F1474 Board issues in Eight Deluxe

Doubled

New Member
Hi,

I am a newly subscribed member having problems with my beloved Eight Deluxe. I had not used it in several months, and when I powered it on it would not energize the protection relay. All fuses are intact. I looked at the service manual and checked for DC on inputs to the protection CB F1475. I found 45VDC on the left channel.

I checked all of the output transistors. They seemed fine. So I swapped the left side driver board for the right side driver board and the problem changed sides. When I remove the problem board, the protection relay will energize.

I have spent a good deal of time doing some on the board checks of at least the discrete components, comparing reading from the good board to the bad board. So far, I have not been able to find any discrepancies.

The service manual seemed to suggest that TR-801, TR-802, TR-803, and TR-807 could be suspect. So I changed these, even though I didn't find any issue with my checks. I also changed the electrolytic caps on the board. Probably not surprisingly, I didn't fix a thing!

Can anyone give me a pointer or two on this? I can send the service manual if needed.

Thanks so much,

DD
Imlay City, MI
 
The setup instructions are fairly complex for this amp - how have you gotten on with the basics - setting DC offset and bias for this board?
 
Well, yes. In fact, that is how I found that there was 45VDC on the left channel. I am fairly inexperienced, but I think that indicates a component failure. I was not able to adjust it.
 
You are starting your adventure with a very complex receiver. As you know by now, the inside of the Deluxe is quite packed and not particularly easy to get around. I salute your progress. I opened mine after I bought it, dead, at a local thrift store. I knew there was no way I was going to jump in there. You have the right idea with testing each component on the bad card. You are going to have to test all of the items properly, one at a time. Using the good card as a comparison is also the way to go. I had one of the AK techs rehab mine, and it is worth all the hassles and cost. Good luck to you.
 
I love my Eight Deluxe also. It played great when I bought it several years ago. I actually tried to find someone who would service it in my area (SE Michigan) to no avail. I'm not keen on trying to ship it across the country. So I decided to take a crack at it. I'm not a complete novice. I have successfully fixed a couple of vintage receivers, but have given up on some others.

I have tried to attach a JPEG of the driver board schematic.
 
Since I had them on hand, I went ahead and changed TR804, TR808, and TR809. These all seemed to check ok while they were on the board. Of course, they didn't change my result either. I still have +45VDC on whichever output I install the driver board in. This only leaves TR805, TR810, and TR811. They all seem to check OK too, so I'm not to keen on heading in that direction. It seems like this should be relatively simple, but I must be missing something.

Can someone recommend a Sansui tech that I can speak with? I wonder I could just send out the driver boards for evaluation and repair?

Thanks!
DD
Michigan
 
I'd be slipping a new ZD801 in there. 13V Zener, probably a 250mW but I'd stick a 500mW one in just in case.
 
Also check your VR801 DC ofset pot. These pots, especially in the older Eight have caused problems. You can probe the ohm reading on the good board's VR801 by measuring on the first two (outer, middle) legs, just leave amp off, then measure on the bad boards SAME two legs and see if you are in the ball park, or are able to match up - if not replace (both)

It can also be something as simple as a cracked solderjoint, that one day just gave in from years of heat stress. So a good visal inspection would be in place using a magnifier checking for discoloration or other anomalies around joints, traces and components. A continuity tester is a fast way to check joints are connected, trace by trace.

Speaking here from bitter experience of wasting hours swapping components when "taking a good look" and "beeping some traces" was all it took.
 
Last edited:
Also check your VR801 DC ofset pot. These pots, especially in the older Eight have caused problems. You can probe the ohm reading on the good board's VR801 by measuring on the first two (outer, middle) legs, just leave amp off, then measure on the bad boards SAME two legs and see if you are in the ball park, or are able to match up - if not replace (both)

It can also be something as simple as a cracked solderjoint, that one day just gave in from years of heat stress. So a good visal inspection would be in place using a magnifier checking for discoloration or other anomalies around joints, traces and components. A continuity tester is a fast way to check joints are connected, trace by trace.

Speaking here from bitter experience of wasting hours swapping components when "taking a good look" and "beeping some traces" was all it took.

Very good advice. My first thought is that it was the pots. I cleaned them thoroughly and checked them many times. They appear to be OK. I have also gone over the tracks with a magnifier, but I have not rung them out with my meter. I will do this.
 
One more question. Does the +45V suggest that there is a problem with the zener controlling voltage?
Its one possibility. There are others as well, but if that Zener is shorted, or not turning on, then it could cause this type of issue.
 
Well the new zener diodes showed up today, and I quickly swapped it out. No joy, :( I still have +45V DC on the output controlled by that driver board. I also spent some time examining the board and ringing out tracks. I didn't find anything there either. Any other thoughts?
 
We need to start at the beginning, list everything you have done, and the symptoms and any other relevant information.
 
We need to start at the beginning, list everything you have done, and the symptoms and any other relevant information.
OK, from the beginning.

1. Protection relay not engaging on start-up. Checked DC voltages on F1475 Protector Unit board. Found 7mV on one channel and +45V on the other.
2. Cleaned VR801 and VR802 pots on suspect F1474 driver board. Issue remains.
3. Changed TR-801, TR-802, TR-803, and TR-807 on suspect F1474. Also change Electrolytic caps, C806, C814 for good measure. No improvement.
4. Checked output Transistors, TR-853, TR-855, TR-857, and TR-859. All seem OK. Swap F1474 boards side to side. +45v reading moves to the other channel. This seems to indicate that the issue is isolated to the F1474 board.
5. Check all remaining discrete components (transistors and diodes) on the board, comparing readings from the good board against the suspect board.I used my Fluke 87 on diode check mode. Found no anomalies.
6. Changed TR-808, and TR-804 just because I had them on hand. No changes.
7. Changed ZD-801. Thoroughly inspected the PCB tracks and rung them out with a meter. Re-soldered any joints looking the least bit suspect. Still no good.

I think that has us up to date.
 
I just remembered one other bit of info. When I remove the suspect F1474 board, the protection relay will engage for normal operation.
 
Okay, so I think what you have to do is follow the 45 volts from the output and back through the driver board and find the source, the smallest and silliest thing can cause something like this.
I had a G9000DB with this problem a few years ago now, and after hours of searching, it turned out to be a crack in the earth trace from the input molex connector, it was not visible to the naked eye, so the front end of the driver had no earth (0V) reference, so the DC voltage would slowly swing up to near maximum rail voltage.
So there may not even be a faulty component in your Eight, it could be something like that.
With what you have done already, its looking very much like something along these lines...
Its there somewhere....you just have to be meticulous, print the circuit diagram off and use your highlighter on the schematic and note down the voltages.....
Let us know how you get on...
 
I think Kevzep is onto something with that ground reference. Took a peek at the schematics. See how the check point (N/M respectively) that connects to the protection board is directly connected to the R841 resistor that then goes to the Ground trace (should be labeled G on the driver board right at the bottom where it connects to the molex. There will be two ground traces next to one another and it will be number 2,3 trace either from the left or the right (depending on how you have the board turned)

At any rate beep from that trace's beginning at the bottom of the board, then just move with it - including touching the legs of the components in its path beginning with the R841 leg that could be suspect. I bet you Kev is right and at some point you will find a break either in the traces or from trace to a component leg, which will indicate it is that legs joint that needs a re-flow.
 
Back
Top Bottom