1. Time for some upgrades in server hardware and software to enhance security and take AK to the next level. Please contribute what you can to sales@audiokarma.org at PayPal.com - Thanks from the AK Team
    Dismiss Notice

Pioneer SX-850 Protection Relay and DC Offset

Discussion in 'Pioneer Audio' started by Brivan, Oct 12, 2018.

  1. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    I have a recently-acquired Pioneer SX-850 with a known protection relay issue. I purchased the receiver with the protection issue as a known problem. I removed the bonnet and base plate and checked power supply board voltages - all within 0.5v - 1v of spec. O-scope probing of all supplies gave me a nice, clean line. I started the power amp calibration per the service manual, but am having issues with the DC offset. My meter reading fluctuated all over the place - especially on pin 25 of the power amp board. The scope showed the same thing, with the average voltage reading being around -6v with VR1 fully counter-clockwise:
    20181012_130800(0) (Small).jpg

    Pin 10 is a little better, but I can't get close to 0v - about 30mV is as close as I can get:
    20181012_130940 (Small).jpg

    Output transistors checked good, as well as the 8 transistors on the power amp board (all were checked while soldered in place). With the power off, I ran VR1 - VR4 back and forth several times, then set them back to calibration position before powering back up. I'm guessing a power amp capacitor replacement is in my future - what do you think?
     

     

    Please register to disable this ad.

  2. markthefixer

    markthefixer On Hiatus, dealing with Dad's estate full time Subscriber

    Messages:
    21,054
    Location:
    Bensenville,Illinois
    put 512-KSA992FBU transistors into q1, q2, q3, q4

    Facts: the power amp runs off of +51.5v, +36.5v and -51.5v regulated.
    The DC offset setting circuits are run off the +36.5v and -51.5v regulated AND ARE CHANNEL INDEPENDENT .
    Pin 25 of the power amp board is the RIGHT channel.
    the power amp uses the
    input:
    2sa726 or 2sa763

    Vas stage:
    2sa818 or 2sa898
    2sc1628 or 2sc1903

    driver stage:
    2sd381 or 2sc1903
    2sb536 no alternate listed

    outout stage:
    2sb531 no alternate listed
    2sd371 no alternate listed

    so:
    2sa726 in the input circuit
    and the collector of one of them on each channel is directly connected to the -51.5v power supply.
    A failing 2sa726 might just jerk around the -51.5v regulated supply.
    That COULD have a lesser effect upon the LEFT channel, pin 10 DC offset.

    low noise caps could have film caps subbed in...

    power amp awh-059
    awh-059 c1 2.2 25 cssa 850 2.2 50 647-UKL1H2R2KDDANA **** low noise ****
    awh-059 c2 2.2 25 cssa 850 2.2 50 647-UKL1H2R2KDDANA **** low noise ****
    awh-059 c3 1 25 cssa 850 1 50 647-UKL1H010KDDANA **** low noise ****
    awh-059 c4 1 25 cssa 850 1 50 647-UKL1H010KDDANA **** low noise ****
    awh-059 c5 33 16 cea 850 33 50 647-upw1v330mdd
    awh-059 c6 33 16 cea 850 33 50 647-upw1v330mdd
    awh-059 c11 330 10 cea 850 330 25 647-UPW1E331MPD6
    awh-059 c12 330 10 cea 850 330 25 647-UPW1E331MPD6
    awh-059 q1 2sa726 512-KSA992FBU
    awh-059 q2 2sa726 512-KSA992FBU
    awh-059 q3 2sa726 512-KSA992FBU
    awh-059 q4 2sa726 512-KSA992FBU
    awh-059 q5 2sa818 512-KSA1381es
    awh-059 q6 2sa818 512-KSA1381es
    awh-059 q7 2sc1628 512-KSC3503estu
    awh-059 q8 2sc1628 512-KSC3503estu
    awh-059 q9 2sd361 512-KSC2073TU
    awh-059 q10 2sd361 512-KSC2073TU
    awh-059 q11 2sb536 512-KSA940
    awh-059 q12 2sb536 512-KSA940
    awh-059 vr1 10k multi 652-3296P-1-103LF
    awh-059 vr2 10k multi 652-3296P-1-103LF
    awh-059 vr3 100 single 652-3386W-1-101LF
    awh-059 vr4 100 single 652-3386W-1-101LF
    awh-059 d7 1s1885 512-UF4004
    awh-059 d8 1s1885 512-UF4004
    awh-059 d9 1s1885 512-UF4004
    awh-059 d10 1s1885 512-UF4004
    output transistor q13 2sd371a 863-MJ21194G
    output transistor q14 2sd371a 863-MJ21194G
    output transistor q15 2sd371a 863-MJ21194G
    output transistor q16 2sd371a 863-MJ21194G
    output transistor q17 2sb531a 863-MJ21193G
    output transistor q18 2sb531a 863-MJ21193G
    output transistor q19 2sb531a 863-MJ21193G
    output transistor q20 2sb531a 863-MJ21193G


    upload_2018-10-12_15-14-12.png
     
    merlynski likes this.
  3. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    Wow. Thank you so much for all the good information! I've said it before - this kind of knowledge is indispensable! I'll get your recommended replacements ordered for Q1 through Q4. I'll report back. I've read through a few other threads on this - it looks like the best access to the main amp board is to separate the back plate from it, remove the plate, then unscrew the output transistor heat sink assembly and swing it up and over the main amp board. Does that sound about right? I'm a little concerned about the rigidity of the wires connecting the main output transistors to the amp board.
     
    merlynski likes this.
  4. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    The replacement transistors should arrive tomorrow and I should be able to get them installed the next day. As has been warned on AK many times, I'll make sure the transistor orientation agrees with with the silkscreen print on the board.

