counterfeit Xilinx ?

Georg Acher wrote:
> Jon Elson <elson@pico-systems.com> writes:
> 
>>Georg Acher wrote:
> 
> 
>>>I have XCS10XL in TQFP100 from around 1999/2000 and they also have printed
>>>labels. They were obtained from the official German distri at that time
>>>(Metronik/Unique). So I guess that white ink labels are no sign of unoffical
>>>chips...
>>>
>>
>>But, I have no other examples of Spartan chips which DON'T have 
>>the Spartan(tm) marking right below the Xilinx(tm) logo.  That 
>>was one of the things that made me curious, although I don't 
>>know why a counterfeiter would miss something that obvious.
> 
> 
> My XCS10XL have 5 lines: Xilinx-logo, type, package and date code, lot code and
> speed grade. No "Spartan". Maybe it's not on the TQFP/VQFP100 package because
> auf the limited area...
> 
OK, well that's good to know.  But, the chips in question are 
TQ144, I have some just 20 weeks newer, and they have the 
Spartan logo on them, the label otherwise looks very similar.
Anyway, I sent photos to Xilinx, and am waiting to hear if they 
can tell anything from a picture.  I'm on pins and needles not 
knowing if I have some kind of process-related problem or a 
batch of bad chips.  I have shipped half of this batch already, 
so if the chips are failing slowly, they are going to be seeing 
problems too.  Not a good feeling!

Jon

On Mar 27, 10:19=A0am, Jon Elson <el...@pico-systems.com> wrote:
> Georg Acher wrote:
> > Jon Elson <el...@pico-systems.com> writes:
>
> >>Georg Acher wrote:
>
> >>>I have XCS10XL in TQFP100 from around 1999/2000 and they also have prin=
ted
> >>>labels. They were obtained from the official German distri at that time=

> >>>(Metronik/Unique). So I guess that white ink labels are no sign of unof=
fical
> >>>chips...
>
> >>But, I have no other examples of Spartan chips which DON'T have
> >>the Spartan(tm) marking right below the Xilinx(tm) logo. =A0That
> >>was one of the things that made me curious, although I don't
> >>know why a counterfeiter would miss something that obvious.
>
> > My XCS10XL have 5 lines: Xilinx-logo, type, package and date code, lot c=
ode and
> > speed grade. No "Spartan". Maybe it's not on the TQFP/VQFP100 package be=
cause
> > auf the limited area...
>
> OK, well that's good to know. =A0But, the chips in question are
> TQ144, I have some just 20 weeks newer, and they have the
> Spartan logo on them, the label otherwise looks very similar.
> Anyway, I sent photos to Xilinx, and am waiting to hear if they
> can tell anything from a picture. =A0I'm on pins and needles not
> knowing if I have some kind of process-related problem or a
> batch of bad chips. =A0I have shipped half of this batch already,
> so if the chips are failing slowly, they are going to be seeing
> problems too. =A0Not a good feeling!
>
> Jon- Hide quoted text -
>
> - Show quoted text -

Hello all,

I am a Quality manager at Xilinx, and I have asked to provide specific
guidance on the question of counterfieting.  I would like to start by
saying that the ONLY way to protect yourself is to purchase your
devices from an authorized Xilinx distributor.  A list can be found at
this link. http://www.xilinx.com/company/sales/ww_disti.htm

If you a buying outside of this channel you are taking on a fair
amount of risk.  Over the past 1.5 years Xilinx along with the Dept of
Homeland Security have become aware of an escalatng issue out of SE
Asia, where Xilinx component are being marked up for sale into grey
market channels.  We currently are looking into how to limit this
activity.

The real simple rule is that if you go to they brokers or independent
distributors you have absolutely no way of validating devices.  Even
if that entity provides what may appear to be legitimate
documentation, the supply chain is broken and the care, custody, and
control of the material is suspect.  In many cases Xilinx will not be
able to assist you in determining if those suspect devices are
legitimate or usable.

We must sustain our valid authorized sale and distribution channels
and in nearly all cases will not be able to assist or lend resources
to unvalidated supply chains.  That is not to say that we do not want
to support our customer base, but this ever escalating pollution of
the supply chain with grey market materials will ultimately strain
available resources that would historically assist in these cases.

Please help eliminate this issue by only purchasing though authorized
Xilinx sources, and you will avoid, completely, these types of issue.

Best regards

craig.taylor@xilinx.com wrote:
> In many cases Xilinx will not be
> able to assist you in determining if those suspect devices are
> legitimate or usable.

Hmmm - so these are no counterfeit at all (in the literal sense)

but relabeled devices ? - Maybe Xilinx die from Easypath,
or do Xilinx have 'creepage' issues on their test rejects ?

