I'm struggling with a mystery issue.

Discussion in 'General Electronics Chat' started by ranatungawk, Aug 4, 2010.

  1. ranatungawk

    Thread Starter Senior Member

    Oct 30, 2008
    Hi friends;:)

    I'm struggling with a mystery issue. Please help me by giving your expert comments on this….:confused:

    I'm troubleshooting somewhat old "YF-1150A" capacitance meter. When I received it didn't work at all. I found some signaling transistors got burnt. Having replaced them, I was able to fix major issues and now it's reading is almost correct. However, after 3,4 hours of repairing , the meter's idle value is changed to -00.2,-00.2…-0.4 (idle value should be 00.0 when any button is not pressed.).

    The mystery I found was this, when the idle value is changed (to -00.4 or -00.6) If I remove some parts (ex: transistor, capacitor..) from the PCB and resold them again as they were on the board, idle value comes to "00.0" again. but after 2,3 hours it go wrong as it was.

    So I thought this can be a result of a "bad capacitor" somewhere in the PCB. so I replaced all caps and checked all transistors. But the issue remains same.

    This meter circuit is based on The ICL7106 A/D converter. Please see the diagram and pics I have attached herewith. Could you please let me know what should I check…:eek:

  2. beenthere

    Retired Moderator

    Apr 20, 2004
    About the only way to run the problem down is to use a meter or oscilloscope to trace the input signal for the ICL7106 back. You can check the '7106 by opening the normal signal path and applying ground to see if the IC is stable.

    With a minor-appearing variance, though, what happens when you actually use the meter? Do you get consistent reading on the same capacitor? Do the values fall within the stated tolerance for the capacitor? If that is the case, the instrument might not need fixing.
  3. Nik

    Well-Known Member

    May 20, 2006
    Uh, could this be thermal drift ? I've vague memories of zeroing a desk DVM prior to high-precision use...

    Ah, in the attached Doc : ) to manually adjust zero, the adjustment range is ± 20pF.
  4. mik3

    Senior Member

    Feb 4, 2008
    Maybe is caused by the integrator circuit (bottom left IC) which drifts with time.
  5. Potato Pudding

    Well-Known Member

    Jun 11, 2010
    If you have had an overvoltage situation with that meter, it could be have scorched the PCB. Look at both sides of the circuit board starting from the lead connectors with a magnifier. See if you can get a bright light to shine through the board and you might notice internal charring that way.

    The char paths can be hair thin but if you look for where the path of least resistance is, the shortest route across the board from lead to lead, that should be where the board is likely to fail. I would look at the fuse holder area just to see if it is the .2Amp 250V from the diagram, in case the fuse had been blown and replaced with too high a current value. That would make this type of damageseem more likely. For charring I would look near C1 (?diagram is unclear?) 47pf connected between the two inputs. Check the whole board even

    This might seem like an unlikely possibility for a capacitance meter but imagine a student or repairman that "tests" a capacitor in a breadboard, or a circuit like a TV or computer monitor by using wires from the meter parallel touch connecting to the leads of a capacitor in circuit. This kind of bad idea has killed more than one capacitance meter.

    Or they put a charged capacitor into the meter after connecting it and pulling it out of a live breadboard or circuit.

    If it is the board and you spot the lead of charring, you just drill a hole through the board (between components and leads) in a spot to break the connection caused by the carbonization. You don't normally need to replace te board.

    Another possibility would be damage to the IC, but look at the board first.

    Your problem sounds like it might be caused by a carbon path especially since you mentioned burnt transistors.

    And your first repair would have been fine if the stray high resistance path created by carbonization caused leakage that changed your zero. Remove the batteries and (short out the battery terminals - sometimes works - or) wait 20 minutes and put the batteries back in. If the zero shows good again then somewheres you have leakage around a capacitor.
  6. ranatungawk

    Thread Starter Senior Member

    Oct 30, 2008
    *** however i found that wrong voltages come to ping 30 and 31 (IN+,IN-) of that ICL7106 A/D converter "+" and "-" is changed!!!!! [​IMG]
  7. Potato Pudding

    Well-Known Member

    Jun 11, 2010
    I would not assume that those voltages are wrong too quickly. And if they do turn out to be backwards then look for a reversed Diode or transistor.