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

fisher MPX decoder troubleshoot

Discussion in 'Fisher' started by vintelectra, Apr 21, 2013.

  1. vintelectra

    vintelectra Active Member

    Messages:
    473
    Dave: Grounding #7 of V102 causes the signal to disappear. This would suggest that 38k subcarrier is leaking thru the left ring, Yes??
     

     

    Please register to disable this ad.

  2. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    OK. You've really got to help me here. If lug 1 is grounded, and in doing so, it eliminates the 38 kHz noise previously seen at the output of the left channel, then how can there be clean 1.0 volt 1 kHz sine at the left channel output, since lug 1 is the input to the sub-chassis and it's grounded??????

    If grounding lug 1 eliminates the left channel noise, then that suggests checking for issues at V100 -- noisy tube, noisy components, etc.

    I'm not there to see what you're seeing, so you've got to paint me a clear pic. Eliminating the 38kHz noise at the output of the left channel was the original problem presented. If grounding the input to the MPX sub-chassis makes the outputs go silent, the the issue has got to be around the input stage.

    Dave
     
  3. vintelectra

    vintelectra Active Member

    Messages:
    473
    Sorry the misunderstanding: Grounding the input does NOT eliminate the signal, but just cleans it from noise. The only thing that eliminates the signal
    is grounding the left grid of V102. At present( and it changed from beginning) the signal is a clean sine 38k 1V, which takes us to your first comment
    "smacks of a bad diode". I have made a new ring, should I place it?
     
  4. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    Thanks for the clarification! Yes, I'd change out the ring if only to eliminate that possibility -- but first, make sure that whatever external cap is decoupling the B+ source to the trigger/oscillator stage is good too. If that cap is bad, any AC reference that the CT of the 38 kHz transformer has to ground is gone, which will then prevent the rings from balancing to zero output -- relative to ground -- when no composite is present.

    Dave
     
  5. vintelectra

    vintelectra Active Member

    Messages:
    473
    Checked B+ by placing a 20Uf electrol in parallel with B+ source of MPX = no influence on spurious signal.
    Replaced left ring with newly made ring = no change. ???
     
  6. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    You've got a good mystery to chase down. It's just going to take good, thorough troubleshooting to find the cause. I suppose the next place I'd go is the 38 kHz transformer itself. If there are shorted turns within the transformer, then that would un-center the quiescent balance of the rings as well.

    Dave
     

     

    Please register to disable this ad.

  7. vintelectra

    vintelectra Active Member

    Messages:
    473
    I will replace it and let you know. Martin
     
  8. vintelectra

    vintelectra Active Member

    Messages:
    473
    However, if the tranformer shows a normal 2:1 lisajous, what are the chances of that being bad?
     
  9. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    That simply means that the pilot and oscillator frequencies are locked 2:1. It doesn't indicate if the output of the transformer is balanced relative to ground. Somewhere you've got an imbalance or short allowing 38 kHz energy to leak out under quiescent conditions. That's what you've got to find.

    Dave
     
  10. vintelectra

    vintelectra Active Member

    Messages:
    473
    Without input drive: Pic#1 is signal on Rt grid of V102 ,2 V. Pic#2 is signal at left grid ,5V , 76K .
    Question: If both rt &left are problematic, that would exclude a diode ring problem?
    Pic #3 is signal at one limb of Z101: clean sine 37K 11V
    Pic#4 is the signal at the other limb, clean sine 13 V 36K, mild difference. Your thoughts?
     

    Attached Files:

  11. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    Not sure how the oscillator can be operating at two different frequencies unless you scope/counter probes are loading the circuit down.

    Try disconnecting the rings from the grids of V-102,and then examining the output of each ring. If there are differences there, it can only be due to the rings.
     

     

    Please register to disable this ad.

  12. vintelectra

    vintelectra Active Member

    Messages:
    473
    Rt ring output: pic#1 7mV
    Left ring out: pic#2 200mV.. Big difference
     

    Attached Files:

  13. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    We're back to where we started. Now, if you flip the input leads to the left ring, then then it will be operating in the same time domain as the right ring. In that scenario then, the output of the two rings should be identical. If not then left the ring is bad. Or conversely, if you flip the input leads to the right ring, then it will be operating in the same time domain as the left ring. Again, the output of the two rings should be identical unless one is damaged. Between this test, and ensuring the same output on each half-cycle of the 38 kHz sub-carrier, this should get to the bottom of things pretty quickly.
     
  14. vintelectra

    vintelectra Active Member

    Messages:
    473
    "flip the input leads" means swapping inputs left to right?
     
  15. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    Flip the 38 kHz input to one ring.
     
  16. vintelectra

    vintelectra Active Member

    Messages:
    473
    Better: Flipped left diode only. Pic #1 right diode out, not flipped 140 mV. Pic #2 left diode out (flipped) 45 mV 38K.

    I dont understand what you mean by diodes operating in time domain, i dont see a fast fourier tranform microprocessor in there.
     

    Attached Files:

     

    Please register to disable this ad.

  17. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    Look at the schematic. The diode rings are identical in every way. But one ring is wired in reverse of the other relative to the regenerated 38 kHz sub-carrier. That means that one ring conducts when the phase of the two 38 kHz legs is one way, and then the other ring conducts when the 38 kHz changes phase 180 degrees a fractional second later. The two rings do not both conduct at the same time. This is precisely why this method of multiplex detection is called a time division design. So, I'm merely suggesting that you flip the 38 kHz leads to one ring, SO THAT BOTH RINGS CONDUCT AT THE SAME TIME. Then, check the balanced output of the rings. They should both be identical. If not, one (or both) rings are defective. After you flip the leads to one ring and check the balance, put them back the way they belong and then flip the leads of the other ring and check the balanced output of the rings again. Now you're checking the opposite phase of the 38 kHz signal. You need to put the status of the rings to bed. This test will do that.
     
  18. vintelectra

    vintelectra Active Member

    Messages:
    473
    Left ring reconnected pic#1 150mV
    Right ring flipped pic #2 only 6 mV
     

    Attached Files:

  19. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,673
    Location:
    Ball Ground, GA
    And there ya go -- With them both connected the same direction across the 38 kHz source, you're getting only 6mV from the right ring, and 150 mV from the left ring. The left ring is bad. The diodes within it may still be good in terms of only providing current flow in one direction as shown with a conventional diode test. But the drop across one or more of them is well out of characteristic, causing the imbalance.

    Dave
     
  20. vintelectra

    vintelectra Active Member

    Messages:
    473
    Great! But the left ring was replaced, so both left rings are bad?? The new ring was made with NTE 109 diodes. I have three choices: 1 ... remake another- 2....replace it with a original from another MPX
    or remake the right ring with nte109 to make it equal. Which one would you do first? Thanks Martin
     

Share This Page