Pioneer SX-737 Distortion left channel

Is there someone who can help me find out what the problem is with this receiver? The left channel is distorted and I can find out where is is coming from. I was going to refurbish it and thought by recapping it and replacing the transistors and a cleaning it would come out of it but after all the work it still has distortion on the left channel.

Here is what I have done.

I followed the posts by MTF and LarryL and have rebuilt the power supply with new caps and transistors. I have also rebuilt the protection board, amp board, replaced the outputs with Onsemi with new insulators and grease, cleaned all the switches and controls, recapped the phono board and the control amp board. The control amp board I replaced the 2sa726 with KSC992's and I tried to replace the 2SC1344's with KSC1845's but when I did I could not get any sound so I checked the 2SC1344's with my PEAK checker and they were good so I put them back in and the sound returned but it still has distortion in the left channel. I also removed and cleaned the speaker relay. I ordered one from Mouser but it just has not come in yet. I also ordered some new pots. I tried setting the voltages according to the updated posts but the left side will not adjust.

I tried tracing the signal with the signal tracer but I can find out were it is coming from.

I have no idea where to go from here. I just like working on this vintage stuff but I am not an expert. Sometimes I can find the problem and sometimes I can't. This one I have been working on for two full days and long nights. Any help would be greatly appreciated.

"The left side will not adjust"

offset voltage wont adjust?
idle current won't adjust?
neither offset voltage, nor idle current will adjust?

post ground referenced voltage readings (e,c,b left to right) for q1, q3, q5, and (bce left to right) q7, q9
and it would be helpful to do the same for q2, q4, q6, q8, q10.

The big question is what it was doing before you started, and any changes noticed?

The original had 2sc1451's (those horrid little round things) in the amp, correct?

Check R21 and R22 (100 ohms, brown black brown gold - right next to the 2sc1451's or where they used to be) with one lead lifted from the pattern.

You speak of "sound", that implies that the protection relay is clicking in after an appropriate time delay, is it? That also implies the amp is not "railed"...

markthefixer said:

"The left side will not adjust"

1. offset voltage wont adjust?
idle current won't adjust?
neither offset voltage, nor idle current will adjust?

2. post ground referenced voltage readings (e,c,b left to right) for q1, q3, q5, and (bce left to right) q7, q9
and it would be helpful to do the same for q2, q4, q6, q8, q10.

3. The big question is what it was doing before you started, and any changes noticed?

4. The original had 2sc1451's (those horrid little round things) in the amp, correct?

5. Check R21 and R22 (100 ohms, brown black brown gold - right next to the 2sc1451's or where they used to be) with one lead lifted from the pattern.

6. You speak of "sound", that implies that the protection relay is clicking in after an appropriate time delay, is it? That also implies the amp is not "railed"...

Click to expand...

Thank you for your response. I will try to answer your questions.

1. VR3 and VR4 will adjust to 0v
VR2 will adjust to 20mA
VR1 will not adjust although it reads around 20mA

2. Will measure and post. which board are we speaking about?

3. It was doing the same as it is now.

4. Original had the 2SC1451's but were changed to 512-KSC3503DSTU

5. Will check and post

6. Protection relay is clicking in after around 4 seconds. Strong sound in right channel, distortion in right. If I turn it up to load say 3/4's it will shut off the sound and relay will restart.

The readings on the amp board for the 100Ω resistors next to where the old 2SC1451's with leg lifted are:

Right side: 100.7Ω
Left side: 100.8Ω

Power amplifier assembly board. Got it. It will take some time. I will post as soon as I can.

I hope I did this right. DC readings. No DBT direct to AC.

Q1: e 0.578v
c -36.25v
b 0.004v

Q2: e 0.570v
c -36.30v
b 0.003v

I could not find Q3 on the board?? or printout. I'm I missing something?

