Sansui G-7500 Fried Resistors F2980 Board

So, can you verify what happens with the Bias when you turn the amplifier on. If the amplifier is oscillating I would expect the bias to be normal for a few seconds, and then suddenly shoot up...
Can you connect the meter, and try this...
[/QUOTE]

When I turn the amplifier on, it starts climbing immediately.
 
How quickly does it climb to 29V?

It always begins around 27-28V and climbs to 29V in about 15-30 seconds. Then it will continue to climb at around the same pace. Seems to slow down at 31V. Highest voltage I’ve see is 31.14 but I always shut it down when it gets up around 30-31V. Also when it gets to around 30V the voltage starts to fluctuate down then up.

The comments about the burned traces are typical of my board. As I stated before, this board had been worked on before. Several pads are lifted or completely gone. I have bridged the connections but maybe (probably) that’s a big part of the problem.

Thanks for the advice on the scope.
 
It always begins around 27-28V and climbs to 29V in about 15-30 seconds. Then it will continue to climb at around the same pace. Seems to slow down at 31V. Highest voltage I’ve see is 31.14 but I always shut it down when it gets up around 30-31V. Also when it gets to around 30V the voltage starts to fluctuate down then up.

The comments about the burned traces are typical of my board. As I stated before, this board had been worked on before. Several pads are lifted or completely gone. I have bridged the connections but maybe (probably) that’s a big part of the problem.

Thanks for the advice on the scope.
At this point, I don't think it's oscillating........I think you are going to have to go meticulously through every trace on that board and verify them. I am thinking this is the problem, until we do this, anything else is just an assumption.
 
At this point, I don't think it's oscillating........I think you are going to have to go meticulously through every trace on that board and verify them. I am thinking this is the problem, until we do this, anything else is just an assumption.

Yeah, I went through it carefully but not meticulously. I need to do that and will spend next week doing so. What are your thoughts on bridging? Best method for bridging? I wish I had a way to repair the traces but I don’t. Would it be of help to list what components have been bridged? I will report back ASAP, probably toward the end of next week.

Thank you for your help and patience.
 
Yeah, I went through it carefully but not meticulously. I need to do that and will spend next week doing so. What are your thoughts on bridging? Best method for bridging? I wish I had a way to repair the traces but I don’t. Would it be of help to list what components have been bridged? I will report back ASAP, probably toward the end of next week.

Thank you for your help and patience.
There are many different ways to repair traces, just depends what works for you.
You always have to scrape the green stuff off, then, you can either use a small piece of a leg from a resistor or diode, chop it to length and solder it over the crack/gap. I mostly do it this way rather than trying to flow the solder over the gap.
If you bridge the gap with solder, it may well crack again as the solder doesn't really have enough tensile strength for that.

I keep a bunch of off-cuts from components for that kind work work...
Use flux as well, it helps the solder flow nicely and make a good bond to the surfaces.

Just go over the board paying attention to the area where the STV joins the rest of the bias circuit.

I am sure you will find something.
 
When i repaired this F-2980 before I used some 24awg solid telephone wire, stripped the insulation, formed the wire as was the original trace, make small circles for eyelets (wrap around the intended lead diam to form) so the leads can poke through. Solder the ends first and then the middle parts. you can use the stripped insulation bits and slide back on, to use as spacing or insulation from other comps. Can glue if needed.
Follow the pcb pattern and the other channel.
While you are at it, you should do an exact 1:1 ohm readings comparing the good channel vs the bad channels. Follow paths like from a supply to a semi junction to compare 1:1.

I re-ran simulations for this design, using modern day parts,
As you see I have the DC ecap(C15 in the sim) installed in the FB loop, you should do the same, as it helps to tame the amp down and may fix your problem.
I believe this design has too much open loop gain for a DC amp.

Good luck
Rick
 

Attachments

  • Sansui G-7500(F-2980) Ltspice sim.pdf
    39.7 KB · Views: 12
Last edited:
When i repaired this F-2980 before I used some 24awg solid telephone wire, stripped the insulation, formed the wire as was the original trace, make small circles for eyelets (wrap around the intended lead diam to form) so the leads can poke through. Solder the ends first and then the middle parts. you can use the stripped insulation bits and slide back on, to use as spacing or insulation from other comps. Can glue if needed.
Follow the pcb pattern and the other channel.
While you are at it, you should do an exact 1:1 ohm readings comparing the good channel vs the bad channels. Follow paths like from a supply to a semi junction to compare 1:1.

I re-ran simulations for this design, using modern day parts,
As you see I have the DC ecap(C15 in the sim) installed in the FB loop, you should do the same, as it helps to tame the amp down and may fix your problem.
I believe this design has too much open loop gain for a DC amp.

Good luck
Rick

Thank you for the information and encouragement.

I mentioned an oscilloscope in a reply. I’ve never used one but would like to learn. I enjoy this hobby so it could, as kevzep said, be a good investment.

Frank
 
When i repaired this F-2980 before I used some 24awg solid telephone wire, stripped the insulation, formed the wire as was the original trace, make small circles for eyelets (wrap around the intended lead diam to form) so the leads can poke through. Solder the ends first and then the middle parts. you can use the stripped insulation bits and slide back on, to use as spacing or insulation from other comps. Can glue if needed.
Follow the pcb pattern and the other channel.
While you are at it, you should do an exact 1:1 ohm readings comparing the good channel vs the bad channels. Follow paths like from a supply to a semi junction to compare 1:1.

