[s-cars] POS Source?

Fri Mar 12 17:19:01 EST 2004

I fail to understand how a wire in a device that "doesn't dissipate much
heat" and has a "low thermal conductivity" case could be thermally
shocked by engine washing, given that the engine should be off anyway,
and the device is partly protected by a plastic cover.

On the other hand, a soft mounted FET tied to a hard mounted connector
via a thin wire might put stress on the wire when vibrated for a long
time.  As the wire structure at the point of constant microbending
changes its metalurgy, high currents accelerate the weak spot.

We also shouldn't ignore the possiblility that the wire from the POS to
a coil was momentarily shorted somehow.

kirby

Vincent Frégeac wrote:
> 
> Hi Mike,
> 
> In fact, the failure analysis you made also prove the failure mode is
> not overheating. An overheating problem would have lead to an internal
> failure in the FET, not a connection failure. The thermal paste may be
> the kind of over engineering which allows some listers to go close to
> 400hp without modifying the internal components of the engine.
> 
> I would rather think about thermal shock as several listers have
> experienced POS failure when washing the engine, me included. Vibrations
> may also be a cause as solder joints do not like them. The paste you've
> seen inside is commonly used on circuit boards to avoid vibration on the
> heavy components like capacitors.
> 
> At the end, I think the thermal paste is a good safety measure. It may
> be overkill but it cost almost nothing and just add a few seconds to the
> POS swap. I was mainly reacting about the heatsink.
> 
> Vincent.
> 
> -----Message d'origine-----
> De : Mike Sylvester [mailto:msylvester at verizon.net]
> Envoyé : 11 mars 2004 22:39
> À : Vincent Frégeac
> Cc : s-car-list at audifans.com
> Objet : Re: [s-cars] POS Source?
> 
> If those temp readings were accurate, why then did Audi even bother with
> the
> paste?
> There must have been some concern.
> A measurement taken with an infrared gun will be reading the plastic
> case
> which doesn't conduct very well.
> A better test would be to unbolt one, run the car and measure the temp
> of
> the metal backing.
> 
> If this failure is not due to overheating, then these are really POS
> because
> too many of these have failed.
> I had one fail at 96k miles.  Most people might think that a failure at
> 96k
> is common, but not for electronic parts.
> It probably is only about 3,000 hours.  The MTBF of this part is
> probably
> something like 200,000 hours.
> 
> I will admit, it doesn't look like a straight up overheating problem.
> Like
> I said in a previous post, it could be fatigue from repeated temp
> extremes.
> I really would like to open up a few more.
> 
> Mike
> 
> P.S.
> I added the FA photos to my site.
> 
> http://mysite.verizon.net/vze6ngsr/S6POS.htm
> 
> ----- Original Message -----
> From: "Vincent Frégeac" <vfregeac at sympatico.ca>
> To: "'Igor Kessel'" <igor at s-cars.org>; "'Mike Sylvester'"
> <msylvester at verizon.net>
> Cc: "'Miller, Casey'" <Casey.Miller at janus.com>;
> <s-car-list at audifans.com>
> Sent: Thursday, March 11, 2004 10:06 PM
> Subject: RE : [s-cars] POS Source?
> 
> >From times to times, the list tends to go a bit crazy about "preventive
> maintenance". So, a few facts to keep your wallet a bit thicker:
> - Temperature on original POS without original crappy thermal paste:
> 107F
> - Temperature on new POS with new thermal paste: 98F
> This was in October, 70F ambient temperature after a 45min drive, engine
> still running, measurement made with the calibrated infrared gun we have
> at the plant. I've yet to see an electronic component with a recommended
> temperature below 160F. So I think adding a heat sink behind the POS is
> a bit overkill. Even the thermal paste seems a bit overkill as the low
> power loss of the POS seems to be already very well dissipated with a
> simple metal to metal contact. We're not talking high power class A
> amplifier here, but just a switching device.
> 
> My 0.02$
> 
> -----Message d'origine-----
> De : s-car-list-bounces at audifans.com
> [mailto:s-car-list-bounces at audifans.com] De la part de Igor Kessel
> Envoyé : 10 mars 2004 10:08
> À : Mike Sylvester
> Cc : Miller, Casey; s-car-list at audifans.com
> Objet : Re: [s-cars] POS Source?
> 
> Mike Sylvester wrote:
> > I just bought one yesterday from The Parts Connection.  It cost me
> $199.
> >
> > When changing your bad POS I recommend applying new thermally
> conductive
> > paste to the other POS as well.
> > Actually, I recommend this as a preventative maintenance on working
> POS.
> > I found that the factory past was dried up and it didn't appear that
> there
> > was good contact between the POS and the bracket.
> >
> > These parts are failing long before they should.  However they are not
> > manufacturing defects since they would have failed much sooner if that
> were
> > true.
> > It is quite possible that they are overheating.  When I replace mine
> > tonight, I also installed a heat sink to the bracket behind each POS.
> > http://mysite.verizon.net/vze6ngsr/S6POS.htm
> >
> > I should note that I sandblasted the bracket and when I repainted it I
> > masked off the contact area of the POS and heatsink.
> > The heatsink may be overkill, but it can't hurt.
> >
> > Mike Sylvester
> > '95 S6 avant w/ POS running cool
> > http://mysite.verizon.net/vze6ngsr/index.htm
> 
> Mike,
> you are a damn GENIUS! Thank you so much for the idea! I have the paste
> and I have a couple of old P-100 processor sinks. And I am just about to
> 
> start fusing the coils individually, having recently witnessed Stefan
> flatbed his car only because of the stupid short in one of the coils.
> I'll do everything in one go.
> 
> Thx again. The timing of your post couldn't be more perfect.
> 
> --
> Igor Kessel
> two turbo quattros
> _______________________________________________
> S-CAR-List mailing list
> S-CAR-List at audifans.com
> http://www.audifans.com/mailman/listinfo/s-car-list
> 
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.594 / Virus Database: 377 - Release Date: 24/02/04
> 
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.594 / Virus Database: 377 - Release Date: 24/02/04
> 
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.594 / Virus Database: 377 - Release Date: 24/02/04
> 
> 
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.594 / Virus Database: 377 - Release Date: 24/02/04
> 
> 
> _______________________________________________
> S-CAR-List mailing list
> S-CAR-List at audifans.com
> http://www.audifans.com/mailman/listinfo/s-car-list