Paul & Sabrina's cheap DIY 144v motor controller

[MPaulHolmes]

The 15v supply would help. Also, you could add a "diode and resistor in series" in parallel with the other gate resistor, so that turn off has a lower resistance than turn on. Turn off gets stuck at the miller knee, and only has like 4v or something to help it turn off, whereas turn on gets stuck at the miller knee, and has 8v to push it out (since it's stuck at around 4v again, but on its way up instead of down).

[voltmatic automaton]

what kind of values do you think would be good for the diode and resistor combination.
Now that i made the changes to the gate resistors the gate resistance is 30 ohms or so (30.5 i think)
so aside from resitor values what kind of diode should i get?

and would i just daisy chain these parts together? seems like a lot of parts just twisted and soldered together....not that i'm not good with that....that describes most of the car

i believe someone on the diyEC forum described the parallel dode and resistor setup you mentioned for this project.

thanks for your help!!

[Weisheimer]

Jack Bauer uses the reverse diode and resistor in parallel with the normal gate drive resistor.
I think that he uses the UF4007 diode (ultra fast) and a 15 ohm resistor in series
and that pair is then placed in parallel with the gate drive resistors that you already have.

Paul discussed it here

http://ecomodder.com/forum/131476-post2353.html

[voltmatic automaton]

okay, i'll give it a go with the diode you specified and 15ohm resistor.
now i do have a question which comes about coz i don't have a full grasp of what i'm doing.

when i reduced the resistance by putting the other resistors in parallel with the original gate resistors it brought the total resistance down a WHOLE lot.

if i place the series combo in parallel with whats there already, i am working on the assumtion that as long as the combo is oriented in the correct direction, the resistanc ewon't drop any further (except in the direction i want yes?)

I will be using a small diagram i think i got from jack bauer to make these,

any suggestions or corrections are more than appreciated,

and Paul

the controller ROCKS!!

imagine, a controller that doesn't want to blow itself up...how novel!

[Cyruscosmo]

Hey Paul

To be honest I am not sure.

What I see in the schematics is that an optical coupler was used to drive the 12 Amp Mosfet Driver MIC4451YN to isolate the output from the high power section. Yet the Opto coupler and the driver are on the main board? If the magic smoke gets loose in there would it not destroy parts of the main board as well?

I have seen concerns expressed over the distance from the driver to the mosfets.
Over the number of mosfets a single driver can switch.
And over how many AMPS a driver can handle.
But I have seen nothing concerning the distance the high speed CMR Optocoupler can be driven from the CPU.

So could the optocoupler and MIC4451YN be placed with the power section and the rest of the controller electronics be in the dash where it could be easily removed for upgrades or simply as a theft deterrent. The car is going nowhere without a brain.

I know that CAT5 and USB cables can transfer data at high rates so why could they not be used to connect between the "Brains" and the "Brawn" parts of the setup?

The reason I ask is I will be building an electric vehicle from the ground up when my shop is finished. I will also be building the power section in modular water-cooled units that can be stacked for higher output. A remote mount controller which can be daisy chained to each module would make more sense cost wise than a controller for each unit.

And as I have said before if looking into this possibility would interest you I am willing to fund your research. ;-)

[sawickm]

Originally Posted by Cyruscosmo

"So could the optocoupler and MIC4451YN be placed with the power section"

Cyruscosmo,

Check out the Driver PCB that adamj12b is working on: http://ecomodder.com/forum/165449-post3138.html

I'm not sure what the maximum distance between Driver PCB and Controller PCB would be ??? for your remote mounted Controller setup.

-Mark

[Cyruscosmo]

Originally Posted by sawickm

Cyruscosmo,

Check out the Driver PCB that adamj12b is working on: http://ecomodder.com/forum/165449-post3138.html

I'm not sure what the maximum distance between Driver PCB and Controller PCB would be ??? for your remote mounted Controller setup.

-Mark

Thank ya Mark that is the direction I am working towards

[adamj12b]

Mark,

How do you find these things?! I posted that 4 months ago, and almost 450 posts back!

Right now I have about 6 inchs of wire between my driver and my logic board. There are a few things you can do to improve however.

If you place the opto on the board with driver, you can set it up as a current loop instead of a voltage loop. This will greatly increase its noise immunity.

For the distance you are talking though, and others have been asking about the same thing, I would use a simplified logic board on each controller, all connected with a high speed CAN bus to the master controller. This would allow for temp sensors to be placed in each module to account for differences from being mounted in different locations. Duty cycle per module would be able to be controlled to keep all modules running at the same temp.

I actually had a plan to do something like this with my control board, in order to parallel my IGBT controller. I was planing 2, 4 module 1200A controllers for my corvette, wired in parallel with high speed bus linking them to make it act as 1. Each controller would be mounted to the inside of a wheel well under the hood.

-Adam

[voltmatic automaton]

okay so i have the parts, just one last question before i put them in, for the inverse parallel will the diode be letting current go to the power stage or should it be turned the other way?

its late so if my technical language gets slurred sorry...

[Weisheimer]

Originally Posted by voltmatic automaton

okay so i have the parts, just one last question before i put them in, for the inverse parallel will the diode be letting current go to the power stage or should it be turned the other way?

its late so if my technical language gets slurred sorry...

hmmmm.....mine is always slurred.
The diode will go cathode (band end) towards the driver chip, and blank
end toward the IGBT gate.
It doesn't really matter as to diode or resistor towards which as
the diode will determine that current flows in that leg anyway.
The diode is allowing current to flow through ALL of the parallel resistors
when the gate driver chip is LOW.