Q4: e 0.570v
c -37.56v
b 0.005v


Q5: e -36.90v
c -1.28v
b -36.30v

Q6: e -36.91v
c -1.068v
b -36.31v

Q7: b 1.198v
c 35.61v
e 0.625v

Q8: b -1.060v
c 35.76v
e 0.700v

Q9: b 1.193v
c 35.55v
e 0.624v

Q10: b -1.060v
c -35.69v
e -0.498v

Q1,Q2,Q3,Q4 are the 2sa726 differential pairs at the beginning of the amp board

Q1,Q3 are paired
Q2,Q4 are paired.

If you split the board in the middle one sie will be EVEN and the Other side ODD. On the Board drawing Q3 is directly above Q1. The designator lools like it says Q8 which is a misprint. It's Q3. Remember these drawings are the Foil or Solder side.

The left channel dc offset and idle current were not adjustable - but WHAT were their readings?

The readings are done almost correctly - format and voltages are ok, but I suspect we're a bit off as to who is q7, q8, q9 and q10.

That said, I can infer quite a bit. But the problem may lay in slightly perilous territory - the 100 ohm vr1/vr2 and STV-4h circuit.

Check - with POWER OFF !!! - vr1 and vr2 for resistance over their rotation range. Remember and return them to the positions they are initially in - go by the resistance reading.

I suspect the idle current adjust pot may be shorted, OR the (shudder) stv-4h diode is damaged (unusual) and not providing enough voltage for the next section ( q7 to q9 base versus q8 to q10 base)

Q7: b 1.198v
c 35.61v
e 0.625v

Q8: b -1.060v
c 35.76v
e 0.700v

Q9: b 1.193v
c 35.55v
e 0.624v

Q10: b -1.060v
c -35.69v
e -0.498v

Q7 and q9 are a pair in the same driver circuit, while q8 and q10 are the pair in the other channel's driver circuit.

Of these four readings, two bases should be + 1.2v , two should be - 1.2v while each of their corresponding emitter should be the same polarity, just 0.6v LESS than the base. Thus while your Q8 data shows a possible problem, it's on the wrong channel.

There was a REASON that I did

post ground referenced voltage readings (e,c,b left to right) for q1, q3, q5, and (bce left to right) q7, q9
and it would be helpful to do the same for q2, q4, q6, q8, q10.

Click to expand...

in THAT order. - and NOT have you jumping from channel to channel taking readings.
q1, q3, q5, q7, q9 are the left channel
q2, q4, q6, q8, q10 are the right channel.

IN those groups, q1 and q3 are a pair, and located quite physically close together. Unfortunately q7 and q9 are quite far apart, although q7 is right next to the VR1 1dle current adjustment pot on one half/channel and q8 is right next to the VR2 idle current adjustment pot on the other half/channel .

HOW do you tell the halves of the board apart - look at the pins - they are numbered - on both sides IIAC, with pins 1 and 33 at one end of the board and pins 19 and 20 at the other.

larryderouin said:

Q1,Q2,Q3,Q4 are the 2sa726 differential pairs at the beginning of the amp board

Q1,Q3 are paired
Q2,Q4 are paired.

If you split the board in the middle one sie will be EVEN and the Other side ODD. On the Board drawing Q3 is directly above Q1. The designator lools like it says Q8 which is a misprint. It's Q3. Remember these drawings are the Foil or Solder side.

Click to expand...

Thanks for the direction. I forgot about it being on the foil side when I took the last readings so I took them again and double checked them. I will make note in my service manual about the misprint. :yes:

Here are the updated voltages:

Q1: e 0.573v
c -38.31v
b 0.002v

Q3: e 0.572
c -38.35v
b 0.007v

Q5: e -37.69v
b -37.03v
c -1.080v

Q7: b 1.183v
c 37.67v
e 0.684v

Q9: b -1.073v
c -36.49v
e -0.512v

Q2: e 0.575v
c -37.12v
b 0.003v

Q4: e 0.574v
c -38.37v
b 0.007v

Q6: e -37.68v
b -37.37v
c -1.080v

Q8: b -1.083v
c 36.67v
e 0.617v

Q10: b -1.176v
c -36.47v
e -0.615v

I did unsolder and remove the VR1 1dle current adjustment pot earlier in my quest to fix this which I did not post. I thought it might be shorted also. I placed my meter on ohms and took note of where it was adjusted so I could put it back where it was when I finished. I was able to change the value in both directions on both posts. Now whether I did just correctly I do not know. I learned how to do this from an older Navy tube guy.

