Paul & Sabrina's cheap DIY 144v motor controller

[flores]

Originally Posted by MPaulHolmes

PLAN B: With 3 feet of 2 gauge cable, you get a voltage drop of 0.193v at 400amp, and a drop of 0.289v at 600amp.

I still don't get it.. .. i'm sure its me.. but..
Isn't the voltage drop only dependent on the resistance of the cable? and thus dependent on the length?

Have the cables different lengths?

[MPaulHolmes]

Same length, but different current. It basically turns each Igbt into A mosfet. Higher current? Fine! But you will have a higher voltage drop, which causes it to use less curtent. Heres the math:

V1 + i1*R = V2 + i2*R. The point above both igbts to the point where both M+ connect is the same drop in voltage. R is the same for both because we are using equal lengths of wire.

So,

V1 - V2 = i2*R - i1* R
V1-V2 = R*(i2-i1)
i2-i1= (V1-V2)/R

So, the difference in current between the 2 IGBTs is smaller the bigger R is. And, its small when V1-V2 is small. But we cant control V1-V2.

In our case R is about 0.00005 Ohms. So,
i2 - i1 = 2000*(v1-v2).
So, if they differ by, say, 0.01v in their drop, the igbts will be within 20 amps of each other. Without the resistance, that 0.01v difference could be catastrophic.

[Astro]

Originally Posted by flores

I still don't get it.. .. i'm sure its me.. but..
Isn't the voltage drop only dependent on the resistance of the cable? and thus dependent on the length?

Have the cables different lengths?

Voltage, current and resistance are all related.

To get the Voltage(V) drop you need current(I) multiplied by resistance(R).
Increase the current or the resistance and the voltage increases.

[flores]

haaaaaaaaaaa.. I get it..

because: V = R * I and R is fixed and I varies V will also vary, and you calculated that 3 feet of 2 gauge wire is enough R to vary V enough at max I, right?

To make the picture complete.. where does this voltage drop occurs? Between the collector and the emitter of the IGBT?

[MPaulHolmes]

From capacitor plus to motor plus is the same voltage drop for all paralleled IGBTs. (From point where all collectors are tied together to where all emitters are tied together). the differerences happen inside each igbt, the drop from each collector to emitter.

[flores]

Originally Posted by MPaulHolmes

From capacitor plus to motor plus is the same voltage drop for all paralleled IGBTs. (From point where all collectors are tied together to where all emitters are tied together). the differerences happen inside each igbt, the drop from each collector to emitter.

Ok..and what are real numbers? 0.9V and 1.0V, that's a marginal difference of 0.1V on a total of 150V, that can't create really large current differences.. right?

I'm betting I'm wrong again?

[MPaulHolmes]

The largest diode variation I saw was 0.02 across all 12. My guess is that the igbt on voltage drop will be similar. 0.1v difference would be very unlikely I think.

[duncan]

Originally Posted by flores

Ok..and what are real numbers? 0.9V and 1.0V, that's a marginal difference of 0.1V on a total of 150V, that can't create really large current differences.. right?

I'm betting I'm wrong again?

Hi flores

My understanding is that the 150v is across the system BUT the total current goes across the IGBT's in parallel so the problem is one could do most of the work - overloading it and under-loading the other

This would not be a problem (0.1v in 150v) if we were driving separate motors - one per IGBT

Paul's neat idea is to separate them out using the resistance in the cables - so they have to play nice together

It sounds really neat

Paul - is this your idea? - or is it a known fix?

If it is your idea you should look at deleting all of these posts and getting a patent

[MPaulHolmes]

haha, no it's not my idea. Lee Hart on the EV Tech list and an application note both suggested adding a resistor. At first that seemed really impractical, since who makes resistors with that power rating?! The application note suggested adding resistance that has around 0.2v drop at the current you will be running it at. That worked out to 3 feet of 2 gauge cable, so I figured, good enough for government work!

[jedsmd]

Originally Posted by MPaulHolmes

Lots of debuggIng this afternoon after work. It's all almost good to go as far as the software and circuit board goes.

Paul
I think you said in an earlier post that RTDexplorer will not be supported by this controller. Will there be some similar user interface for setting variables and monitoring the controller?
Jerald