Hi TechNetters
We hope that the following may be of some assitance in this issue.
If you want more, let us know.
Conformal Coatings - Adhesion and Reliability?
Many producers of high reliability electronic equipment consider that
adhesion is an important criterion for conformal coating selection.
This report is intended to show that the issue of adhesion is not only a
complex one, but in fact bears little real correlation to the resultant
reliability of the circuit.
When a Printed Circuit Assembly (PCA) has been manufactured, the surface
properties or characteristics, are dependant upon several things:
? Solder masks or resists
? Flux residues
? Temporary solder masks
? Contamination from elsewhere (e.g. Finger salts and acids, mould release
agents etc.)
Solder masks are formulated to allow circuits to be mass soldered and are
available with matt, semi-matt, semi-gloss or gloss finishes. They contain
matting agents, thixotropes and flow or levelling agents, referred to as
tensides or surfactants.
These agents are the components which reduce the surface energy of the cured
solder resist film which, without them, could not be applied to a substrate
by standard techniques and remain defect free.
Flux residues and other surface contamination will effect adhesion in its
own right, but their largest effect is in terms of reliability.
Testing for Adhesion
The most popular method for testing adhesion is the cross hatch test. In
this type of testing a ‘cross hatch’ pattern is scored into the coating
surface with a constant and specified spacing between the cuts, generally
0.010 inch x 0.010 inch over > 0.5 inch2. Next an adhesive tape of known
adhesion is applied to the hatched area, and firmly lifted.
The greater the amount of coating removed, supposedly, the worse the
adhesion.
The cross hatch test is fundamentally flawed, as the percentage of coating
lifted by the tape depends on several other factors, not just adhesion of
the coating to the PCB. i.e.:
? The adhesion between the tape and the coating.
For example a silicone based conformal coating may perform excellently in
cross hatch adhesion testing yet silicone coating generally has very bad
adhesion to the PCB and this can be seen by lifting it with a scalpel. The
reason the coating stays on the PCB is that the tape will not stick to it
either. Every coating will have a different adhesion to the tape, and so the
results are effected
? Thickness of the coating
? Softness of the coating
? The pressure used to apply the tape
? The rate and angle at which the tape is pulled
? The depth of scoring and sharpness of the blade used.
If a blunt blade is used, the result is a greater lifting of the coating
either side of the score lines and this provides a lifting point when the
tape is used.
Nevertheless, if efforts are made to keep everything constant and if
silicones are excluded, sensible (if not particularly quantitative) results
can sometimes be obtained. It is, of course, important that the test coupons
are subjected to exactly the same solder resist, flux, solder paste,
soldering technique and cleaning method as the production PCA’s, for the
test to be meaningful. Heating during soldering can open up pores in epoxy
substrates and release substances, which would otherwise remain in the
laminate bulk. This can also grossly effect adhesion.
Testing for Reliability
Reliability in harsh environments is generally the goal when a conformal
coating is used. All conformal coatings have a finite permeability to water
vapour and do not work by excluding this from the vicinity of the PCB
surface.
Surface Insulation Resistance (SIR) is an indication of reliability and is
becoming an important test method for validating individual solder resists,
fluxes, pastes, cleaning solvents and conformal coatings, as well as total
process validation.
The following is a simple experiment using SIR which shows how a conformal
coating works, and why cleanliness is important:
1 Two SIR comb pattern coupons (A and B) are used in the test. They are both
put through the manufacturing process, components soldered to them, well
cleaned and then only A is conformally coated.
2 A SIR test at high humidity (90%RH) gives the following results:
Coupon A. 10 to the 9 Ohms
Coupon B. 10 to the 10 Ohms
The conformal coating under high humidity has a lower resistance than the
air above coupon B under high humidity. The resistance between the tracks is
still high for coupon A however, and would not produce board failure.
3 The coupons are now exposed to the environment i.e. contamination from
fingers and atmospheric dust etc.
The test is repeated with the following results:
Coupon A. 10 to the 9 Ohms
Coupon B. 10 to the 4 Ohms = PROBLEM!
The uncoated coupon B shows a significant drop in resistance. This is
because contaminants that are exposed to high humidity become mobile and
conductive. This is applicable to some organic as well as inorganic
substances. It is the presence of the humidity and the contamination, which
causes a potential low resistance and board failure condition.
The conformally coated Coupon A remains unchanged. This is because the
coating ‘filters’ and only allows water vapour to slowly permeate to the PCB
surface and the contamination is kept above the coating. This unit will be
reliable.
Conclusion: Is adhesion important?
Evidence suggests that there is no correlation between adhesion and
reliability.
