Help with an H-bridge circuit

Discussion in 'The Projects Forum' started by Soulfly188, Apr 27, 2011.

  1. Soulfly188

    Thread Starter New Member

    Dec 28, 2010
    4
    0
    I designed this circuit for a TEM(peltier). I have measured the gate voltages of each Mosfet and they seem to be at the correct levels. Any help would be great.
     
  2. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    3,762
    924
    Sounds like you are doing fine. Anything specific we could help you with?
     
  3. Soulfly188

    Thread Starter New Member

    Dec 28, 2010
    4
    0
    It works well until I attach a load. I measure the voltages at the top gate at 7V(-5V of the source) and the bottom gate at 3.9V(+3.9V of source).
    The transistor controlling the bottom FET is dropping 1V from CE, but I don't think that should matter since the FET is a voltage controlled device.
     
  4. Soulfly188

    Thread Starter New Member

    Dec 28, 2010
    4
    0
    I adjusted some of the resistors and according to the simulator it can handle more of a load. I have yet to put it in practice. I will post the results.
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Referring to this schematic:
    [​IMG]
    You have not specified what you are using for Q1 or Q2 besides showing that they are P-ch MOSFETs, however it is likely that they are not logic-level MOSFETs - so in order for them to be considered fully turned on, Vgs (voltage on the gate, referenced to the source terminal) must be -10v. Your upper MOSFET's Vgs is only -5v, which is not turning the MOSFET on fully.

    The IRFZ44N MOSFETs are standard-level N-ch MOSFETs, which require a Vgs of 10v in order to be considered fully turned on.
    You show B5 as being 5v. Q8 is an emitter follower. Thus, the emitter of Q8 will never be higher than about V(B5) - 0.63v, or 4.37v; you are barely above the Vgs(th) (gate threshold voltage rage), where the MOSFET might only be conducting at around 250uA and Vds=Vgs. You must look at the Id specification to determine Rds(on) at the specified Vgs; in this case Vgs is specified at 10v.

    You have further complicated the situation by the voltage divider made up of R7, Q8, and R8 in parallel with R33. This will result in very slow gate charge/discharge times, resulting in the MOSFETs operating in the linear region where they dissipate power as heat.
     
  6. Soulfly188

    Thread Starter New Member

    Dec 28, 2010
    4
    0
    I designed a transistor gate for the two inputs. I think I am using unnecessary parts, but the bridge is functioning. It switches 5.5A without any problem. Let me know if you have any ideas on simplifying the inputs.
     
Loading...