SX-880 Restoration

Discussion in 'Pioneer Audio' started by Ross Henning, May 10, 2018.

  1. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Hi Guys,
    I am now working on an SX-880. First thing, the protection relay won't engage. I pulled out the manual and started following the DC offset and Bias adjustments on Page 17, and found that I have 40 volts potential at pins 5 and 12. Not good obviously. I've been having some really good luck restoring these Pioneers thus far, but I'm admittedly still a newbie, and this is the first one that I've had with this type of problem.

    I've seen posts before where it's recommended to replace all transistors in the power amp chain when one or more has been blown. I've not gone to check the transistors, but with 40 volts at pins 5 and 12, I'm guess that's probably the situation. Am I wrong to think that? If that's probably right, is it right to replace the whole kit-n-kaboodle, or should I pull each one, test it, and replace the bad ones?
     

     

    Please register to disable this ad.

  2. john stumpf

    john stumpf AK Subscriber Subscriber

    Messages:
    608
    Location:
    philippines
    i did a 880 and a 890,they are vrtiually identical, i suppose you know about the power supply re-solder job due to heat. if not there's heap's about it here. in that vein, id like to point out that there are some things wrong from the get go, you will note a ceramic wire wound resistor attached to the heat sink. as i recall its 5 ohm,5watt. double it at 10ohm 10 watt. i used a dale w/w of that value if you do just cut the ears off and you can use the original mounting clip with some heat compound. it dramatically reduces the heat load on the supply transistors. there's also a cap in backwards thats in the protection circuit. i've got to go to the dentist!
     
    Last edited: May 11, 2018
  3. Watthour

    Watthour Electron Rancher - JS3600

    While Mr. Stumpf is out visiting Dr. Phil Andbill...

    If you measured near rail voltage at Pins 5 & 12, there is a good chance at least one of the outputs in each channel is shorted E-C. Unplug/remove them, and test again. It is also possible that their drivers (Q15 through Q18) and emitter resistors (R275 through R280) suffered damage. Check all that as well.

    Search the board for reference, but some of our more learned members would advocate replacing all the outputs and provide the reasoning for doing so. This would apply to the pairs of drivers as well, if those are damaged.
     
  4. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    I've pulled Q15 - 18 off the board and tested them. They're testing fine, as are the 2SD745 and 2SB605 transistors (Q1 - Q4 off-board). I've also checked Q19 and 20, as well as D3 and D4. With everything back on the board, voltages are all out of whack in comparison to the voltages listed on the schematic. It's not limited to one channel, but given that both channels have thus far had the same components check out, would it be more likely that the problem is something that feeds both sides? Again, I'm really new to trying to track down a problem such as this, so any pointers you can give, I would really appreciate.
     
  5. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    I've been going through the power regulation section to see if I could find anything wrong. I've replaced all of the transistors in this section, with the exception of Q24 an Q25 using recommended replacements from a list that I found in another thread that Mark the Fixer had come up with. I didn't test any bad transistors, but I was thinking that maybe one or more was weak or leaking or something. That didn't make any difference in the main issue, but I did notice a couple of things that I think are weird, and might help point to what's wrong. First, measuring the voltage at C422 and C423 (A pair of 100uF 50V electrolytics), the voltage is around 2 volts different between the two sides.I'm pretty sure that these are supposed to be the same. I also thought this was weird, but maybe there's a reason for it: D29 tested in-circuit failed as current would flow backwards, or so it appeared. I pulled it out and tested it, and it was fine. Looking at the schematic, I realized that D29 is connected across the base and emitter of Q28, and Q28 being an NPN transistor allows the current to flow in the reverse direction across these points. Again, I'm a newbie, but I'm not understanding what the purpose of a diode is here if the transistor can completely counteract what it does? I'm guessing there's a purpose, and I'm hoping someone can explain that one to me. All other diodes check out fine.
     
  6. Watthour

    Watthour Electron Rancher - JS3600

    Are the emitter resistors intact?

    As for the differences at C422/423, what are the voltages at the main bus, Pins 9 & 11 on the amp board?
     

     

    Please register to disable this ad.

  7. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Pin 9 is +46.7v
    Pin 11 is -46.7v

    Emitter resistors all read properly.
     
  8. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    I had a weird, temporary, breakthrough this evening. I was running through, checking voltages again, and I noticed that one channel had suddenly "fixed" itself, and the output transistors no longer had negative voltages where they shouldn't, not high voltages anyway. Instead of -37 volts at the base of the main output transistors, I had like -1 to -2 volts. The other channel still had the -37 volts at the base, so I decided to take Q16 and Q18 off the board. I fired it up again, and the protection relay engaged. I started back-tracing, trying to find the source of the negative voltage in the one channel. I didn't have any success in locating it. After a while, the other channel went back to the -37 volts at the base of the main output transistors again. Is it possible that when I test these transistors with low voltages, they don't leak or short, but when the rail voltages are applied, they do? With Q16 and Q18 out of circuit, I get the -37 volts at the collectors. Same story on Q15 and Q17 when the first channel started acting up again.
     
  9. Watthour

    Watthour Electron Rancher - JS3600

    Out of curiosity, are the in-circuit resistances of R420, 422, and 423 close to spec? Are all the ground connections in the amp and protection sections solid, with good solder joints? Mr. Stumpf alluded to the power supply solder problems common to these (and 780s) and this may carry over to the other area of the board.
     