I re-ran simulations for this design, using modern day parts,
As you see I have the DC ecap(C15 in the sim) installed in the FB loop, you should do the same, as it helps to tame the amp down and may fix your problem.
I believe this design has too much open loop gain for a DC amp.

Good luck
Rick
Some good stuff there Rick, I am actually in the middle of rebuilding my G7700, I might just apply some of this whilst I am in there.....
 
Thank you, I will. I took a chance on the Hitachi V223. It powered up and I could adjust on xy axis. It is in very good condition and had two new probes.
Nice one Frank, take your time with it and look at the tutorials, basic scope operation is not that complicated.
 
Sorry I haven't responded in a while. I am going over the board and did some repair/patching on the board but nothing seems to have changed. I continue to inch forward though at a slow pace. It has been really hot here and my work area is not air conditioned. Please be patient.

Thanks
 
Sorry I haven't responded in a while. I am going over the board and did some repair/patching on the board but nothing seems to have changed. I continue to inch forward though at a slow pace. It has been really hot here and my work area is not air conditioned. Please be patient.

Thanks
All good!! We all get busy with other things....... We are here when you are ready.
 
All good!! We all get busy with other things....... We are here when you are ready.
Yesterday I decided to give up on trying to repair the 7500. I want to thank all those who helped me. Ghazzer, Kevzep especially for their patience and not giving up on me. I’m just a novice and should have realized earlier that I bit off more than I could chew with this one. If I didn’t mention you by name, it doesn’t mean I didn’t appreciate the comments, help, and support.
 
Yesterday I decided to give up on trying to repair the 7500. I want to thank all those who helped me. Ghazzer, Kevzep especially for their patience and not giving up on me. I’m just a novice and should have realized earlier that I bit off more than I could chew with this one. If I didn’t mention you by name, it doesn’t mean I didn’t appreciate the comments, help, and support.
Takes real character to realise you're in over your head, and I commend you for that.

Its a shame I do not live closer, I'd love to come and help you get that thing going....
 
Hi AK'ers, I'd like to reopen this cold case wrt to Young46 attempt to determine the the cause of this run away audio driver circuit. I have a friends G-7500 that has nearly identical symptoms and much too high voltages. I do have and know how to use test equipment as I've repaired many Sansui 8080's, 9090's and many of the Pioneer SX 80 and 50 series receivers. Prior to getting into hifi repair I worked on many amateur transceiver rigs, and before that aviation navigation equipment. Needless to say, this G7500 has me stumped. Below are the steps and conditions where I'm at now. Keep in mind that it's the right channel side that is inoperable.
- Replaced all destroyed fusible resistors. All shorted transistors including the both OPT, driver and bias transistors with new recommended replacements. Only the dual FET is original and tested OK.
- The STV3 tests good with about 1.4v forward voltage on my DVM in semiconductor test mode, also 1.4V when in circuit and rig powered up. Continuity tested all circuit paths into the STV and it components and are good.
- Reflowed all solder connections and continuity tested every circuit path I could find.
- Retested all replaced transistors and resistors numerous times to ensure no new burnouts and retested all other resistors on the F-2980 board.

Current (no pun intended) conditions at this point:
- Rig is being powered through DBT using 100W bulb.
- The two .33 ohm OPT emitter resistors replaced with 100 ohm 10W resistors.
- Per suggested, added two 47pf caps across C18/19 and per RCS suggestion, 220uf NP in series with 1K R20 to reduce any chance of oscillation. I have a 200mhz Tektronix scope and I have not seen any indication of oscillation with or w/o the caps.
- Both trimmer resistors for speaker 0VDC and bias current replaced with Bourne trimmer pots. The bias current pots VR3 and VR4 set to minimum resistance (fully CCW) as suggested by AK members.

When power on bulb bright for a moment, dims down only a small amount. Voltage readings below are taken with 100W DBT.
- 23.6VDC and -23.6VDC on large main PS filter caps
- Plus 24VDC across R78 (100ohm)
- Negative 23.1VDC across R72 (!00 ohm)
- Output (point B prime) to ground reads +.475VDC
- No change in current flow through R78 or R72 when bias pot is moved in CW direction.
- All other PS DCV's are lower than expected due to DBT limiting PS to operate fully.
This is basically the same point with the nearly same voltages and conditions as Young46 rig was producing in 2018 when further TS was halted.

My guess is something is making both driver transistors to turn on hard which then makes the OPT turn on hard. Is there a point in the first (FET), or second differential amp (TR02 & TR04) to safely switch them off to see if that turns off the driver stage?
One TS trick for Pioneer's with STK modules is that they can be replaced with 2K -4K resistors across the collector to emitter connections in effect reducing the current flow to such a low level that would allow all other PS voltage to come up in order to T/S the amp circuit. Anyone tried this technique with success?

On a side note, I've noticed that my g7500, Young46's, several others in AK and all I've seen in a popular auction site that show the burned out section is always the same right channel, (right hand side looking down at component side to F-2980). Not saying the left side of these rigs never burn up but sure a lot of right side burn outs.

Regards,
Mark
 
The last time I checked with the owner of the one I was working on, it was still not fixed. After he picked it up from me he took it to another friend (experienced tech) who still has it and has not been successful in making it hit on all cylinders again. I hope your thread will bring an answer to the problem. :lurk:
 
Back
Top Bottom