Fault finding old amp that blows fuses

Discussion in 'General Electronics Chat' started by RobeeJ, Aug 25, 2013.

  1. RobeeJ

    Thread Starter New Member

    Mar 19, 2013
    24
    1
    Hi everyone!

    I'm trying to fault find in an old amp from the very early 80s, but I'm struggling a bit because of the fact that the fault instantly blows the two 4 amp fuses on the main power board for the amp.

    I've disconnected everything I can, tested as much as I can with a meter, and even tried to check the ESR of the capacitors using a function gen + scope (though the exact method of how to do this seems badly described everywhere I found, so I'm not entirely sure if I was doing it right).

    I think I could probably find the fault quite easily if I could just power the board and trace it (I have the schematic), but obviously I can't do that because the fuses instantly blow.

    Does anyone have some good advice on how best to proceed, without unsoldering every single component and testing it?

    TIA
     
  2. MrChips

    Moderator

    Oct 2, 2009
    12,452
    3,371
    You can begin by telling us the make and model of the amp.
    If there are no schematics to be found post some pictures, inside and outside.
     
  3. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    I personally would start by checking the power transistors, as they are most likely shorted.
     
  4. #12

    Expert

    Nov 30, 2010
    16,355
    6,852
    From the way you talk, it sounds like we can trust you to know what a bad filter capacitor looks like. Still, I've been fooled so thoroughly at times that...Let's not rule them out yet.
     
  5. LDC3

    Active Member

    Apr 27, 2013
    920
    160
    Since the fuses blow right away, I would say that the transformer windings on the primary side are shorted.
     
  6. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    This tells me that those are fuses on the secondary, which like I said before points to shorted power transistors or maybe the bridge rectifer.
     
  7. sheldons

    Well-Known Member

    Oct 26, 2011
    616
    101
    post the schematic so we have a better chance to go through fault finding with you.....
     
  8. RobeeJ

    Thread Starter New Member

    Mar 19, 2013
    24
    1
    Firstly thanks for all the replies!


    Sorry, I wasn't clear about what I was after. :( I'm not really looking for specific "check A and B" more a generic suggestions as to how one goes about fault finding a circuit without having to remove components individually, and yet the fuses blow instantly so you can't test it whilst live.


    That was one of my early thoughts, so I removed the two power amps early on, still blows. :/


    No leakage or doming, I did find one that looked different when feeding a 100Khz signal through it that looked like a short, alas this is on a different supply off its own winding and isn't blowing that fuse. :/

    I agree about not ruling them out, I've actually systematically begun removing any smoothing cap starting with the biggest downwards. But at some point I'll run out of fuses.


    Kubeek is correct, the fuses are all off the secondary. Being an old amp there are actually about 3 or 4 secondary windings all feeding their own supplies with their own bridges. Only the ones for the power amp are blowing.


    Anyway, I guess my real questions which I probably should have been clearer about at the start are...

    1. Has anyone tested the ESR of caps in circuit without an ESR meter, but with a function generator and scope?

    2. Can I wire up my bench supply (which outputs +/- 16V 1A max rather than the 33V this circuit wants) to parts of this circuit and use that to maybe let me trace for faults? Or is that likely to cook components just as much as whatever is causing the 4A fuses to blow?

    3. Could I use a function generator to feed 50Hz into the diode bridges if the voltage is high enough to get over the diode drops, and trace faults that way?
     
  9. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    Usually its one or more of the output power transistors - usually its easier to remove all of them and test for; *ANY LEAKAGE AT ALL!*, with the transistors out you could apply power and see if the fuses still blow, if they do its probably the rectifier.

    If the output transistors failed - there may have been a cause in the driver stage - check the driver transistors for leakage and any resistors for risen value/ O/C. There may also be a bias adjustment to set the current in the output transistors, too much current burns them out, too little causes crossover distortion - a bias preset pot can also have a loose wiper, with nasty consequences!
     
  10. ramancini8

    Member

    Jul 18, 2012
    442
    118
    I would troubleshoot this amp with an ohmmeter. Start at the fuse terminal; you should see a low resistance to ground. Disconnect each circuit/component in the circuit until the short goes away. Then chase the bad circuit. If this sounds like too much work throw the amp away because you are most likely wasting your and our time.
     
  11. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    I will also encourage you to post pictures, or make/model of this unit.
     
