Si9410BDY mosfets - are they particularly ESD sensitive?

[Den]

I'm having some probs repairing some gear with Si9410BDY SO-8 pkg mosfets in
them.

The FET will appear to be faulty - replaced it - still no go......replace it
again - it works - WTF!

It's happened too many times to be finger trouble or gremlins. I thought
maybe it was ESD - even though I'm working on a grounded mat with a grounded
iron. Are they particularly sensitive??


Data sheet: http://www.vishay.com/docs/72269/72269.pdf


cheers.
D.

 

[MooseFET]

I'm having some probs repairing some gear with Si9410BDY SO-8 pkg mosfetsin
them.

The FET will appear to be faulty - replaced it - still no go......replaceit
again - it works - WTF!

It's happened too many times to be finger trouble or gremlins. I thought
maybe it was ESD - even though I'm working on a grounded mat with a grounded
iron. Are they particularly sensitive??

Data sheet:http://www.vishay.com/docs/72269/72269.pdf

The faster and smaller the MOSFET the more static sensitive it will be
but I don't think that that is your problem.

Try putting your DVM on both AC and DC between various "ground"
connections on your bench. Any reading over 0.1V is cause for
concern.

 

[Archimedes' Lever]

The faster and smaller the MOSFET the more static sensitive it will be
but I don't think that that is your problem.

Try putting your DVM on both AC and DC between various "ground"
connections on your bench. Any reading over 0.1V is cause for
concern.

The very word MOSFET alone should say to anyone in the industry "very
ESD susceptible". Remember folks, it boils down to the potential
required to breach a single pn junction. These features are pretty
damned small.

ESD is now, has been for decades, and should be, a major part of the
electronics industry, and I find it ludicrous that some engineers
downplay susceptibilities. I have seen the micrographs.. parts that
didn't even fail... right away. It looked like an Iraqi roadway.

Proper ESD workstations and handling procedures should always be
followed... as a rule... not merely when handling known to be
susceptible devices. Even fields can cause failures on parts that haven't
been placed yet. Even non volatile parts can build up and keep a charge,
and discharge into a susceptible part, so proper procedures should be
used at all times to ensure that every step possible was taken to keep
all parts at as close as can be to a net zero stored electrostatic
charge.

Such tiny stored charges can be all the energy it takes to kill a pin on
a sensitive device.

 

[HarryD]

The very word MOSFET alone should say to anyone in the industry "very
ESD susceptible". Remember folks, it boils down to the potential
required to breach a single pn junction. These features are pretty
damned small.

ESD is now, has been for decades, and should be, a major part of the
electronics industry, and I find it ludicrous that some engineers
downplay susceptibilities. I have seen the micrographs.. parts that
didn't even fail... right away. It looked like an Iraqi roadway.

Proper ESD workstations and handling procedures should always be
followed... as a rule... not merely when handling known to be
susceptible devices. Even fields can cause failures on parts that haven't
been placed yet. Even non volatile parts can build up and keep a charge,
and discharge into a susceptible part, so proper procedures should be
used at all times to ensure that every step possible was taken to keep
all parts at as close as can be to a net zero stored electrostatic
charge.

Such tiny stored charges can be all the energy it takes to kill a pin on
a sensitive device.

Hey Arch, your living in the dark ages, BGW. You are correct about some
MOSFET devices will blow if you just get near them, especially if they have
unprotected inputs. The gate on a MOSFET is the most sensitive pin and on
these larger devices they look like 10nF to the other nodes. They are rated
at +/-20V but can withstand >40V. So your 100pF body must be charged to >3KV
to blow the junction. The OP sounds very careful so I am betting, like most
others that ESD is not the problem.
Cheers,

Harry

 

[Joerg]

The very word MOSFET alone should say to anyone in the industry "very
ESD susceptible". Remember folks, it boils down to the potential
required to breach a single pn junction. These features are pretty
damned small.

ESD is now, has been for decades, and should be, a major part of the
electronics industry, and I find it ludicrous that some engineers
downplay susceptibilities. I have seen the micrographs.. parts that
didn't even fail... right away. It looked like an Iraqi roadway.

Proper ESD workstations and handling procedures should always be
followed... as a rule... not merely when handling known to be
susceptible devices. Even fields can cause failures on parts that haven't
been placed yet. Even non volatile parts can build up and keep a charge,
and discharge into a susceptible part, so proper procedures should be
used at all times to ensure that every step possible was taken to keep
all parts at as close as can be to a net zero stored electrostatic
charge.

