Ben,

I am familiar with the Agilent family of ICT, AOI, and X-ray
Laminography systems.  The three systems, although each expensive in
their own right, are not mutually exclusive nor totally redundant in
test coverage.  All three are complimentry and many test plans include
two of the three and even all three based on the shop's capabilities,
the product requirements, and ultimately what degree of coverage the
customer needs and what level of cost the customer is willing to bear.

ICT tests the board for electrical opens and shorts, checks analog
values of nets with set tolerances where applicable, can validate
checksums of programmables (I think?), and can program eeproms (I'm sure
there's more but these are the biggies).  ICT does not validate product
functionality or performance.  There is some advanced features such as
test jet, JTAG, and boundary scan that can help when conventional ICT
access for parts of the board is nonexistant.

The growth of X-ray laminography has gained popularity for a few
reasons.  Although still expensive relative to the other two systems
mentioned above, the cost of ownership of 5DX has reduced considerably
over the last 5 years or so primarily due to increased speeds of the
system, better programming tools which have substantially reduced
programming develpoment time, complexity, and cost, and longer lasting
x-ray tubes.  I don't think the machines themselves have got much less
expensive.  In some people's minds (not necessarily mine), the
proliferation of BGA components begs an automated x-ray inspection
solution.  Laminography paints a more complete picture about what is
going on at the board/ball interface than 2D x-ray but is still not 100%
definitive (i.e. will not detect a collapsed ball on a nonsolderable
pad).  Due to the ever increasing trends of greater circuit density and
node count ICT test coverage is shrinking.  5DX (and AOI) is a means to
increase test coverage.  RF products as a family are a classic example
of products which have poor test coverage where 5DX can provide valuable
process feedback to correct manufacturing problems and of course detect
failures that are reworked prior to shipment.  Not everyone can afford
5DX and obviously the industry has built billions of assemblies through
the years without it.  It is my humble opinion that the main reason you
design 5DX into your test plan is to meet stringent product reliability
requirements and/or to compensate for poor ICT coverage.  ICT will tell
you that your circuit is wired up but it won't tell you if you have low
volume joints on your QFP's or BGA's that may result in field failures.
Another benefit to the manufacturer is that 5DX can provide "process
indicator" feedback up your manufacturing line by setting solder joint
thresholds inside workmanship standards but just outside of preferred
limits, just like the IPC 610's "Acceptable", "Process Indicator", and
"Rejectable" criteria system.

Since I'm writing a book here anyway......I know you didn't ask about
AOI but I can't help myself :) .  AOI of course can only provide a
line-of-sight inspection.  It has the benfit of being able to provide
manufacturing quality data up the line earlier in the process than the
other two.  Depending on your floor's flows there could only be minutes
between AOI and 5DX/ICT or it could be hours or even days.  Most AOI
systems provide the benefit of being able to actually check part marks
and find misloaded parts (primarily actives), backwards parts, misplaced
parts, and missing parts.  Many are selling their AOI system's solder
joint inspection capability but it certainly is not as robust as the
5DX's and can't deal with array packages and SOJ/PLCC's where the joints
are obscured by the package.  Depending on the AOI machine and where you
place it in your line you likely won't be looking at through hole
devices and joints with AOI where 5DX looks at everything.

Agilent publishes a really nice graphic that shows three
overlapping/interlocking rings and the fault coverage each of the three
systems offers.  It's a nice way to visualize how the systems compliment
each other and where they overlap each other.  One thing that would make
their graphic more useful is if they identified the reliability risk
natured faults versus the purely functional related faults but you can
likely figure this out when you see the graphic (i.e. low and no volume
solder joints).  Contact your friendly local Agilent salesman for this
slide.

Good Luck,

Glenn Woodhouse

-------------------

Date:    Fri, 1 Feb 2002 14:43:17 -0500
From:    "Royal, Ben" <[log in to unmask]>
Subject: 5DX vs. ICT

I am a newbie to the SMT world and would appreciate some assistance on a

question that came up this afternoon. I hope I get the question right; I

used to build rockets.

One of our customers wanted to know if we had any experience re: which
is
better - a in-line 5DX-to-FCT process or an ICT-to-FCT process. Not a
lot of
response around the table from the more experience hands. I assume
"which is
better" means which process is a more efficient inspection setup for
locating nonconformances.

Would appreciate any comments. Now if you would excuse me, I am going
out on
the floor to find a 5DX and figure out what the devil that is.

Ben Royal

---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------