ir2110 test for driving h bridge

Discussion in 'The Projects Forum' started by etiquoe, Jul 27, 2010.

  1. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    hi,
    i'm gonna use ir2110 for driving my bldc h-bridge.
    so for testing, i connected it with mosfet like this:

    [​IMG]

    i fed the Lin pin with 60Hz noninverting PWM signal, and 60Hz inverting PWM signal for Hin pin.

    for bootstrap capacitor, i use 0.1uF ceramic and 47uF electrolytic in parallel.

    but both of the mosfets got pretty warm in few seconds, and then damaged.

    need solution :confused:

    anyone help please...
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    It sounds like you are not providing any "dead time".

    You must provide a period of time where both MOSFETs are in the OFF state. If you fail to do so, you will wind up with the dreaded "shoot-through" condition, where both MOSFETs are conducting simultaneously, which creates a dead short across the power supply. This means that you cannot simply use an inverter aka "NOT" gate to provide opposing signals.

    Unfortunately, you cannot use the SD input to achieve this, as it shuts the driver down completely.
     
    etiquoe likes this.
  3. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    can i provide the dead time from hardware/circuit configuration, instead of providing it from software?
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I suppose you could, but if you inserted the dead time via software, you would keep your parts count low.

    One item I forgot to mention in my initial reply, is you must initially charge the boost cap by making certain the upper MOSFET is off, then turn ON the lower MOSFET for a period of time. If your load is non-inductive, you will also need to perform this recharge periodically, as the cap charge will bleed off rather quickly when the upper MOSFET is turned on.
     
    Last edited: Jul 27, 2010
    etiquoe likes this.
  5. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    thanks SgtWookie,
    i increased the gate resistor value and connected a diode parallel to the gate resistor to provide additional dead time as an-978 mentions.
    it worked
    :D
     
  6. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    i have a question, again.
    when Hin and Lin logic are low, high side mosfet gate voltage is about 3.9v. the output voltage of the hbridge circuit is about 3.9v too. it should be 0v. shouldn't it?
     
  7. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    i have a question, again.
    when Hin and Lin logic are low, high side mosfet gate voltage is about 3.5v, the output voltage of the hbridge circuit is about 3.5v too, and 0v in fast switching. it should be 12v (half of VDD). shouldn't it?
     
  8. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    I have a question on your circuit. Why do you have the source of the upper mosfet connected to the drain of the lower mosfet?

    I've looked at a lot of IR2110 circuits, trying to learn how to use the IR2110, and have never seen this configuration. Am I missing something on their use?

    Also on the first page of the data sheet http://www.irf.com/product-info/datasheets/data/ir2110.pdf they show a cap betwwen the drain on the upper mosfet and ground.

    [​IMG]
     
  9. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    i'm using this testing circuit because i'm gonna use ir2110 for my bldc motor driver bridge. i got this configuration from many application notes, one of them is AN-978 (page 26)

    the source of upper mosfet is connected to the drain of lower mosfet.
    Vs and Rload in my circuit also connected to that point.
     
    Last edited: Aug 20, 2010
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, it should be 0v.

    Did you charge the boost cap by turning the high-side MOSFET off, and turning the low-side MOSFET on for a period of time?

    If your load is non-inductive, you will have to repeat that sequence periodically. If it is inductive (motor, relay, etc.) the inductive kick you get when you turn off the high-side MOSFET will keep the boost cap charged.
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    shortbus,
    There should be a dot where the wire to Rload crosses the MOSFET drain to source connection. That was an omission in the original datasheet that our OP copied from.

    It's a bit easier to visualize when the complete H-bridge is shown. Just seeing a half-H bridge often doesn't make much sense, particularly if you're a n00b.

    It is typical and customary to have a bypass cap across the entire bridge.

    Have a look at the attached .pdf. It starts off with the IR2110 datasheet, and then an application note for the Harris HIP2500, which is quite similar, along with more handwritten notes (not mine). It will help you a great deal to understand the IR2110 better.
     
    Last edited: Aug 20, 2010
    shortbus likes this.
  12. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    Sgt., didn't mean to high jack this post, just trying to learn. Thanks for the link and answer.
     
  13. etiquoe

    Thread Starter New Member

    Jul 11, 2010
    27
    0
    Yes

    the load for this testing circuit is noninductive

    in fast switching frequency, it is 0v indeed.
    but in low switching frequency (<10 Hz), it's about 3.5
     
Loading...