Such tiny stored charges can be all the energy it takes to kill a pin on
a sensitive device.

However, those are pretty fat devices. Ok, tennis shoes and non-treated
tile could still do ESD damage.

SD5400 and similar low capacitance devices are another matter. I had a
tech almost throw his coffee mug at me when he swapped a SD5400 four
times and I told him it's still damaged. He was sure it was my design
until I sat down, soldered one in and everything worked.

What really helps is to make sure that the air in the room doesn't
become bone dry. I always make sure I wear natural fiber clothing,
cotton and stuff, nothing with mixed in acrylic fibers etc.

 

[Archimedes' Lever]

Hey Arch, your living in the dark ages, BGW. You are correct about some
MOSFET devices will blow if you just get near them, especially if they have
unprotected inputs. The gate on a MOSFET is the most sensitive pin and on
these larger devices they look like 10nF to the other nodes. They are rated
at +/-20V but can withstand >40V. So your 100pF body must be charged to >3KV
to blow the junction. The OP sounds very careful so I am betting, like most
others that ESD is not the problem.
Cheers,

Harry

"your"? "BGW"? Bwuahahahahah!

As little as 20V electrostatic charge on a person can blow just about
any modern chip made these days. 3kV WILL blow anything. The chip makers
that tout resistance to such voltages refer to parts IN CIRCUIT, wired in
a specific manner. This discussion is about raw parts, as assemblies
are being populated, etc.

Even a "High Voltage Diode" is actually a STACK of pn junctions. A
single one of those pn pairs would be susceptible. Being in a stack is
what makes it an HV Diode.

In a chip, however, the features are very, very tiny and very, very
frail. Far more so than they were back at the advent of CMOS and ESD
concerns.

Also, before an FET goes in, especially with a high sensitivity device,
merely touching the pins with fingers, grounded or not, and solder iron
tips can blow them.

If his iron is not a modern, ESD compliant type, there is no way to
know if it is grounded at the tip or carrying a floating AC potential. If
he grounds the tip, is he also incorporating the 1 MegOhm series limiter?

We had some transducers that were FET included that had the leads wired
together for handling prior to installation, and installation required a
specific procedure. Yes, many blew before we decided that indeed, it IS a
concern.

 

[Archimedes' Lever]

However, those are pretty fat devices. Ok, tennis shoes and non-treated
tile could still do ESD damage.

OK big parts? Then they are a bit more hardy, but I would still not
stress any segment of one.

SD5400 and similar low capacitance devices are another matter. I had a
tech almost throw his coffee mug at me when he swapped a SD5400 four
times and I told him it's still damaged. He was sure it was my design
until I sat down, soldered one in and everything worked.

What really helps is to make sure that the air in the room doesn't
become bone dry. I always make sure I wear natural fiber clothing,
cotton and stuff, nothing with mixed in acrylic fibers etc.

Even though it is the best choice, cotton and especially the
artificial fibers can hold a charge. Cotton simply is not a great
generator, whereas the synthetics are huge generators.

When a body is electrostatically charged, the electrons sit on the
surface of the insulator. A conducting body, like a person's hand,can
build up charge by peeling electrons off of charged bodies. This is all
with respect to the "ground" one's ESD workbench and hopefully, the bulk
remainder of one's assembly and parts sits at.

One does not have to actually see (read witness) a direct failure off
the line to have ESD damage cause a failure later on when the product is
in
service.

Yes, cotton, and at least 50% RH. Our lab hovers around 60% actually.
Smocks abate one's field however, and should also be considered when
sensitive devices are utilized. Our smocks have ground leads to the
bench, and our smock sleeves connect to our body and that drains us. So
we do not need wrist straps, and even our smocks are grounded.

 

[qrk]

However, those are pretty fat devices. Ok, tennis shoes and non-treated
tile could still do ESD damage.

SD5400 and similar low capacitance devices are another matter. I had a
tech almost throw his coffee mug at me when he swapped a SD5400 four
times and I told him it's still damaged. He was sure it was my design
until I sat down, soldered one in and everything worked.

What really helps is to make sure that the air in the room doesn't
become bone dry. I always make sure I wear natural fiber clothing,
cotton and stuff, nothing with mixed in acrylic fibers etc.