-jg

On Apr 2, 3:08=A0pm, Jim Granville <no.s...@designtools.maps.co.nz>
wrote:
> craig.tay...@xilinx.com wrote:
> > In many cases Xilinx will not be
> > able to assist you in determining if those suspect devices are
> > legitimate or usable.
>
> Hmmm - so these are no counterfeit at all (in the literal sense)
>
> but relabeled devices ? - Maybe Xilinx die from Easypath,
> or do Xilinx have 'creepage' issues on their test rejects ?
>
> -jg
Jim, this is all idle speculation. At least, Jon was desrcribing his
dubious source openly.
When it comes to things that you cannot easily evaluate (like FPGAs,
microprocessors, food, medicine etc) you do not want to buy them at
the flea market.
The fact that the parts look like the real thing is incidental.
I was wrong in offering help (but nobody in Xilinx has chewed me out
for it. We are a friendly company,) but I now realize the mistake).
Peter Alfke

Peter Alfke wrote:
> On Apr 2, 3:08 pm, Jim Granville <no.s...@designtools.maps.co.nz>
> wrote:
> 
>>craig.tay...@xilinx.com wrote:
>>
>>>In many cases Xilinx will not be
>>>able to assist you in determining if those suspect devices are
>>>legitimate or usable.
>>
>>Hmmm - so these are no counterfeit at all (in the literal sense)
>>
>>but relabeled devices ? - Maybe Xilinx die from Easypath,
>>or do Xilinx have 'creepage' issues on their test rejects ?
>>
>>-jg
> 
> Jim, this is all idle speculation. At least, Jon was desrcribing his
> dubious source openly.
> When it comes to things that you cannot easily evaluate (like FPGAs,
> microprocessors, food, medicine etc) you do not want to buy them at
> the flea market.

Agreed

> The fact that the parts look like the real thing is incidental.

Yes and no. If it were my company, I'd be interested in just
HOW the parts got to be there.

Do I need to tighten up the Testers, or Fabs, or do I need
to laser mark EasyPath devices ? - or are they
someone elses pulls, nicely refurbished
(possible, but rather unlikely, due to erratic volumes)

Grey market devices are usually 'easy dollars',
so a speed-grade hike is one easy action.
Intel has struggled with this.

Xilinx's call - I guess this must be in the noise floor for them.

-jg

On Apr 2, 4:59=A0pm, Jim Granville <no.s...@designtools.maps.co.nz>
wrote:
> Peter Alfke wrote:
> > On Apr 2, 3:08 pm, Jim Granville <no.s...@designtools.maps.co.nz>
> > wrote:
>
> >>craig.tay...@xilinx.com wrote:
>
> >>>In many cases Xilinx will not be
> >>>able to assist you in determining if those suspect devices are
> >>>legitimate or usable.
>
> >>Hmmm - so these are no counterfeit at all (in the literal sense)
>
> >>but relabeled devices ? - Maybe Xilinx die from Easypath,
> >>or do Xilinx have 'creepage' issues on their test rejects ?
>
> >>-jg
>
> > Jim, this is all idle speculation. At least, Jon was desrcribing his
> > dubious source openly.
> > When it comes to things that you cannot easily evaluate (like FPGAs,
> > microprocessors, food, medicine etc) you do not want to buy them at
> > the flea market.
>
> Agreed
>
> > The fact that the parts look like the real thing is incidental.
>
> Yes and no. If it were my company, I'd be interested in just
> HOW the parts got to be there.
>
> Do I need to tighten up the Testers, or Fabs, or do I need
> to laser mark EasyPath devices ? - or are they
> someone elses pulls, nicely refurbished
> (possible, but rather unlikely, due to erratic volumes)
>
> Grey market devices are usually 'easy dollars',
> so a speed-grade hike is one easy action.
> Intel has struggled with this.
>
> Xilinx's call - I guess this must be in the noise floor for them.
>
> -jg- Hide quoted text -
>
> - Show quoted text -

Let me say this
Xilinx is investigating the issue of counterfeit devices.  We have
taken steps to increase awarness of the issue with other companies in
the industry.  We are working with several government agencies on the
issue.  Many time we see that these parts are remarks.  In some cases
they are remarked devices that originally were not even Xilinx
devices.  Often they are really old devices that are marked to look
young, faster than they are, or of a higher grade then they were.  I
have not seen pollution of reject materials from our supply base.
There are very tight controls on reject and scrap materials, with
several levels of validation.  Xilinx has personnel (XILINX Employees)
in each of our factory partners.

We are also looking at the issue proactively.  We are investigating
several new marking protocols and security features that would be
nearly impossible to duplicate or replicate. Our intent is to increase
the amount of time and money it would take to remark devices to look
like true Xilinx, but this is still off in the future.

