Opposite of MOV for BEMF Spike Protection?

[Jdo300]

Hello All,

I am designing a pulse circuit output stage which uses BEMF to produce high voltage spikes (see attached circuit).

I understand that MOVs work by loosing their resistance when HV transients are applied to them. But I am wondering if there is anything out there which does exactly the opposite. So when a high voltage pulse is applied, the resistance increases?

The reason I am asking is because I want to come up with a simple way to protect my MOSFET from the BEMF spike produced in the inductor. I want to be able to produce BEMF spikes that are higher in voltage than the MOSFET's breakdown voltage; so I reasoned that if I could have some type of passive device which has negligible resistance while the switch is on, but increases in resistance after turn-off as the BEMF voltage rises, that all the voltage would be dropped across the device, thus saving the MOSFET from destruction. Does anyone know of such a thing?

Thanks,
Jason O

 

[NoTime]

I never heard of anything like that.

The usual ways of dealing with this are
Capacitor across the switch.
Diode on the inductor.
Dissipating the stored energy on the output side of the inductor.

You might want to look into automotive ignition circuits.
That sounds similar to what you want to do.

 

[berkeman]

Yeah, just put a diode from the drain up to +V. That's the traditional way to catch the transient on the primary side.

 

[Jdo300]

Hi Guys,

Thanks for the suggestions. I have heard of using a diode across the coil, or using a cap across the FET (I think you are referring to a snubber circuit?). But I don't actually want to suppress the spike itself, I simply want to stop all of the voltage in the spike from dropping across the MOSFET alone.

So far, I have only found one or two possible solutions, both of which are questionable. I found this site here which talks about using a second MOSFET as a type of 'voltage controlled resistor':

http://www.google.com/patents?id=i04fAAAAEBAJ&dq=5264785

and also:

http://graffiti.virgin.net/ljmayes.mal/comp/vcr.htm

However, these are both active devices and I was hoping to find a passive device that could accomplish this.

The Voltage-controlled MOSFET resistor would be cool if I could rig it up to automatically turn off during the transient voltage. However, my concern there would be how fast it could actually respond to the quick voltage change. But then again, if I want to go that route, It may be better to just go with my second idea below.

My other thought was simply to have two MOSFETs in series that could switch on and off together at the same time, so the voltage could at least divide across both of them together. But my concern then is how to ensure that both FETs will turn off at the same time even with the same drive signal going into both? I also know that MOSFETs are notoriously known for having fast turn on times but slow turn off times, (even when using a driver) so I'm trying to think of how to combat that issue as well.

- Jason O

 

[NoTime]

FWIW the cap version can work as a boost, with the main output pulse coming off the charge on the cap.
di/dt.

Here is a quick blurb on automotive ignition.
http://home.earthlink.net/~jimlux/hv/igncoil.htm

 

[Phrak]

What voltage is V+? What is your transformer turns ratio, and how many volts do you wish to produce acroos the secondary?

 

[Phrak]

What voltage is V+? What is your transformer turns ratio, and how many volts do you wish to produce acroos the secondary?

I ask, because there might be you could use a different topology or perhaps you don't have th problem you think you do.