• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

A troubled Pioneer SA-8500

MHR

New Member
First of all, hi to everyone who’s interested to take time to read my problems and questions. This post is quite extensive, but it’s better to be complete, to rule out the obvious. At this point I had to swallow my pride and ask for help.

I’m not a complete idiot in the electronics department, as I successfully revised quite a few amplifiers up until this point.

Before I got my hands on this one the amp worked, but had the infamous crackle on the speaker-output caused by 2SA725’s gone bad on the power-amp board.

I decided the amp was well worth it to make it future proof so I have fully recapped my SA-8500, and all SA725/6’s are replaced by KSA992 and KSC1845. Also the 2SC1451 were replaced by KSC3503. The primary power-ELCO’s have been replaced from 50V/15000uF to 63V/15000uF.

I also did a balance and BIAS calibration.

Now, first of all let me explain the symptoms my SA-8500 now has after done all this work:

When fully hooked up, so speakers on terminals A and AUX in use from streamer, when i up the volume to like a quarter of the way, the relay switches off. When I decrease volume the relay jumps back on and the sound comes back. When I unhook everything, I can crank the volume up to max if I wanted to but the relay does not switch off.

So I did a little investigating myself:

Note: All these tests I did behind a dim-bulb-tester, just to be safe.

1. I first put on a signal generator on the right channel with a 1kHz/300mV sinus wave. Then I hooked up a 200W 8 Ohm power-resistor to the terminals to mimic speaker load. I hooked up my oscilloscope to the power resistor to visualize output.

Result: Same symptom on both left and right channel, no matter what input, Tuner, TAPE1/2. The little sinus wave I saw was good, no weird artifacts in the wave.


2. Then I checked the voltages on EQ-assembly (Volume, input selector and Phono stage) as well as Control amp (tone-control board) and primary DC input from just behind the rectifier from the Switch circuit board.

Result: All voltages were in-spec.


3. Now… I did the same test when hooking the AUX input, put a 1kHz/300mV sinus signal and the power resistor tot the speaker terminal. There were no dramatic voltage drops across any of the known measure points, as can be found in the manual.


4. Then I noticed some very faint hum coming off the transformer (ATT-233B EU-model). I realized that I never heard hum coming from a transformer in a power-amplifier before. So I decided to investigate on Ohm’s and continuity within and between phases.

Result: Nothing out of the ordinary, resistance well between 0 and 10 Ohms. No contact between phases (no short)

OK, this bit may be the important bit:


5. OK… Then I decided to bypass the pre-amp section by removing the RCA short bridges and put the same signal on there. Because I had to rule out which one is faulty, the pre-amp section or the power-amp section. I put the same signal on the pre-out as before turned on the amp, no issue. I cranked the signal up to 1.6V and I measured 20.4V when the amp started clipping, so with 8 Ohm this equals (20.4x20.4) / 8 = 52 Watts. Specs are 60Watts/channel so I didn’t know why it was under performing. But… of course, the dim- bulb-tester. I took the amp off the dim-bulb and tested again. However this time after switching on the amplifier with the signal already on the relay switch didn’t come on. So I turned off the signal and the relay clicked back on. So I reduced the signal too 100mV, then the relay turned on, when I upped the signal amplitude to 200mV it switched off again.


So I hooked the amplifier back on to the dim-bulb tester, which had a 300W light bulb in it and started the procedure again. I noticed that when I got above 1V the bulb started to brighten. So I’m thinking there is a fault in the power-amp section drawing excessive current, thus maybe an intermittent fault in one of the components causing a short above a certain power-level.



At this point, this is as far my knowledge goes. Maybe that’s the reason the hum of the transformer is audible, because it has to work hard to overcome some threshold and eventually gives up and the whole amp goes in protection mode.


If hou have managed to read my story all the way up to here, I commend you for it ;)

If anyone has a suggestion how to proceed with determining what component might be at fault I’d be happy to receive any suggestion you might have. I'd be happy to provide pictures if necessary and any additional information you might need.
 
Last edited:
Register to hide this ad
I use this post to update you on the suggestions given and their results

Thank for the quick replies, Aud10file and markthefixer. I did bias and balance without the DBT. I did replace those little pesky 0,22uF caps. So it could also be a fault in the protection circuit. I will elaborate soon on the choice of transistors I used to replace the old ones.
 
Last edited:
Testing for out power output under dim bulb may yield unexpected results. The dim bulb is only good as a safety against amp killing shorts.

Use a 100 to150W bulb for this 60W per channel amp. Set your bias and dc offset. No need to turn up volume.
If the above goes ok without dim bulb going bright, you can assume there are no shorts. Now take it off the dim bulb and plug into ac. Check and adjust your bias and dc offset once more.

Now test for output from speakers.
 
sounds like you set your idle current with the dim bulb tester operative. which strangles many aspects of the unit.

get it off the DBT,

before powering up dial the idle currents to minimum.

then power up STRAIGHT from the wall - NO DBT!!!

NOW set the dc offset and idle currents.

reevaluate.