Going barefoot really helps to reduce the ESD issue. In the previous
company I worked for, we had a charge meter. It was interesting
looking at different things like people and plastic bags. Found a
batch of pink poly bags that had static cling.

 

[Den]

And only when you drag race? ;-)

they'd be high heels of course.

 

[Den]

Thanks for the many replies guys. I have no control over the design (which
seems reliable in service), the faults I've got at the moment are the down
side of production yeild.

My ESD procedures are reasonable but I'll take a bit more care with the
handling and see if that makes any difference. I might make a few
measurments as well and see if there are any stray potentials "floating"
around.


thanks.
D.

 

[Joerg]

OK big parts? Then they are a bit more hardy, but I would still not
stress any segment of one.

Even though it is the best choice, cotton and especially the
artificial fibers can hold a charge. Cotton simply is not a great
generator, whereas the synthetics are huge generators.

When a body is electrostatically charged, the electrons sit on the
surface of the insulator. A conducting body, like a person's hand,can
build up charge by peeling electrons off of charged bodies. This is all
with respect to the "ground" one's ESD workbench and hopefully, the bulk
remainder of one's assembly and parts sits at.

One does not have to actually see (read witness) a direct failure off
the line to have ESD damage cause a failure later on when the product is
in
service.

Yes, cotton, and at least 50% RH. Our lab hovers around 60% actually.
Smocks abate one's field however, and should also be considered when
sensitive devices are utilized. Our smocks have ground leads to the
bench, and our smock sleeves connect to our body and that drains us. So
we do not need wrist straps, and even our smocks are grounded.

And watch the shoes. I found that some brands of sneakers can cause a
lot of ESD grief so we had straps that go over the heel and bleed charge
off into the (treated) VCT floor. Didn't work too well but at least it
helped.

 

[Den]

And watch the shoes. I found that some brands of sneakers can cause a lot
of ESD grief so we had straps that go over the heel and bleed charge off
into the (treated) VCT floor. Didn't work too well but at least it helped.

And those track pants made from parachute type material that young guys like
to wear.....
Spraying chair squabs / backs with a squirt bottle mix of water & fabric
softener every now and again seems to help as well.

 

[Archimedes' Lever]

Going barefoot really helps to reduce the ESD issue. In the previous
company I worked for, we had a charge meter. It was interesting
looking at different things like people and plastic bags. Found a
batch of pink poly bags that had static cling.

Pink poly only works in properly humidified environments. In dry air, it
will not generate a charge, but it doesn't dissipate one either.

They are pretty useless (especially for circuitry) other than that the
alternative is the bad poly, so I guess they have their place. I prefer
black, dissipative media, or the metallized bags with the bubble linings.

With the new carbon nano-tube technologies being what they are, I am
surprised we do not have perfect packaging films and table top mats, etc.
for this industry. The old, powder form carbonized media is past its
prime.

 

[Joerg]

And those track pants made from parachute type material that young guys like
to wear.....
Spraying chair squabs / backs with a squirt bottle mix of water & fabric
softener every now and again seems to help as well.

I prefer parachutes above my head, plus one spare just in case

 

[neon]

On Aug 11, 10:40*pm, "Den" <[email protected]> wrote:
> I'm having some probs repairing some gear with Si9410BDY SO-8 pkg mosfetsin
> them.
>
> The FET will appear to be faulty - replaced it - still no go......replaceit
> again - it works - WTF!
>
> It's happened too many times to be finger trouble or gremlins. I thought
> maybe it was ESD - even though I'm working on a grounded mat with a grounded
> iron. Are they particularly sensitive??
>
> Data sheet:http://www.vishay.com/docs/72269/72269.pdf

The faster and smaller the MOSFET the more static sensitive it will be
but I don't think that that is your problem.

Try putting your DVM on both AC and DC between various "ground"
connections on your bench. Any reading over 0.1V is cause for
concern.


>
> cheers.
> D.

what is that going to do to the ESD? The problem is that when you get to the bench you must ground yourself to it before you do any measuraments. and voltage of 1.v is not going to blow anything but a ESD of a couple of kv will. FETS are hi inpedance devices.

 

[Joerg]

Breach a single PN junction? Fets die from blown gate oxide.

I found that most FETs die from "ballistic activity". A loud bang,
followed by little chunks of plastic flying about, smoke wafting through
the room and so on

The best ones are those which blow part of their TO220 package off and
still somehow work.

[...]