Protect Relay issues - Sansui 881

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Hello,
I just registered to explore some opinions. I've been on this project for a few months now and have made some progress. My apologies that this may get a bit long.... Sansui 881 - Owned since new / Owned since new and played well until October 2020 / Protect relay stopped closing / then the fun started. I probably should have measured the DC offset right away but wasn't educated enough then, to do that.

I found a darkened area on the Power Supply board around one of the TO-220 type transistors and associated Caps. So my take was that the system was 45 years old and needed new everything. I've only addressed the power supply board and have replaced most of the caps with newer type aluminum polymer, then the two TO-220 type transistors and the three TO-92 transistors with replacements recommended from Digi-Key. Nothing improved, so I decided to immerse myself in the operation of the board and started taking voltage measurements at every test point defined in the service manual and at other component connections trying to assess why the amp wasn't getting power out. I'm using a DBT at 100 watts, it goes out when the system is powered up / The entire board as been gone over quite a few times, measuring components in circuit and a few other components were swapped out for newer, ie; metal film resistors plus anything that didn't pass muster on measurement. Some resistors were replaced when they didn't pass on the tests, only to find they were good off the board . However, there are correct plus and minus voltages at many of the correct points, but the driver board is not getting pass thru on all the slot points. ( if anyone is familiar with a Sansui)
Today I went back and checked the old TO-92 transistors and found them good, so I put them back on the board and went about checking voltages again. To my surprise, as I proceeded down one rail, I placed the probe on a point and heard a click from the Relay. This would recur at more than a half dozen places on the board.

This has never happened before, so I thought now was a good time to check the DC offset and found pretty high readings on all three speaker set ports. These were something like 34.5Mv R and 13.5Mv L or similar on all three. Given the info I have now, that's definitely a reason for the relay to stay in protect, but the values are a quite a bit higher than I would have even anticipated. What I expected was more like 0.05 - 1.00 mv difference. Would the variable resistors will even address a difference this high? Is there anyone that can guide me thru the fine points of adjusting this and is this a reasonable approach?

thanks for your time
 

sghioto

Joined Dec 31, 2017
5,380
Welcome to AAC.
Can you post the schematic?
"I'm using a DBT at 100 watts, it goes out when the system is powered up". What is that or what does that mean?
 

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Welcome to AAC.
Can you post the schematic?
"I'm using a DBT at 100 watts, it goes out when the system is powered up". What is that or what does that mean?
A Dim Bulb Tester is a device that uses an incandescent bulb as a current limiter, used between the wall outlet and the system. The wall current is run through the bulb and acts like a choke to prevent a large current surge if there's a short somewhere that could severely damage the system. If the bulb goes out when you turn the system on, it's a good sign that the issues are likely minor or at least manageable.
 

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Here's a circuit of the Power supply board

F2437.jpg
View attachment 226442The rail I was testing when the relay clicked on, was from the F02 (fuse 02) to diode 01 and beyond. With the DMM negative lead grounded on the chassis, I was checking the test points, outlined at the lower edge of the schematic. I think I was up to 23/W following the rail down to R12 and found that when probing any point after D01, the relay would click on. So the system was obviously being grounded by the circuit through the DMM. This never happened before, so now that I see the relay works, I'm interested in correcting the DC offset to see if that's what is keeping the relay in "Protect".
 

Tonyr1084

Joined Sep 24, 2015
7,853
Protect relay stopped closing / then the fun started.
Have you checked the relay coil? If you can, does the armature move? Or is it welded in one position?

Some resistors were replaced when they didn't pass on the tests, only to find they were good off the board .
It's difficult to test resistors in circuit. There may be other parts of the circuit causing the reading to change; likely lower. If you saw a lower value than expressed on the resistor then it could be in parallel with another resistor or some other component may be at play. Caps will cause readings to change, ever growing until the meter exceeds it's capability to read out a value.

I'm using a DBT at 100 watts, it goes out when the system is powered up
Is it lit before you power up the board? If so then something may be shorted. Or when you turn something on, something switches off the power; possibly because it's supposed to ? ? ? IDK!

These were something like 34.5Mv R and 13.5Mv L or similar on all three.
MEGA Volts?

What I expected was more like 0.05 - 1.00 mv difference.
Please clarify what you mean. Use of capital "M" and small "m" are hugely different. "M" means Meg and "m" means Micro.
 

Tonyr1084

Joined Sep 24, 2015
7,853
Oh, have you checked for burned out traces?

Touching the leads of a meter causing the relay to click in - well, that's an indication the relay works. But it could mean something simple may be wrong. Possibly a diode has gone bad. They not only "Open" they also "Short". Or maybe a fuse is blown??? I'm assuming you checked those. Right? They're good?
 

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Have you checked the relay coil? If you can, does the armature move? Or is it welded in one position?

The Relay is new, NOS - never used until I installed it a few weeks ago. I read that it's possible to have fused like you mentioned, so instead of going on another tangent, just replaced it. I've never heard it operate until yesterday. I thought about shunting the circuit, so it would stay engaged that would allow me to check the voltages on the relay pad, but thought that move my generate smoke and I decided not to .