ESPECIALLY the protect circuits, did you replace all those damn Sanyo sky blue guaranteed failure e-caps?
 
sounds like you set your idle current with the dim bulb tester operative. which strangles many aspects of the unit.

get it off the DBT,

before powering up dial the idle currents to minimum.

then power up STRAIGHT from the wall - NO DBT!!!

NOW set the dc offset and idle currents.

reevaluate.

ESPECIALLY the protect circuits, did you replace all those damn Sanyo sky blue guaranteed failure e-caps?

I did the bias and balance without the DBT and I already did replace those little pesky 0,22uF caps. Still the same result. So it could also be a fault in the protection circuit.
 
briefly: at 1/4 volume the unit goes into protect, ONLY when speaker loads are present.. C16 & C17 are newer 0.22uf caps. Correct??

Setting idle current implies that the emitter resistors are ok.

Q5 & Q6 are overload detectors for the channels. Lift one leg of D17 to interrupt the protect signal from them. If that gets the problem suppressed,
replace Q5 & Q6 (ksc1845) and try again. Check the diodes and resistors around Q5 & Q6 too.

Otherwise lift one leg each of D15 & D16 to suppress the dc voltage sense. If that gets the problem suppressed,
replace Q7 & Q8 (ksc1845).

if those three diodes don't suppress the protect, start looking for shorts on the board.

upload_2022-6-28_21-35-13.png
 
briefly: at 1/4 volume the unit goes into protect, ONLY when speaker loads are present.. C16 & C17 are newer 0.22uf caps. Correct??

Setting idle current implies that the emitter resistors are ok.

Q5 & Q6 are overload detectors for the channels. Lift one leg of D17 to interrupt the protect signal from them. If that gets the problem suppressed,
replace Q5 & Q6 (ksc1845) and try again. Check the diodes and resistors around Q5 & Q6 too.

Otherwise lift one leg each of D15 & D16 to suppress the dc voltage sense. If that gets the problem suppressed,
replace Q7 & Q8 (ksc1845).

if those three diodes don't suppress the protect, start looking for shorts on the board.

View attachment 2603547

Many thanks for your insight, I will look into it this weekend and try as you suggested. I will keep you posted on the outcome.
 
briefly: at 1/4 volume the unit goes into protect, ONLY when speaker loads are present.. C16 & C17 are newer 0.22uf caps. Correct??

Setting idle current implies that the emitter resistors are ok.

Q5 & Q6 are overload detectors for the channels. Lift one leg of D17 to interrupt the protect signal from them. If that gets the problem suppressed,
replace Q5 & Q6 (ksc1845) and try again. Check the diodes and resistors around Q5 & Q6 too.

Otherwise lift one leg each of D15 & D16 to suppress the dc voltage sense. If that gets the problem suppressed,
replace Q7 & Q8 (ksc1845).

if those three diodes don't suppress the protect, start looking for shorts on the board.

View attachment 2603547

Well, I've read that the 2SC869 should be replaced with KSC2310Y, that's what on the board right now.
Q7 and Q8 are the original 945's, should I replace them with KSC1845 ?
Q9, the 2SA733 I replaced with the KSA992.

Any thoughts on my replacements?

I've, yet again, replaced the 0.22uF/25V with 0.22uF/50V electrolytic.

I've ran the test again, and the funny thing is..... so far so good. With a 600mV@1kHz I can crank up the volume to 25.4V peak to peak on both channels, that's 80Watts per channel into 8 Ohms.
 
I use this post to update you on the suggestions given and their results

Thank for the quick replies, Aud10file and markthefixer. I did bias and balance without the DBT. I did replace those little pesky 0,22uF caps. So it could also be a fault in the protection circuit. I will elaborate soon on the choice of transistors I used to replace the old ones.

I've, yet again, replaced the 0.22uF/25V with 0.22uF/50V electrolytic.

I've ran the test again, and the funny thing is..... so far so good. With a 600mV@1kHz I can crank up the volume to 25.4V peak to peak on both channels, that's 80Watts per channel into 8 Ohms.

I will do BIAS and balance once more and will do some extensive testing to see if the amp stays stable.
 
Q7 and Q8 are the original 945's, should I replace them with KSC1845 ? yes .
Q9, the 2SA733 I replaced with the KSA992. yes .
 
Q7 and Q8 are the original 945's, should I replace them with KSC1845 ? yes .
Q9, the 2SA733 I replaced with the KSA992. yes .

And what about the KSC2310Y, the amp seems to look pretty happy with them in the circuit.
Will replace the 945's. Thank you for your help.

I will call this issue SOLVED.

Thank you all for all the suggestions! This amp is now ready to be sold.
 
And what about the KSC2310Y, the amp seems to look pretty happy with them in the circuit.
Will replace the 945's. Thank you for your help.

I will call this issue SOLVED.

Thank you all for all the suggestions! This amp is now ready to be sold.

as you did... leave them there... future replacement if needed is ksc1845
 
Back
Top Bottom