I recieved a phone call from someone reading my last post.  They were
not very happy about my authorized channel statement.  I provide this
information for peice of mind.  If you buy parts from independent
distributors that is not my issue.  If the independent Disti is
legitimately buying for an authorized Xilinx channel and can
demonstrate a clean line of custody you may be Okay.  It is your own
comfort level with that transaction.  However, some folks choose to
buy parts from the broker market.  There are some good resources in
this market, but the devices that are coming in an out of this have a
much higher rate of legitimacy issues.  How many hands to you want
your parts going through before you put them in your end product?  It
is really a question of Care, Custody, & Control.  If you are buying
for the Xilinx Disti, the CCC line is clear and easily supportable buy
Xilinx warranty.  Then when you get to 3rd, 4th, 5th partys, you have
lost whatever CCC you might have warrented with the 2nd party
purchase.

I hope that this help further clarify.

Craig wrote:
<snip>

I recieved a phone call from someone reading my last post.  They were
not very happy about my authorized channel statement.  I provide this
information for peice of mind.  If you buy parts from independent
distributors that is not my issue.  If the independent Disti is
legitimately buying for an authorized Xilinx channel and can
demonstrate a clean line of custody you may be Okay.  It is your own
comfort level with that transaction.  However, some folks choose to
buy parts from the broker market.
Well, the problem is this board was originally designed with a part I 
could get from an authorized distributor, I believe for about $15.  (I 
could be mis-remembering this price.)  Now, I am faced with $60 price 
and multi-hundred piece minimums from all auth. dists!  I am a small 
manufacturer, just trying to stave off disaster while I'm redesigning
with a newer part.  So, I am caught between a rock and a hard place.

Jon

On Mar 23, 6:40 pm, Jon Elson <el...@pico-systems.com> wrote:
> I got a batch of "Xilinx" Spartan XCS30 FPGAs from a Chinese
> seller, and am having problems with random failures at first
> power up.  Sometimes it is a stuck I/O pin, sometimes a failure
> to configure.  I first thought maybe we had an ESD problem, but
> I'm now thinking these may be counterfeit.  They have white ink
> printed labels on the front, whereas other Xilinx chips have
> laser-etched labels.  Also, these Spartan chips don't have the
> Spartan logo just below the Xilinx logo, like my other Xilinx
> chips.  Anyone have any comments on this?
>
> Jon


While I have never had to use them for counterfeit detection, I have
use Process Sciences for process control, inspection and rework for
several years now. I have always been very happy with their work. They
have some information about their counterfeit detection services here:

http://www.process-sciences.com/services/counterfeit_detection.asp

One of the trade journals I get has been running a series of articles
about counterfeit parts. I think it was either EE Times, or EDN.  Some
of the things that they mentioned include remarking to a better speed
grade, a lower power version, or a larger memory size.

Regards,

John McCaskill
www.FasterTechnology.com

John McCaskill wrote:
> On Mar 23, 6:40 pm, Jon Elson <el...@pico-systems.com> wrote:
> 
>>I got a batch of "Xilinx" Spartan XCS30 FPGAs from a Chinese
>>seller, and am having problems with random failures at first
>>power up.  Sometimes it is a stuck I/O pin, sometimes a failure
>>to configure.  I first thought maybe we had an ESD problem, but
>>I'm now thinking these may be counterfeit.  They have white ink
>>printed labels on the front, whereas other Xilinx chips have
>>laser-etched labels.  Also, these Spartan chips don't have the
>>Spartan logo just below the Xilinx logo, like my other Xilinx
>>chips.  Anyone have any comments on this?
>>
>>Jon
> 
> 
> 
> While I have never had to use them for counterfeit detection, I have
> use Process Sciences for process control, inspection and rework for
> several years now. I have always been very happy with their work. They
> have some information about their counterfeit detection services here:
> 
> http://www.process-sciences.com/services/counterfeit_detection.asp
> 
> One of the trade journals I get has been running a series of articles
> about counterfeit parts. I think it was either EE Times, or EDN.  Some
> of the things that they mentioned include remarking to a better speed
> grade, a lower power version, or a larger memory size.


This was a very small batch of parts, so it doesn't make much sense to
spend a lot of money on it.  I still have no idea whether there was any
funny business, or these parts were just mishandled in some way to cause 
them to deteriorate.  A number of them would not do the master mode 
configuration, so that can't be a speed grade problem.  But, half of 
them work!  I still don't know if I have some kind of process problem 
here in my shop, but I am coming around to think it is not something I
caused here.

Jon

Jon Elson wrote:
>
> This was a very small batch of parts, so it doesn't make much sense to
> spend a lot of money on it.  I still have no idea whether there was any
> funny business, or these parts were just mishandled in some way to cause
> them to deteriorate.  A number of them would not do the master mode
> configuration, so that can't be a speed grade problem.  But, half of
> them work!  I still don't know if I have some kind of process problem
> here in my shop, but I am coming around to think it is not something I
> caused here.
>
> Jon

Some ideas:
Did you bake the parts before assembly?
Are you hand-soldering the deviecs?