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

Restoring uncle's Sansui 9090DB

Discussion in 'Exclusively Sansui' started by Juan9568, May 5, 2018.

  1. tnsilver

    tnsilver AK Subscriber Subscriber

    Messages:
    2,993
    Location:
    Boston
    That would be interesting. +/-200mV is borderline, but shouldn't trip the relay. Are you sure you're measuring DC with the correct scale? In a worst case scenario you're gonna get rails voltage, so set the meter to high enough scale (one that can handle at least 63V).
     

     

    Please register to disable this ad.

  2. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Yes..DC Voltage
    Yes DC Voltage I got around -200 when I put it in 2000m scale

     
  3. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
  4. tnsilver

    tnsilver AK Subscriber Subscriber

    Messages:
    2,993
    Location:
    Boston
    2000m as in 2 volts tops? Suppose you had 10VDC, what would the meter show? I'm not convinced that measurement is right, it doesn't make sense. The -200mV may be a false reading and all that's needed is a little more trimming. I'd set an appropriate scale for this measurement first (like 200V), make sure about DC levels in the unit, then move on to troubleshooting the driver amp, if still required.
     
  5. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Yes, 2000m as 2volts tops. The meter wouldn't give me any reading with that scale if I had 10VDC.
    I'm gonna try 200V setting, hope I get something out of range so I don't have to troubleshoot the driver board
     
  6. DougBrewster

    DougBrewster AK Subscriber Subscriber

    Messages:
    535
    you need to get some confidence in that reading; do you have another meter to do a cross check?
     

     

    Please register to disable this ad.

  7. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Yes...got a new meter, I will cross check everything at night
     
  8. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
  9. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
  10. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Well, the manual is wrong again here..It says:

    F-2656

    R13,14---4.7Ohm 1/2W Fuse R

    R25, 26 ---10Ohm 3W M.R.


    In my circuit I got it the other way around.

    R13,14---10Ohm 3W Ceramic Resistor

    R25, 26 ---4.7Ohm Fuse resistor


    All of them read OK. I'm checking all the resistors on the driver board and see if I find something out of range
     
  11. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Yes this is stock relay..

    The old driver board caught fire and lots of the components burned (mostly on the lelft channel) so it never came out of protection but I know It worked before that...I don't know what really happened cause I didn't have the amp with me at that time...

    Also, all the left channel output transistors where shorted that's why I decided to replace everything.

    On power board I replaced:

    R30, R31, R32
    All ceramic and electrolytic caps
     

     

    Please register to disable this ad.

  12. tnsilver

    tnsilver AK Subscriber Subscriber

    Messages:
    2,993
    Location:
    Boston
    IIRC there's a black grounding wire from the F2657 protection board that hooks to a terminal in the upper chassis RF power supply PCB. However, I don't know if it existed on this version. Can you spot it? Is it grounded?
     
  13. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
  14. tnsilver

    tnsilver AK Subscriber Subscriber

    Messages:
    2,993
    Location:
    Boston
    I meant a ground connection to the tuner power supply board and I can't really see where the one you're holding is connected, but suppose it's grounded nonetheless.
    Let's assume the protection board is intact (but I suggest you overhaul it anyway) and go back to square one.

    Here's what I suggest:
    1. Swap the DBT light bulb to 100W b/c the 70W is too bright for me (maybe others too) to get an indication from the pix.
    2. You have a mixture of old Sanken and new On Semi output transistors pairs in the back of the unit - swap them all to match.
    3. Make sure the connections at the bottom of the driver amp Molex footer are all springy and make good contact with the pins. No solder blobs should be in there restraining the contacts.
    4. The only components I could not check on the excellent pictures of your driver amp are the LED's orientation and the STV-3HY's polarity. LED's go anodes (+) to the outside, cathodes (-) facing each other. STV-3HY's go yellow dots (cathodes) facing the outside. Can you verify?
    5. Make sure to hook up the top Molex header (2 channels and ground) to the driver amp. Set the bias trimmers on the driver amp to minimum and measure bias current. You should have about 17mA. If you get less, set them to 17mA which should be about a quarter turn. Notice the DBT and the LED's and make sure they do not light up. If they do, it's a too-much-bias-current warning.
    6. See if you can get the protection circuit to click the relay and exit protection mode. If not, measure DC voltage on the gray/blue wires coming out of the bottom of the PS board near the driver amp, with, and then without, the driver amp. Put the F04/F05 fuses back in when you remove the driver amp. Leave them there while not measuring bias.
    I hope we see some progress b/c there's no apparent reason for the protection circuit to protect with nearly zero VDC input to it.
    Let us know how it goes.
     
