2

We got a bulk of stepper motors, you can see specs on the following image:

enter image description here

I'm doing some work on my school laboratory, and they want me to test this motors to see which ones work and which ones doesn't. As of now, I have used a Arduino micro and an Easy Driver(Allegro A3967), just STP and DIR signals, after 10 motors, the driver died, I found a short between A- signal and V+.

Reading the complete datasheet I have now realized that we are not suppose to "hot swap" the motors, we have to turn off the unit and swap, but that is slow and does not work for me, so now I connected a Relay to an arduino output (Darlington Driver), that Relay turns ON and OFF the easy driver board, (Opening and closing V+ Line), after 60 motors, again the driver died, again short from 1 output terminal to V+.

My next move is to put some schotky diodes to clip transient voltages on the A+, A-, B+ and B- terminals and see how the board responds.

My question is: Do you have any idea how to make a Hot swappable driver for a stepper motor?, Do you know any IC that can do that? Do I have any more options? is this really possible?

AKR
  • 2,273
  • 4
  • 22
  • 34
user2982010
  • 67
  • 3
  • If you don't mind a bit of heat. Permanently connect a load resistor across each coil terminal (and driver supply) so that transients are dampened. Select a bigger controller if you can but even if not size the resistor to waste half of the controller current and then use the other half to test the motors, you may find that disabling the driver before you switch off is a good idea. you can have the controller hold a relay closed when powered on (you a push button to bootstrap it) and then stop motor and disable driver and release relay, moments later safe to swap motor. – KalleMP Oct 27 '15 at 19:48

2 Answers2

2

Since the drive is bipolar, single diodes won't work. What you should use is a pair of series-opposed zeners across each of the coils with the zener voltage a little higher than the drive voltage across the coils, like this:

enter image description here

EM Fields
  • 17,377
  • 2
  • 19
  • 23
  • The emf induced by disconnecting one side of the coil will not be a potential difference between the two sides of the zeners, though, will it? As one side of the Coil has no connection anymore to the driver, the coil will just present a floating potential to the still-connected side, thus not trigger the zeners - or where am i going wrong? – loonquawl Oct 28 '15 at 08:07
  • @loonquawl: Part 1. At the instant of disconnection, the current through the coil, from the driver, will end. When that happens, the magnetic field about the coil will quickly collapse, which will induce a voltage across the coil opposite in polarity from that supplied by the driver, and with a magnitude equal to \$ E = L\frac{dI}{dt} \$, so it's possible to build up a very high voltage in a very short time. – EM Fields Oct 28 '15 at 09:38
  • @loonquawl: Part 2. That quickly rising voltage will continue to increase until it arcs across the gap created when the motor is disconnected from the driver, the result being that a high voltage will propagate through the plasma which is the arc and into the driver, destroying it. So, putting back-to-back Zeners across the driver will clamp the high voltage - regardless of its polarity - to one Zener forward voltage plus one Zener reverse voltage, saving the driver. – EM Fields Oct 28 '15 at 09:48
0

The windings in a motor, when disconnected work just about the same way as an inductance in an electric fence high-voltage generator. That is to say, disconnecting one end of a powered inductance will generate a sharp voltage spike. You'll probably not be fast enough to diconnect simultaneously, but if you write the test-program so that the windings are unpowered & the motor unmoving at certain times, the swap should be no problem NOTE: Unmoving motor does not mean unpowered coils, as steppers can be in 'hold' mode. So make sure to have your program put the driver into 'freewheel'. Note that hot-adding a motor is theoretically possible, but even a slight tremor will connect-disconnect-connect, and the diconnect will still work as a source of high voltage, so have the unpowered state extend to adding the next motor too.

loonquawl
  • 220
  • 1
  • 8