New approach to brushed motor controller

[safe]

There are many ways to model controller behavior and I'm not sure exactly what is the actual way that they work. For now let's just talk about "Power Input":

The "naive" (most natural) first idea is like the first chart which simply caps the current to a fixed value and then as the voltage goes down the power input goes down accordingly.

The second chart incorporates the idea of discontinuous current and whether this is "real" or not depends on the values that you plug in. It's very possible that with bad values I've assumed that discontinuous mode is more usable than it really is. (it might be real at 1/10 throttle and not real everywhere else)

The third chart would be what Power Input might look like if the current increased as the voltage decreased. Since no one seems to ever publish actual real world charts for motors with their controllers I'm wondering if this is real or not?

If today is "April Fools Day" is the idea of full power across the entire spectrum (like in chart three) all just a joke?

Which is it: 1, 2, or 3?

These are hand drawn charts that are supposed to depict Power Input (volts times amps) Watts...

 

[Mathurin]

Full throttle power in for NW250B + Kingsmotorbikes hubmotor w/36v & 26" wheel follows something like this:

Note how both the X and Y axis are conveniantly labeled with units and increments.

So it starts out low and past walking speed hits the amp limiter at high 600w to low 700w's depending on battery sag, and when it reaches 25km/h it drops below the amp limiter, then it keeps giving juice upto around 35km/h.

Safe, what you're doing is trying to hit a target by shooting a gun in the dark. A DrainBrain costs around 6-7oz, you won't even feel that over your 140lb motorbike.

 

[safe]

Well this is useful information... I seriously wish that more people would post charts because you can learn more in one chart than a thousands words. In fact, they should even invent a saying... something like:

"A picture is worth a thousand words"

...how's that for an original idea! :wink:

The equations for the motor are published and well known. The equations for a buck converter are published and well known. What I've realized is the the way the current limiting mechanism works is not easy to discover and I've never seen someone work the equations of current limit. I've looked into my spreadsheets and I've been calculating the IR value as being:

Current Limit * Resistance

...but I now think that's probably wrong. The actual current is larger than the current limit and that creates a circular dependency if you attach it to voltage. (the idea being that as voltage decreases the current limit increases)

So I'm going to go and see the effect of changing that once I figure it out. I'm confident that the motor equations and buck converter equations are correct. But the "motor side" current limit... well that might be simply too flat when it actually needs to increase.

It's a good day to discover mistakes... after all... the day is made in honor of fools!

 

[fechter]

Thanks Mathurin,
Your graphing software is more sophisticated than mine. Otherwise I might try posting some graphs myself. I think the curve is accurate in general. Now safe has to figure out the equation that generates the right graph.

 

[Mathurin]

One thing I'd like to know is why it didn't start at 20A, even with the wire on the shunt it did this kinda ramp up thing from low speeds. Once I'd bypassed the current limiter, I don't recall how it acted, but it seems like it just started high and dropped from there, for sure if I had made of vid of the DB screen at that point then I could have plotted it pretty accurately, say per 2km/h steps or so.

 

[fechter]

At really low speeds the limiter circuit might be "fooled" by being in the discontinuous mode, where it might be responding to the peak, which is much higher than the average. I'm not really sure.