moisture ingress through thermoplastics 1

[dgowans]

Can anyone point me in the direction of any research that has been done on moisture ingress through thermoplastic materials?

We've got existing products that consist of a PCB assembly sealed in a thermoplastic enclosure via a hot plate welding process. As we look to expand our product line, new enclosures are being developed and a comment was made about the research that was done (long before my time at this company) on moisture ingress through various thermoplastics and that there was a volumetric range of the enclosure for which the moisture ingress over time would be insignificant.

These devices are installed outside and have an expected field life of 20 years. The concern is (was?) that significant moisture ingress over time could lead to condensation on the electronics and cause a short. For the purpose of discussion let's assume that the seal produced with the welding process is the same as the rest of the enclosure as far as ingress is concerned. We do extensive leak testing on these enclosures and I'm not worried about an inadequate seal being a leak path.

The questions I have are:

Is the concern over moisture ingress valid?

If so:

What is the relationship between enclosure volume and acceptable moisture ingress?

What other variables factor into how much moisture will pass through the enclosure over time? Material selection seems to be obvious, although I keep going back to the validity of the concern overall.

I imagine that there will be numerous questions as to the specifics and I'll be happy to answer them as best I can. I'd be speculating about what information is relevant at this point and don't want to randomly post what I think are the key points.

 

[patprimmer]

I don't know any specific papers on the subject, but I am sure there are many. Moisture diffusion rates will be studied for many plastics in relation to:

Food and beverage packaging.
General clothing.
Surgical gowns.
Water filters.
Wound dressings.
Personal hygiene wear.
Application of dyes by wet processing.
Wet weather wear.

Remember, synthetic fibres are plastics.

Off the top of my head, critical factors will be:

Affinity for water to the surface.
Water diffusion rates or permeability.
Concentration gradient.
Temperatures.
Surface area to volume ratio.
Relative humidity differences.
Time of exposure.
Film thickness.
In multi layer films, effective %age cover and thickness of the barrier layer.




Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[dgowans]

Pat,

Thanks for the input. Informative response from you, as always. I was keeping you in mind as I formulated the initial post - didn't want to get blasted (and rightly so) for being vague.

I would interpret your response to mean that you think the concern over moisture ingress is valid, it's just the degree of concern that is warranted that is yet to be determined.

In any case, it sounds like I have no trivial task in front of me. My first step will be to contact our material supplier to see what information they might be able to provide. I'm also going to touch base with the person who did the initial research to see what approach was taken. I would have done that right out of the chute but he's out of the country. Any other obvious sources of information I should try to tap into other than internet searches?

 

[rconner]

No specific expertise claimed here, and of course perhaps not the case in your application, but I was just curious also if the rate of ingress (or "permeation", as it were?) might be some different if the environment/atmosphere contained only pure water vapor, or if there were also some sort of cocktail mix of different permeants (including fuel vapors, hydrocarbons etc.?)

 

[Demon3]

Interesting. As Pat says, there's tons of work done on permeability of plastics so you should be able to find something. I know that Professor Ulf Gedde at the KTH in Stockholm and his group has done plenty of work and can model diffusion. He is a nice guy and fluent in English, he might be able to do a rough calculation to estimate whether you could have a problem or not.

You say "We've got existing products that consist of a PCB assembly sealed in a thermoplastic enclosure via a hot plate welding process". My first question is to ask what thermoplastic you use.

Other comments would be that there are commercially available water scavengers that you could put inside your part to adsorb any water that gets through. That will be way cheaper, quicker and easier than long term measurements and calculations. Some water adsorbing compounds include anhydrous copper sulfate, zeolites and silica gel.

Also, is it possible to completely cover the parts so there is no air space and therefore no chance of water droplets forming? I have seen this done for some electrical parts I have bought (an AC filter/surge protector) the whole part was filled with potting compound, presumably epoxy based.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[dgowans]

Demon3,

The material is a calcium-filled polypropylene copolymer.

From what I've learned by digging up bodies here in the last couple of days, the original enclosure was analyzed by a PhD level person who performed a bunch of calculations that were boiled down by the founder of the company (the person I'd like to talk to who is out of the country) to give a range of either enclosure volume or surface area/volume ratio which would not promote an unacceptable level of diffusion into the enclosure.

