Searching \ for '[EE] CMOS and Latent ESD Damage' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=cmos+latent+esd
Search entire site for: 'CMOS and Latent ESD Damage'.

Exact match. Not showing close matches.
PICList Thread
'[EE] CMOS and Latent ESD Damage'
2008\01\07@090951 by Mike Hord

picon face
Happy New Year to everyone!

What is everyone's opinion of latent defects caused by ESD in
CMOS ICs?  We have a couple of failure modes for two different
products, both involving analog CMOS ICs (one a MUX, the other
a POR/BOR supervisor).

The generalities of the failures are the same (previously reliable,
same component on several boards which has never failed before,
failure in the field but never in the factory).  The biggest question
mark is that all of the boards which have failed were built on the
same line, same machine, within a couple of weeks span.

During the winter.

In the arid northern US.

The bugaboo is this: IF these are ESD failures at assembly time,
why are these making it through production and out to the field
before failing?  Corollary: why are only these components failing?

Failure mode of the IC is reflowed silicon due to high current.
FA lab says electrical overstress: high voltage applied for a long
time.  They CAN'T tell us why the other components on the bus
aren't dead (or even damaged, per their own package removal
studies).

Mike H.

2008\01\07@094905 by Bob Axtell

face picon face
I just went thru a plethora of the same thing here in Arizona. We had
designed
a system with the PLC located one meter from the sensor control PCB,
which has
some HC logic and a big PIC.

The PLC was connected with two IDC26 cables, unshielded. We were repeatedly
killing 74HC86s by Fairchild and TI; using  transorbers stopped the
failures but as
a safety measure, we replaced the HC86s with line receivers DS14C89A.

I  believe that a die shrinkage has occurred in the  HC world AGAIN, and
some
designs have had to be re-examined, because as dies shrink they become more
responsive to noise.

--Bob A

Mike Hord wrote:
{Quote hidden}

2008\01\07@110655 by Peter P.

picon face

Bob Axtell <engineer <at> cotse.net> writes:
> I  believe that a die shrinkage has occurred in the  HC world AGAIN, and
> some designs have had to be re-examined, because as dies shrink they become
> more responsive to noise.

I have tried to analyze the problem represented by the 'shrunk' dies as applied
to pics and other chips. I am not an engineer and some of the following may be
bull from the technical point of view. However, it is important enough to
discuss it, imho.

Die shrinks are usually accompanied by doping strength increases. To maintain
the conductivity (esp. of outputs) the doping density is simultaneously
increased. This causes un-important features (n and p etc wells vs. substrate
for example) to become deeper and thus increases the capacitive coupling
between the theoretically uncoupled wells on a die. Also the abrupt transitions
in doping should be able to produce hotter injected carriers in the substrate
and the better insulation should keep them longer once trapped. At the same time
the vertical 'reverse' diodes in cmos devices become worse and worse, becoming
smaller in area and longer in length. Also the speed of the 'wanted' features
increases with lower area and higher doping concentrations.

I believe that the resulting high sensitivity to esd is due to the sum of these
factors, and that it affects all shrunk devices, including opamps, voltage
regulators (oscillation-prone 780x clones for example) and even audio amplifiers
like the lm386 and tda200x series which I have personally seen degrade from
benign, easy to use chips, to oscillation-prone 'hair trigger' devices that
require costly VHF grade pcb routing and strong decoupling and low pass
filtering otherwise seen only in video circuits and above. Concluding, I would
say that if there will be no breakthrough in technology, things will continue to
become worse and worse in this direction. By breakthrough I would see perhaps a
modified output cmos transistor structure that incorporates active turn-on when
the attached pin exceeds the supply voltage in the relevant direction or other
such measures. I cannot see that happening anytime soon, though.

All of these things raise the implementation cost in ways that are sometimes
unpredictable. I recently saw a simple design that involved a pic and had a
total (and critical) parts cost of under $2 that had to have two costly Schottky
diodes added (normal diodes would not do!) after the pcbs were designed and
manufactured to prevent a current injection related malfunction that did not
occur with any previous series of chips. That was a 10-15% parts cost increase,
manual labor and ugliness of piggybacked parts not accounted for. There was not
even a suffix change between the parts that needed the mod and those which did
not.

