SA-8100 protection circuit issue

Discussion in 'Pioneer Audio' started by KeithD, Jul 2, 2017.

  1. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    Out of curiosity, I reconnected Q5 base. Readings

    E C B
    Q3: -0.2 +33.1 0
    Q4: -0.4 +0.6 0
    Q5: +16.9 -1.1 +25

    I was thinking next to connect D7 and take readings and then if all is well, D6.
     
  2. petehall347

    petehall347 the brandy coffee man

    Messages:
    23,506
    Location:
    uk.. the middle bit
    q4 seems to be having problems .. say if you lifted q4 collector the voltage has to rise .
     
  3. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    Ok, I will try that this afternoon (I know it is already afternoon where you are!)

    Another question came to mind. I put in the two spare KSC1845FTA I had for Q3 and Q4. Those were recommended replacements for Q1,2 (2SC869 - though I had 2SC857 in Q1,2 since I have an older board).

    Original Q3,4 were 2Sc945. Is KSC1845FTA a good or recommended replacement, or is there a better match for 2SC945? Just wondering if that may be a source of problem.
     
  4. QSilver

    QSilver Super Member

    Messages:
    1,094
    What is your board part number?
     
  5. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    AWM 027 A
     
  6. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    SA8100 protection data.jpg SA8100 protection data.jpg
    I agree Q4 C seems to be the problem. I did a series of tests this afternoon summarized above.
     
  7. petehall347

    petehall347 the brandy coffee man

    Messages:
    23,506
    Location:
    uk.. the middle bit
    q4 is leaky or is in wrong way . well that is how it looks from here .
     
  8. petehall347

    petehall347 the brandy coffee man

    Messages:
    23,506
    Location:
    uk.. the middle bit
    i have looked for an A revision diagram but haven't found it as yet . i see the board layout is different from pics . no idea if circuit is different though . if it is we might be missing something .
     
  9. QSilver

    QSilver Super Member

    Messages:
    1,094
    I've uploaded some pictures of the protection board in my SA-8100. I think it is the same... AWM-027.... EDIT: Seemingly mine is the D revision?

    I've tried to photograph the transistors I used to rebuild it too. Mine seems to be working just fine.

    Q1+Q2: KSC1845FTA

    Q3+Q4 KSC945CGTA (I think i replaced these too... and it helps if they are gain matched I think?)

    Q5 for me is the original 2SA733 (compliment to 2SC945)

    Q6 is 2SC945 (or KSC945CGTA)

    Q7 is KSC2690AYS

    You could be right.... I think I have had problems sometimes when replacing 945's with 1845's...

    Not sure if you checked yet... but Pin 6 should have -13V on it?
     

    Attached Files:

  10. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    Not in wrong. I double checked schematic and what is printed on the board. The 1845 is ECB from face.

    The schematics for the A version is in the service manual that can be downloaded at HiFi Engine. Do you have access to that? There is another document on there that is just the schematics, but that one has the revision D board.

    Yes, you have revision D, you can see it printed on the board after the board number.

    Yes, pin 6 shows -13. All pins have correct voltages coming in according to the schematic. I listed them in post 104.
     
  11. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    I have looked at the A and D board board images, and it seems to me the circuits are different. I don't have the circuit diagram for revision D to clearly see it, so I was just going off the image in the schematics document I mentioned that showed revision D.

    In all of my work and my posts, I have been working off the revision A diagram, since that is the board I have. Hopefully we have not been working at cross purposes.
     
  12. petehall347

    petehall347 the brandy coffee man

    Messages:
    23,506
    Location:
    uk.. the middle bit
    just in case did we cover r15 r16 r17 ? i suspect they are fine . but something is not correct .
     
  13. petehall347

    petehall347 the brandy coffee man

    Messages:
    23,506
    Location:
    uk.. the middle bit
    ok i now have the correct schematic , it is similar but not same as 027 .
    be back in a few hours ..
     
  14. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    Just so we are on the same page, here is the schematic I have.
     

    Attached Files:

  15. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    All three are good.
     
  16. QSilver

    QSilver Super Member

    Messages:
    1,094
    Do you have a scope to probe Q3 and Q4 for noise? Have you tried putting the old 945's back in here?

    The 945's are less "sensitive" than the 1845's so perhaps chaning back to some 945's would get it functioning normally. Data sheet seems to suggest that 945's base typically saturates at a higher voltage than the 1845's.

    I've re-done a few protection circuits in this and the SA-9100. If I changed Q3 and Q4 to 1845's it didn't work properly and I think its down to this difference in sensitivity. Might be worth a try?
     
  17. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    I do have access to an oscilloscope. What is the procedure for testing a transistor with an oscilloscope?

    I do not have new 945s to put in for q3,4. I replaced them with the spare 1845s I had because the opinion was the 945s might be the issue. If I put the old ones back in, won't I have the same issue I started with?
     
  18. KeithD

    KeithD Active Member

    Messages:
    248
    Location:
    La Crosse, WI
    Out of curiosity, what function does d6 and q4 serve in the overall "protection" function?
     
  19. LesE

    LesE 110284 Subscriber

    Messages:
    614
    Location:
    Ontario, Canada
    This is a very good question. Q3 and Q4 work together as a differential amplifier in order to detect the presence of excessive DC voltage at the amplifier output. The differential amplifier translates a small difference in the base voltages of Q3 and Q4 into a large difference in collector voltage.

    The power amplifier audio output is applied to pins 2 & 4 and are summed by R9 and R10 and the summed signal is applied to the base of Q3 and Q4. The signal path to the base of Q3 is through C3 and C4 which will pass the normal AC audio signal but block any DC voltage that may appear. So, any DC voltage present in the amplifier output will appear on the base of Q4 only. If a large positive DC voltage appears on the amplifier output, Q4 will conduct more current and Q3 will conduct less. The result will have the collector voltage of Q4 to go low while the Q3 collector goes high. The low collector voltage on Q4 then pulls down the base voltage of Q5 which turns it on and triggers protection.

    When a negative DC offset is present, the situation reverses and the amplifier balance swings over so that Q3 collector is low and Q4 is high. So why are D6 and D7 required? They are essential as they isolate the collectors of Q3 and Q4 which are connected together at the base of Q5. Without these diodes, the collectors of Q3 and Q4 would always be the same voltage and the DC protection would not work.
     
    Last edited: Aug 11, 2017
    QSilver likes this.
  20. LesE

    LesE 110284 Subscriber

    Messages:
    614
    Location:
    Ontario, Canada
    To illustrate somewhat, here is the simulation data for an amplifier DC offset of +4V and -4V.

    upload_2017-8-11_12-18-6.png
     
    QSilver likes this.

Share This Page