diodes on the output of IRLZ44N?

Discussion in 'General Electronics Chat' started by tracecom, Aug 17, 2011.

  1. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,871
    1,393
    I built a unipolar stepper motor driver circuit using 4 IRLZ44N MOSFETS to switch the 12vdc to the motor windings. The circuit is working fine, but I just realized that I didn't include any counter-emf diodes on the outputs. Should I add them?
     
  2. debjit625

    Well-Known Member

    Apr 17, 2010
    790
    186
    Their is already an internal clamping diode inside IRFZ44N ,but don't be too much dependent on it rather use an external freewheeling (flyback) diode protection like I use IN4007 in most case.

    Here is a usefull application note from IRF may some of you get help from this...
    http://www.irf.com/technical-info/appnotes/an-936.pdf

    Good Luck
     
    tracecom likes this.
  3. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    I'm missing something obvious-
    The diode in a MOSFET (really a transistor with a shorted junction) will not conduct any current from drain to source because of its orientation except in reverse breakdown.

    When an inductor is switched off, current will continue to try and flow towards the drain and the back biased cathode of the body diode. How is back emf being suppressed? You cannot push current back through the diode's cathode.

    Figure 8 in AN-936 shows a zener across D-S. That makes sense and works well.
     
  4. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    In a H-bridge with an inductor between the legs they conduct, yes. Let's say the upper left and the lower right switch is ON. On the inductor you have + on the left and - on the right. Turn off the switches , the inductor generates an inverted voltage that can be higher than the DC bus. Now all switches are off but since you'd have a (big) positive voltage on the right of the inductor and negative on the left the lower left mosfet diode and upper right mosfet diode will conduct. They will be forward biased.
     
    Last edited: Aug 17, 2011
  5. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    I forgot the H bridge part. I should of realized that when he said 4 MOSFET's. Kinda obvious.
     
  6. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,871
    1,393
    I don't think it's an H-bridge; it's four individual MOSFETs, each on a separate microcontroller output, and each controlling a separate winding of the stepper. I would post a schematic, but I don't have one since I sort of based the design on previous projects I have done.

    I apologize for the confusion.

    ETA: See post 9 for the schematic.
     
    Last edited: Aug 18, 2011
  7. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    Well in this case I think, yes, you should add diodes on the windings, since the mosfet body diode will never conduct. (I assume something like positive voltage --> motor winding --> drain of an N-FET --> source goes to GND) diode on the motor winding pointing to positive.
     
    Last edited: Aug 18, 2011
  8. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    Diodes are good as long as you are not in a hurry. They are the slowest of all back emf suppressors because of their 0.7v clamp.
     
  9. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,871
    1,393
    Here is the schematic with the diodes shown where I think they go. I drew the schematic after building the circuit, so the schematic may have errors.

    I have not added the diodes yet, so if there are schematic errors or better options, I would be glad to hear them. Thanks.
     
  10. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    If the MOSFETs aren't turning on fast enough and warming up, you could lower the 4 gate resistors to around 33 ohms or so, that will help. But if its working fine the way it is, I wouldn't bother.

    The diodes are where they should be.

    What pgm did you use to generate the schematic?
     
  11. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,871
    1,393
    I assume the MOSFETs are turning on fast enough, although I do have to put a 1 msec pause after each high command from the PICAXE. The MOSFETs do not get hot at all.

    What would be better than the 1N4001 diodes? Zeners? What voltage? Same polarity?

    DipTrace. Thanks.
     
    Last edited: Aug 18, 2011
  12. debjit625

    Well-Known Member

    Apr 17, 2010
    790
    186
    IN4007,I already said that on my first post anyway that gives you :
    Maximum repetitive peak reverse voltage : 1000V
    Maximum RMS voltage : 700V
    Maximum DC blocking voltage : 1000V

    Good Luck​
     
  13. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    You won't need higher voltage diodes. They are either reverse biased , when the mosfet is conducting (reverse voltage is DC-bus voltage) or forward biased right after the mosfet opened.

    You should be more worried about their current carrying capability; it depends on the motor how much energy these diodes have to dissipate. Do some tests to see if they get hot.

    If one of these diodes fails THEN you may worry about the body diode of the mosfet whose maximum breakdown voltage may be exceeded.
     
    tracecom likes this.
Loading...