SA-8100 protection circuit issue

KeithD

AK Subscriber
Subscriber
So, I picked up a very nice SA-8100, and I absolutely love it. Sound is great, and I really like the three range volume control. However, I have noticed two issues at this point.

1. The amp will play fine, but somewhere between one and five minutes of play time and the protection circuit will engage for 2-3 seconds cutting the speaker output. After that everything is fine and plays without problems. I'd say this happens about 80% of the time after turning the amp on. Once it a while it does not happen. Does not seem to matter what input I am using. However, it won't happen if no inputs are on.

2. There is also a hum heard in the speakers when I have a phonograph connected to Phono 1. Tried two different turntables, but with same result. However, I do not have this issue on Phono 2. So, at this point I just use Phono 2, but would like to solve this issue on Phono 1. The hum only occurs when a turntable is connected to Phono 1. When nothing is connected to it, no hum.

Any ideas on solving these issues would be appreciated.
 
Measure the DC offset across the speaker terminals while the relay is engaged. Use a DMM on millivolt setting - report back.

Check the RCA connections to the chassis for the Phono 1 input. Make sure the "shield/ground" is properly connected to the cable shield and/or the amplifier chassis.
 
OK, did a quick test. Turned the speaker setting to off and listened through headphones (is the equal to physically disconnecting the speakers?). Protection circuit came on after a few minutes as before. Turned the amp off and got the multimeter out and connected to speaker a L output. Turned the amp back on and listened for about 10 minutes, but protection circuit did not come back on. Readings were steady at 0.01 V. At some point they started jumping around 0.01 to 0.1 V. Looks like once the unit is "warmed up" the circuit does not come on even if it has been turned off and back on. So I will try this again cold.

I assume I need to open up the case to check the ground connection inside for Phono 1. Will try to do that tomorrow.

Thanks,
Keith
 
I would separate the pre amp from the main amp. I see it has jumpers but there is a switching circuit that connects them as well.
Try to isolate by separating the pre and main amp by reading the owners manual. The pre amp could be causing the main amp to go into protection.
Once you have the units separated, retake the amp settings (Idles and DC Balance). see if they remain stable. The 0.100mv Balance is high. .020v or -+20mV's is the max the SM allows.
The settings in both amps must be stable within a few mV's.
The maps use the dreaded 726's for the diff pair and also in the control amp. This too could be an issue.
I bet if you replace the 2SA726's with the KSA992's it will be fixed. A sound but somewhat wild guess.
Regardless, use proper troubleshooting technique.
 
I was able to connect a DMM across the A speaker output on the L channel this morning. Readings were very stable at -0.007V. After about 14 minutes the protection circuit tripped and the reading when to +0.007V for about 2 seconds. Came out of protection and went back to -0.007V and is very stable there, no jumping around. At about 18 minutes, the reading started jumping around a lot. Ranging from -0.009V to -0.058 V. At about the 20 minute mark the readings settled down to -0.009 V. At 23 minutes things started moving up again -0.07 to- 0.02. That lasted for only a minute before settling back to -0.009 V. Stopped measuring after that.

Just for reference I have KLH model 6 speakers on A and KLH model 17 speakers on B.
 
That is the DC Balance and it does look good.
What about the idle settings?
I would remove the pre amp connection to the main amp and perform the checks as stated in the above post.
Disconnect all speakers.
 
reading suddenly jumping could be something externally (voltage change?) or something like that. if not, watch for deteriorating transistors focusing on the two Zebulon1 stated.

From the info I still have on my own SA8100 restore, on my protection circuit "one of the transistors was not good anymore and it was not the one steering the relay" (which might be a replacement candidate too and additionally you might want a protection diode over the relay coil).
So, apart from other possibilities, you may have a protection circuit problem as well as a spurious DC instability problem, but nothing is sure, yet.
As they get old, having one problem at a time only is not guaranteed....
 
Unless I'm missing something, the DC offset measurements above, at the speaker output terminals, at actually meaningless during the fault phase - once the amp goes into protection, then the speaker terminals are isolated, and you are no longer measuring a true voltage value. The values while the amp is operating normally are fine, but as soon as the protection trips, the circuit is broken and the values will jump all over the place - to get accurate DC offset values, while the amp is in protection, you need to monitor the audio inputs going into the relay. I doubt that's the problem, but that's where to measure them.

One of the other possibilities not mentioned so far could be some of the electrolytic caps in the protection circuit itself - I think it's C6 which primarily controls the start-up delay. If that's leaking / failing, it could be tripping the relay at unpredictable intervals. Several of the other electro caps in the protection circuit are also low Voltage rated, and could possibly need replacement.
 
That is the DC Balance and it does look good.
What about the idle settings?
I would remove the pre amp connection to the main amp and perform the checks as stated in the above post.
Disconnect all speakers.
Sorry about the newbie question, but how do you measure "idle"?

I have yet to take the cover off the 8100 to look inside, but will do that later today. I will check the Phono 1 grounding and look for any leaking caps.

I do not have the pre out/power in jacks connected at this point since the separation switch is set to Off. If I switch it to On, then should I be measuring the Pre Out settings? With them separated there shouldn't be any signal at the speakers to measure, correct?

Thanks,
Keith
 
Here are the results of further tests:

Separated the Pre out/Power in and measured the voltage difference on Pre out. Here are the readings as a function of time:

minutes mV

0 190
1 79
2 37
3 17
4 9

Eventually falls to a steady value at 1 mV. At the 15 minute mark the protection circuit engaged and the reading when to - 1mV or possible higher, it was very short, 1 or 2 seconds, then went back to 1 mV.

Also checked the ground on Phono 1, and left channel ground is open, so that looks like the fix I need to make.
 
You'll have to read the service manual to check the idles.
Just looked at the service manual, and it gives a procedure to adjust the idle current, but not just "check" it. The procedure requires at 4.7 kohm resistor, which I will need to pick up as I don't have one. When it says "Remove output load" I assume that means disconnect the speakers?

This does seem like a temperature problem to me. This morning I turned on the 8100 with with Pre out/Power in separated and no inputs on. After about 10 - 15 minutes the protection circuit tripped for just a few seconds, so it does not seem to need any input signal for this to happen, and after that one trip of the circuit it does not happen again no matter how long I have the unit on or what the inputs.
 
You don't need the resistors to check /adjust the idles. Volume must be at min (Zero). Report back the idle readings you see for both channels. Be careful and do not speed through the procedure. Only move the meter clips when the power is off. Monitor the idle readings on one side/channel to warm-up watching for the protection to engage. Then move to the other channel.
You are following the correct troubleshooting procedure. It may end up as a protection circuit fault but you have to verify the amps operation first. The idle circuit has a relationship related to the heat generated from the outputs.
 
Here are the idle readings. At startup pins 1-2 were 15.8 mV. At 9 minutes the protection circuit tripped, and at 11 minutes the reading was 23.8 mV. Pins 24-24 had 16.6 mV. I will follow the procedure to adjust idle and neutral potential as given in the manual to get the readings where they should be.
 
As Goldie99 stated you will have to get the schematic and check dc offset in both channels prior to the relay so you have valid figures when going into protection. I think you only mentioned checking the left channel offset?

It will likely be either an offset issue or a faulty component or bad solder joint on the protection board itself, l replaced all caps and transistors on the protection board on my SA-8100 when l had it for a similar symptom although the relay on mine would just randomly cycle.
 
I would separate the pre amp from the main amp. I see it has jumpers but there is a switching circuit that connects them as well.
Try to isolate by separating the pre and main amp by reading the owners manual. The pre amp could be causing the main amp to go into protection.
Once you have the units separated, retake the amp settings (Idles and DC Balance). see if they remain stable. The 0.100mv Balance is high. .020v or -+20mV's is the max the SM allows.
The settings in both amps must be stable within a few mV's.
The maps use the dreaded 726's for the diff pair and also in the control amp. This too could be an issue.
I bet if you replace the 2SA726's with the KSA992's it will be fixed. A sound but somewhat wild guess.
Regardless, use proper troubleshooting technique.

This. Stop measuring and get those 726's out of there. You want them out anyway. 90% chance it will fix the issue. If it doesn't, you're still ahead of the game.
 
This. Stop measuring and get those 726's out of there. You want them out anyway. 90% chance it will fix the issue. If it doesn't, you're still ahead of the game.
OK, took a look at the schematic and I see 6 2SA726 transistors, four in the power amp and two in the control amp.

The protection circuit engages even when the Power out/Power in are separated. When they are separated looks like the only thing feeding into the protection relay is the power amp, so it makes sense that the problem must be in the power supply, power amp, or the protection relay itself. Looks like Mouser has the KSA992, so I could go ahead and order a batch of those.

Anything else I should check to potentially narrow the problem down?
 
Back
Top Bottom