The main task for a conformal coating is to prevent contamination reaching
the board surface in the field. If the PCB’s are not cleaned properly, or
not at all, then no coating will remove any problems caused by flux residues
(remember no-clean does not mean no-residue) or finger salts etc.
The following extracts are taken from reports produced by the Swedish
Institute for Production Research, IVF.
Evaluation of Various Combination of Solder Masks and Flux Residues on the
reliability of Conformally Coated Circuit Assemblies. Dr. Per Erik Tegehall.
IVF Report 94/28 dated Jan 1996.
The study involved many different conformal coatings from several suppliers.
5.3 Influences of Solder masks. “There is a strong indication that flux
residues from the manufacturing of the PCB’s have a large influence on the
SIR and the formation of dendrites.”
5.4 Influences of conformal coatings. “All conformal coatings have given
very good protection against migration and formation of dendrites.
...The adhesion of the conformal coatings to the various solder masks vary
considerably”
Elavuation of the Compatibility of Solder Masks and Conformal Coatings. Dr.
Per Erik Tegehall. IVF Report 93810. 1993.
4.4 Adhesion of the Conformal Coatings. “The results of cross hatch adhesion
depend upon the thickness and softness of the conformal coatings. Therefore,
the figures are not directly correlated to the adhesion strength of the
conformal coatings to the underlying surface.”
4.5 Optical Inspection after the tests. “Very few changes were observed
after the humidity environmental test……even fewer changes were observed
after the thermal cycling test. No peeling or blistering of the conformal
coatings was observed on any of the boards. Not even at those places where a
scalpel had been used to cut through the coatings before testing.”
5.1 Correlation between adhesion and performance. “Even though the adhesion
of some conformal coatings was quite poor to several of the solder masks, no
blistering or peeling of any of the conformal coatings occurred during the
environmental tests. ……above a certain level of humidity the ionic
contaminant will absorb water, through the conformal coating, forming a
saturated solution……..causing pressures up to several hundred bar. Since no
conformal coating can be expected to withstand such pressures without
blistering, it is doubtful if the adhesion of the conformal coating is
important for preventing blistering caused by contamination. ….both ionic
and non-ionic contamination may cause blistering……..However that has nothing
to do with how good the adhesion of the conformal coatings are to the solder
masks under clean conditions.”
This is a subject that has confused and baffled many people within the
industry and it is hoped that this information will help to de-mystify the
issue. Selecting a coating because of its ability to adhere to a “problem”
surface is, at best, most unwise and at worst, commercial suicide.
Coatings adhere perfectly well to copper, tin/lead, FR4 etc., yet does not
adhere to a solder resist that may or may not, have been subjected to
soldering. It is a known fact that coatings are not intelligent, so how does
it know the difference?
Selecting a coating that sticks better to certain surfaces should only be
done AFTER thorough reliability testing – else you may be simply masking a
serious reliability problem.
We may well be able to get you more specific data but would be accused
(rightly) of spamming. Please request off line we will not send unless
requested so to do.
Regards, Graham Naisbitt and Alan Brewin
Concoat Ltd
Alasan House
Albany Park
CAMBERLEY GU15 2PL UK
Phone +44 1276 691100
Fax +44 1276 691227
[log in to unmask]
-----Original Message-----
From: Mike Waters <[log in to unmask]>
To: [log in to unmask] <[log in to unmask]>
Date: 03 August 1998 15:21
Subject: [TN] HELP! Conformally coating on no clean flux pcb
>TechNet,
>I need advice badly. In our wave solder process we use no clean flux.
>We have an application where the assembled pcb has to be conformally
>coated to withstand up to 100% condensing humidity and 95% non
>condensing humidity in ambient temp ranges from -20deg C to +85 deg. C.
>A concern has come up with the long term reliability of the conformal
>coating based on cleanliness of the pcb prior to spraying the silicone
>or polyureathene coating (from previous posts, I am assuming silicone
>will be the best of the two solutions). It has been suggested that we
>water wash the pcb after solder before application of the coatings. The
>board is double sided with larger magnetics on the component side and
>includes SMT. In your experience, has anyone had problems using no
>clean flux in conjunction with silicone coatings with respect to field
>reliability?
>
>Regards,
>
>Mike Waters
>Phone: 919-767-3815
>Fax: 919-767-3966
>
>################################################################
>TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
>################################################################
>To subscribe/unsubscribe, send a message to [log in to unmask] with following
text in the body:
>To subscribe: SUBSCRIBE TechNet <your full name>
>To unsubscribe: SIGNOFF TechNet
>################################################################
>Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section
for additional information.
>For technical support contact Hugo Scaramuzza at [log in to unmask] or
847-509-9700 ext.312
>################################################################
>
>
################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe: SUBSCRIBE TechNet <your full name>
To unsubscribe: SIGNOFF TechNet
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################
|
