soldermask thickness control on PCB's

[Higgler]

We have a company that uses a silkscreen process to apply a soldermask Taiyo BN4000. They state thickness control from 1/2 mil to 1.5 mil due to the silkscreen process. We'd like a tighter control (I know, that's picky, read on).

Question 1) Does this sound like a normal thickness control range for a silk screening process?

Question 2) Does anyone have experience in thickness consistency of dry film solder mask. I've looked at dry film specs but don't see any tolerance, only that they are 2,3,4 mil options.

Explanation why we're so picky; We have a patch antenna design whereby we have a very thick high dielectric radome atop our antenna and the spacing of this radome from the copper of the antenna shifts the frequency greatly. 0.5 to 1.5 mil spacing range can change the frequency 25%.

Thanks,

kch

 

[itsmoked]

I can't answer that.

How about skipping the solder mask?

Do your own conformal coating after assembly? (or not)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Higgler]

Can't do that, need some layer of protection on the copper (corrosion) and we need the spacing.

One other option is bonding a Kapton film of about 2 or 3 mils. then the tolerance buildup is Kapton + 1 mil +/- 0.5 mils if we silkscreen the adhesive. Better tolerances variation.

I do know that parylene coating can be vapor deposited with approx. 0.1 mils accuracy. Just a time thing, longer time in the vapor builds up more coating. We may just resort to that. It's an extra step in the build process though.

We could just make a lot of circuit cards with the standard soldermask process and pick the ones that have the proper coating thickness, but if we don't have to, it's best to avoid. That does get you spares though, which is always nice.

kch

 

[MacGyverS2000]

What about immersion gold plating for corrosion and skip the soldermask?

If the thickness of the soldermask itself is not a frequency-determining factor, place some non-soldermasked mounting holes around the antenna and use known-thickness washers on those mounting holes. A 3-mil piece of kapton tape on those mounting holes will mean your radome is 3 mils off of the trace, regardless of soldermask thickness.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[itsmoked]

How about standard 'whatever' treatment and make the mounting adjustable. Just use various thickness washers or cheap lasered sheetmetal spacers to make any board of any thickness always work? Or height screws? Or?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Comcokid]

If your patch antenna to the radome is so sensitive that a mil difference affects it, then I hope your final product is not used outside. If so, then just a thin layer of dew on a cold morning is going to have a bad effect.

 

[Higgler]

good suggestions, but nocando on the gold plating, we actually need a dielectric spacer above our copper layer. The radome has to attach to this coating without air gaps, so no spacers can be used.

Seems like parylene coating can be applied with 5% accuracy and may work. It's low dielectric too, about 2.8, and pretty low loss.
Typical deposition thickness rate is 0.2 mils per hour and that's adjustable and somewhat measurable.

10% tolerance on parylene thickness is easy per one of the coating houses.

still researching.

kch

 

[Higgler]

The affect from water depends on how thick the radome is.

I've made a plexiglas radome about 0.25" thick for a narrow band 439 Mhz patch antenna and pooling up water on the radome didn't change the frequency but a few MHz, so it depends on how far away the water is.

Radomes can do more then just keep the rain off,they can stop the antenna from detuning.

If the radome or antenna is waxed/hydrophobic coated to ball up the water droplets, the effect of rain or dew can be reduced. Sheets of water are much worse than streams or droplets of water running down a radome.

You are correct that if water touches the patch antenna metal, it will drop in frequency.

kch

 

[sreid]

Have you tried a dry film solder mask? This should have much better thickness control.

 

[itsmoked]

There's a board house that doesn't use dry film?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[sreid]

The OP said "Silk Screen Process."

 

[Compositepro]

To obtain the precision in coating thickness you are seeking you should apply excess coating and lap it down to the desired thickness using an appropriate measurement technique for feedback control. One measurent could be the tuned frequency of the antenna.

 

[GonzaloEE]

Maybe some kind of thermal control would expand the mask to keep the antenna always tuned.

 

[Higgler]

I've suggested dry film solder mask to the CCA vendor and he says they can do that.
The films I've seen have 2,3,4 mil thickness options, it does make sense these are more controlled, but unless I have some tolerance data, both initial thickness and final thickness, I'd be guessing and hoping it's accurate and unchanging. I wish I didn't need this control, but our computer and our results so far need the control.

Seems parylene is a better option. Plus dielectric constant is 2.8 instead of the 4.5 dielectric constant for soldermask I've looked up (Taiyo BN4000).

kch