those are not used in the controllers we are talking about at this point.

IR is a nice company.

There is one big problem with small FET's...
As long as we are familiar with FET's behavior under high temperature, there is one component which still acts as old transistor. It is built in diode, which protect FET, but in TO-220 package it is serious diode while in SOIC tiny one and not matched to the rest of the diodes. So when it goes it either short or opens, in first case meltdown in second delayed eventual meltdown.
Solution - serious diodes 6 way H-bridge on the motor leads with output directly to the battery. You will be impressed how much heat you will take from ESC.

Artur

P.S. Good low voltage drop high-current diodes.

Serious diodes indeed! How about simply add a slightly bigger (but much smaller than you'd use on the motor leads) low v-drop diode across the source-drain on each FET? Or, just use bigger (while low Rdson) FETs? :)

There is no reason to add FET's if you cannot control them. (If you can built Power FET's replacement board and re-use existing brains. Then no further questions.) While 6 diodes is not a big broblem, it is like old schottky diode, only in BL configuration you need 6 not one.

Artur

That's kinda what I was thinking; use the brains from a 9920 or something and create an outboard "power board" where all the high current (and heat) is. Keep the surface mount where it should be: low power, noise sensitive devices - and keep the through-hole devices where they should be: high power applications. It's not we're starved for room on a 1/8th-1/10th scale R/C...

About using a schottkey diode: Some ESCs specifically mention to NOT use them or it voids the warranty. I guess this is for people who don't know how to properly select the right device, or how to hook them up.

Brain, I like the idea of this 2-stage controller. Would be nice. And you're right, in our application the extra size would be no big problem.

Yeah, I have to agree on the BKs. Their workmanship is somewhat cheap. With more and more big sensorless ESCs coming out, I foresee a hard future for them.
At least they could have packed them in a case.

Well, with the increasingly stiff competition out there, BK will have to improve their product and/or lower their prices or they will be phased out. Competition is good (for us): It increases product features and preformance while reducing prices. Gotta love it. :)

/\ /\ /\ Not to mention service.. schulze

im suprized schulze is still in buisness with their crappy customer service

Well I can't say nothing about schulze cause that esc is still going. Got that on a truck right now.

mgm24160
 

I got my mgm in. I hooked it up and tried it out. Can some one tell me the cut off on the lipo voltage? I can't seem to find info on it. I thought it was great esc. It seems alot smoother than a bk. Have to give it further testing though.

It cuts Lipo at 3.1v/cell. You can do hard cutoff (stops the motor) or soft cutoff (decreases motor RPM).

I think i missed the cut off setting. Have to go back and look for that. i set it up for 6s and it seemed to draw my cells down farther than wanted. I didn't check them but with the charger when hooked them up. Thanks metal man.

What size daughter board would be Ok?

I am assuming 24 D2Pak7 (4 HI side 4 LO side times 3) 160A FET's per phase with up to 40V (75V later) would be good start.

So, to put all this on "small" PCB would be a fun job, but it would be nice to get dimensions...

As for parts, we can start from simple 24 x IRF2804S-7P and 1 x IR2131S combo.

Artur

The size of the daughterboard doesn't matter too much, as long as it stays smaller than the big schulzes.

Or do the same thing that MGM does.

The controller board in the center and the two powerboards outside (with a heatsink)