Freewheel diodes

8 posts / 0 new
Last post
Patrick
Offline
Last seen: 9 years 5 months ago
Joined: Tuesday, November 21, 2006 - 17:31
Points: 55
Freewheel diodes

I'd like to drive a standard DC motor (Perm or Etek) with a controller that can't tolerate CEMF. I'm thinking I can solve the problem by putting a freewheel, or snubber, diode across the outputs of the MOSFETs. I haven't measured the back EMF of these motors, but I typically see regen currents of up to 50 amps max. So I have two questions:

How best to measure CEMF? I do have access to some test bikes. Will I see the CEMF using a non-regen controller like the Alltrax? I admit to some laziness here - I should measure first, ask questions later.

What parameters should I look for in the diode? Clearly voltage and current capabilties - anything else? Switching times? I don't think heat dissipation will be much of an issue, as regen doesn't happen for very long. I suppose coming down a long steep hill might be a problem, and I should plan for that.

Any thoughts?

Thanks,

Patrick

Fechter
Fechter's picture
Offline
Last seen: 15 years 2 months ago
Joined: Friday, November 17, 2006 - 07:01
Points: 199
I'm not sure what you're

I'm not sure what you're asking. Why would a controller not tolerate back emf?

The normal (brushed) configuration has the freewheel diode across the motor reverse biased. The FETs usually also have an internal diode across the switch. When the FET switch opens, the magnetic field collapses in the armature generating a voltage that is conducted by the freewheel diode to sustain the current in the armature. Without a freewheel diode, the collapsing field generates a deadly voltage spike that can blow the FETs.

If you want to prevent regen, you could put a diode in series between the controller and the motor, and you would need another diode across the motor for the freewheel. This adds extra loss due to the voltage drop in the series diode.

With a non-regen controller, like an Alltrax, you would never get regen unless you were going down a really steep hill where the rolling speed exceeds the no-load speed of the motor.

Freewheel diodes are usually rated about the same as the FETs for voltage and current. Heat dissipation can be significant.

Patrick
Offline
Last seen: 9 years 5 months ago
Joined: Tuesday, November 21, 2006 - 17:31
Points: 55
Old Alltrax

Actually, it is an Alltrax - old style. The story I've heard is that these earlier models, meant for series wound motors, get blown by PMDC motors. I'm just guessing that they don't have freewheel diodes, or if they do, they aren't sufficient to deal with the back emf. So I thought I'd outboard one, reverse biased, across the controller output. I'm not getting regen anyway, so that's not an issue.

So it's an experiment, if not a solution. I'm not sure where I can find diodes that can handle 50-100 amps - maybe I'll end up with more than one. Any ideas on sourcing?

Of course I need to measure. We'll see what happens - I'm just going to capture (reverse) volts and amps on a 48V Etek with a current Alltrax. Presumably it will have the freewheel, and will sink whatever is thrown at it. I'll post results when I have them.

Patrick

Fechter
Fechter's picture
Offline
Last seen: 15 years 2 months ago
Joined: Friday, November 17, 2006 - 07:01
Points: 199
DigiKey

Here's one: 43CTQ100-1PBF 100v, 40amp. These can be ganged up in parallel to increase current.
http://rocky.digikey.com/scripts/ProductInfo.dll?Site=US&V=21&M=43CTQ100-1PBF

I've heard about the earlier Alltrax controllers blowing up and there was something they did to change the design that apparently corrected the problem. The failures I heard about were FET failures, not freewheel diode failures (I suppose one could lead to another). I suspected the main capacitors were under sized. Somebody at Alltrax knows exactly what this is.

Patrick
Offline
Last seen: 9 years 5 months ago
Joined: Tuesday, November 21, 2006 - 17:31
Points: 55
Yep

Well, it seemed like a good idea. I did find an 80A, 400V diode. But your point was well taken, and I called the designer at Alltrax before i spent the $10. He kinda hemmed and hawed, and said it would work, but that it would need to go in series with the diode inside the controller - so no go, as they're potted.

But it was worth asking, and thanks for your help!

Patrick

Fechter
Fechter's picture
Offline
Last seen: 15 years 2 months ago
Joined: Friday, November 17, 2006 - 07:01
Points: 199
In series?

Hmmm... sounds like what he's saying is the stock freewheel diode needs a higher voltage rating.
Let me think about that. I'm not sure if there's a way to help that from the outside.
Seems like a stiffer main capacitor would help the most.
BTW, I have one of those early Alltrax controllers in my garage that has the weakness.

Patrick
Offline
Last seen: 9 years 5 months ago
Joined: Tuesday, November 21, 2006 - 17:31
Points: 55
Yes, that puzzled me too

He wasn't hemming so much as thinking out load - you know the kind of thing "Well, the diodes are still connected to the bus bars, and . . . ." but he was perfectly congenial about it, no hiding propiatary stuff or anything like that. But I didn't get the series bit - but then i don't really know anything about this stuff. We've got a bunch of those taking up shelf space, who knows if they're blown or what, i just thought it would be good to use them if possible. They aren't even that old - 2004 vintage.

Patrick

Fechter
Fechter's picture
Offline
Last seen: 15 years 2 months ago
Joined: Friday, November 17, 2006 - 07:01
Points: 199
Epoxy

I've also heard of people running them for a long time with no problems.
I guess it would be nice to know exactly what Alltrax changed in the design.
Epoxy potting isn't the end of the world, just a pain in the butt to dig out.

I have a test fixure I use to check controllers. I have a bench power supply, a throttle pot mounted in a small box (with no spring) and a small 24v motor. The motor only draws about 2 amps, so my bench power supply can handle it. If there's something blown in the controller, the current limiting on the bench supply prevents major smoke.

Log in or register to post comments


Who's online

There are currently 0 users online.

Who's new

  • Juli76
  • xovacharging
  • stuuno
  • marce002
  • Heiwarsot

Support V is for Voltage