A few questions about potting

Thread Starter

strantor

Joined Oct 3, 2010
6,781
Question #1, and most time-critical question:
Could I fill this signal conditioner with potting to make it effectively water proof?





It is mounted in a box outdoors and every time the weather changes, the box sweats, and the conditioner quits working. It is a load cell strain gauge amplifier. I got a quote for a "fully potted, waterproof" conditioner to replace it but it's real expensive and I've already got plenty of these non-potted units I could pot. I was thinking just shoot the plastic case full of potting and cover everything except for the ajustment pot screws. What'ya think?

Question #2 - just out of curiosity - Does potting things thermally insulate them? For example, If I filled an entire Variable Freq Drive with potting, would it cause it to overheat? I mean, all of the heat generating components are already attached to a giant heatsink which wouldn't be covered in potting... or how about if I snaked a tube all around through it and then potted, and then ran coolant through it... just brainstorming

...but why the heck would you want to do that?... which leads me to
question #3 - Does potting things protect them against physical shock (vibration, g-forces)? At some point I want to mount a VFD in a car, but they have a shock rating which is lower than what you'd find on a bumpy street, so could I effectively "ruggedize" a drive by shooting it full of potting?

Thanks,
Charlie
 

Thread Starter

strantor

Joined Oct 3, 2010
6,781
It is weather proof, and it is also the second type of weather proof enclose we've tried, and it's full of dessicant bags. It doesn't leak, it sweats. Can't seem to beat the sweat. Thanks, I'll try potting it and see how it works out.
 

t06afre

Joined May 11, 2009
5,934
Does your device have any IP coding http://en.wikipedia.org/wiki/IP_Code To me it does not look like it is very protected against water. It would probably be much simpler and better to find box that is more protected than the one you have now. Put the all cable terminations inside the box. And use a Cable gland like this one to feed cables in and out of the box. Done correctly it can work well for you.
 

Thread Starter

strantor

Joined Oct 3, 2010
6,781
Does your device have any IP coding http://en.wikipedia.org/wiki/IP_Code To me it does not look like it is very protected against water. It would probably be much simpler and better to find box that is more protected than the one you have now. Put the all cable terminations inside the box. And use a Cable gland like this one to feed cables in and out of the box. Done correctly it can work well for you.
That's a good point. I don't remember seeing an IP rating on that box & it's definately not a watertight gland like that (it's a regular gland with epoxy all over it). I will raise the flag about the box but I think it'll probably fall on deaf ears, as previous techs have already replaced it and the general consensus (among my supervisor and the people he's consulting with) is that the box can't be the problem since "we've already tried that." - if you know what I mean....EDIT:...That, and when I showed him the really expensive potted amplifiers he got all excited. People seem to think that the more you spend, the more sure of a bet it is.
 
Last edited:

Thread Starter

strantor

Joined Oct 3, 2010
6,781
On the other topics, e.g. filling a VFD with potting & possible fluid cooling... any ideas? I know most people won't comment if they don't KNOW an answer, but I'm only looking for speculation here, as I'm sure this sounds bizzare.
Look for warm spots, without air cooling, they could become hot spots.
So Bill's telling me that potting will make air cooled components overheat, which what I figured, but the other side of my mind keeps saying "but all the heat generating components are attached to the heat sink, which is still air cooled." My suspicion is that it's a stupid idea but I'm waiting for someone to breathe some glimmer of hope into it.

p.s. I know alot people encase their entire project in potting to prevent it being reverse engineered - how to they prevent overheating?
 

Thread Starter

strantor

Joined Oct 3, 2010
6,781
Sorry, scattered thoughts... About "ruggedizing" with potting: if I made a simple device which didn't generate too much heat and encapsulated it in something like this, with no pcb inside, all components suspended in the epoxy connected with air wires, would it be possible to make a "nearly indestructible electronic device"? Such that you could run over it with a truck, play baseball with it, sink it to the bottom of the sea, etc?
 

GetDeviceInfo

Joined Jun 7, 2009
2,192
I'd speculate that the VFD wouldn't do well with a general potting. I've worked with liquid cooled inverters, but convection was still required.
 

