For what it's worth, here are my two cents worth.

<< > 
 > What are expirences with the test time length when testing boards with =
 > only pattern and with soldermask and after mounting with components.
 >10-20 minutes depending on the size of the boards and the components.
 > 
  > Here are some of my questions.
 > 
 > A.
 > The mil spec. requires a minimum of one minute testtime, but this is =
 > normally not enough.
 > What are the correct test time ? Or rather.. How du you determine the =
 > right test time ??
 > 			   	-XX minutes by experience ?
 > 				-time until stable reading ?  YES.
 > 				-time until raise in conductivity per =
 > minute is                                                    under some =
 > sort of limit.?
 > 				- Or WHAT ???
 
The MIL-SPECs that I am most familiar with have a minimum test time of 10
minutes.  In practice, solution conductivity increases with time due to two
effects: from the board contributions, both ionic contamination and mask
breakdown; and from carbon dioxide from the air mixing in with the solution.
 The latter effect is more prevalent on ionic cleanliness testers that use
heated solutions and spray agitation.  Therefore, no one tests longer than
they have to in a production setting, since readings would only rise.  If you
are doing materials or process evaluations, then perhaps a 15-20 minute test
time would be acceptable.  You might also use the autoterminate function (no
change in readings over 5 minutes) to save time.  I have known some
evaluations that went for an hour or longer.

> Our problem is, that some soldermasks start to decompose when testing time
=
 > is more than 5 minutes. And by decomposing, the soldermask adds to the =
 > conductivity, witch in that way continue to raise ....... and so forth =
 > until the test reatch the fail limit.
 
* Are you sure that the mask is actually decomposing or that the garbage,
* i.e. ionic contamination, incorporated into the board is slowly leaching
 *out?  The later is generally my experience.
 
I would tend to agree with Karen, that contamination is more often due to
surface or absorbed contaminants (e.g. flux) than to mask decomposition.  You
do need to be aware that both can occur.  Mask decomposition, in my
experience, happens when you A) have an incomplete cure of the mask; B) are
using a dry-film mask that is undercured, or C) when you have chosen a mask
with very poor thermal resistance.  My upcoming column in Circuits Assembly
magazine discusses our "Mask from Hell" project in which a cheap mask was so
degraded by the fabrication HASL process that it ceased to be effective.

One of the characteristics that I see with a decomposing mask is that the
ionic levels will keep rising with time.  If you suspect a decomposing mask,
run a 60 minute ionic cleanliness test.  If the levels keep climbing, it is a
good indication that you may have a decomposing mask.  In this case, the mask
is probably emitting a conductive organic material, rather than a halide.  Be
careful though, this could also be indicative of a high level of absorbed
HASL flux.  Ion Chromatography would be useful in determining which was the
case.

 > Do you have any experience in different soldermask systems resistance =
 > against the solvents used in cleanliness testing.

There are so many masks on the market now that it would be difficult to say.
 In general, I have found dry films to be more susceptible to solvents than
LPIs.  I think success or failure with your mask is more dependent on your
skill in processing than in the inherent materials characteristics.  
First, I would recommend using solder masks qualified to IPC-SM-840.  
Second, look at how many passes to soldering temperature the mask can
withstand without degradation.  Five passes or more, use it.  3-4 passes,
treat with caution.  Less than three, shoot the sales rep trying to foist off
this garbage on you.

 > Fr-4 (GF) polyimide (GI) bare boards:
 > 
Again, more process related than materials related.

 > We have seen some examples, that IPA to some extent atacks the lamminate =
 > When testing  the test does not come to a stable level, but are continuing
=
 > to raise, but in a decreasingly speed rate.
 > GF laminate is seen to be more vulnerable than GI laminate.
 > 
 > Have you seen laminate beeing *affected* in the described way, during =
 > cleanliness testing ?
 
 *When I see this sort of behavior, I attribute it to ionics leaching
 *slowly out of the laminate or mask.  Do you have any IR data to indicate
 *that your mask or board is being attacked.
 
I would tend to agree.  If you had the steadily increasing case, rather than
the approach to steady state, I would say a problem with the mask.  Since you
do approach a stable final reading, you have a case of ionic contamination,
either on the surface or in the subsurface.  Keep in mind that you have
solubility concerns as well.  Room temperature tests tend to bring ionic
materials into solution at a slower rate than using heated solutions.  You
may have a contaminant that is slow to dissolve, but I think it more a case
of absorbed flux.  GI laminate has a higher glass transition temperature than
does GF.  When a laminate is raised above it's glass transition temperature
(Tg) it tends to absorb surface residues, such as flux during a HASL
operation.  Since GI has a higher Tg, it absorbs less, so your effect is less
pronounced.

 > What are your experience with the in MIL-P-55110 listed different =
 > equipments for cleanliness testing.
 > 
 > Can you bring me in contact with the companies selling the in MIL-P-55110E
=
 > table IX listed equipments?
 > 
 > E
 > Does any one know some good litterature on this subjekt ?
 
* Jack Brous' White Paper, that I will post separately.  It has good 
* references also.

In addition, anyone who does ionic cleanliness testing needs to get EMPF
Report RR0013: An In-Depth Look at Ionic Cleanliness Testing.  Contact Toni
O'Connor (317) 655-3673 X123 for a copy.

 > What are your experience with testing boards after mounting with
components=

Mounting components is a good thing.  Bare boards function much better with
components installed.

 >  ?
 > What spec no. talks about testing after mounting?
 > 
Depends on what kind of testing you are talking about.  Most burn in tests
will be determined by the assembler or by agreement between customer and
vendor.  If you are looking for environmental test methods, I suggest looking
at MIL-STD-202, or MIL-STD-883.

* I hope I have helped some.  I think it would be useful if Dave Hillman
* and Doug Pauls threw their two bits in also, I might be biased since
 *we make Omegameters.

Hey, my pleasure.  Dr. Tellefsen is obviously a brilliant person; she agrees
with me <grin>.

Doug Pauls
CSL
[log in to unmask]

***************************************************************************
* 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.        *
***************************************************************************