Paul & Sabrina's cheap DIY 144v motor controller

[jyanof]

Originally Posted by voltmatic automaton

the way i have it now, the control board (revolt), i use one of the GR pads to run a line from the control board to the LS power stage. I did this because the Gate Resistors are in the power stage of the LS controller.
For the LS, there is a single pin that this line was soldered into.

If i go with the IGBTs, should i just make a small board to insert between the control and power stage so that it will produce the same signal effect?

a intermediary drive board might be better. You'll want the gate drive loop to be as short as possible to ensure that the inductance in the wiring doesn't interfere with turning the switches on and off. Not sure what it looks like now, but it sounds like the control board and driver are pretty far from the gates.

[jackbauer]

You can drive an igbt with 12v (i am). It will turn fully on just a bit slower which will increase switching losses and heat. My best advice would be to scope the gate emitter of one of the devices and post a pic on here. If you look at some of my videos you will see my early experiments. The purpose of the fast recovery diode and extra resistor is to aid the turn off (another area of potential loss).

[voltmatic automaton]

well, i don't have a scope so thats difficult.
the previous post mentioned an intermediate board, that sounds promising.

heres a better picture and explaination of my current setup.
if you guys are sure i can stick with 12 volts as a driver then ok, but if 15v would be more definite that would be completely do-able.

[Cyruscosmo]

Hello Paul!

Hope you remember me as I have been away for a bit. A few months back my Grandma took a down turn in her health and passed away a few weeks ago. She was twelve days from her 97th birthday.

I was reading the forums and seen that you have also recently lost a loved one. My heart goes out to you Paul I'll say a prayer for you too. It's all we can do now is remember the good times and wait for the day we will see them again.


I got a question for ya if you are up for it. I am getting closer to finishing other projects that have put an electric vehicle on the back burner for me till now.

One of the ideas I have had running around in my head is a modular approach to the power section where you would stack on as many amps as you need for whatever motor.

The other is a smaller control board say 2.5 by 3 or so with surface mount components and mount it inside the car in the dash area.Then use a CAT5 cable to connect them together.

Anyway let me know if you are interested in this and if so I would like to get some money in your hands and chat a bit about some other thoughts on the controller.

Take Care

[electrowizard]

Hi! I've built a controller inspired by this thread, after reading the entire thing about a month ago.

I've been blowing mosfets and was hoping to get some expertise from this thread to help me out. I would really appreciate it; I'm fresh out of ideas.

I've made a new thread in "DIY / How-to" so I don't hijack this thread.
It's called "Why is my homebrew controller blowing mosfets?"

jackbauer, I would really value your input as well, even though you use IGBTs. I am testing with a starter motor just like you did in your videos; those were a real inspiration to me

[jyanof]

Originally Posted by jackbauer

You can drive an igbt with 12v (i am). It will turn fully on just a bit slower which will increase switching losses and heat.

hm, I took a look at some data sheets and it looks like there really isn't a difference in saturation voltage for 12v and 15v, though I don't know what are exactly in the LS power board. But, they were likely driven with 15v before, so the gate resistors are probably sized as such. As you say, they're likely turning on quite a bit slower and increasing the switching losses.

Also, it would be good to know what the values of the gate resistors are and determine if the current driver can even supply the required drive current.

[voltmatic automaton]

ok so the IGBTs are the following part number GB30B60K the values according to digikey are as follows: IGBT 600V 78A TO-220AB, 2.35V @ 15V, 30A Max Power 370W

The Gate Resistors are difficult coz the color bands are hard to decipher, its a 5 band code; Red,Black,Black,Black,Brown. If its the other direction just reverse, the values either way seem to be either 10 ohm, 16 ohm, 20 ohm.

So it looks like i do need a 15 Volt supply for max performance.

The gate resistors don't help me much, maybe you guys can help.

[Weisheimer]

Originally Posted by voltmatic automaton

ok so the IGBTs are the following part number GB30B60K the values according to digikey are as follows: IGBT 600V 78A TO-220AB, 2.35V @ 15V, 30A Max Power 370W

The Gate Resistors are difficult coz the color bands are hard to decipher, its a 5 band code; Red,Black,Black,Black,Brown. If its the other direction just reverse, the values either way seem to be either 10 ohm, 16 ohm, 20 ohm.

So it looks like i do need a 15 Volt supply for max performance.

The gate resistors don't help me much, maybe you guys can help.

Hi VA,

I think that the resistor is 200 ohm if the red band is closest
to an end or 100 ohm if the brown band is closer to an end.
Do you have a multimeter to be able to measure it?

I suspect that you will want to use a lower value gate
drive resistor if they above 100 ohm. Do you have a supply
of parts available? If so, I'd start by replacing them all with
20 ohm resistors.

How many IGBT's are used on that power board?

And I agree that a 15 volt supply is better in a perfect world,
but I don't think that is what is causing the heat issue.


Cheers

Mark Weisheimer

[voltmatic automaton]

Hi there,

i counted 20 IGBTs, there are 12 center tap recovery diodes.
there are also 1 of each of the following IRFI620G, TIP41C.

i will try desoldering a gate resistor and let you know the value,

thanks very much for your help

[Weisheimer]

Originally Posted by voltmatic automaton

Hi there,
.....
i will try desoldering a gate resistor and let you know the value,

Hi VA,

No need to do that now.
I read your DIYEC post and found that the gate drive
resistors are 200 ohms. Quasar showed tacking another
200 ohm resistor onto the bottom of the original resistors.
That would make them 100 ohm, so you might want to
tack on an even lower value, such as 50 or 75 ohm.
I only suggest the tacking on as I sometimes find that
boards can only handle so much soldering/de-soldering
and doing it several times can lead to lifted circuit
traces. Who needs that problem?

I'd also recommend that you use twisted pair wire
to connect the gate drive from the Revolt control board
to the LogiSystem power board. (from a CAT5/6 cable)

The LS design has quite a lot in common with the Curtis controllers.
So, the TIP41C might be part of a voltage regulator to power the logic board.

I'm wondering if the IRFI620G is being used as a gate driver.
Can you tell where the circuit traces go from it?
It might also be part of a switch or etc that is no longer used
when the control board is removed.

My belief is that the Revolt control board has enough gate drive
to work with your LS power board.

Another thought here too....The heat could be coming from the
diodes. You could use a non-contact laser thermometer
to aim it at the bodies of the IGBT's and diodes to see where the
actual heat is coming from. Do you have one or do you know anyone
who could lend you one to test with? Or even better, use
an infrared camera if you have a relative who is the Fire Chief!
You did say that it got too warm when driving above 45 mph
so it is most likely the IGBT's, but I'd check it out if it was not
too difficult.

Mark