Thread Starter

strantor

Joined Oct 3, 2010
6,781
Have you considered conformal coating both sides of the board instead of encapsulating?

creakndale
going back to the signal conditioners? (...I remember now, why it's not a good idea to present multiple questions in one thread...) No I didn't know about it until you mentioned it and I looked it up. Looks like something that should have already been on the boards since they left the factory. Is it clear? how can I tell if it's there or not?
 

creakndale

Joined Mar 13, 2009
68
There are different types of Conformal Coatings and some are clear. Generally, they are used on high reliability electronics or electronics used in harsh environments.

One way to test: Ohm a trace on the PC board with standard (non-piercing) multimeter leads. If it shows continuity then there's no conformal coat. Typically, you need to use sharp, piercing probes to penetrate the coating in order to get a reading.

MG Chemicals has a variety of conformal coatings.

I've purchased some from here:
http://www.web-tronics.com/electronic-chemicals-protective-coatings.html

creakndale
 
I think it would be cheaper and probably more relyable to add a PTC panel heater to the enclosure which would keep the temperature from swinging .....
The sweat you mention is probably actually condensation formed after rapid cooling.
http://www.stego.de/fileadmin/Bilder/PDF/English_US/HGK_047_2010_EN_US.pdf
this sort of thing .... the low temp versions dont even need a stat and they are available for low DC voltages if you look round a bit.
Al
 

Thread Starter

strantor

Joined Oct 3, 2010
6,781
I think it would be cheaper and probably more relyable to add a PTC panel heater to the enclosure which would keep the temperature from swinging .....
The sweat you mention is probably actually condensation formed after rapid cooling.
http://www.stego.de/fileadmin/Bilder/PDF/English_US/HGK_047_2010_EN_US.pdf
this sort of thing .... the low temp versions dont even need a stat and they are available for low DC voltages if you look round a bit.
Al
Now there's an idea. I've seen those used before in outdoor boxes but it never occurred to me. Thanks; I'll bring that up
 

someonesdad

Joined Jul 7, 2009
1,583
Question #1, and most time-critical question:
Could I fill this signal conditioner with potting to make it effectively water proof?

It is mounted in a box outdoors and every time the weather changes, the box sweats, and the conditioner quits working. It is a load cell strain gauge amplifier. I got a quote for a "fully potted, waterproof" conditioner to replace it but it's real expensive and I've already got plenty of these non-potted units I could pot. I was thinking just shoot the plastic case full of potting and cover everything except for the ajustment pot screws. What'ya think?

Question #2 - just out of curiosity - Does potting things thermally insulate them? For example, If I filled an entire Variable Freq Drive with potting, would it cause it to overheat? I mean, all of the heat generating components are already attached to a giant heatsink which wouldn't be covered in potting... or how about if I snaked a tube all around through it and then potted, and then ran coolant through it... just brainstorming

...but why the heck would you want to do that?... which leads me to
question #3 - Does potting things protect them against physical shock (vibration, g-forces)? At some point I want to mount a VFD in a car, but they have a shock rating which is lower than what you'd find on a bumpy street, so could I effectively "ruggedize" a drive by shooting it full of potting?

Thanks,
Charlie
If you put it into a box with desiccant and things don't stay dry, either the desiccant isn't working, you have a leak, or it's still reaching temperatures below the dew point. All things being equal, I'd vote for an air leak first.

One thing I'd consider trying would be to dip the electronics side in molten paraffin (wax). You shouldn't hurt the parts and it should waterproof things. Plus, if you have a hot spot, you might be able to see where the wax melted (scribe a fine pattern into the surface first).

Yes, such potting will cut down on the effectiveness of convective cooling. You'll have to test to see if it's a problem. If there's a hot spot, the wax coating could be scraped off. Don't coat any heat sinks though.

Potting something can reduce some failure modes (think of a can capacitor on leads that now can't move around from vibration), but other failure modes such as breaking a bonding wire inside of an IC won't change.
 
Top