Davey..check V8 voltages again. You indicated the base was -1.083 yet the emitter was +0.617. Possible the Base is actually + ???????

fact one - in this amp all transistors are biased on all the time.

fact two - biased on means there is 0.6v difference between the emitter and base voltages.

fact three - idle current means biased on, and if there is no idle current there is substantial "zero crossing" distortion where some of the signal is lost because until the signal rises to a particular level and turning on the transistor - the transistor is NOT reacting to the signal and more or less passing it through. We call it distortion.

That +1.2v and -1.2v (2.4v total) provided by the stv-4h diode and vr1 is extremely critical. The circuit is touchy because IF it is broken (or even the diode installed in reverse), the voltage rises to destructive levels (+38 and -38 for example), turning on the drivers and outputs HARD - SMOKE hard.

Plus the current THROUGH the STV-4h diode determines the voltage across the STV-4h diode. Thus R11 (1.8k) and R19(3.9k) are also critical as they set the current flowing through the stv-4h diode.

WHY 1.2 volts? There are output transistors and they need to be turned on, thus they need 0.6v on their bases.
Next there are the driver transistors. They put out the 0.6v on their emitters for the outputs transistor bases. BUT the drivers NEED their OWN 0.6v to turn on... thus 0.6 + 0.6 = tadaa... 1.2 volts.

continued:


Q1: e 0.573v (0.573 - 0.002 = 0.571) possibly on
c -38.31v
b 0.002v

Q3: e 0.572 (0.573 - 0.002 = 0.571) possibly on
c -38.35v
b 0.007v

Q5: e -37.69v (37.69 - 37.03 = .63) definitely on
b -37.03v
c -1.080v

Q7: b 1.183v (1.183 - 0.684 = 0.499) having a hard time believing this is on, which would mean NO idle current is flowing.
c 37.67v
e 0.684v

Q9: b -1.073v (1.073 - 0.512 = 0.561) possibly on
c -36.49v
e -0.512v

0.684 + .512 = 1.196 , so driving the outputs: 1.196 / 2 = 0.598 with an offset voltage of about -0.086 volt and a strong possibility of idle current being measurable.
becomes
(.684 - 0.086 = 0.598 ) + ( .512 + 0.086 = 0.598)

now, correcting for offset voltage:
Q7: b 1.183v (1.183 - 0.684 = 0.499 plus the 0.086 offset = 0.585v it is plausible that it is turned on
c 37.67v
e 0.684v

but what one side giveth, the other side taketh, so:

Q9: b -1.073v (1.073 - 0.512 = 0.561) we LOSE 0.086v here... sigh...
c -36.49v
e -0.512v


----------------------- other supposedly good channel ------------------------

Q2: e 0.575v (0.575 - 003 = 0.572 ) possibly on
c -37.12v
b 0.003v

Q4: e 0.574v (0.574 - 0.007 = 0.567 ) possibly on
c -38.37v
b 0.007v

Q6: e -37.68v (37.68 - 37.37 = 0.31) this one is definitely NOT turned on!!
b -37.37v
c -1.080v

Q8: b -1.083v (1.083 - 0.617 = 0.466) this one is definitely NOT turned on!!
c 36.67v
e 0.617v

Q10: b -1.176v (1.176 - 0.615 = 0.561 ) this on is possibly trying to turn on
c -36.47v
e -0.615v


now +0.617 and -0.615 is a GOOD base drive voltage for the outputs, with idle current certainly flowing and about a 0.002 volt offset.


-----------------------------------------

The way these readings of on and not on is mixed I have a hard time believing any channels are working. Now if your transistor locations as to what readings are which transistor, that makes it more believable.

