Help needed for H-Bridge circuit

Discussion in 'The Projects Forum' started by xyz9915, May 4, 2008.

  1. xyz9915

    Thread Starter Member

    Feb 24, 2008
    Hi! The attached file is a schematic of 50 Hz H-Bridge which I want to use in the final stage of an inverter. In this respect please help me either any modification in the circuit is required? the power of inverter is 500 Watts.
  2. Audioguru

    New Member

    Dec 20, 2007
    Where can you buy a package of 240VDC/2.3A?
  3. Wendy


    Mar 24, 2008
    40 6V batteries in series? I've seen UPSes with something similar on an industrial scale, worked on it as a matter of fact. It was for emergency lighting of the factory floor. I approached this one with extreme caution.
  4. SgtWookie


    Jul 17, 2007
    I'm afraid it isn't going to work the way you have it wired.

    You don't have a high-side driver for the upper N-ch MOSFETS. The instant you get them turned ON, Vss is going to climb towards 220V. However, as soon as Vss gets to Vg (which is limited to 12v the way you have it set up) the MOSFET will turn OFF.

    In order to use N-ch MOSFETS on the high side, you need to use a charge pump or the like to get Vg at least 10v above the maximum Vss you're expecting to see. There are driver ICs made specifically for this purpose.

    Also, the IRF840s are capable of a continuous drain current of 8A. For a 500W 220V inverter, that means 2.3A on a <50% duty cycle per each leg of the H-bridge, or an average current < 1.15A. Even a single IRF840 has six times the capacity that you require. The problem here is that adding additional MOSFETS increases the gate drive current required.

    There's another problem; from the brief description of your 50Hz drive, I have to assume that Q and Q\ are complementary outputs from a gate or perhaps a flip-flop.
    Turn-On Delay Time: 14nS Rise Time: 23nS Tot:37nS
    Turn-Off Delay Time: 49nS Fall Time: 20nA Tot:69nS
    So, the MOSFETS take nearly twice as long to turn off as they do to turn on. If you ever wind up with both the high and low sides of an H-bridge on simultaneously, you wind up with a "shoot-through" condition, or a direct short across the power supply.

    You're turning the gates on and off with a single transistor that has a 4.7K resistor on the collector. While turn-off times may be pretty decent, the 4.7k resistor will cause the turn-on times to be quite slow in comparison. You may need to go to a totem-pole configuration.

    You're using 2.2k resistors on the bases, two bases in parallel per side, 1.1k in parallel. If you're trying to drive the transistors using CMOS, you're severely overloading those ICs.

    This whole circuit could be simplified a great deal if you used a driver IC that had a high-side floating driver in it. Some of the driver ICs have circuitry that prevent shoot-through.
  5. thingmaker3

    Retired Moderator

    May 16, 2005
    On-line, of course!

    If you are unable or unwilling to find a high-side low-side driver, try an internet search on "H bridge bootstraping." Half-bridge high-side low-side drivers can be had for a buck or two apiece, making the cost of driving the bridge two to four dollars.