Plastic Enclosure for Autoclave Environment: Moisture Barrier

[lukeseed]

Hi there,

I have a design problem with sealing out moisture during autoclaving.

We currently are using a ferrite core wrapped in magnet wire with a cap in series to make a rf excitable lcr circuit. When exposed to rf energy this circuit rings at a certain frequency and q factor.

The problem we are having is that it is a medical device and we need to sterilize it. The chosen (and only) option for this sterilization is autoclaving which is a steam bath at 250deg F at 21psi for 20min. This process causes water vapor to invade the lcr circuit and changes the dielectric constant of the air coil/ferrite thus dropping the q and shifting the frequency.

To combat this shift we are attempting to seal the circuit in plastic. As you can imagine, the autoclave process is very hot and very likely to damage the plastic and invade the enclosure. Two things we have tried are insert molding the circuit in plastic and molding a bottom case and overmolding a cap over the case and the tag. Neither of these has worked. The insert molding pushes the tag to the side and breaks through the capsule (we are attempting to combat this by using self retracting pins to hold the circuit in place during molding) and the overmolded cap still leaks.

The plastics we have tried are PP, PE, HIPS, ABS, PC, Nylon 12, PEEK, Ryton and LCP. Of these, only PEEK, Ryton and LCP held up to the autoclave environment and all have let in moisture and an associated, unacceptable shift.

Does anyone have any ideas on this? Has anyone else been able to seal out an autoclave process and how was that done?

Thanks,

Luke

 

[dgowans]

Can you mold a 2-piece enclosure out of one of the autoclave-friendly materials? I envision one of the enclosure halves having a well that would encapsulate your electronics in potting compound and the plastic enclosure would be for aesthetics only. If you adequately seal the electronics in potting compound the enclosure can let moisture in and not cause problems, correct?

Alternatively, take a look at the encapsulation process by Cavist. I don't know how this would stand up to the autoclave, but it's worth a look at least.

 

[lukeseed]

Do you have a suggestion for a potting compound? We have also begun to look at this and have the wells already.

 

[dgowans]

I'd rather not recommend anything specific, but I would definitely recommend that you talk with someone at Ellsworth Adhesives. They supply the potting compounds we use and our technical sales rep has been very helpful - hopefully you have the same experience.

 

[itsmoked]

Or mold an enclosure with a proper O-ring groove on both sides.

I have reached for potting compounds many times and it has mostly been a very,(nay shockingly), expensive, messy, giant hassle. Picture a nice clean electronics assembly shop with someone cooking pancakes in the middle of it.

You need gloves, vacuum pumps, bell jars or equivalent, table covers, strict time frames based around something with a wildly varying viscosity and temperature, volume affected curing rates, etc ,etc. Potting most electronics is like jumping into the deep end.

Then try to rev your product!

Top it off with someone wanting to ETO sterilize your device. The potting material has all sorts of headaches with that! Gas release times, potting breakdown, etc.

I now see potting as the v e r y last resort.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ajamnia]

I just read your post and was rather amazed that peek did not work. Peek has a very small water retention property. I am curious to know the percent shift in the frequency that you experinced. Are you sure you have not damaged the wire insulation during the over-molding process?

One more question: how many cycles of autoclaving are you subjecting your device?

In addition to PEEK and LCP, Ultem and Noryl would also survive the autoclaving conditions.

Ali

 

[lukeseed]

Thanks everyone for your suggestions. We have come across a couple of things since the last time I wrote.

For the capsule, several materials weren't overmolding to themeselves very well (PEEK and LCP mainly) and were either pushing the part through the cap or not sealing off correctly. This lead to wild and varying results that were misleading. We have since corrected the molding problems and are seeing very good results with PEEK and Ryton. The LCP will probably never work without extensive and expensive insert molding as it does not like to overmold to itself AT ALL. :0( (For your information ajamnia we have thought of ultem and noryl but haven't got around to trying them yet, thanks for the suggestion)

Currently it looks like both PEEK and Ryton in a capsule with overmolded cap are working on the short term. We have only concluded about 10 Autoclave cycles while we are looking at a product life of 200 cycles.

In addition to thermoplastic molding we have also begun to look at other alternatives, mostly towards completely hermetic materials such as metal and glass.

The glass looks promising. We have sealed the tags inside glass tubes now and they worked beautifully, with no degredation to signal strength or frequency shift for the few tests that we did with them. We are now looking at high volume automated processes. Things we worry about with the glass is the annealing process burning the tags and if we don't anneal the glass, whether we can protect the glass enough to keep it from breaking.

While the vapor barrier foil worked wonderfully at keeping out moisture, it interferes with the q, effectively knocking it to zero. We are still up in the air with mylar vapor barrier films. They seem to be working but the film looks like it warps through the autoclave cycle and may not be a reliable material for a high volume product.