I feel that there is a limit beyond which I can see no point in ordering 'brand
name' parts as opposed to using directly counterfeit c**p and designing
accordingly from start to catch any 'unspecified behavior' that should not
happen [tm]. I am not sure whether this limit has been reached yet. In any case
I have no intention to whine against pics, which are usually good and rarely
display undesirable 'improvements' (silicon erratas are not covered by this
message - they represent a different chapter). As to the verosimility of
counterfeit c**p, that it the one thing I am not afraid of. Unfortunately the
largest improvements in counterfeit parts are seen in the exterior part marking.
Only last year I went through three sets of power transistors for an amplifier,
all being grossly counterfeit and of low quality, blowing up after 10 minutes to
6 hours. The fourth set was good because a) I massively overspecified it (this
was a repair, not a new design!!) and b) I made sure I bought them from a
reputable source (as opposed to the usual shops where the other three came
from).

Peter P.


2008\01\07@122805 by Michael Rigby-Jones

picon face


> -----Original Message-----
> From: spam_OUTpiclist-bouncesTakeThisOuTspammit.edu [.....piclist-bouncesKILLspamspam@spam@mit.edu] On
Behalf
> Of Peter P.
> Sent: 07 January 2008 16:06
> To: piclistspamKILLspammit.edu
> Subject: Re: [EE] CMOS and Latent ESD Damage
>
> unpredictable. I recently saw a simple design that involved a pic and
had
> a
> total (and critical) parts cost of under $2 that had to have two
costly
> Schottky
> diodes added (normal diodes would not do!) after the pcbs were
designed
> and
> manufactured to prevent a current injection related malfunction that
did
> not
> occur with any previous series of chips. That was a 10-15% parts cost
> increase,
> manual labor and ugliness of piggybacked parts not accounted for.
There
> was not
> even a suffix change between the parts that needed the mod and those
which
> did
> not.

Was the device being used outside of the specifications originally
though?  Lots of history on the list to suggest the injecting any amount
of current into PIC pins is a bad idea and not within the operational
spec.

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2008\01\07@132528 by Marcel Duchamp

picon face
Michael Rigby-Jones wrote:
> Was the device being used outside of the specifications originally
> though?  Lots of history on the list to suggest the injecting any amount
> of current into PIC pins is a bad idea and not within the operational
> spec.
>
> Mike

Regarding latent damage, I once saw a cmos part get damaged and appear
to work but later go to a hard short on the power supply.  A data logger
in a box with a gel cell battery was experiencing strange failures
involving an Intersil 7660 charge pump.  When a new chip was put in and
power cycled a few times, a puff of smoke would sometimes be emitted but
the part would seem to keep working.  Then a little later, it would go
to a hard short.

The circuit design was fine; the problem came from a non-electrical
person who did the wiring plan and believed that wire had no resistance
and no inductance.  In this case it certainly did. A ground line that
should have gone direct from the battery to the electronics went instead
out 3 feet and back to power a solenoid.  A turn on current surge caused
one of the I/O lines to go below ground 5 to 10 volts and the 7660 was
the victim.

A little o'scope work spotted the culprit and a slight re-wiring of the
main wiring harness fixed it.

2008\01\07@153347 by Spehro Pefhany

picon face
Quoting Mike Hord <.....mike.hordKILLspamspam.....gmail.com>:

{Quote hidden}

Suspicious, IMHO, that you've seen *no* outright failures from
this assembly house & run, yet apparently several "latent" failures.
It's *possible*, of course, but I'd tend to suspect something else
first and maybe second too.

Was it the same customer as well?

Symptoms.. latchup, perhaps? Is the power supply entirely internal or
is it supplied externally or brought out to the outside world? Does this
mux deal with external signals?

Best regards,
Spehro Pefhany
--
"it's the network..."                          "The Journey is the reward"
EraseMEs...spam_OUTspamTakeThisOuTinterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com


2008\01\07@160057 by Brent Brown

picon face
> The bugaboo is this: IF these are ESD failures at assembly time,
> why are these making it through production and out to the field
> before failing?  Corollary: why are only these components failing?

When I was a student we were told this is exactly the nature of ESD damage. The
device is "weakened" by ESD and will function initially but fail after some time under
the stress of normal use (by Murphy's law not untill the customer is using it).

Why only that component failing in your case? It may be the only component
subject to ESD damage during assembly, more susceptable to ESD than other
parts, there may be other parts damaged by ESD but this one fails sooner, or a
dodgy batch of parts more susceptable than normal.

