Motor controller with full bridge

Discussion in 'The Projects Forum' started by mejlby, Oct 8, 2014.

  1. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1
    Hello everyone

    I have previously made a motor controller with a TL494 and uses a half bridge L6384. it can handle 40Amp without mosfet get hot. I do not use heatsink.

    Now I will try to make a motor controller with a full bridge so that I can run the motor forward and reverse I have found the attached circuit. Do I understand it right, the PWM signal is either set to high or low so that the motor changes direction ?

    I I hope someone can help me with this question

    best regards
    Mejlby
     
  2. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,507
    2,367
    Yes, with a full bridge shown, the top semi on one side will conduct with the lower semi on the opposite side, applies to each direction selected.
    Max.
     
  3. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    Why duplicate the circuit? Can you drive Q1 & Q4 together from the same gate signal, and do the same with Q2 & Q3?
     
  4. RamaD

    Active Member

    Dec 4, 2009
    254
    33
    The charging path for the bootstrap cap needs to be checked.
    The Left High side bootstrap cap gets charged during the left low side MOSFET turn-on and vice versa. During any one direction movement, the upper turning on mosfet's driver's bootstrap cap loses charge driving the gate. During this time, only the lower mosfet of the opposite side turns on. To be a little more clear, if
    Left Top MOSFET on and Right Bottom MOSFET on - Right Top MOSFET's bootstrap cap charges and let top mosfet's bootstrap is not charged and is losing charge driving the gate!
    It may be necessary to provide 2 isolated supplies for Vb. Or, Turn on both the low side MOSFETs during the off time.
    Sorry, if I have not communicated well.
     
  5. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1
    Thank you Max

    Then I have understand it correctly and now I will try too make a new motor controller.

    Attached circuit shows how I make the half-bridge controller. Do you think I can use the same Mosfet driver L6384 to a full bridge if i use 2 x L6384 ? It has worked 100% correct and as mentioned it can handle at least 40Amp without heatsink. I use it for an electric go-kart that I made for my daughter. I use a relay to reverse.


    ( see on youtupe: )

    Regards Mejlby
     
    Last edited: Oct 8, 2014
    MrChips likes this.
  6. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1

    Hi RamaD

    Sorry i don´t understand: " It may be necessary to provide 2 isolated supplies for Vb. Or, Turn on both the low side MOSFETs during the off time."
    do you have the opportunity to draw it on the circuit so I can see what you mean?
     
  7. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    Tonny, in your half-bridge circuit, why do you need the lower pair of MOSFETs?
     
  8. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1
    Hi MrChips

    This link explain the whole theory behind the moefet driver much better than I can explain it. I have make some experimenting with other forms of driver - it does not work - mosfet gets so hot that they require a very large cooling or simply burn out. This circuit I made WORKS 100%.- no problem with hot mosfet

    the reason I use a L6384 mosfet driver is this driver has a Intregrated bootstrap diode and can replace an external bootstrap diode

    http://www.microsofttranslator.com/...1/using-high-low-side-driver-ir2110-with.html
     
    Last edited: Oct 8, 2014
  9. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    I do not doubt that your circuit works 100%.

    What happens if you disconnect the lower MOSFETs?

    I suspect the author of the article does not fully understand the operation of high-side, low-side drivers.
     
  10. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1


    Now you made me curious !!!

    Tomorrow I will try to remove the 2 lower Mosfet and see what happens

    I think the bootstrap capacitor vill not discharged and can not feed the upper Mosfet

    What do you think ?
     
  11. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,507
    2,367
  12. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1
    Hi MrChips

    Now I already have an answer for you!

    In my workshop, I had a prototype of the controller and the mosfet is mounted in some terminals. I removed the 2 lower Mosfet and immediately became the 2 upper MOSFET very hot. When I mounted the two lower Mosfet again and the controller was broken.

    Tomorrow I have to find the error. I think the upper mosfet is burnt of or the capacitor / diode is burnt of ...or the driver L6384 is broken

    ( shit - I should never have listened to you)

    So now you have the answer why the 2 lower Mosfet is mounted

    Let me hear your answar

    correctly Max .... mr MrChips should have read tahmids blog !!
     
    Last edited: Oct 8, 2014
  13. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    Why did the MOSFETs get hot and die?
    What was the load on your test circuit?
     
  14. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    I think I follow what you are saying. I have been reading up on how the MOSFET drivers operate. Without the lower MOSFET, the charge on the bootstrap cap blows the gate on the upper MOSFET. Is that correct?

    So I eat humble pie tonight!:(
     
    Last edited: Oct 8, 2014
  15. mejlby

    Thread Starter New Member

    Apr 26, 2013
    28
    1
    Yes you are absolutely right
    I do not know why I tried it - in fact I already knew the outcome.
    I only managed to overload the controller with about 2-amp and then the damage was done
    Tomorrow I will fix it again. :mad:
     
  16. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    The best guide I have located that describes the problem is:

    Fairchild Application Note AN-6076

    As I understand it, the fault was caused by the omission of the freewheeling diode that is usually a part of the MOSFET.
     
  17. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,281
    1,232
    There are no clamp diodes in either circuit. In the half bridge the bottom FETs perform that function. I think in the full bridge you will need some, but I don't know your whole timing diagram.
     
  18. MrChips

    Moderator

    Oct 2, 2009
    12,418
    3,355
    The clamp diodes are in the MOSFET package. When the lower MOSFETs were removed the clamp diodes went with them.
     
  19. Sensacell

    Well-Known Member

    Jun 19, 2012
    1,127
    266
    The design needs the lower mosfets only to charge the bootstrap capacitor- this cap provides the higher voltage to drive the high side mosfets.
    Without the low side, the upper mosfets do not see enough gate voltage to turn on all the way.

    The low side also shorts the back EMF of the motor! not a good idea!

    This design is very inappropriate for driving a motor in one direction, a single low side switch would be much better and simpler.
     
  20. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,281
    1,232
    No, it is a good circuit. It provides regen braking which is probably good for a little girl on a scooter.
    The cap would charge thru the motor without it, but then you need a big diode.
    http://www.4qd.co.uk/fea/half.html
     
Loading...