Need help with Mitsubishi DA-A30 repair

Discussion in 'Solid State' started by kotofei, Aug 31, 2017.

  1. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    EDITED: I think you need a brief test at full power, to check voltage drop at R 146.

    Something is wrong at Q114 base. It could be shorted or those -34V coming from somewhere. Please check Q114 or replace it before doing a full power test. Q112, even if not perfectly symm to Q111, has -19V at the emitter and Q111 has 17.8, I don't see a short there.

    You could also check in this same condition (with DBT), the good channel at that same Resistor and Q114 equivalent.
    Last edited: Sep 11, 2017
  2. kotofei

    kotofei permanent gear evolution Subscriber

    Tested R146 out of circuit and it's fine, 152 Ohm. Soldered out Q114 and tested with diode test. It's shorted C to B.

    That's strange since I removed it working from the other channel and it was good before measurements. Furthermore, all measurements were done with DBT i.e. not at full voltage. The Q114 is 2SB536, Vcbo =130V, Vceo=130V, Vebo=5V, Ic=1.5A (pulse 3A) and Ib=0.3A. That was only around +/- 34V in the amp at the moment of testing and so the only way to short it was to apply more than 5V between base and emitter or more than 0.3 A to the base.

    I don't have anymore Q114 substitute so I need to get some to continue tests; however, why this Q114 failed three times?
  3. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    Perhaps it took the current when the output was shorted. ..

    I think you can try another transistor in that place, the 130V rating I guess is to hold the full PS voltage (like 118V) in case of a short. If you have some other transistors at hand post the numbers, perhaps you have a replacement.
  4. kotofei

    kotofei permanent gear evolution Subscriber

    They (drivers) were installed with no outputs present and tested also without outputs.

    The complete storyline is as follows: a pair of original drivers was gone when I "tuned-up" the working amp in the first place. Then I removed them and replaced with MJE15030/15031 which I had on hands. These were gone when I tested the amp without outputs and accidentally plugged it into 117V AC. THEN I removed working drivers from another channel, installed them in the wrong channel and tested voltages-and one of them, Q114, is gone now. Again, no outputs were present. So now I need to find a new output(s) to continue work.

  5. kotofei

    kotofei permanent gear evolution Subscriber

    Q114 is 2SB 537 PNP. I have 2SA1370, FJN4303RTA and MJE253G. The two former are small signal transistors, the MJE253G is referred as "power transistors" by Onsemi: Vceo=Vcbo=100V, Veb=7V, Ic=4A (peak 8A), Ib=1A, Pd=15 watts, Fr=40 mHz, TO-225 case.

    I think these could be used as Q114.

  6. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    I suspect you had a physical short when testing. I think if you redo all from scratch it will be fine, but you need to be obsessive, make sure nothing is shorted before powering up again.

    BTW, it's not a good practice to swap parts from a working unit/channel to a non working one, since you can end with 2 non working units/channels, and you loose the reference of a good channel to take measurements or compare readings. As you see, sometimes there is a 3rd thing making the suspect component to blow, so when replacing the component it blows again.

    Can you use 2sa1370 at q114?

    Edited: I see your original transistor is 1.5A, I think you can try the MJE253 (4A) but just for a test. I wouldn't increase the Ampere rating that much in that application, some transistors act like fuses in some parts of the circuit. Up-sizing them can make other things to blow under a short circuit.
    Last edited: Sep 13, 2017
  7. kotofei

    kotofei permanent gear evolution Subscriber

    Soldered in new MJE253 in Q114 place. Powered up the amp using DBT and checked voltages at Q114. In the very beginning B was +44V, C was - 34.4 V and E was - 34.4V. Then re-measured the voltage at base and it became -20.3V. Voltage around R146 was 0.0 mV.

    I stopped testing and soldered out Q114. Diode test revealed E-C junction is shorted, 0.00 mV drop. E-C and E-B are OK. Another transistor lost.

    So I will solder back original Q113 in the good channel and 2SA1370 as Q114 in the good channel to have it working. That good channel will be used for voltage testing and in the bad channel I'll be doing MASSIVE shortgun change of all transistors and diodes I can get my hands on and testing all passive components in between.
  8. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    I don't think the 2sa1370 can be used. It's 100mA.

    Be careful q113 can be burned too. We need to see what is doing the transistors to blow in that position. If base was +44V something is shorted in the positive rail, doing that +V to reach that point.
  9. kotofei

    kotofei permanent gear evolution Subscriber

    Schematics of the amp is wrong! It doesn't correspond to the actual connections on the PCB.

    Specifically, point 3 which is the + rail or + of filtering C505/C605 is connected to collector of Q114 through C507. According to the schematics C507 is connected to collector of Q112.

    In addition point 2 is connected to bases of Q113 and Q114 through R143, R145 and R184. According to the schematics, it's point 3 i.e. + rail that should be connected to Q113/114 bases through these resistors.

    This is true for both channels looks like factory setup. Checked several times.

    Waiting for replacements to come from Mouser...
  10. kotofei

    kotofei permanent gear evolution Subscriber

    Replacements came and were installed in the amp as follows:

    Q109 - original 2SA999 - replaced with KSA992FBTA
    Q111, Q120 - 2SC1904 - KSC2690 AYS
    Q112, Q117 - 2SA899 - KSA1220AYS
    Q115, Q119 - 2SB516 - KSA992FBTA
    Q116, Q118 - 2SD756 - KSC2383 YTA
    Q125 - 2SA693 - 2SA1370
    Q503 - 2SD381 - KSC2073 TU
    Q506 - 2SB536 - KSA940 TU

    D112- MZ306 zener 6v - 1N5233B zener 6V
    D117, D120 - 1S188 - BAV20-TAP

    Drivers and outputs are not installed yet.

    Also replaced all resistors that were exposed to high voltages and burned: R145, R184, R156, R144. Cleaned all soldering points with acetone including output and driver ones.

    Plugged the amp through DBT with drivers and outputs not installed. The bulb was bright for a second or so than turned very dim and stayed this way.

    Measured voltages against ground. Numbers in volts unless shown:

    transistor E C B

    Q109 2.17 -1.91 1.86
    Q110 -1.92 1.60 -0.69
    Q111 4.22 52.06 2.12
    Q112 -4.0 -52.07 -1.85
    Q113 0.81 52.11 3.48 no transistor installed, tested at terminals
    Q114 -0.71 -52.19 -3.20 no transistor installed, tested at terminals
    Q115 59.09 1.76 58.45
    Q116 1.22 57.6 1.76
    Q117 58.13 0.8 57.61
    Q118 -59.01 -1.67 -58.46
    Q119 -1.13 -57.66 -1.67
    Q120 -58.2 -0.71 -57.61
    Q125 22 mV -260 mV 83 mV

    Outputs, no transistor installed, tested at terminals:

    Emitters: all 48 to 50 mV
    Collectors: Q121, Q122 + 51.86V; Q123, Q124 -51.97 V
    Bases: Q121, Q122 +0.81 V; Q123, Q124 -0.71 V

    Voltage across D112 Zener: +/- 1.512 V depending on probe orientation.

    I'm going to install drivers ( Q113, Q114) and remeasure voltages.
  11. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    Base collector and emitter look OK at the output.

    Q114 now looks OK .

    Double check against schematic the bases with high voltages are OK that way.

    Also check your replacements have similar or higher voltage rating.

    Voltage across a 6V zener must be 6V if current is flowing and there are more than 6V between the leads. Check again when drivers are installed. Check the current path in the schematic.
    Last edited: Oct 1, 2017
  12. kotofei

    kotofei permanent gear evolution Subscriber

    Install drivers and fired the amp through DBT. The bulb went bright then dimmed.

    Voltages at Q113 was at E initially 2.92 V then went down to +1,5 V after few on-off power cycles ( I measured voltages for a few seconds after power up then turned amp off). Voltages at C are +53. 4V and B is +1.61V.

    The Q114 had E voltage -2.42V and -0.71V after few power on-off cycles. THe C voltage is -53.4V and B is -1.64V.

    At the outputs, still not installed, all E at 38 to 50 mV. Collectors are at + or - rail voltages same as before and B at Q121, 122 are +0.92 V and at Q1123, Q124 are +0.92V.

    Voltage across D112 Zener is 5.24V, across R146 is 5.21V.

    I'm going to install outputs and test with DBT again.
  13. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    Everything seems to go fine. Remember that OUTPUT transistors will start to conduct when the base is at more than 0.6 / 0.7 V. If you can, adjust the bias trimpots to have less than 0.7V at the output tr bases. So they are off, and you can start adjusting bias current.
  14. kotofei

    kotofei permanent gear evolution Subscriber

    Tried this and it didn't work. Rotation of VR104 and 105 completely counterclockwise didn't lower the voltage at the base of Q121 or Q122 lower than 0.83 V and Q123, 124 higher than -0.65 V.

    Perhaps new transistors I installed have higher hFE than old ones ( measured for some) and now some resistors should be changed?
  15. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    I see some feedback from D118 D121 back to Q115 Q118, so perhaps the circuit is not complete without the outputs.

    Try with the outputs, a brief test with DBT, then a brief test to quickly check bias (voltage at emitter resistors).

    Do your math before to see the value to expect, and power off immediately if too far from that range.
  16. kotofei

    kotofei permanent gear evolution Subscriber

    Installed outputs. Turned VR104 and VR105 into middle position.

    Powered thru DBT- bulb stays bright for 3-4 second and doesn't dim. Voltage between emitters of Q121 and Q123 is -190 mV and stays same regardless VR104 or VR105 rotation ( I turned amp off, rotate pots than turn on). Checked filtering capacitors- they don't charge.

    Apparently fixing of this amp is way too above my amp-fixing abilities. So I stop here and call it quit. After all, this hobby should be fun; last month while working on this amp it was anything but.

    Elnaldo and Slimecity, thank you for your help. Unfortunately it was not enough but it's not your fault.

    If anyone interested in getting this amp with the intention to fix it OR knows some simple amp design its chassis could be used for, please let me know.
    Last edited: Oct 3, 2017
  17. elnaldo

    elnaldo Addicted Member

    Buenos Aires GMT-3
    I still think in something simple we are not seeing.

    Just let it aside and if you don't sell it or give it away, try again next winter.

    I think the key is around that bias you can't adjust.
    slimecity likes this.
  18. vince_pgh

    vince_pgh AK Subscriber Subscriber

    agree that smth overlooked.
    Kotofei, I guess that might be a good idea to start the new thread on vegalab forum.
  19. slimecity

    slimecity Super Member

    New Zealand
    Its fun until it gets frustrating. Then consider it a challenge,

    From Elnaldo -
    There arent many things that keep the DBT bright. simple things like a dead short, blown output, shorted track, shorted rectifier are the most common causes.

    Best thing you can do is put it aside for a while and come back to it later. I was pulling my hair out for days once when i realised i had put a powersupply cap in backwards.....
  20. kotofei

    kotofei permanent gear evolution Subscriber

    Vegalab? What's it and why there?


Share This Page