DMOS handling precautions

P

Paul Burridge

Jan 1, 1970
0
Hi all,

ISTR reading somewhere that DMOS devices don't require as much care
with handling WRT ESD as do the more common CMOS types. Having just
bought a bunch of 2N7000s for use as MOSFET drivers, I was hoping
someone could confirm or deny this...

Thanks,

p.
 
K

Ken Smith

Jan 1, 1970
0
Hi all,

ISTR reading somewhere that DMOS devices don't require as much care
with handling WRT ESD as do the more common CMOS types. Having just
bought a bunch of 2N7000s for use as MOSFET drivers, I was hoping
someone could confirm or deny this...

I'll deny it.

Almost all CMOS chips have ESD protection. Many DMOS parts have no ESD
protection because they don't want to give up any performance.
 
W

Winfield Hill

Jan 1, 1970
0
Ken Smith wrote...
I'll deny it.

Almost all CMOS chips have ESD protection. Many DMOS parts have no
ESD protection because they don't want to give up any performance.

Right, 2n7000 are not protected, and are probably more fragile than
a typical CMOS logic chip. However the gate's input capacitance is
much higher than that of a logic gate, and provides some protection.
For example a typical logic gate has 2 to 5pF of capacitance, which
compares to 40 to 50pF for a 2n7000 at zero volts bias. Nonetheless,
you can easily zap the 2n7000 straight to kingdom come, as they say.

In the early 70s I always used a Supertex VN01 type that came with a
gate zener. When they discontinued the zener version, I called and
complained. An engineer there told me they had more failures of the
protective zener than they had from the unprotected gate alone, so
they discontinued it. There are still MOSFETs available with zeners,
but in considering them I think back to the Supertex guy's claim. In
using about 3000 pieces of 2n7000 in recent years, I have encountered
only about 4 damaged ones. This is a bit on the high side compared
to CMOS gates, but it's not too bad. If one uses good static-control
procedures, a much lower failure rate should ensue.
 
P

Paul Burridge

Jan 1, 1970
0
Right, 2n7000 are not protected, and are probably more fragile than
a typical CMOS logic chip. However the gate's input capacitance is
much higher than that of a logic gate, and provides some protection.
For example a typical logic gate has 2 to 5pF of capacitance, which
compares to 40 to 50pF for a 2n7000 at zero volts bias. Nonetheless,
you can easily zap the 2n7000 straight to kingdom come, as they say.

In the early 70s I always used a Supertex VN01 type that came with a
gate zener. When they discontinued the zener version, I called and
complained. An engineer there told me they had more failures of the
protective zener than they had from the unprotected gate alone, so
they discontinued it. There are still MOSFETs available with zeners,
but in considering them I think back to the Supertex guy's claim. In
using about 3000 pieces of 2n7000 in recent years, I have encountered
only about 4 damaged ones. This is a bit on the high side compared
to CMOS gates, but it's not too bad. If one uses good static-control
procedures, a much lower failure rate should ensue.

Thanks Win (and Ken). The datasheet I've got seems to imply these ones
*may* have a zener. Doesn't that screw-up the input capacitance to
*some* degree? And what would be the likely avalanche voltage of these
build-in diodes? I'd like to test my batch to see if they're protected
or not.

p.
 
K

Ken Smith

Jan 1, 1970
0
Paul Burridge said:
Thanks Win (and Ken). The datasheet I've got seems to imply these ones
*may* have a zener. Doesn't that screw-up the input capacitance to
*some* degree? And what would be the likely avalanche voltage of these
build-in diodes? I'd like to test my batch to see if they're protected
or not.

If the zener is there, you will often see a "gate current" absolute max
rating on the data sheet.

The zener will have its knee above the rated gate voltage so testing for
it puts the part at risk if it isn't there.
 
M

Mook Johnson

Jan 1, 1970
0
Right on

ESD tolerant or not, we make manufacturing work at ESD protected
workstations and follow ful ESD prvention procedures when handling eny
electrical part. That way there are no OOPS, I thought you didn't need that
for these excuses. And it makes it easy for the reliability auditor find
slip-ups.
 
Top