Flyback Diodes

[CalgaryPT]

Hi all.

I am building an RC Lawnmower from scratch. It's almost finished and I am preparing to test. The motors are two 24VDC wheelchair motors. Traditionally people who build these remove the mag brakes from the motors. The motors are 250W each, and require 24VDC @ 30mA to each magbrake to disengage it. The reason they remove them (I suspect) is (1) they aren't usually needed and (2) to save on power consumption.

However, I am keeping mine installed because we have some hills in our lawn and a runaway mower seems like a bad idea. Anyways, I have an email into the supplier of the motors (SuperDroid Robots) to see if the brakes have built in flyback diodes or not.

If not, I will add my own. But I am wondering...when doing this should I put a diode across EACH motor or will just one across both suffice? The mag brakes are wired in parallel and both will be engaged simultaneously.

 

[Harald Kapp]

The mag brakes are wired in parallel and both will be engaged simultaneously.

When the brakes are paralleled, a single diode suffices.

 

[duke37]

If diodes are fitted, then the resistance will differ depending on the polarity of the meter. An extra diode will do no harm but must be connected the right way riound as must be the power supply.

 

[CalgaryPT]

If diodes are fitted, then the resistance will differ depending on the polarity of the meter. An extra diode will do no harm but must be connected the right way round as must be the power supply.

I see your point, but do you think this test would be a real indication of diode being present though? (My decades old electronic knowledge is failing me perhaps.) The coil on each mag brake is 48.2 ohms, measured either polarity. So supposedly this means there is no diode across the coil. Yet if I temporarily connect a diode such as a 1N4004 across the mag brake coil, and crunch the numbers, wouldn't this be the math:

A 1N4004 has a infinite resistance (for all practical purposes) when reversed bias. When forward biased its resistance is 2.4M. So if I calculate the resistance (R1 x R2 / R1 + R2) when reverse biased I'll get 48.2 ohms (rounded), and when forward biased 48.19 ohms (48.2 ohms rounded)...putting both at the same value given the margin of error. So I don't understand how this measurement is significant enough to say weather or not a flyback diode is present. It could be either meter or operator error with that small of a difference I would have thought. Squeezing the alligator clips of my Fluke meter could itself reduce the resistance 0.01 ohm I'm sure.

Something that maybe more indicative is that there is no polarity to the mag brake leads, which may be a clue there isn't a diode present?

Your thoughts? Maybe a scope across the leads and watch for an inductive spike when I shut off power?

Thanks.

- Peter

 

[Harald Kapp]

Maybe a scope across the leads and watch for an inductive spike when I shut off power?

That could work.
Or put a small signal sinusoidal AC source ("tone generator") at 1 kHz with a voltage around 1 V or more across the brakes. Without diode the current will be sinusoidal. With diode you wil see the rectifying effect on the current.

 

[duke37]

I think you are using too posh a meter. For this sort of test I use a 3V Taylor analog meter which shows about half range when a silicon diode is tested in the forward direction, a germanium diode gives a greater deflection. The resistance range setting makes little difference and there is obviously no true measurement of resistance,
I do not know what a Fluke does, I have never found one in a skip
You will need a voltage supply which will turn on the diode. 12v and a series bulb would do.

 

[CalgaryPT]

Thanks everyone. I will likely add my own diode but the thread really got me thinking. I will try both these methods and see what kind of results I get. That's part of the fun for me.