Anyone else noticed local electronics stores becoming more ESD ignorant? Couple
of examples: Jaycar here has no anti-static straps or packaging for chips. Dick
Smith offered to install extra laptop RAM at the counter, no ESD protection, I said
no thanks. The typical mindset is "Well I've never damaged anything with static so it
must be a myth." As described above, failure can occur a long time after the
damage is done and so does not get attributed to the actual cause.

--
Brent Brown, Electronic Design Solutions
16 English Street, St Andrews,
Hamilton 3200, New Zealand
Ph: +64 7 849 0069
Fax: +64 7 849 0071
Cell: +64 27 433 4069
eMail:  brent.brownspamspam_OUTclear.net.nz


2008\01\07@163550 by James Newton

face picon face
I hate to say this, but it sounds to me like either the customer had a
learning curve on how NOT to use the product that burned through a few
boards or production did not actually test a batch of outgoing boards.

E.g. it smells like a social engineering problem.

--
James.

{Original Message removed}

2008\01\07@174100 by Mike Hord

picon face
I like Gmail; it's easy to quote several e-mails at once...

James:

> I hate to say this, but it sounds to me like either the customer had a
> learning curve on how NOT to use the product that burned through a few
> boards or production did not actually test a batch of outgoing boards.
>
> E.g. it smells like a social engineering problem.

Possible but unlikely.  I've attempted to duplicate this damage through
conditions possible under normal use to no avail.  Plus, both of these
products are fairly mature (one has been shipping since late 2005, the
other since 2001).

Spehro:
>Suspicious, IMHO, that you've seen *no* outright failures from
>this assembly house & run, yet apparently several "latent" failures.
>It's *possible*, of course, but I'd tend to suspect something else
>first and maybe second too.

This was NOT my first choice.  I'm coming to this after about six
weeks of hypothesize/evaluate/discard.  And yes, I do find that
part troubling as well.

>Was it the same customer as well?

There are two distinct products in question here, and two different
OEMs, with failures occurring at many of their customers.

>Symptoms.. latchup, perhaps? Is the power supply entirely internal or
>is it supplied externally or brought out to the outside world? Does this
>mux deal with external signals?

Power supply in both cases is over IEEE 1394.  In one case, 24V, the
other, 12V.  Transient testing to see if I could force a transient through
the input protections bore no fruit.  Symptoms are a 20-40 ohm
short from pin to pin, always from Vdd to Vss and usually from one
pin to another as well.

The mux is completely internal- it takes a voltage from a DAC and
connects it to a 5k resistor divider, OR leaves the high side of the
divider floating.

We have something like 12-13k of these MUX circuits in the field with
no prior reported failures.  Certainly not on this scale.

Mike H.

2008\01\08@040606 by Alan B. Pearce

face picon face
>> The bugaboo is this: IF these are ESD failures at assembly time,
>> why are these making it through production and out to the field
>> before failing?  Corollary: why are only these components failing?
>
>When I was a student we were told this is exactly the nature of ESD
>damage. The device is "weakened" by ESD and will function initially
>but fail after some time under the stress of normal use (by Murphy's
>law not untill the customer is using it).

A company I worked for went through a loop of this when they had memory
cards from a word processing system fail on a regular basis in the field.
They imposed proper ESD systems in the workshop, and when fitting cards in
the customer premises, and in time all the chip failures went away, as they
replaced the weakened chips. Typically before the ESD regime came into
operation the board would be repaired in the workshop, and test fine, then
fail at some non-determinate time later in the field.

2008\01\08@115145 by Mike Hord

picon face
On Jan 8, 2008 3:05 AM, Alan B. Pearce <@spam@A.B.PearceKILLspamspamrl.ac.uk> wrote:
> >> The bugaboo is this: IF these are ESD failures at assembly time,
> >> why are these making it through production and out to the field
> >> before failing?  Corollary: why are only these components failing?
> >
> >When I was a student we were told this is exactly the nature of ESD
> >damage. The device is "weakened" by ESD and will function initially
> >but fail after some time under the stress of normal use (by Murphy's
> >law not untill the customer is using it).
>
> A company I worked for went through a loop of this when they had memory
> cards from a word processing system fail on a regular basis in the field.
> They imposed proper ESD systems in the workshop, and when fitting cards in
> the customer premises, and in time all the chip failures went away, as they
> replaced the weakened chips. Typically before the ESD regime came into
> operation the board would be repaired in the workshop, and test fine, then
> fail at some non-determinate time later in the field.

