DC Protection Circuit

Discussion in 'The Projects Forum' started by Skeebopstop, Apr 18, 2010.

  1. Skeebopstop

    Thread Starter Active Member

    Jan 9, 2009
    358
    3
    Dear All,

    I have designed the attached DC protection circuit for a Class D audio amplifier to disconnect the speaker via a relay when either +33V or -33V is found to be stuck on the speaker output.

    I was testing it last night and the negative protection worked great but the positive on the other hand did nothing at first, but after holding it for 2-3 seconds damaged the transistor devices and latched the relay open. This condition now persists.

    I then came back to my simulations, noting that the transistors are arbitrarily chosen, and I'm noticing a similar behaviour but I can't get my head around it. I think I'm just missing something silly and hoping an independent reviewer can knock some sense into me.

    Attached is the schematic (LTSpice IV file included) and the trace results. Even after the 33V DC on DC-IN is cleared back to 0V, it holds up?!?

    Many thanks in advance.

    James
     
  2. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi James,
    1. When Q1 turns on via current from R4 it turns on Q2 also.
    2. The collector of Q2 is also connected to the collector of Q3, which gets pulled positive also.
    3. Since Q3 is PNP, its C-B junction is now reversed bias and that pulls the Q1 base up positive also. This forms an un-intentional latch circuit (like an SCR).
    4. The Q1 base won't un-conduct because R4 can no longer supply enough currrent. Since Q1, Q2 and Q3 are shorted across the +33V supply... they burn up:eek:.
    Regards,
    Ifixit
     
  3. Skeebopstop

    Thread Starter Active Member

    Jan 9, 2009
    358
    3
    Thanks FixIt, you've fixed another one!

    I've made a modification which seems to be working in my simulation now. Just posting it so maybe you might have a one over to ensure I don't etch another board and have it fizzle like the last one :)

    Cheers,

    James
     
    • DCP.JPG
      DCP.JPG
      File size:
      122.9 KB
      Views:
      47
  4. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi James,

    Your circuit seems to basically work now, however, Q4 doesn’t saturate during the negative over-voltage detection phase. This is not a good design practice. The trouble is Q3 can’t drive Q4 base positive enough, because of the drop across R3 & R9.

    You need to change the circuit so that Q4 is fully turned on when the relay is activated. Q4 could overheat and/or the relay might not get enough current to pull-in properly.

    A second relay and PNP driver could be used to solve this problem. You could also change the circuit design and use more transistors. See if you can come up with a solution using LTspice first.




    Questions:
    • What is the relays DC resistance? Equivalent to R8 at 500ohms?
    • What is the relays recommended operating voltage?
    • Do you want to use an SCR in place of Q4 to hold the relay on until manually reset by a push button, or the power is removed? It may not be safe to re-connect speakers back on to an active power amp. Perhaps one of our audio experts will comment on this.
    Tip: Have Fun, and always keep a fire extinguisher handy in your workshop.:D
    Ifixit
     
Loading...