    Additionally, I removed the back plate from the amp board, then removed the screws from the output transistor heat sink and swung it up 90 degrees. It looked like it was still going to suck trying to solder with that heat sink in the way, so I photographed and labeled the output transistors, then removed them, which allowed me to remove the heat sink. I replaced all the old electrolytic caps on the amp board.
     
    rcs16 likes this.
  5. rcs16

    rcs16 Super Member

    Messages:
    3,909
    Location:
    Baldwin, Ontario, Canada
    See if you can match the a992 for Vbe, a good match is <2mV
     
  6. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    Good news this morning - after replacing Q1 through Q4 (as well as Q9 through Q12, since I had it apart), I was able to go through the amplifier calibration and get everything set according to spec. I ran a 100Hz tone through it and the outputted sinewave looks just fine. Audio sounds good, as well. One problem remains that wasn't detectable previously - I have a loud rumbly, grating noise when selecting either phono 1 or 2 as an input. The sound is mostly coming from the right channel, but is also present in the left at lesser volume and the volume in both varies a little. I think the noise is telling me I have some crappy caps in the phono preamp section. I'll check on that shortly.
     

     

    Please register to disable this ad.

  7. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    I replaced all the caps on the phono preamp "equalizer" board, but still was getting the same noises. Some research on AK revealed that the transistors on that board - 2SA725/2SA726 and 2SC1313 are notorious for causing problems as they age. This was apparent when I replaced some of those same transistors on the amp board, per MTF's suggestion. So, I have replaced the two 2SA725s (2SA726 on the actual transistor) with KSA992 transistros and still have the same noises.

    I'm now ready to replace the 2SC1313s, but have a question: In some of the AK forums, the 2SC1313 has a replacement of BC337. I have some BC33740BU transistors in stock, but not sure if I should use them. When comparing the 2SC1313, KSC1815 (more-common substitution), BC337 and BC33740BU transistors, it's like comparing apples to oranges to grapefruit to watermelons - there doesn't seem to be a lot of likeness between the four. Can someone shed some light on this for me?
     
  8. Watthour

    Watthour Electron Rancher - JS3600

    From what I can determine from published data, the BC337 and BC337-40 types are not considered low-noise devices (not less than 4 dB noise figure) and may nit be suitable for the application. A BC550 might be a better choice if the KSC1815Y or KSC1845F is not available. The other operational characteristics would probably allow it, but it wouldn't be my first choice.
     
  9. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    Okay. That's the kind of info I was looking for. I'll order a few of KSC1845, as well as more KSA992 since these are to be replacements for common-failure transistors. Thanks for your input!
     
    Last edited: Oct 20, 2018 at 6:17 AM
  10. rcs16

    rcs16 Super Member

    Messages:
    3,909
    Location:
    Baldwin, Ontario, Canada
    BC337 was not intended for low noise operation, meant as a high current low Vce saturated switch. i looked at Onsemi and NXP data, they do not have noise specs for this device. I know of others actually testing them for noise.
    It inherently has low noise due to its construction, similar to a 2n4401. In the past designers, have used 2n4401 series in MC head amp applications due to their inherent LN.
    I'd still use A992/C1845 as they have more applications and use BC337 in a pinch.
     
  11. Brivan

    Brivan Active Member

    Messages:
    193
    Location:
    Sioux Falls, SD
    Thanks for the info! I'm not in a big hurry to get this receiver completed, so I'm fine with waiting for the correct parts to arrive - Wednesday of next week.

    For the roughly two years I've been working on receivers, I've never managed to install a cap backwards - until yesterday on the SX-850 equalizer (phono preamp) board. Don't know how I missed it, but I actually felt it getting hot - before it exploded, thankfully. That bad part is, at the same time, I saw light smoke wafting up from the power supply board. Great. After quickly powering down, I gingerly started feeling components on the PSB for heat, but didn't feel anything. Turns out Q6 that supplies -19vdc to that capacitor I installed backwards is a smaller TO-92, which I skipped over, figuring the heat-sunk TO220s were one of the victims. Anyway, the capacitor has been replaced and Q6 on the PSB is still providing -19vdc, but I have a replacement coming for it, as well - just to be safe.
     
    merlynski likes this.

     

    Please register to disable this ad.

  12. Watthour

    Watthour Electron Rancher - JS3600

    That's a good discovery. It should reduce confusion later.
     

Share This Page