Maker Pro
Maker Pro

12V 400A mosfet motor control

P

Pat Ford

Hi All;
I'm looking at replacing the relay controls on my warn winch with a mosfet
based system ( like the new winches have). Where would I find a schematic
for such a beast?
Thanks
Pat
 
F

Fritz Schlunder

Pat Ford said:
Hi All;
I'm looking at replacing the relay controls on my warn winch with a mosfet
based system ( like the new winches have). Where would I find a schematic
for such a beast?
Thanks
Pat



Is your winch a Warn Works 4700?

http://www.warn.com/works/dcwinches/DC4700.shtml

A tasty beast to be sure. 400Amps though... Ouch.

Is 400 Amps the stall current? The website doesn't seem to specific about
that.

I assume the other winches you speak of probably use ordinary N-channel
devices and switch the motor's ground lead in ordinary fashion. There are a
number of caveats you should be aware of though before starting this
project.

The first thing to be concerned about is saftey. Most semiconductor devices
normally fail short circuit (including MOSFETs). If your MOSFETs fail for
any reason (perhaps inductive avalanche wasn't properly addressed in your
design for instance, or maybe they got too hot) then there exists a
potentially dangerous situation if the MOSFETs fail short circuit. Without
any other circuit interrupter in place the winch will be powered on
continuously and the user will be powerless to stop it. If this happens at
a very in-opportune moment while the operator is standing say between the
winch and the 2 ton load it is pulling, the operator has a potential of
getting squished.

Of course, besides personal saftey a winch is not meant to operate at 100%
duty cycle for any length of time at all. As such unless a human can
manually shut the device down somehow by mechanically disconnecting the
power, smoke might start coming out of it and it may never function again.

That said, there are a number of technical problems that also need to be
addressed. The inductive energy storage of the motor is surely very large
and it must be dealt with. A simple large diode anti-parallel to the motor
might be sufficient. The diode would need large surge current handling
capability however. A couple of P600A or similar devices in parallel might
be adequate if you don't switch the winch on and off too frequently.
Datasheet at:

http://www.vishay.com/docs/88692/p600a.pdf

You probably don't need to worry about current sharing resistors for this
type of application. At very high current densities diodes have quite good
current sharing since the forward voltages get quite large.

With that out of the way the next real problem is driving the MOSFET gates
properly. The MOSFETs must be either fully on or fully off at all times,
and that isn't exactly as easy as it sounds. Since the winch user interface
is presumably some kind of mechanical switch you would need some kind of
switch debouncing circuit or else the MOSFET would likely turn on and off
many times during each transition which would possibly overheat the MOSFETs
themselves or the antiparallel diode(s). Presumably you will need to
parallel multiple devices to obtain this kind of current handling
capability. This will add some complication, make sure you understand the
implications of that. International Rectifier's application notes are a
good place to start.

http://www.irf.com/technical-info/appnotes/an-941.pdf


The next rather potentially serious problem is getting adequate voltage for
gate enhancement. While a 12V battery measures a good 12.4V or more without
any load, when you start cranking the 400 Amps through it things start
changing. If the battery is in good condition, fully charged, and new it
may sag only maybe say a couple of volts under that kind of load. A "bad"
battery (although still may be able to start a vehicle just fine) could
easily sag much lower perhaps under 8V or even less than 6V. 6V is not good
enough to provide adequate gate drive for standard gate MOSFETs. As such
under heavy load the MOSFETs may enter the linear region and start heating
up drastically, well beyond their capability and will subsequently fail.

Yet more to consider on this problem is the effects of the winch's wiring.
The Warn Works 4700 claims to have 6' long 6 gauge cabling. Since the
current must return to the battery it has two wires, thus the total high
current wiring length is 3.6 meters. The resistivity of 6 guage wire is
about 0.00132 Ohms/meter. So the resistance of the total wire is around 5
milliohms. At 400 amps this works out to a drop of about 2V along the
entire length of the cable. This voltage drop as well as the wiring's
inductance must be considered for designing the MOSFET gate drive circuit.

Basically you should not use the winch battery as your power source for the
MOSFET gate drive circuit. You should probably use a totally separate
isolated and well regulated supply. A second battery could be used for
this, so long as you have some sort of undervoltage lockout on the gate
drive circuit to make sure it will never turn on if the battery voltage ever
gets too low. The gate driver should also have relatively fast transitions
(especially turn off). Turn off in less than say 10us would be desireable.

That said, several IRF2804 MOSFETs would be an excellent choice for this
application (DigiKey sells them).

http://www.irf.com/product-info/datasheets/data/irf2804.pdf

Though the die claims to be capable of 280 Amps, you should probably not use
them at more than 75 Amps due to package limitations. For short periods it
may be acceptable to exceed this somewhat, but my suspicion is that winch
duty would not be short enough to obtain substantial benefit.

Not to discourage you in any way, but there may be other potential pitfalls
not mentioned here. I hope this helps. Good luck.
 
P

Paul Burridge

Is your winch a Warn Works 4700?

http://www.warn.com/works/dcwinches/DC4700.shtml

A tasty beast to be sure. 400Amps though... Ouch.

Is 400 Amps the stall current? The website doesn't seem to specific about
that.

I assume the other winches you speak of probably use ordinary N-channel
devices and switch the motor's ground lead in ordinary fashion. There are a
number of caveats you should be aware of though before starting this
project.

[caveats snipped]

Well done, Fritz! A highly informative posting I must say. Pity some
of our other contributors aren't disposed towards such similar
sterling efforts.
 
F

Fritz Schlunder

[caveats snipped]

Well done, Fritz! A highly informative posting I must say. Pity some
of our other contributors aren't disposed towards such similar
sterling efforts.
Winston Churchill


Thanks, I appreciate the praise.
 
Top