Paul & Sabrina's cheap DIY 144v motor controller

[harlequin2]

OK, the motor will still see 400V pulses, so if its an insulation breakdown that is the problem, that will still occur even though the average voltage is only 200. The motor current will still be right up there, so will need heavy cables but the battery current will be half so you can use smaller cables for those connections.
ie if your controller limits the motor current to 500A, your battery current will be 250A (ignoring losses, etc) and the motor will be seeing an average voltage of 200v so the power input will be 200x500 = 100 kw.
Is that what you are thinking?
I wonder what the armature sparking will be like.

[isaac_alaska]

i'm fairly certain that the motor voltage limit is only because of the rpm limit of the motor (higher voltage=higher rpm.) running much over 200 could lead to the motor over-revving and damaging itself i think. i'm not POSITIVE on this though, i need to contact the motor manufacturer and see what they say. We've added different brushes and bearings to the motor already that increased the voltage limit a little from stock, but i'll need to check and see what the insulation breakdown voltage is.

but yeah, everything you said is correct.

If the voltage pulses are a problem i could probably come up with a low pass filter circuit that would average the voltage...essentially the same thing that happens in a rectifier. i'm not sure how pre-charge would be taken care of though.

[Greg Fordyce]

Originally Posted by isaac_alaska

driving a netgain warp7 motor with it. i was planning to limit the duty cycle to 50% to limit the voltage to the motor to half the pack voltage. Current limit would be handled by the controller.

The controller already has a motor overspeed threshold that you can configure in software for your setup. This has the effect of limiting motor voltage when properly setup.

[isaac_alaska]

Originally Posted by Greg Fordyce

The controller already has a motor overspeed threshold that you can configure in software for your setup. This has the effect of limiting motor voltage when properly setup.

oh. i wasn't sure how this works, but it sounds pretty handy. how hard is it to set it up? and what other sensors do i need to make it work? it must take some kind of encoded signal from the motor?

[DJBecker]

Originally Posted by isaac_alaska

i'm fairly certain that the motor voltage limit is only because of the rpm limit of the motor (higher voltage=higher rpm.)

You can over-rev a typical EV motor with modest voltage. Perhaps with 24V if there is no load. Certainly with 48V.

The motor voltage limit is a combination of insulation breakdown, especially over time+heat+voltage, and commutator/brush arcing.

Originally Posted by isaac_alaska

If the voltage pulses are a problem i could probably come up with a low pass filter circuit that would average the voltage...essentially the same thing that happens in a rectifier. i'm not sure how pre-charge would be taken care of though.

There isn't a reasonable way to do a low pass filter. The currents are too large, and the load varies too much. You would end up with something that is inefficient or huge, likely both.

In particular, think about current multiplication and what happens with the freewheel current.

[MPaulHolmes]

You don't need any sensor for the over-rev protection to work. It's just setting a couple variables. It's pretty cool actually.

[isaac_alaska]

Originally Posted by DJBecker

You can over-rev a typical EV motor with modest voltage. Perhaps with 24V if there is no load. Certainly with 48V.

The motor voltage limit is a combination of insulation breakdown, especially over time+heat+voltage, and commutator/brush arcing.
In particular, think about current multiplication and what happens with the freewheel current.

True. However, dc motors are essentially no different from a generator. . as the rpm increases, the back emf from the motor begins to balance the forward voltage from the battery supply. when they're equal, the motor can't speed up any faster (unless you're doing something like going down a really steep hill, in which case it could actually charge your batteries.)

i think the fact that you're actually hitting the motor with full voltage spikes, instead of a steady voltage, counteracts this though.

Originally Posted by DJBecker

There isn't a reasonable way to do a low pass filter. The currents are too large, and the load varies too much. You would end up with something that is inefficient or huge, likely both.

In particular, think about current multiplication and what happens with the freewheel current.

it could certainly be done with a low pass filter. the variable load isn't an issue, you would just size it for a fifty percent duty cyle. any other speed would require less work from the capacitor filter bank. On the other hand, you're right that it would end up being quite large. not enormous though. i'll see if i can set up a pspice simulation of this and stick the results on here. for worst-case scenario i'll model the motor as a constant current of 1000 amps in series with 1/5 ohm to make it show up as 200 volts to the outside components.

[DJBecker]

Calculate the size and inrush current for a capacitor bank with a low enough ESR to feed the motor.

If you use a L/C filter to avoid the huge inrush current, the inductor needs to be massive.

[isaac_alaska]

i'll see what i can come up with. i've built rc filters before but something that can handle a thousand amps or so is bigger than anything iv'e worked with yet. i'll do a few calculations and see if i can get a meaningful graph (:

[harlequin2]

You are wasting your time!