  10. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    I had what I believe is a major breakthrough today. I kept tracing the schematic and measuring voltages, looking for places where the -37v could be crossing over. I started pulling parts here and there to eliminate possible paths. Even though all of the output transistors test fine out of circuit, I found that when I pull the leads of one of the 2SB705s, I no longer get the -37 volts at the pad where the base of that transistor connects. It appears that the rail voltage is partially feeding back from the collector through the base of the 2SB705. With not transistor soldered in place, that pad now has -3V, the other channel has around -1.3V on the same pad with it's transistor in place. Not sure, but I'm theorizing that maybe having a good 2SB705 soldered in place may pull that pad for the base down to the -1.2V spec from the voltages on the schematic. I have a set of 2SB705's coming in on Monday, so that will definitively tell me if I'm right. The protection relay is not clicking at this point, but I'm guessing not having the transistor in place may be causing some other imbalance that's preventing the protection circuit from opening up. In the meantime, I replaced all of the electrolytics on both circuit boards (amp and tuner) and cleaned all of the pots and switches. I had already also replaced the main filter caps.
     
  11. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Well the 2SB705s arrived today, and no luck. I replaced the old ones and get the same -37 volts at the base. Back to the schematic to try to figure out where the heck this voltage is coming across.
     

     

    Please register to disable this ad.

  12. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Watthour, I've checked all of the resistors you mention, plus some. They all check out. I've now replaced most of the transistors on the board in hopes that one of them was leaking at a higher voltage and I just wasn't seeing it. I've looked all over for a cracked trace, but I'm not seeing one. I'm simply baffled as to what it could be, or how to find the source.
     
  13. Watthour

    Watthour Electron Rancher - JS3600

    It certainly is an odd one. The fact that it "fixed" itself temporarily really screams "connection" but it could be internal to a component as well.

    Have you checked continuity in the bias and idle current potentiometers? VR5/6 and VR7/8 need to have continuity from the ends to the wipers. If a bias trim pot goes open circuit, the driver can just run away to full ON state. Similar things can happen if the idle current trimmers go open circuit.
     
    The Fuxtor likes this.
  14. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Yes, I did check that. In fact, I just replaced VR5 and VR6 with nice multi-turn pots because I found something else strange. I found that if I remove Q13 & Q14 (the first stage of what looks like a Darlington pair to me), the voltages past that point are fine and the protection relay kicks in. I checked these two transistors, and they are fine. I've also tried replacing them anyway for good measure. That made me think that the problem must be prior to this point. I get 0V at the base of Q19 and Q20 as I should, but what's very odd is that on one side of VR5 and VR6 I'm getting -25V, and on the other side, -.5V. Maybe that's correct, but the schematic shows -16.6V between R233 and VR5, so I would think that I should measure more than -5V at the emitter of Q19 and Q20. Am I wrong on that? Seeing that odd voltage, I decided to replace VR5 and VR6 as I had intended to do so later anyway. The voltage at the base of Q5 and Q6 isn't exactly .6V as it shows on the schematic, but it's in the vicinity. I want to say somewhere around .8V if memory serves. Unfortunately, the schematic doesn't show the proper voltages around Q5, Q6, Q7, and Q8. I checked both of these out of circuit and they tested fine, but I'm beginning to wonder if they aren't involved? I don't have direct replacements, but I could put some matching pairs of KSC1845s and KSA992s to replace them if I have to.
     
  15. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    If I want to try replacing the 5 legged transistors with a pair of KSC1845 and KSA992 respectively, do I only need to compare the hfe, and if so, how close do they need to be? Do the NPN/PNP “mates” need to be matched in any way as well?
     
  16. Watthour

    Watthour Electron Rancher - JS3600

    Match the HFe as closely as you can between pairs. It isn't as critical to try matching them across sets (NPN pairs to PNP pairs), only the actual connected pairs. Try to thermally bond the pair as well so their junction temperatures remain as close as possible.
     

     

    Please register to disable this ad.

  17. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Well, the 5 legged transistors were not the source of my stray -37volts. However, I checked their hfe, and the 2SA979s were at an hfe of 5 on both sides, and the 2SC2291s were an hfe of 7. According to specs on these parts, they should have a minimum hfe of 250. I kept the old parts in case I need to solder them back in, but I've matched two pairs of each KSC1845 and KSA992 and shrink-wrapped them together as replacements for now. With the 2SA979s pulled, the negative voltage didn't appear at the base of the output transistors, so the protection relay activated. However, if I pull the 2SC2291s and leave the 2SA979s in place, the negative voltage makes it to the output transistor bases. I have a feeling that maybe the problem has something to do with Q19 and Q20 as it looks to me that somehow those balance the difference between the negative and positive voltage, and I'm only getting -0.5v at the emitter of both of them while there's around +40V at the collector. That theory could be bunk too...I'm just guessing.
     
  18. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    I'm just about at my wit's end. I feel like I'm just going in circles on this thing. I'm trying to fix this one up for my son who lives across the country from me, so it really bugs me that I haven't been successful with this one. Would one of you guys who really knows what you're doing (unlike me) be willing to get on the phone with me and guide me through what to look at and see if we can't figure it out together?
     
  19. Watthour

    Watthour Electron Rancher - JS3600

    Can you list the voltages at Q5/7/9/11/13/15/17 and 19?
     
  20. Ross Henning

    Ross Henning AK Subscriber Subscriber

    Messages:
    106
    Location:
    Pinehurst, TX (Houston Area)
    Sure thing. Unfortunately I won’t be able to get to it until tomorrow evening, but I will check them and post as soon as I get home from work tomorrow.
     

Share This Page