Back EMF Diodes

Discussion in 'General Electronics Chat' started by bytraper, Oct 20, 2010.

  1. bytraper

    Thread Starter Active Member

    Sep 28, 2010
    Hi Guys and Girls,

    I have a problem with my DC Speed controller. To give bit of background, Ive been building high current DC controllers for 24v 750w e Bike motors (60A).

    I haven't had a problem with any of them up until one customer with a Bosch 750w 24v motor. Every time he connected it to the speed control, it will work fine with a small load, but as soon as he would try and ride around on his bike it would fry the controller within seconds.

    So after replacing the controller 3 times, he sent the motor to me.

    I played with the motor on the controller and it ran fine, but when i stalled the motor and began to turn the controller speed up, it quickly kills the diode.

    And once the suppression diode is dead it literally fries the controller.

    The controller uses a 10A SR1060 Schottky barrier diode because I've had no problems with these before and they are quite high current for these types of applications.

    Anyway, to try and solve the problem I then ran a 40epf06pbf (40a) diode on the motor to suppress the back emf from there, and again i tested, and the sr1060 blew (though slower this time) and the bigger diode got hot to the point where it began to smoke.

    Somethings wrong here, the motor should not be putting this much current back to the suppression diodes.

    If i run the motor without load the diodes don't get hot at all and it runs fine, it always happens as soon as there is a load on the motor (and only this particular motor)

    Is it possible the motor has shorts between the internal coils? What else could it be causing this much back current?

    Hope this is understandable, and hope maybe I can get some help or advice!

    Just to add to this, if i run the motor at full speed and begin to stall it from there, theres no problem, its only from when the motor is under load as it is starting up.

  2. marshallf3

    Well-Known Member

    Jul 26, 2010
    An SR1060 is only rated for 42V forward and 60V peak inverse. Back EMF can be as high as 10 times the voltage applied at the time the current is interrupted.

    It's not so much the switching time or a low voltage drop you're looking for, it's the peak current and voltage the diode can handle.

    Look for a fast or ultra fast recovery rectifier that's capable of high current peaks but also that can handle at least 240V.

    This may be a good possibility
  3. bytraper

    Thread Starter Active Member

    Sep 28, 2010
    Any idea why it would just be this motor?

    All the other motors are fine and work with no problem. i thought it may have something to do with the windings breaking down :/
  4. SgtWookie


    Jul 17, 2007
    If the windings were breaking down, the smoke would come from the motor, not the diodes.

    You do need ultrafast recovery diodes, preferably with a very low Vf - and a means to keep them cool.

    A more modern approach would be akin to synchronous rectification; using high and low MOSFETs as ideal switches. However, this would require a re-design of the controller to incorporate the necessary changes.
  5. marshallf3

    Well-Known Member

    Jul 26, 2010
    Actually the forward voltage only needs to be low enough to protect the associated circuitry, that's why zeners are sometimes used. Fast yes, but not instant, there's a slight delay before the circuit realizes it's become open ended and the reverse EMF reaches full potential.

    I used to think the same way too until I did some more reading on the subject.

    As to that motor, it probably has a more substantial amount of magnetic core giving it the potential to zap out a lot more when the current is cut.
    bytraper likes this.
  6. ifixit

    AAC Fanatic!

    Nov 20, 2008
    Hi bytraper,

    Is it possible this motor has a problem with the brushes, or the contact pressure with the commutator that causes excessive arcing? The spikes produced are very, very fast and a diode can't turn on fast enough to suppress them. Check the brushes; that they are is good shape, and move freely in the holder. Commutator should be clean and shinny.

    A 0.1uF to 1uF 600V cap similar to that used on older car ignition systems should be in place across the armature to slow down, and suppress the arcing caused by brush-commutator action.

  7. marshallf3

    Well-Known Member

    Jul 26, 2010
    On more question - when he cuts the power is the motor still being driven by the momentum of the bike? If so, it's probably acting as a generator during that time.

    I'll also agree with the idea of putting the cap in there.
  8. bytraper

    Thread Starter Active Member

    Sep 28, 2010
    Good point, I think I'll try the cap first and see if that makes any difference.

    Yes, when he cuts power I'd assume its actibe as a generator, brushes look ok though armature is a little dirty