Searching \ for '[OT] Failure Analysis Help??' 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=failure+analysis
Search entire site for: 'Failure Analysis Help??'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Failure Analysis Help??'
1999\03\18@213359 by Ed Edmondson

picon face
I have a child's toy which has a 50% failure rate.  I am trying to determine
the reason for the failure.  Why?  Something to do.... My four children want
to know why the toys are dying like fly's.

I have two separate integrated circuits contained on two COB's (Chip On
Board).  A COB actually has the integrated circuit die soldered to a small PCB
and covered with some form of molding compound (usually black).  I need to
dissolve this compound to see the die under it.  Then I can get
photomicrographs made of each die to determine if ESD or some other cause
killed the toy.  Anyone know what this compound might be?  How about
suggestions on what solvent might dissolve the compound without damaging the
device under it?  Any suggestions??

I also have several transistors (??) with what appear to be possibly house
markings.  They have the following numbers; S9012, S8050, S9013, and SS9015.
They are in leaded TO-92 packages.  The S9013 is also marked KSP2222A in
another toy of the same make.  This indicates these are probably standard
devices.  The only other markings are the date code on the devices (D-847,
H-844, D-839, ect.)

Any suggestions or directions where to get my questions answered would be
appreciated.

Thanks,
Ed

1999\03\18@225151 by Mike Keitz

picon face
On Thu, 18 Mar 1999 21:29:33 EST Ed Edmondson <spam_OUTEaejrphdTakeThisOuTspamAOL.COM> writes:
>I have a child's toy which has a 50% failure rate.  I am trying to
>determine
>the reason for the failure.  Why?  Something to do.... My four
>children want
>to know why the toys are dying like fly's.


>  They have the following numbers; S9012, S8050, S9013, and
>SS9015.
>They are in leaded TO-92 packages.  The S9013 is also marked KSP2222A
>in
>another toy of the same make.  This indicates these are probably
>standard
>devices.  The only other markings are the date code on the devices
>(D-847,
>H-844, D-839, ect.)

The S numbers are likely the CS-series made by National Semiconductor, or
a second source.  They are ordinary NPN and PNP transistors.

I suspect most toy malfunctions are related to broken wires, bad solder
connections, cracked circuit boards, broken LCDs, waterlogged speakers,
etc. (and of course dead batteries).  Did you look for the obvious first?
If the COB's weren't assembled properly, they could fail but if done
right they should be more reliable than most of the other parts.  One
possible COB failure is broken bond wires.  You can check for those by
looking for the parasitic / protection diodes with an ohmmeter without
removing the plastic.



___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]

1999\03\19@050229 by g.daniel.invent.design

flavicon
face
Don't buy any more.
Copying them (pretty obvious considering the lengths you want to go to)
will be illegal.

Reverse engineering them is a different story however...


Ed Edmondson wrote:
{Quote hidden}

--
Steam engines may be out of fashion, but when you consider that an
internal combustion engine would require recovery of waste heat by
transfer just before top dead centre then fashion becomes rather
redundant, USE STRATIFIED HEAT EXCHANGERS ! and external combustion.

You heard it first from: Graham Daniel, managing director of Electronic
Product Enhancements.
Phone NZ 04 387 4347, Fax NZ 04 3874348, Cellular NZ 021 954 196.

1999\03\19@063144 by paulb

flavicon
face
Ed Edmondson wrote:

> I have a child's toy which has a 50% failure rate.  I am trying to
> determine the reason for the failure.

 Those Furbies look more and more like a rip-off!
--
 Cheers,
       Paul B.

1999\03\19@080332 by Nicholas Uloth

flavicon
face
At 10:30 PM 3/19/99 +1000, you wrote:
>Ed Edmondson wrote:
>
>> I have a child's toy which has a 50% failure rate.  I am trying to
>> determine the reason for the failure.
I'm not sure tracing the paths on a microlithigrapgh is the first place to
look if I wanted to find out why they died. Much more likely to look at
those places
that are hand soldered.

Maybe if you payed extortionate rates for furbies you might be provoked
into doing this :-)

If you wanted to copy it then reverse engineering is going to be cheaper.

>  Those Furbies look more and more like a rip-off!
They are giving them away with mobile phones here in Australia, apparently
they arent such a
hot seller here.


Nic

1999\03\19@110329 by Ed Edmondson

picon face
Furby's?? Did anyone say Furby's?? ;-)

Yes, that is what they are.  No, I didn't pay more than the retail price of
$27.99 US (that is times 4).  And yes it seems they are a rip off.  My current
mortality rate is 50%.  A little high??

I have completely dismantled (I call it an autopsy) the first fur-ball and the
workmanship and quality of parts is really pretty good.  For a toy that is...

I still want to finish my autopsy though! I think I may have uncovered a small
design problem??

If they are giving these away with phones, I would be buying a lot of phones
and exporting these fur-balls to the US.  People are still paying through the
nose for these things.  The lowest price (not in a retail store) is between
$60.00 and $80.00 US dollars.

Ed

1999\03\19@113014 by John Mitchell

flavicon
face
On Fri, 19 Mar 1999, Ed Edmondson wrote:

> Furby's?? Did anyone say Furby's?? ;-)

check out this Furby Autopsy page:
       http://www.phobe.com/furby/


- j

1999\03\19@150233 by Ed Edmondson

picon face
Been there! Done that! Got a T-shirt too! :-)  I have already been trading
info with them.  However, they are no longer making any progress.  I want to
go further.  I want to know why they are on the endangered species list. ;-)

Ed

1999\03\20@121048 by Al Williams

flavicon
face
>Ed Edmondson wrote:
>>
>> I have a child's toy which has a 50% failure rate.  I am trying to
determine
>> the reason for the failure.  Why?  Something to do.... My four children
want
>> to know why the toys are dying like fly's.

I used to do FA for Motorola professionally. Let me caution you not to
attempt this without a real lab. All the chemicals you
need to use to successfully decapsulate a part are nasty, toxic, and in some
cases illegal for you to posses without permits.

There are many things this type of analysis won't tell you anyway -- like
what charge are in EEPROM cells, for example. Now electron beam stroboscopy
(something we were working on as early as 1985) might be a better bet for
that. That's big bucks, however.

Search for "hack furby" on the Web and you will see there is a nice site
done with schematics and everything people have figured out about these.

Regards,

Al Williams
AWC
*Floating point math for Stamps, PICs, or any microcontroller at
http://www.al-williams.com/awce.htm

1999\03\20@150426 by Ed Edmondson

picon face
Al,

I also worked for TI, Rockwell, BEI, ect.  I just can't remember what the
chemicals are.  We used to do RE and sometimes the device part number is on
the die mask.  There are a lot of us who really want to know what form of
intelligence this critter contains.  Is it custom, ASIC, DSP (My Vote) or a
low end CPU?

I have all the info available from the web (that I could find).  Most of the
info and articles are non technical though.  I am trying to tap the expertise
of all the other techs and engineers to get opinions, help, suggestions,
comments, ect.

I realize this is probably not PIC related and I apologize for the OT nature
of the subject.  But it is people like the PIC listers who have the bulk of
the curiosity and intellect.

Regards,
Ed

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...