Darlington and Field Effect Question!

Discussion in 'General Electronics Chat' started by NM2008, Apr 22, 2008.

  1. NM2008

    Thread Starter Senior Member

    Feb 9, 2008
    135
    0
    Hi,
    I have a diagram of a 5A H bridge, in it the transistors used are Tip122 and Tip127's Darlington's.

    What I am wondering is, why field effect transistors are not used here?
    Do field effect transistors switch high currents or are they generally used for low voltage/current high frequency switching?

    I am only in the early stages of learning about transistors, so hence the question.

    Thanks for your time,
    Regards NM
     
  2. Caveman

    Active Member

    Apr 15, 2008
    471
    0
    They will work, but sometimes you don't have enough voltage to fully turn them on. Look at some datasheets for FETs and look at the Vgs and rdson specification. Vgs tells when it comes on and rdson tells the resistance of a FET when it is on. Of course, this is a sliding scale where as the voltage from gate to source rises, the rdson will decrease, to a point.
     
  3. NM2008

    Thread Starter Senior Member

    Feb 9, 2008
    135
    0
    Caveman thanks for the info!

    But in practice would FET's be used in H bridges which switch heavy current say 5-20A?

    NM
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Following on what Caveman said, many common N-channel enhanced power MOSFETS need Vgs=10 to fully turn on, and Vgs=0 to fully turn off. The gates can have a fair amount of capacitance, which combined with the "Miller charge" ultimately limits how fast you can switch them on and off. Trying to run them at too high a frequency will result in them spending much of their duty cycle in the transition state, which will cause them to dissipate power and heat up.

    Darlingtons aren't particularly "zippy" speed-wise, but they're faster than MOSFETS.
     
  5. Caveman

    Active Member

    Apr 15, 2008
    471
    0
    You can however find many logic level FETs as well. But even they will turn on harder at higher voltages.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Power MOSFETS for lower voltages (say, below 200v) or IGBT's for higher voltages would be likely candidates, yes.

    BJTs have a minimum voltage drop across the emitter of around 0.63. MOSFETs don't have that fixed voltage drop when they're fully ON; the drop will depend on the load impedance vs RdsON.
     
  7. Caveman

    Active Member

    Apr 15, 2008
    471
    0
    The trick with handling a lot of current in an H bridge is making sure that the switching elements (FET, IGBT, BJT) do not drop a lot of power across themselves. The power across a component is:
    P = V*I, where V is the voltage across the switch.

    If you want to drop 20Amps across a switch, you need to make sure that its voltage is low. N-channel FETs with RdsOn < 10mOhms are available. This means that only 2mW is dropped at 20Amps! This is when the switch is on.
    When the switch is off, it should be completely off, so the voltage across the FET is high but there is no current, so very little power is consumed by the FET.

    The main issue is transition. During transition, you are increasing either V or I and decreasing the other. So more power is burned in the switch. This generates heat and takes from the load. The heat may damage the switch causing a failure or reduced lifetime.

    BJTs generally saturate at about 0.3V from collector to emitter. This is due to the fact that the collector and base form a junction and it begins to turn on a bit. That means that at 20Amps you will drop 6Watts! This is why FETs are used more in high power H-bridges.

    I'm not as familiar with IGBTs but I do know they handle heavy currents well and are sometimes used in place of FETs
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    FETs have some significant efficiency advantages over using BJT's, but the control circuit can be more complex.

    For one thing, N-ch devices are more efficient than P-ch devices; but you still need to get that Vds=10 to turn them on fully. This can be an inconvenience if you only have a single supply. Also, power MOSFETS seldom have identical turn-on and turn-off times. This means that timing between the upper side and the lower side is critical; you never want both of them to be ON at the same time, or you will have a "shoot-through" condition where there is a dead short across the power suppy rails.

    There are ICs available now that will take care of the +10 boost supply, along with preventing both upper and lower sides from being on simultaneously.
     
Loading...