That's the sort of experience I was looking for, although it still
doesn't answer
these questions:
Why only these two components on each board, of all the components?
Why latent failures only?
Why these two products of all the products built at that facility?
Why doesn't heat accelerate this failure rate?

Mike H.

2008\01\08@171531 by Bob Axtell

face picon face
Mike Hord wrote:
{Quote hidden}

Mike, these mysteries will be answered by that Great Engineer in the Sky
when our time comes.
Before then, its just a guess.

--Bob A

2008\01\08@190516 by Paul Hutchinson

picon face
> -----Original Message-----
> From: RemoveMEpiclist-bouncesTakeThisOuTspammit.edu On Behalf Of Mike Hord
> Sent: Tuesday, January 08, 2008 11:51 AM
>
> That's the sort of experience I was looking for, although it still
> doesn't answer these questions:
> Why only these two components on each board, of all the components?
> Why latent failures only?
> Why these two products of all the products built at that facility?
> Why doesn't heat accelerate this failure rate?

If you can afford to spend 10's to 100's of thousands of dollars (possibly
millions) then there are firms that can diagnose the failures for you. They
can use X-ray machines, remove the packaging and inspect with a scanning
electron microscope, as well as other exotic and expensive procedures to
determine the precise nature of failure. For an example of the services
offered by this type of company see:
http://www.muanalysis.com/services/failure_analysis.php and
http://www.bridgept.com/reliability-failure-analysis.htm

Here's a good article from the "Journal of Semiconductor Technology and
Science", "Electrostatic Discharge (ESD) and Failure Analysis: Models,
Methodologies and Mechanisms for CMOS, Silicon On Insulator and Silicon
Germanium Technologies":
http://www.jsts.org/html/journal/journal_files/2003/9/2003_vol3_no3_153.pdf

Also check out this "ESD Myths and the Latency Controversy" article from
Compliance Engineering magazine: (also check around their site for more
information)
http://www.ce-mag.com/archive/02/Spring/dangelmayer.html

Bad ESD handling causing weakening of semiconductors is extremely difficult
to diagnose. Failure can be so long after production and so impractical to
detect in production test, that I put in place a special rule at my employer
decades ago. "When in doubt (about how an ESD sensitive component has been
handled), just throw it out". It isn't fun to trash hundreds or thousands of
dollars of inventory but it is the only safe thing to do when in doubt. When
it comes to ESD sensitive component handling procedures, a gram of
prevention is worth a kilogram of cure :-).

Hope this helps,
Paul Hutch

>
> Mike H.

2008\01\08@223029 by Sean Breheny

face picon face
I had a case where the manufacturer did this for me free (it was a
Lattice FPGA with an IO port which didn't work - it was determined to
be due to a static hit). They even sent me a full report with color
photos (under an optical microscope). If you buy enough of them (and
in this case it was a part which is $16 each in quantity), they may do
this for you.

Sean


On Jan 8, 2008 7:04 PM, Paul Hutchinson <spamBeGonepaullhutchinsonspamBeGonespamyahoo.com> wrote:
> If you can afford to spend 10's to 100's of thousands of dollars (possibly
> millions) then there are firms that can diagnose the failures for you. They
> can use X-ray machines, remove the packaging and inspect with a scanning
> electron microscope, as well as other exotic and expensive procedures to
> determine the precise nature of failure. For an example of the services
> offered by this type of company see:
> http://www.muanalysis.com/services/failure_analysis.php and
> http://www.bridgept.com/reliability-failure-analysis.htm
>

2008\01\09@042704 by Alan B. Pearce

face picon face
>That's the sort of experience I was looking for, although it still
>doesn't answer these questions:
>Why only these two components on each board, of all the components?

Possibly these are the most sensitive ESD wise, and because they take the
brunt the other components don't get harmed? Or maybe the other components
do get harmed, but to a lesser degree, and failure may take longer.

>Why latent failures only?

Probably the ESD level is 'only just enough' to damage them.

>Why these two products of all the products built at that facility?

Are these the only products using this component?

>Why doesn't heat accelerate this failure rate?

One would need to investigate the failure mode to find out about this I
guess.


As a side note, we have a CCD camera that was developed in house for
satellite use. It uses LVDS interface chips. We had the cameras returned to
us to get the LVDS chips changed, as there is a suspicion of ESD damage, but
just where the damage occurred is the question. The most likely suspect is
apparently the integration facility we sent them to, as they are reputed to
have been hot plugging the devices. The problems were found by doing
resistance measurements on the LVDS lines. I do not know what caused the
suspicion of ESD damage as I believe all cameras are still working.