  12. nigelwright7557

    Senior Member

    May 10, 2008
    487
    71
    You need to isolate the problem.
    Start with just the transformer connected.
    Then add rectifier, then smoothing capacitors.
    Then check output transistors for shorts.
     
  13. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    They taught me 50 years ago:

    RULE #1: always assume the power supply is the problem

    RULE #2: always assume any old electrolytic capacitor is bad
     
    Metalmann likes this.
  14. RobeeJ

    Thread Starter New Member

    Mar 19, 2013
    24
    1
    Okay, I'm ready to post a schematic, but only because I've found some faulty components and am trying to understand something.

    Firstly, here's the schematic:
    http://images.robeesworld.com/sharp-sg-500h.png

    The area of the PSU which is causing the problem is on the right about midway up.

    F305 and F306 are the fuses that blow.

    All the caps are fine, as far as I can tell. Most of them I've tested out of circuit.

    The diodes at D305 and D306 are bad, they are shorts both ways. So obviously I need to replace those. I found that out by putting a 20V sine wave through it and tracing around with my oscilloscope, the moment I found AC there I knew something was up and started checking those diodes.

    But I couldn't quite understand how that would lead to blown fuses, so I started mapping the circuit out in LTSpice.

    Ignoring everything I'd disconnected, this the circuit:

    [​IMG]

    You can download the file here if you want to follow along at home:
    http://images.robeesworld.com/sharp.asc

    So according to the Spice simulation, which admittedly isn't going to be 100% accurate because I've chosen components as close to the originals as possible, on start up there is a huge current flow through C339 (yes I know current doesn't flow through a cap!) until that cap is charged.

    This spikes to around 23A which would cause a 4A fuse to blow at the least.

    [​IMG]

    Am I missing something here? I appreciate in the actual circuit it might not be able to draw that much, but unless I've made a mistake, would that not cause a fast-blow fuse to go?

    The service manual specifies slow blow fuses, would that be for this reason?

    And is this a design flaw? And should I look at limiting the current through D305/D306 (the ones that blew) because they are only spec'd for 1 amp? Would a short spike like this everytime the unit was powered on lead to them dying over time?

    (If you've spotted the voltage doubler in the circuit, well done! CNP302 requires about 104 volts)
     
  15. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    While you're at it, measure the current on power up through C344 and C345.
    (those two large flux capacitors).
     
    Last edited: Aug 31, 2013
  16. #12

    Expert

    Nov 30, 2010
    16,355
    6,852
    Unplug CNS904 "to player".
    Q307 is a 25 volt regulator that is fed by D305/306.
     
  17. RobeeJ

    Thread Starter New Member

    Mar 19, 2013
    24
    1
    LOL

    So busy staring at the tuna, I totally missed the shark.

    Is LTSpice accurately representing those current spikes? Or is the current limited on the mains side of the transformer so it's actually not as bad?

    I can't see anything on the schematic to limit it.
     
  18. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    The filter caps are basically a short (very low ohms) at power-up, and then begin to charge up. The transformer provides a small impedance to limit the in-rush current. LT spice is basing the current spike mainly on the parameters of the selected transformer parts. But, the simulation is similar to the actual circuit. That is why slo-blo fuses are specified.

    The in-rush can be limited, but the amp was working OK for decades, correct ? ;)

    How is the trouble-shooting coming along ?
     
    Last edited: Aug 31, 2013
  19. RobeeJ

    Thread Starter New Member

    Mar 19, 2013
    24
    1
    Thanks, that's what I thought.


    No, it's been blowing fuses for decades. :) Before we got hold of it, it had an original owner who emigrated, and a second owner a TV mast engineer who never had the time to fix it.

    I suspect though with the two faulty diodes replaced, it will be fine.

    The original owners just looked for obvious faults like bad caps, it's what I did. I only found the faulty diodes by throwing my signal generator at it. :/


    As far as I can tell, D305 and D306 are the only bad components. If both are shorting (which they are) then that's certainly a short between the + and - rails.

    D309-D312 are much bigger diodes, and I think they are rated for a much higher reverse voltage breakdown. So maybe over time with the turn on cycle, D305/306 just wore out.

    Seems like bad design to me. :/

    I feel like I should fix it!



    Thanks for your reply btw!
     
  20. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    I would suggest replacing the 2200uf cap also. If R330 is OK, then everything downstream is likely OK.
     
Loading...