As an employee of a power company, am very 'sensitive' to control of
procurement process. Build to print is not a big issue, but component
procurement has always been a concern. Have just returned from some of
employer's Asia factories where a considerable amount of time was spent
going through supply bins. 

While at the Shenzhen site, observed a CAB's factory audit. What a load of
doo-doo. The only time the auditors do anything in an Asian country (other
than Japan) is when the agency wants to bring in more revenue - which
results in some rather creative variation notices.

But the TJ, Mexico factory, where most of the custom stuff is done, is less
problematic because:
1. They are next door to engineering.
2. The culture supports a different set of ethics.
3. Agency FUS audits tend to be more legitimate.

Moral of the story -> know your suppliers and be aware of the culture of
both the engineering group and the production sites. If your parts change
structure or behavior - contact the engineering group first, because the
factory has little control and even less interest in fixing the customer's
problem. The designers care - it is their reputation and their baby.

Brian 

-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of
McInturff, Gary
Sent: Wednesday, October 31, 2012 7:47 AM
To: 'John Woodgate'; '[email protected]'
Subject: RE: [PSES] Possible Counterfeit EMC Components?

Morning Ed, John has it correct. A company that produces thousands or even
hundreds of thousands generally scales up its builds and inventory of parts
as the design matures, but nothing can go to production release unless the
company is sure that the design meets its goals. You wouldn't want to have
order 10,000 input line filters only to find out that the compliance test
identified a problem with that part and it needed a change. The logistics of
returning and purchasing the correct parts is quite expensive. Obviously the
design team tries to verify as much as early as possible, but it can't (or
certainly shouldn't) go full blown production until after Engineering design
verification is complete. There should be no known issues on a design when
its thrown over the wall to mass production - and manufacturing shouldn't
accept it until they are convinced that the design is complete. As I read it
- the designer did exactly what he should have done, his manufacturing arm
doesn't!
 
  seem tied into his quality or design process or Design change request
process and changed his design to reduce cost with no regard to performance.
This happens a lot with off shore suppliers that aren't poorly monitored. I
had a SMPS from a very large vendor that for the first two years we used the
supply performed well. It had been used in several designs and had gone
through a series of compliance tests successful. A new end product design
which also used the power supply was going through final tests and the
supply failed an immunity test. Since the supply had been used and verified
in many other designs I assume a bad unit. After swapping it out a couple of
more times it was obvious that something had changed. When I called the
supplier to inquire I got "Oh you wanted one of those supplies".  They had
made some cost reductions without telling the customer base and they knew
the supply no longer met the requirements but shipped it without changing
the part number etc. T!
 
 here was no way that our purchasing department could tell anything was
 different. The vendor knew it but not his customers, and we certainly
weren't the only customer of this "off the shelf" supply. 
The cost of purging inventory and the line shutdown while correct power
supplies were shipped was very expensive for us. It might be noted that the
vendor saved themselves lots of money, but they certainly didn't pass on the
cost savings nor did they reimburse us for the cost of the inventory purge
and lots production time. 
 Some countries are almost famous for making cost reductions without
verifying them. I think someone earlier pointed out that their supplier
thought they could cut the cost of a motherboard if they took off all of
those pesky "bypass" caps etc. As a military guy you likely familiar with
MIL-TFP-41 (Make it like the F*&#$ print for once!)
Sometimes you don't get what you pay for.
To amplify Johns comments just a bit, one should do production audits but
that can be prohibitively costly to do effectively when lots of product
lines are involved. 
Gary


-----Original Message-----
From: John Woodgate [mailto:[email protected]] 
Sent: Wednesday, October 31, 2012 3:55 AM
To: [email protected]
Subject: Re: [PSES] Possible Counterfeit EMC Components?

In message <[email protected]>, dated Wed, 31 Oct 
2012, Ed Price <[email protected]> writes:

>I thought that compliance testing was supposed to be done on the 
>"as-shipped" product, not the designer's prototype.

Neither: the prototype is not representative and 'as shipped' is too 
late. Maybe the 'pre-production' phase has been eliminated from modern 
manufacturing, but it used to produce 5 to 10 sample products using the 
purchased parts and built under as close to production conditions as 
possible. In those days, some moulded parts might be replaced by 
fabricated parts, but with 3D printing....

These samples not only increase confidence in the design, and eliminate 
bugs, but can be used for compliance testing.

-
----------------------------------------------------------------
This message is from the IEEE Product Safety Engineering Society emc-pstc 
discussion list. To post a message to the list, send your e-mail to 
<[email protected]>

All emc-pstc postings are archived and searchable on the web at:
http://www.ieee-pses.org/emc-pstc.html

Attachments are not permitted but the IEEE PSES Online Communities site at 
http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used 
formats), large files, etc.

Website:  http://www.ieee-pses.org/
Instructions:  http://listserv.ieee.org/request/user-guide.html
List rules: http://www.ieee-pses.org/listrules.html

For help, send mail to the list administrators:
Scott Douglas <[email protected]>
Mike Cantwell <[email protected]>

For policy questions, send mail to:
Jim Bacher:  <[email protected]>
David Heald: <[email protected]>

Reply via email to