Mosfets as freewheel diodes

Thread Starter

bytraper

Joined Sep 28, 2010
127
Hi Everyone,
I've got a question that's been bugging me for a while now.

I want to build another dc motor control, and this time, rather than using diodes for the freewheel clamp I want to use a mosfet.

The controller will use a HV regulator on the board (good for about 60V) and 75V mosfets that I'll use for the drive side of the circuit. This means I can run either a 12v motor or a 48v motor. I was thinking to use the IRF1405 as the freewheel mosfet.

Now, when I look at this:



It shows the source and the gate connected, but my question is, if I am going to be running a 48v motor, do I need to put some sort of zener on the gate to protect it or keep the voltage at an acceptable level?

Thanks for any and all advice
 

Jaguarjoe

Joined Apr 7, 2010
767
.





It shows the source and the gate connected,

It is showing the source and substrate tied together. This negates the effect of the built-in transistor inside the MOSFET.

The gate is the long bar on the left.

What's wrong with a 5 cent diode or a 10 cent zener?
 

crutschow

Joined Mar 14, 2008
24,912
The MOSFET is more expensive and has no advantage over a diode for the freewheeling function. You are just using the MOSFET substrate diode for conduction. The voltage drop when conducting will be very similar to a standard diode. If you want a lower drop you could use a Schottky diode.
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
I've been using Schottky Diodes, but it seems if a motor has big back current (not voltage, the snubber and cap take care of those) it can destroy the diodes (literally) and then take out the Mosfets.

On the current controller I have 4x 40A Schottky diodes (VTS40100CT), but some beefy 12V 50A motors with big windings can blow the crap out of all 4.

Being that a Mosfet can handle much greater currents I thought it might be a good option.

I've bought some used industrial DC motor controllers (Leeson Speedmaster and Dayton) and both of them use mosfets for the clamping, so it can't be that bad, but the models I have are a 12V and a 24V, no 48v models, so that's why I thought I'd ask the question about the zener.

If anyone can point me to a circuit that can handle greater clamping currents for a lower price than the 4 diodes, I'd be interested to see.

Just so people can see the current setup:


It's just not up to it.
 
Last edited:

#12

Joined Nov 30, 2010
18,210
Senior moment..I can't remember what an ECU is besides a device that uses leftover Freon to make domestic hot water. (Side effect of the work I do.)

Little help?
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
Engine control unit... I had a look at it, its for firing off the solenoids in the injectors, but the current they put back is low, so its pretty useless for what I'm trying to acheive.
 

THE_RB

Joined Feb 11, 2008
5,438
I've been using Schottky Diodes, but it seems if a motor has big back current (not voltage, the snubber and cap take care of those) it can destroy the diodes (literally) and then take out the Mosfets.
...
The freewheel current is exactly the same as the forward current, and TO220 power schottky diode packs are standard. Your schematic shows a snubber across the FETs but nothing across the motor or schottkys. Maybe you have been getting some motor generated spikes that are killing your schottkys?

The thing that kills schottkys is voltage, have you checked your PSU voltage with a 'scope to check for voltage peaks? Snubbers/transorbs etc across the schottkys is definitely recommended.

I'm also assuming you have ruled out heat too and have some good solid heatsinking in place running a 50A motor.
 

Thread Starter

bytraper

Joined Sep 28, 2010
127
Hi THE_RB,
I drew the schematic up last night, Its just a mistake in the diagram. I also accidently put an electrolytic in the diagram at the base of the snubber) and also the caps are LOW esr not high esr... I was pretty tired when I drew it, but I figured people would pick it up and know it was a mistake.

The snubber sits across the Power + lines and the drain of the mosfet (Where the motor + and neg go) So across the motor.

But yep, checked for spikes, they get caught by the caps and the snubber. It seems to be the current from the windings of big motors. They do get slightly warm, but well within their allowable figures.

As an afterthought, I might try and put a TVS in series with the schottkys..
 
Last edited:
Top