Bill Bromley asked:

>We have been tenting via holes (dia. .014-.020) with dry film solder
>resist per IPC-SM-840, type B, Class 3, for the following reasons:
>
>- increases usable area for silkscreening
>
>- eliminates solder fill inspection in the via-hole, thus enabling vias
>to be placed under components
>
>On the down side, tenting does introduce the risk of contaminating the
>plated-through via hole if the hole does not get fully tented.
>
>
>
>Our PWB fabrication vendors are showing an increasing reluctance to use
>this type of solder mask coating, claiming that it is expensive, wasteful
>panel-wise, and has adhesion problems. In order to get the dry film to
>adhere better, they are forced to do another process (oxide treatment).

Yes, the film type mask is very expensive to use. Vaccuum-type sheet
lamination panel-by-panel is necessary, slower than hell, very troublesome
for the fab people. Probably costing you an arm and leg indirectly. Be sure
and ask for a price adjustment!

I never liked the stuff as it seems to me solder sticks to the developed
margins of dry film mask, and IMO solder mask to which solder will stick is
less than useless.

>They would rather use a liquid photo-imagable (LPI) coating conforming to
>the same spec, since the process takes less time, with less waste of
>material, better adhesion, and finer pattern definition. Unfortunately,
>LPI coatings don't tent vias reliably under normal processing.

True. I think LPI has taken over the market pretty nearly 100%, and it
won't "tent" reliably, as it is not too viscous when applied, drops into
the hole.

>My question:
>
>
>
>What material and processing can tent  via holes (dia. .014-.020) in a
>reliable and cost-effective way?

We have smaller vias - finished at 0.010" with 0.030" dia. pads. Our
original design Gerber data shows the mask "tenting" on both sides, but
what is _actually_ done:

The fab working film has the via pads opened to about 0.016" dia. both
sides. This results in the LPI mask covering most of the via pad on either
side. On the "solder" side, the boards are "plugged" with screened liquid
mask (I believe this is just conventional theramlly cured "PC401") as a
secondary process after HASL. This keeps solder from coming through the
hole at wave solder; which could result in shorts to SMT pads located
0.021" from the hole center, worst case; and permits the HASL tin-lead to
coat the copper in the via.

It would seem likely to me that they could do 0.014" dia holes in the same
fashion.

My understanding is that if the LPI mask covers either or both sides, at
HASL you can entrap flux or air pockets in the solder, and later at
assembly reflow this will possibly blow out.

After a couple of years, our vendors seem comfortable with this, they tell
me this is pretty common today.


cheers,

Jerry Cupples
Interphase Corporation
Dallas, TX