OK, now for the non-smart-ass answer, oops, uhhhhh smart-alec answer (don't
want to get myself into MoonMan class trouble).
Ioan,
You ask all the right questions, but it is an extremely board topic. For
each question you ask, we would need to ask several questions to narrow the
field for answers.
1. what reliability concerns should I have with automotive applications
Any reliability question should start with answering the basic questions of
how long must it last (design life), in what kind of environment (under the
hood or inside the car), and what are the consequences of failure (annoyed
people vs dead people)?
A proper reliability statement would be that the unit must last for five
years in an under-the hood environment, Upper Midwest (snow, salt, cold),
with no failures (MTBF of XXX hours), based on 8 hours per day use.
2. any special coating material more suited for automotive
That depends. Will the unit be in a hermetic enclosure or will conformal
coating be the only line of defense against the environment. Will you be
doing rework and repair on the unit or will it be throw away? If so, go
with an acrylic. If not, consider a urethane or silicone. But if you go
silicone, you really need to optimize your other materials around silicone
chemistry. Silicones do not play well with the other childern. Will the
coating have to be protection against fossil fuel combustion products?
Silicones tend to be porous to H2S and SOx. You have to answer the
question of exactly what do you expect the coating to protect against, in
what concentrations, and for how long? Get a copy of the most recent draft
of IPCs Coating Handbook (nearing publication). That goes into great depth
on coating issues.
3. except for the dendritic growth, any other hazard
Ionic residues in general. Unactivated flux, or inadequately actived flux
can be conductive, and so a problem. See below.
4. the Nortel analysis bulletins for both materials show the residues are
inert. should I worry about the flux residues or not? Can the assembly
process make them less inert?
I would give the caveat that they are inert if properly activated and
processed. Process these fluxes (or any of the low residues for that
matter) too cold or too fast (inadequate activation time or temp) and you
have a problem. Process them too hot or too slow, and you can chemically
degrade the flux residue and get pyrolysis residues that are corrosive.
Process them in the window. Fine. Process them out of the window,
problems. The fun is finding the windows and the extent of the windows.
5. if the flux residues are inert, what could screw up the ionic tests?
Dave Hillman breathing into the Omegameter would be one source, at least
here. Provided that you can keep Dave out of your manufacturing floor, you
have to look at the ionic test, and I presume you mean the ROSE/SEC test,
and look at what your signal is telling you. Many low solids flux residues
will dissolve in hot isopropanol / water solution, making the solution more
conductive. They are inert on the board, but your signal indicates a dirty
surface. If you use the ROSE test with low solids fluxes, throw out all
old pass fail numbers in all the old specs, and figure out for yourself
what a reasonable number is. Determining what that reasonable number may
be is hard, but it can be done. I have probably discussed it before (ad
nauseum).
6. what cleanliness tests to do and when in the process
That is a tough one. I use ion chromatography here at Rockwell Collins,
but it is not amenable to process control due to the times. IC is becoming
more available as a test service. I would use IC testing, together with an
accelerated stress screening protocol, together with ROSE testing, to
determine what good and bad levels of residues are, and then use ROSE as a
process control tool.
7. if fails the test, how to clean
Depends on the flux. In my experience, water is not enough. You need to
pick a chemistry that works with the fluxes you have chosen. Ask your flux
vendor how to clean his no-clean flux. They love that question and can't
hear it enough. Really. Trust me.
8. how to spec the thickness of coating, how to check it?
Most specifications will give a range of 1-5 mils. Thinner and you get
insufficient coverage, thicker and you have coating with less integrity.
Most coating thickness is measured on a flat unencumbered surface. It is
VERY hard to measure coating thickness on a resistor lead or IC lead. Most
specs I see use a flat coupon for thickness and is used as a process
control tool.
9. what to expect from a good coating house
Well, that's an easy one. Good coating. Done now. For free. In a more
real world, one not foggy from Mt. Dew haze, you should look for a
consistent process, with a Cpk of 1.33 or better, that can give you a
uniform film (free of voids or pinholes) that goes where you say and
doesn't go where it shouldn't.
10. what is a good test to do for qualification of the sub-contractor.
Take a look at what they do for other people, ask for references of their
customers and talk to them, or have them spray some evaluation panels for
you and put them through environmental stress screening.
10a. Since it's the only one in town, how to make him do better?
Well, a call from Guido or Nunzio always seems to work. Mobsters in Iowa,
what a concept. How do you normally work with your suppliers? Carrot or
Stick? Do you command enough business to force their compliance? Do you
have expertise in the area to help them improve their process? What kinds
of incentives can you offer them to get them to play ball in your court?
Any particular reason you have to stay in town?
11. except Concoat and Humiseal, any other major players in the fiels?
None. Concoat is the only one (Graham, that's another beer you owe me....)
worth looking into. For an answer that is slightly less graft-laden, I
suggest that you look at the Defense Supply Center Columbus web page and
look at the Qualified Products List (QPL) for MIL-I-46058, the military
spec for comformal coatings. This shows most of the players that sell to
the class 3, high reliability crowd (like us), and have high quality
coatings. While I am sure that there are acceptable coatings out there
that are not on the QPL, for high rel coatings, that is the place to go.
Well, enough for one day.
Doug Pauls
Rockwell Collins
( I REALLY have to start working with Pepsi on a Dew endorsement deal.....)
"Tempea, Ioan" <[log in to unmask]>@ipc.org> on 09/10/2002 08:06:48 AM
Please respond to "TechNet E-Mail Forum." <[log in to unmask]>; Please respond
to "Tempea, Ioan" <[log in to unmask]>
Sent by: TechNet <[log in to unmask]>
To: [log in to unmask]
cc:
Subject: [TN] Cleanliness of no-clean before conformal coating
Hi Technos,
badly need your wisdom on this. We have an automotive application, a simple
assembly (single sided reflow + wave soldering) that needs conformal
coating. We are on no-clean CP36 paste from Multicore and X33 no-clean flux
from the same company.
After having browsed last night through the archives, I still have a lot of
questions:
1. what reliability concerns should I have with automotive applications
2. any special coating material more suited for automotive
3. except for the dendritic growth, any other hazard
4. the Nortel analysis bulletins for both materials show the residues are
inert. should I worry about the flux residues or not? Can the assembly
process make them less inert?
5. if the flux residues are inert, what could screw up the ionic tests?
6. what cleanliness tests to do and when in the process
7. if fails the test, how to clean
8. how to spec the thickness of coating, how to check it?
9. what to expect from a good coating house
10. what is a good test to do for qualification of the sub-contractor.
Since
it's the only one in town, how to make him do better?
11. except Concoat and Humiseal, any other major players in the fiels?
Thank you SIRs and any other contributor,
Ioan
---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/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
---------------------------------------------------------------------------------
---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/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
---------------------------------------------------------------------------------
|