After checking the variable resistors, you DID try to get them back to their pre-test values of earlier today?

WHAT WERE their resistance values in their as-found state?

I can't keep this up much longer this morning. Just no tthe time - hopefully Larry can also use these posts to help you.


Sometimes it is easier to state an offset voltage relative to ground, then put the black dmm lead on the output instead of ground, and then read the transistor voltages.
This automatically corrects for large offsets in a malfunctioning channel.

larryderouin said:

Davey..check V8 voltages again. You indicated the base was -1.083 yet the emitter was +0.617. Possible the Base is actually + ???????

Click to expand...

Sorry it took awhile to get back with you. I had some honey do's to do.

Here are the current voltages on Q8:

b 1.209v
c 34.80v
e 0.623v

markthefixer said:

continued:


I can't keep this up much longer this morning. Just no tthe time - hopefully Larry can also use these posts to help you.
.

Click to expand...

Mark. I understand time constraints. I do appreciate all your work. You are so thorough in your explanation, me not being a tech, sometimes it overwhelms me. I apologize for my lack of knowledge. Everyday I learn something new.

Q: After checking the variable resistors, you DID try to get them back to their pre-test values of earlier today? -

A: Yes

Q: WHAT WERE their resistance values in their as-found state?

A: I know I should have but I did not write them down. I just remember 0v on both pots on bad channel. I could not set VR1 at all. Somehow it now reads 20mV but it will not adjust up or down. Left channel still plays fine. Right channel distortion.

Should I pull and check the stv-4h diode for that channel. I do not have anyone to replace it.

When you are adjusting the Variable resistors power off and use the ohmmeter across the center terminal and one end. Measure OHMS.

I'd pull them and WITHOUT touching knob, test for total ohms across pins 1 and 3. Then WITHOUT adjusting write down the values in OHMS between 1 and 2 and again for 2 and 3. Do this for both.

Use mini clips now. on pin 1 and 2 or 2 & 3, using the ohm meter, SLOWLY rotate the knob and watch for smooth action on the numbers, or any drop off's. If you have problems with them, replace. I don't have the part #'s right off hand, but they should be in one of the lists. I'd just go ahead and replace them anyway due to past history, and the fact that they are crummy trimmers to begin with. 3362 series bourn's come to mind but it's a good possibility I'm wrong.

As for the transistors, Mark lost me with all the math. I'm better with point ot point and tubes. Solid state still gives me the willie's at times. The new numbers look better and it looks like Q8 is either turning on or trying to.

STV-4H's. These are VERY VERY BRITTLE within 1" of the block. If you can desolder the leads, CAREFULLY do so with an ABSOLUTE MINIMUM of LEAD Movement. You just need one disconnected to test. DANGER WILL ROBINSON!!! It don't take much to break them off at the block, but if they do break, then Mark's got you by the short Hair's. If you ask nicely, beg and grovel at his feet, he'll probably fix them for you.

Once you have the lead out switch your meter to DIODE and test them with the meter. If it DIODE Test's ok, reinstall the lead and don't mess with them. Write down the results and post.

Quote: "As for the transistors, Mark lost me with all the math".

He did with me also. He knows his stuff. There are a lot of smart people when it comes to fixing this vintage stuff on this forum and I have read many of your posts that have also helped me in the past. Thank you for taking the time to help me and others.

That being said, I went ahead and ordered the new trimmers for both channels. I agree with you as far as them being "crummy trimmers" so I took the parts number off MTF's Pioneer SX-737 Master list and ordered them. They should be in soon.

The ones that were listed are:

3386F-1-101LF

3296Y-1-103LF

I also removed one side of the STV-4H's and here are the readings.

The side the VR1 is on. Manual reads LEFT channel is:

1.814 and OL

The side the VR2 is on. Manual reads RIGHT channel is:

1.861 and OL

I will let you know when the parts arrive and I have them installed.

Sorry,,,,,,,,,,,,,,,