>From this, a heap of information was learnt from the chip manufacturer.
There are something like 9 different failure modes that could occur in these
particular chips, from ESD damage. Some modes can be detected by doing the
resistance measurement, and some cannot.

And then to cap it all off, the 5V supply version of the chip has normal
CMOS style ESD protection diodes - BUT the 3V version of the chip, which is
the one we are using, does not have the ESD protection diodes ....

2008\01\09@043554 by Alan B. Pearce

face picon face
>If you can afford to spend 10's to 100's of thousands of dollars
>(possibly millions) then there are firms that can diagnose the
>failures for you. They can use X-ray machines, remove the packaging
>and inspect with a scanning electron microscope, as well as other
>exotic and expensive procedures to determine the precise nature
>of failure.

Doesn't have to be that expensive. Look for labs that can do PIND testing,
which is what we use for failed space qualified devices. The only part I
don't know is how they would get on de-encapsulating plastic devices, as all
our stuff is ceramic or metal cans which are easily opened.

2008\01\09@092722 by Mike Hord

picon face
Sean AND Paul-

On Jan 8, 2008 9:30 PM, Sean Breheny <TakeThisOuTshb7EraseMEspamspam_OUTcornell.edu> wrote:
> I had a case where the manufacturer did this for me free (it was a
> Lattice FPGA with an IO port which didn't work - it was determined to
> be due to a static hit). They even sent me a full report with color
> photos (under an optical microscope). If you buy enough of them (and
> in this case it was a part which is $16 each in quantity), they may do
> this for you.

One of the manufacturers has already been gracious enough to do this
for us.  We may contact the other (there are two chips by two different
companies involved here), if we get more failures on the second product.

> On Jan 8, 2008 7:04 PM, Paul Hutchinson <RemoveMEpaullhutchinsonspamTakeThisOuTyahoo.com> wrote:
> > If you can afford to spend 10's to 100's of thousands of dollars (possibly
> > millions) then there are firms that can diagnose the failures for you. They
> > can use X-ray machines, remove the packaging and inspect with a scanning
> > electron microscope, as well as other exotic and expensive procedures to
> > determine the precise nature of failure.

We've already been in touch with one of those companies- they have done
quite a bit (though not enough) for us already.  THEY claim that this was a
simple case of overvoltage- our power supply was too hot, the chips fried,
end of story.  Nothing I have done has repeated that, however, so I'm VERY
skeptical about it.

Price has been in the thousands, but could easily creep up past that 5-figure
mark.

Mike H.

2008\01\09@102948 by Spehro Pefhany

picon face
Quoting Mike Hord <mike.hordEraseMEspam.....gmail.com>:

{Quote hidden}

Yet it fits the described failure situation (not just the condition of  
the parts) pretty well. Possibly there is something you have missed  
with the distances involved etc.?


Best regards,
Spehro Pefhany
--
"it's the network..."                          "The Journey is the reward"
RemoveMEs...spam_OUTspamKILLspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com


2008\01\09@105942 by Alan B. Pearce

face picon face
> We've already been in touch with one of those companies- they have done
> quite a bit (though not enough) for us already.  THEY claim that this was
> a
> simple case of overvoltage- our power supply was too hot, the chips fried,
> end of story.  Nothing I have done has repeated that, however, so I'm VERY
> skeptical about it.

I guess there is always the possibility that your customers have had a
failure on their test station that is stressing your module in some way. It
may not be apparent to them if the failures don't happen until the units are
out in the field. For two different customers this is a bit of a stretch,
but ....

2008\01\09@162628 by Sean Breheny

face picon face
Hi Mike,

Have you tried ESD zapping the various inputs and outputs of the
device to see what effects you can elicit? We had a situation where an
instrumentation amp which did not connect to a user-touchable part
would latch up under a relatively mild ESD zap through the ground of
the PCB it was on. The part would then destroy itself thermally within
seconds.

Perhaps you could also try zapping some of these ICs to see if you can
get them to fail latently. An ESD test gun is about $5000 to buy, and
$500 per month to rent.

Sean


On Jan 9, 2008 9:27 AM, Mike Hord <RemoveMEmike.hordTakeThisOuTspamspamgmail.com> wrote:
{Quote hidden}

> -

More... (looser matching)
- Last day of these posts
- In 2008 , 2009 only
- Today
- New search...