  15. Overundr1

    Overundr1 AK Subscriber Subscriber

    Messages:
    2,514
    Location:
    Mississippi
    Tom is walking you through this, mind if I share a few thoughts?
    It is very unusual to see both offset values on a new card so close, lets see if those values are true or just spurious readings. Install only one of the two fuses and try adjusting offset on the channel you inserted the fuse in while watching your meter. If you see the value change then repeat with one fuse in on the other channel. My thought here is one channel offset is actually out of range thus tripping the protect circuit (as it should) and you might find it by trying this method. Normal caveats about guarding those probe tips apply and for goodness sake don't crank on the bias trimmers by mistake.
    Btw those cards once installed are a great reliability upgrade for the 9090db's, I thought they had all been used up and no longer around.
    -Lee
     
    tnsilver likes this.
  16. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    This is getting weird....While I was reading your guide I got the green light 3 times but I haven't done anything different...just reconnected the board, fuses and it worked.
    I was able to bias the right channel, waited 20 minutes, changed the fuse to F04 and meter on F05 to bias the left channel and again I can't get the green light.
    Tried everything again but still nothing...Re-checked DC offset on the right channel and it was at -15mv, took it down to almost 0...DC on the left is at -1, 0mv

    I thought that maybe the fuse holders weren't making good contact, checked that but nothing.

    Could it be possible that the 70W DBT doesn't let the amp power correctly?? Remember I'm on 220v grid not 120v...

    1. Gonna try with a 100W tomorrow

    2. My outputs are all new. Here a pic

    3. Connections checked, everything is tight and makes good contact, no solder blobs

    4. LEDS are ok, cathodes facing each oder. STV-3HY yellow dot facing outside


    5. With the bias trimmers at minimum I got 2mA. When I got the green light I started trimming the right channel slowly within the 4 minutes till 28mA (2 quarters turn aprox). The LEDs didn't light up and the DBT was barely on
     

     

    Please register to disable this ad.

  17. tnsilver

    tnsilver AK Subscriber Subscriber

    Messages:
    2,993
    Location:
    Boston
    It's not really weird, just some flaky component somewhere. I can think up 3 options and I think you will need to address them all to clear out all the variables out of the equation. On the bright side, I think your driver amp is A-OK.
    • Faulty solder joints on the PS board - reflow / renew all solder pads on the PS board. Look for cracks in that PCB and try to get clean shiny soldering on your newly installed components. You could improve on the ceramic caps soldering and if it represents the rest of the soldering... well, try to improve. Pay attention to the area bellow the F2624 driver amp board. It's prone to mechanical stress. Could be a cracked trace that's randomly conducts based on the driver amp insertion pressure.
    • Flaky F2627 protection board components. It looks real old... I am not a big fan of the shotgun approach but some of the electrolytics on that board couldn't possibly stood the test of time and I doubt all resistors are up to specs.
    • Bad filter caps... These could be doing a bad job of eliminating DC and potentially pollute the input to the protection board. They are kinda expensive, so you'd want to test them first, but they are a likely candidate to cause such issues and it would be darn hard to catch on the meter if they really act up
    You may wanna utilize Lee's tip above. This may work real good in spotting the bad channel (assuming the problem is limited to just one).

    BTW: The DBT limits current, it's not really voltage dependent. Some of our friends here, from the UK use a 70W bulb as well and they have a similar grid voltage to yours. The only reason I asked for this is b/c the 70W bulb looks too bright in your video and I can't tell if it's too bright or dimmed down. You haven't displayed it from the fire up stage and there's no reference to the inrush behavior of the bulb so it's hard to tell. A 100W bulb is dimmer and easier to figure out.
     
  18. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
  19. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Still testing with a 70w bulb with Lee's instructions. My brain is about to explode.
    Check this out!

    Testing only Right Channel DC



    Testing only Left Channel DC. I start getting that buzzing sound



    While testing Left Channel the sound gets louder



    After 3 minutes of that sound, relay clicks, green light (love when I hear that click...it gives me hope haha). Only left channel fuse connected



    I turn the unit off and start again but this time with both fuses. Still testing DC on the left channel. Nothing happens...

     
  20. Juan9568

    Juan9568 Member

    Messages:
    71
    Location:
    Buenos Aires, Argentina
    Now I check again if relay clicks. This time without the fuses but driver board still connected. Nothing happens



    Now I check if relay clicks without driver board....and it works. It also clicks with the fuses connected.




    Sounds to me this is a driver board issue...

    Why would it be main filter caps if relay clicks without the driver board??
     

Share This Page