Rahman,
have you solved your problem? If not, I have an article that may interest
you. Before reading I want give you a homework:
1. Run SEM with much higher magnification, Backscatter detector, 20 kV, 45
deg tilt, 20 um aperture, low probe current, magn 1k and 5k.
2.Strip mask followed by 1.
3. Send images
Now to the short article.
Best wishes / Inge
I assume that you are using dry-film soldermask to encapsulate your circuit
after etching. During dry-film lamination, air can become trapped between
circuit elements as the board passes between the rollers. This is
especially true if the circuit pattern is fairly dense, with long traces
that run parallel to the lamination rollers. This trapped air manifests
itself as bubbles as the film is imaged, developed, and cured.
Another source of bubbles is routinely encountered when using dry-film
photopolymer. The rheology of so-called dry-films (actually super viscous
liquids) during lamination prevents the perfect conformance of the
photopolymer to the sharp corners that form between the copper pattern and
the substrate in an etched circuit. This results in corner gaps that can
run the entire length of a trace and totally surround a pad or via. This
type of defect is especially troubling since it can form capillary paths
that allow fluxes and other liquids to wick along circuit elements leading
to shorts, leakage, and even circuit failure.
The conventional answer to both problems is to use a vacuum laminator which
eliminates the trapped air that is at the heart of the problem. Being
cheap, and reluctant to spend money if we do not have to, we have been
reasonably successful in eliminating these defects in our own shop by using
a modification of the wet lamination technique first introduced by DuPont a
number of years ago. Using a "plant mister" with a 5% solution of methanol
(wood alcohol):
1. evenly moisten the panel immediately prior to lamination. The panel
should not be dripping wet, just moist enough to change the surface color
2. make sure that your feed table is totally free of dust and other
contamination
3. if possible, orient your panel so that no long circuit features
(traces or ground planes) are parallel to the lamination rollers
4. laminate your board with your laminator feed set to 0.25 to 0.50 feet
per minute. I know that this seems a bit slow but it gives the photopolymer
time to absorb the solution and flow around the etched circuit elements
(see below).
5. let your board sit for *at least 15 minutes* after lamination before
proceeding with further processing
The benefits of this approach are two fold. First of all. the water fills
in the gaps and edges of the circuit features. If you watch closely as
lamination proceeds, you will see that a small bead of water forms just
before the nip region and "runs" in front of the lamination rollers. This
pushes air out of areas where it would otherwise be trapped and acts as a
moving seal to prevent the air from seeping back in.
The methanol acts as a wetting agent to allow the water to penetrate all of
the little nooks and crannies in your circuit pattern. In low
concentrations it will not affect the curing properties of the dry film
and, with the water, will reduce the viscosity of a thin layer of dry film.
This "thinning" allows the otherwise super viscous liquid to flow into the
gaps as the water (and alcohol) are absorbed into the bulk of the
photopolymer.
On 3 April 2013 03:15, SARAVANAN R <[log in to unmask]> wrote:
> Hi,
>
> We have carried out EDS on the corrosion sample. It was found copper and
> oxygen contents in the corroded area.
>
> No sulphates,chlorides , bromides, nitrites etc.
>
> It the corrosion is copper oxides only.
>
> i am clue less about the root cause. Since after assembly the pcbs were
> inspected prior and after conformal coating. No corrosion
> was found. It is latent development only. Whether entrapped moisture under
> the SM can cause this?. No additional contact with air is not possible
> since the PCB was conformal coated.
>
> regards,
>
> R.Saravanan
>
>
>
> ------------------------------
> *From:* Inge Hernefjord <[log in to unmask]>
> *To:* [log in to unmask]; [log in to unmask]
> *Sent:* Monday, 4 March 2013 3:08 AM
> *Subject:* Re: [TN] corrosion under Solder Mask
>
>
>
> On 3 March 2013 22:31, Inge Hernefjord <[log in to unmask]> wrote:
>
>
> Vladimir is right, make a SEM investigation to see if it really is
> corrosion and not something else e.g. contaminated water or crystalized
> solvent residues etc. And take some high resolution photos at varying
> magnification.
>
> I've been analyzing such things for years and I could guide you, as well
> as Vladimir, but you'll certainly find assistance at some uni or
> independent lab in your area. I am lazy, so instead of using own words, I
> copied a page from a book, written by a guy in your country. I placed the
> page in Dropbox/Reports../Analysis../Ramakrishnan. You have admission
> already.
>
> Good Luck
>
> Inge
>
>
>
> On 3 March 2013 00:29, Vladimir Igoshev <[log in to unmask]>wrote:
>
> It could be zillions of reasons and unless proper FA is done you are
> NEVER going to figure out what exactly caused the problem.
> ------Original Message------
> From: Ramakrishnan Saravanan
> Sender: TechNet
> To: [log in to unmask]
> ReplyTo: TechNet E-Mail Forum
> ReplyTo: Ramakrishnan Saravanan
> Subject: [TN] corrosion under Solder Mask
> Sent: Mar 2, 2013 18:13
>
> Dear all,
>
> We have received few MLBs from a contract supplier. The board was
> inspected and accepted for wiring. The assembly is for high reliability
> applications. We conformal coated, tested,thermal cycled and closed the
> package. We opened the package for some rework, we found corrosion under
> solder mask on some tracks.
>
> The corrosion is in the form of blisters and localized. Center of the
> tracks or edge or localized and not covering the entire width of the tracks.
>
> What could be the reason for corrosion under solder mask ?. How to
> diagnosed for root cause analysis ? regards,
>
> R.Saravanan
>
>
> Regards,
>
> Vladimir
>
> SENTEC Testing Laboratory Inc.
> 11 Canadian Road, Unit 7.
> Scarborough, ON M1R 5G1
> Tel: (647) 495-8727
> Cell: (416) 899-1882
> www.sentec.ca
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please contact helpdesk at x2960 or [log in to unmask]
> ______________________________________________________________________
>
>
>
>
>
>
______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
|
