Page 1 of 2 12 LastLast
Results 1 to 10 of 13

Thread: prob in H bridge

  1. #1
    Senior Member T-1000
    Join Date
    Jun 2005
    Location
    bangalore
    Posts
    325

    Default prob in H bridge

    hi,
    i was trying a h bridge config for driving a wiper motor...
    current rating is around 4 amps.

    its working but the upper arm mosfet is getting hot like anything....

    anyone has any idea how to solve this prob?
    if u thought u r the only mad person on the planet.......u got serious competition here

  2. #2
    Senior Member T-1000
    Join Date
    Jun 2004
    Location
    Bombay/Mumbai
    Posts
    498

    Default

    Hey,
    Looks like your using discrete components.

    If you used an H-bridge driver chip, you could put two in parallel.

    The reson why its getting so hot is because your motor is drawing too much current and is heating up.

    Use a Mosfet. with a higher current rating.

  3. #3
    Senior Member T-1000
    Join Date
    Jun 2005
    Location
    bangalore
    Posts
    325

    Default

    hey hi dude...

    the mos which i m using is rated for 49 amps cont. drain current...
    motors current rating is around 4 -6 amps...
    so tat shud not be a prob.

    connecting two drivers in parallel wont really be a soln.
    for 6 amps or may be a little more, i ll have to put around 2 or 3 lmd 18200 in parallel for one motor... l293d just cant be used(i ll need around 12 channels )...
    if u thought u r the only mad person on the planet.......u got serious competition here

  4. #4

    Default

    Is the input to the gate of the topside MOSFET low enough to turn it on fully ?

  5. #5
    Senior Member T-1000
    Join Date
    Jun 2005
    Location
    bangalore
    Posts
    325

    Default

    yes it is....
    infact the motor is running very well....
    even switching direction.....
    just the upper pmos is heating up....
    if u thought u r the only mad person on the planet.......u got serious competition here

  6. #6
    Senior Member Cyborg
    Join Date
    Sep 2004
    Posts
    162

    Default

    PMOS will generally be hotter than NMOS. (remember!: conductivty is lower..)

    Are you supplying PWM to the MOSFETs ?

    If you are just keeping them ON, and still they are hot, all you can do is add heatsink.

    But if you are supplying PWM (which you would, sooner or later) then you need to see if the rise/fall time of the signals on gate are not slow, which would cause switching loss, and excessive heating.

    Which chip is driving those Mosfets ?
    What is H-bridge supply Voltage ?

    hope the chip driving Mosfets has output voltage which swings upto the Bridge supply, so the upper Mosfets are totally Off.

    cheers!
    Vivek
    --
    " To get working ideas effectively, I try to fail as fast as I can. "
    Richard Feynman
    --
    http://www.youtube.com/watch?v=JqlawTD_9B0

  7. #7
    Senior Member T-1000
    Join Date
    Jun 2005
    Location
    bangalore
    Posts
    325

    Default

    Are you supplying PWM to the MOSFETs ?
    no, not as of now...

    If you are just keeping them ON, and still they are hot, all you can do is add heatsink.
    the heat rise is very sharp and heat sink ll simply not work there man...

    PMOS will generally be hotter than NMOS. (remember!: conductivty is lower..)
    yes, true but the prob persists even if i use a nmos there...

    What is H-bridge supply Voltage ?
    .
    +12v

    Which chip is driving those Mosfets ?
    not using any driver. am using a bjt on top to switch the pmos....
    have yu ever worked with any mos driver which can solve this prob.?
    if u thought u r the only mad person on the planet.......u got serious competition here

  8. #8
    Super Moderator T-1000 docel's Avatar
    Join Date
    Sep 2005
    Location
    BANGALORE
    Posts
    1,441
    Blog Entries
    3

    Default

    I hope you have installed pull up and pull down bias resisters at the Gates.
    The wiopers are notoriously lop-sided in current drain. Will sink more current in one direction than the other.
    docel
    ^Anything is possible..
    after YOU prove it!!

    http://bangalorerobotics.tripod.com

  9. #9
    Senior Member T-1000
    Join Date
    Jun 2005
    Location
    bangalore
    Posts
    325

    Default

    I hope you have installed pull up and pull down bias resisters at the Gates.
    yes i ve done tat for the upper pmos.... not though for the lower nmos{which is not getting heated anyway after i ve connected a resistor and a cap between its gate and gnd..}

    Will sink more current in one direction than the other.
    right now i m having this heating prob also wen i rigged up just one arm of the hbridge... so this may not be the reason for it...
    if u thought u r the only mad person on the planet.......u got serious competition here

  10. #10
    Senior Member Cyborg
    Join Date
    Sep 2004
    Posts
    162

    Default

    PMOS will generally be hotter than NMOS. (remember!: conductivty is lower..)
    yes, true but the prob persists even if i use a nmos there...
    That is very Incorrect way to use a Nch. Mosfet in High-side.
    In this case the Nch gets heated up because it does not get sufficient gate drive.

    To use Nch in high-side you need a boosted gate drive, (atleast Vth_mosfet + Vmotor_under_source)


    have yu ever worked with any mos driver which can solve this prob.?
    I usually use MC33883 for an H-Bridge of Nch Mosfets.
    (there are plenty others of the same kind).

    However, you do not need any complex gate driver chip because you plan to use Pch Mosfets at high-side.

    I suggest the following steps -

    1. Is the Pmos getting hot even while bridge is off ?
    if No, then we know the Mos can switch OFF completely.

    2. What is the Vds drop across Pmos when the bridge is On ?
    if this drop is > I*Rdson(max) means that the Pmos is switching On in Linear region, which i suspect is the main cause of such high heating.
    (So you can see the motor run, but Pmos may not be perfectly On).

    In step 2, needless to say, current has to be practically measured and Rdson value to be seen from datasheet.


    Are you using a PNP or NPN bjt to switch On Pmos ?

    I suggest using a totem pole pnp-npn bjt driver to drive the PMos as it is much more reliable & better way to drive gate.

    Moreover, later in case if you goto PWM based drive, the totem pole driver will generate better response from mosfets.

    I can suggest totem pole drive for the gate by following methods-

    1. CMOS inverter/buffer (CD4050 or CD4049, 04, 14)
    -put buffers in parallel to drive the gate ...if it works, reduce buffers to 3 and see if you can use 1 chip for both Pmos.
    - connect its supply to your bridge supply

    2. Op-amps with totem pole output. (LM324 is not good)
    - needs to be rail-to-rail output types

    3. totem pole using discrete npn and pnp. (simplest and best)


    If you still face problems, please post your complete circuit.

    After all, maybe just experimenting with the bias resistors in your present circuit could result in "working" circuit.

    cheers!
    Vivek
    --
    " To get working ideas effectively, I try to fail as fast as I can. "
    Richard Feynman
    --
    http://www.youtube.com/watch?v=JqlawTD_9B0

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •