EMRFD Message Archive 5417

Message Date From Subject
5417 2010-11-13 04:52:36 Jim Miller 2N7000 failure mode?
I've got an application using a 2n7000 to drive a small valve. If the
FET fails with a drain to source short the valve will remain open
which is not desirable.

Any idea if this is a likely failure mode?

Pointers to better locations to ask this question welcome.


Jim ab3cv
5418 2010-11-13 11:23:51 kb1gmx Re: 2N7000 failure mode?
By valve I presume a solenoid operated valve. this would apply to a relay coil too.

Likely two thins excessive current or lack of protection
from the solenoids Kickback voltage killing the fet.

the 2n7000 is only a 60V device so a solinoid with a
12V supply can induce enough kickback to kill it without
a protective network.

the Solenoid should have a protective network minimmally
a diode to prevent the kick back voltage.

The other is either the fet is not rated for the current
or the gate drive is insufficient to fully saturate the
fet and increasing the dissipation. it's a 2A peak current
is only attainable with a 6-8V gate drive. If your using logic levels to drive it (or a cpu that runs
5419 2010-11-13 11:37:21 Jim Miller Re: 2N7000 failure mode?
Hi Allison

Thanks for the info. The valve driver coil is rated for 55ma which I think
the 2n7000 should handle with just 5v drive. The duty cycle will be very
low. The 2n7000 does show a shunt diode in the schematic but it sounds like
I should add an addtional diode in parallel, perhaps a schottky to ensure it
turns on before the FET diode. Is that just a substrate parasitic device or
something intentionally designed in?

BTW, the valve actually does operate as a solenoid but the moving bit is a
tiny diaphragm. It's a Clippard Mouse:


jim ab3cv

5420 2010-11-13 13:39:07 Jim Miller Re: 2N7000 failure mode?
I see that IRF makes a line called "Direct FETs" which have a built in
schottky combined with logic level drive. At $1.63 qty 1 from Digikey which
includes the diode that's good enough for me. Other suggestions for devices

I was thinking of running the Mouse valve off the wall wart unregulated
supply but perhaps running from the regulated 12v might be an additional
margin source. The 12v linear regulator has plenty of drive left and the
duty cycle is nil.


jim ab3cv (who is re-adding planted aquariums to his list of hobbies for
which he has no time...)

[Non-text portions of this message have been removed]
5421 2010-11-13 13:47:04 Jim Miller Re: 2N7000 failure mode?
forgot to add the part number : IRFU2607Z


5422 2010-11-13 14:33:23 k5nwa Re: 2N7000 failure mode?
That is the typical failure mode with the
transistor shorting out. I've had plenty of shorts, not once did it open up.

A safe design will not depend on the driver to
bias the tube as a driver failure will also cause
final failure, but should be either self biased
or use an external power supply to bias the
final so it operates in a safe area.

At 06:52 AM 11/13/2010, you wrote:
5423 2010-11-13 15:51:09 ehydra Re: 2N7000 failure mode?
Hi Jim -

How did you find a MOSFET with integrated reverse schottky-diode at IRF
website? If I try parametric source it fails. If I try search it fails.
If I poke the part number it finds it but displays no alternative parts.


- Henry

Jim Miller schrieb:
> forgot to add the part number : IRFU2607Z
> jim
5424 2010-11-13 16:30:31 Jim Miller Re: 2N7000 failure mode?
Hi All

The application is a 100psi max gas valve operating at about 15psi. Not
valve as in glowing bottle. Clippard, based in Ohio, makes them.

I asked here since even though it isn't an RF issue there are lots of smart
people many of whom are still active in the electronics industry in one way
or another. Being retired I've lost contact with a lot of what I used to
spend my time on.

BTW, found another perhaps better choice: IRL520.


jim ab3cv

[Non-text portions of this message have been removed]
5425 2010-11-13 18:50:21 kb1gmx Re: 2N7000 failure mode?
5426 2010-11-13 18:57:29 Jim Miller Re: 2N7000 failure mode?
i'll do some snubber research to ensure the IRL510 or IRL520 stays alive. a
diode for sure across the coil. if you have some good reading on additional
snubber stuff i'd love to see it.


jim ab3cv

[Non-text portions of this message have been removed]
5428 2010-11-15 08:09:11 timshoppa Re: 2N7000 failure mode?
5430 2010-11-18 02:57:18 ha5rxz Re: 2N7000 failure mode?
I would be tempted to add a small capacitor from gate to ground, say 0.1uF. This way the coil will be driven by a pulse with softer edges. This experience comes from debugging an electronic door lock that was blowing the IC driving the solenoid.


5432 2010-11-18 07:02:01 Tim Re: 2N7000 failure mode?