Because I wasn't around in those days, I have to assume that these calculations were performed on the different varieties of enclosures that exist and the enclosures were designed accordingly. I plan on talking with the founder to get some more information on what was done and how to apply it going forward but wanted to do some research on my own to better understand the issues that may arise.

We don't have a problem with long term moisture ingress on any current products and I'd like to keep it that way on future products, of which there will be many.

Regarding the adsorbing compounds like silica gel, etc: would the presence of a substance like this inside the enclosure tend to promote greater diffusion toward the electronics, similar to having a pressure differential across the enclosure walls? Not a totally appropriate analogy, I suspect.

Encapsulating the electronics is possible - we do this on some of our products. In the products I'm particularly concerned with, however, this is not a desirable solution as it would not only add a rather significant cost it would also affect functionality of the device. Good thought though.

 

[IRstuff]

You could see:

http://www.ptonline.com/Scripts/SearchMainPT04.asp?qt=thermoplastic+AND+moisture

TTFN

FAQ731-376

 

[Demon3]

Whoops, I have to red flag myself for posting an e-mail address.

Interesting. As Pat says, there's tons of work done on permeability of plastics so you should be able to find something. I know that Professor Ulf Gedde at the KTH in Stockholm and his group has done plenty of work and can model diffusion. He is a nice guy and fluent in English, he might be able to do a rough calculation to estimate whether you could have a problem or not. You can find him easily with Google.

You say "We've got existing products that consist of a PCB assembly sealed in a thermoplastic enclosure via a hot plate welding process". My first question is to ask what thermoplastic you use.

Other comments would be that there are commercially available water scavengers that you could put inside your part to adsorb any water that gets through. That will be way cheaper, quicker and easier than long term measurements and calculations. Some water adsorbing compounds include anhydrous copper sulfate, zeolites and silica gel.

Also, is it possible to completely cover the parts so there is no air space and therefore no chance of water droplets forming? I have seen this done for some electrical parts I have bought (an AC filter/surge protector) the whole part was filled with potting compound, presumably epoxy based.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[IRstuff]

Air spaces are not necessarily required for moisture intrusion. Plastic IC parts used to suffer intrusion through direct diffusion through the plastic. The corrosion problems eventually were solved with better passivation on the ICs and better plastics.

TTFN

FAQ731-376

 

[patprimmer]

I would expect that water vapour would not hurt your PCB, but liquid water will.

In that case you need to keep the relative humidity (RH) inside the bag less than 100%

Water will permeate away from the high RH air toward the low RH air. The film will only transmit it very slowly.

If the RH inside the bag is moderately high and the temperature drops enough, the RH inside the bag will reach 100% even though water vapour is permeating out of the bag.

I see your problem more aligned with condensation and temperature change than permeability.

If the bag were stored under water, water permeating into the bag to reach 100% RH is going to happen, but if stored in normal atmosphere, the inside will be pretty close to outside average RH at average temperatures.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Demon3]

PP filled with CaCO3 is used in tens of thousands of tons per year inside washing machines (the tub that holds the water in), dishwasher, clothes driers etc. As long as it's formulated properly the ingress of water is not going to be high.

Something that should slow water ingress even more would be to use a stearic acid coated CaCO3 (e.g. Imerys Carbital 110S or Omyacard 2TAV) instead of a non surface treated CaCO3. The surface treatment makes the filler hydrophobic and helps it disperse, using a treated grade has been shown the reduce water permeability of polyolefins.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[dgowans]

Pat,

You're correct - liquid water would be the issue, not water vapor.

I'm now back to questioning the validity of the concern. Over time, the RH inside the enclosure will essentially match that of the RH outside the enclosure. A better question would be "What happens during extreme temperature changes?" We do thermal cycle testing and haven't seen any issues when enclosures are properly sealed, even when completely submerged. I'm beginning to suspect that I'm on a wild goose chase - it wouldn't be the first time that a past concern was misrepresented.

 

[IRstuff]

Most military conformal coatings are tested at 95% RH and high temperature. Once the moisture gets in, whether or not it was liquid prior to ingress is irrelevant. Most of the classic IC lead problems with tin whiskers only needed moisture, not liquid.

TTFN

FAQ731-376

 

[Demon3]

I read there is a new coating that eliminates formation of tin whiskers. Maybe that would be a good insurance against potential problems.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[IRstuff]

Not so much a coating as changes to the processing.

TTFN

FAQ731-376