It's difficult to test resistors in circuit. There may be other parts of the circuit causing the reading to change; likely lower. If you saw a lower value than expressed on the resistor then it could be in parallel with another resistor or some other component may be at play. Caps will cause readings to change, ever growing until the meter exceeds it's capability to read out a value.

I've tested most of the caps with an ESR meter. A few were good, many were marginal, 5 had bad readings. Many of the resistors that terminate on the ground trace did read correctly, but you'e correct about the ones in the more complex circuitry being elusive. I pulled and replaced a couple that had good values, out of the circuit. Lesson learned.
As to the Diodes - yes, one was recently replaced after I did a mass component check last week. That one was located close to where I first noticed the board was 'cooked' a bit, but not actually right at the site.


Is it lit before you power up the board? If so then something may be shorted. Or when you turn something on, something switches off the power; possibly because it's supposed to ? ? ? IDK!

The bulb is out until the system power is turned on. At that point the surge makes the bulb quickly glow bright, but not as much as if it were being used as a lamp. After a half second, it dims and goes out completely. After 5 minutes or so, even tho there's no incandescence, the bulb is warm, not hot to the touch

MEGA Volts?
No, Mea Culpa microvolts ( the DMM setting is at the 200m detent)

Please clarify what you mean. Use of capital "M" and small "m" are hugely different. "M" means Meg and "m" means Micro.
Thanks for your interest. Today I'm setting up to try the DC offset adjustment. The Service Manual is very specific, but assumes the user is an E-Tech, which I'm not. I have a good deal of knowledge, but no experience and IMO, things can go awry here, when you change adjustments that have stood for many years. The higher end equipment needed to check wave forms, is not in my tool chest. There are 4 variable resistors on the driver board, two of them are for the bias adjustment. Watch this space..
 

sghioto

Joined Dec 31, 2017
5,380
Today I'm setting up to try the DC offset adjustment.
Personally I don't believe it's the offsets. Those voltages seem way too low to be of concern.
I would be checking voltages in and around the relay circuit.
Specifically voltages on transistors TR04 and TR05.
Connect negative lead anywhere on the circuit path where the black circle is shown and check voltages at the red squares.
EEE Sansui 881 power supply.png
 

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Personally I don't believe it's the offsets. Those voltages seem way too low to be of concern.
I would be checking voltages in and around the relay circuit.
Specifically voltages on transistors TR04 and TR05.
Connect negative lead anywhere on the circuit path where the black circle is shown and check voltages at the red squares.
View attachment 226495
Personally I don't believe it's the offsets. Those voltages seem way too low to be of concern.
I would be checking voltages in and around the relay circuit.
Specifically voltages on transistors TR04 and TR05.
Connect negative lead anywhere on the circuit path where the black circle is shown and check voltages at the red squares.
View attachment 226495
HI, Many thanks.. I'll get this tomorrow.
 

sghioto

Joined Dec 31, 2017
5,380
This is telling me that C12 or R12 is possibly at fault.
When power is first turned on C12 should charge through R12 to provide a short delay before activating the speaker relay.
Check C12 and R12.
 

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Here we are....
R12 at the junction of D01 is 30.1 Vdc
R12 at the junction of C12 is 0.1 Vdc
The (-) side of C12 is 0.0
 
Solid polymer capacitors have high leakage currents and would not work well in circuits needing low leakage curent, such as the Protect relay on-delay timer C12. Polymers are best in SMPS when low ESR is important, but not so much here in low frequency/DC use. They're also expensive, and if you bought them on the cheap they might have even more leakage. edit: C12 looks shorted.
The Protect function DC-offset usually needs a few volts of offset to trip. 34.5mV is nothing to worry about. Best if it can be zero'd out using VR01, VR02.
 
Last edited:

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
I thought I sent that last one earlier....

now 4PM EST
So I checked C12 and R12 in circuit and after D01, the R12 pin had 30.2 Vdc, then over to C12(+) was at 0.1Vdc. I pulled them both and R12 measured ok.
C12, which was replaced earlier on a sweep of caps, came in with a higher than expected ESR, so I installed a new one.
The odd thing about these parts I'm finding is that the order of (10) 220uF/6.3V caps, none of them even approach 220Uf, one was 194uF, 3 others were at 205uF. I finally found one that was 213Uf and installed it...
I also ordered some aluminum polymer to check out and they are all AT 245uF. what goes on here?

Anyway, the bottom line now is that using the DBT, the system clicks on after 5-6 seconds and stays on. It will cycle as I probe various points, but comes right back. So I tried it on normal house current and it clicks on in 1 second and stays on ( with no smoke) . I think we're out of the woods. Tomorrow, I'll attempt the DC Offset adjustment. The manual says to let the system warm up for 5-10 minutes and zero out the Vr's on the driver board. We're discussing the BIAS current, correct? on Vr03/Vr04?

Like I said, this will be a spacewalk for me, so stand by... I really appreciate your assist
thanks again
 

Thread Starter

rfjohnsto

Joined Dec 31, 2020
15
Here is the page from the Manual and the schematic of the board. This board (2436)plugs into the power supply board (2437) very much like a PCI board in a computer slot.
sansui881p04.jpgf2436power.jpg
 
Top