Hi Dave and Werner:

Thanks for your clarifications re. the standards issue.
All right, no "BLIND" implementation. But is "a well-informed"
one more appropriate? I'd say MIL-STD-2000 did not make it because
no one could say confidently why a thruhole with 60% of 
the pad wetted with solder should be any worse than
the one with solder all around its periphery. I am only
citing a single instance where cost and cycle time could
have been reduced with some information, but was not.

My stand is this: If reliability criteria have been so well
understood as to come up with D-279, then why arent they 
incorporated into the IPC-610/J-STD-001 more obviously? 

Let us face the fact that these documents are being used
extensively as guidelines for inspection. A handbook is being
prepared about how to interpret IPC-610, and lots of inspectors
have attended short courses at EMPF about the guidelines.
When a work such as the 610 or 001 is brought into mainstream
manufacturing, it is natural for the individuals using it, 
to assume that it has been carefully documented and to place
some faith in it. To tell them that it is upto their
abilities and resources to determine whether the guideline is
applicable to them is only unfair. 

Companies are often faced with this question:
How do we ensure that the manufactured parts 
at least meet the expected operating conditions?
This then mushrooms into an argument of conservativeness 
vs. quality. Another real issue is:
With rapid design changes and shorter product lifecycles
becoming the norm, are we spending a lot of money by being
over-conservative?

About the only things that everyone agrees upon are obvious
defects such as bridging, tombstoning, misalignment,...
The real culprits that cost money to verify and are the
bigger headaches are those that affect reliability and
field performance. Every company can come out with nice
shiny joints that conform to guidelines (well not every one)
but are they reliable? This is the determining factor for
product superiority in the current marketplace. Hence, we
must not neglect it.

Every time someone out there asks a question about how to judge
conformance to specifications, it indicates lack of adequate
information. Maybe that's why TechNet exists, but there has 
to be a better way. I hope I have made the case for reliable
solder joints quite clearly. In the following para, I have
included some of my thoughts about how to meld this into
the guidelines. This is what I personally would like to see 
happening and I am eager to assist in whatever way I can.

Regardless of the particular geometry (GullWing, J-Lead, 
Leadless, BGA ...) and loading conditions, the solder 
fillet remains the only sign of electrical and mechanical 
contact. If the post-reflow appearance of the solder joint 
can be correlated with post-reliability testing results, 
there will be a solid case for the guidelines. If this 
has been incorporated into the current guidelines, then 
it is clear that the definition of Assembly Classes is 
deficient. The particular range of reliability conditions 
that a guideline applies to must be specified 
(Temperature cycling -20 to 55, 75% humidity, etc.) so that 
the users of the guideline can obtain a better idea. 
Manufacturing practices can be stated to be viable only
to the extent that product lifecycle considerations can be met.
This of course assumes that the design is flawless.
Now how often does one see that?!!!  So some confidence limits
need to be placed on the guideline. 

I am not disparaging the excellent cooperation that has produced
both IPC-610 and J-STD-001. I am only trying to say my
2 cents worth to improve its applicability. I am sure
that Solderability, because of its very definition, will 
come to be better documented than inspection criteria.
And I am thankful that this is not a flamethrowing match.

Best Regards,

Vijay Sankaran
Research Associate
Center for Integrated Electronics and Electronics Manufacturing 
Rensselaer Polytechnic Institute
Troy, NY 12180
Ph: (518) 276-2721
Fax:(518) 276-2990

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************
* If you are having a problem with the IPC TechNet forum please contact   *
* Dmitriy Sklyar at 847-509-9700 ext. 311 or email at [log in to unmask]      *
***************************************************************************