed. Should I be doing it a different way?

- Pete

 

[EchoWars]

Well, it's not the actual value of the measurement I was referring to, but the removal of the fuse to use the meter as a current measuring device.

I'd suggest that, with power off, get back there to the actual transistors themselves and for each channel, clip a meter lead to the emitter of the output transistors and measure the voltage from emitter-to-emitter and calculate the bias current.

Thing is, if you were measuring correctly, then you will likely read a smaller value, since the fuse method shows current to the entire driver/output stage. If that is the case, you got problems.

 

[arfz28]

Well I got the DC offset set perty close, those pots are about some touchy DEVELS, now I know why people change these pots out with better ones. I let the receiver warm up for about a half an hour before adjustment. I got the right channel at around +2mv and the left at about +4mv took me a many a try to get the channels this closely matched. The meter seems to jump around about 2 to 3 mv per side I guess this is normal? One thing I notice is that after the receiver is on for about 4 or 5 hours the dc offset numbers go up by about 5mv per side. So after about 5 hours I have a reading of +7 mv left side and +10mv right side. I did this yesterday, this morning I wanted to see if when I started the receiver up if the numbers would be lower and then go higher as the receiver was on longer. Sure enough after unit was one for about 15 minutes I had readings around 1mv even going maybe -2mv left channel, right channel about 4mv and going down to around 1mv. After about 4 hours my readings were up again to about the same numbers as yeasterday give or take a mv. I think I will let well enough alone, its close enough for me.

I decided to check out the dc offset on my 8080 sansui also, it read 41mv right channel, 22mv left channel, looks like I need to fine tune this unit a tad bit also.

Thanks for all the help people, couldnt of did it without your help and it really wasnt that hard, just knowing were to begin and how to do it is 99% of the job. Thanks again all.

 

[EchoWars]

The meter seems to jump around about 2 to 3 mv per side I guess this is normal?

Purty much.

Nice job dude. :thmbsp: