Author Topic: potting compound choices  (Read 8647 times)

0 Members and 1 Guest are viewing this topic.

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
potting compound choices
« on: April 23, 2014, 10:03:09 am »
I need to put a board.

Now it won't be in a case we will pot it in a mold and then take the mold off so that we have a lump and drill holes through the potting to bolt it down. it needs to be good for -45/+125 and not shatter. Any ideas ?

Materials that seem to be readily available are: Epoxy, Nylon (very expensive), Poly Urethane, Urethane, Silicone, Silicone RTV.

i have no experience in potting.
 

Offline lewis

  • Frequent Contributor
  • **
  • Posts: 704
  • Country: gb
  • Nullius in verba
Re: potting compound choices
« Reply #1 on: April 23, 2014, 10:19:18 am »
The softer resins like the some urethanes and silicone will probably not be suitable for moulding, they will be too flexible. Epoxy is your best bet. I use Robnor PX700K for pretty much everything, Rapid part number 87-0220. Very easy to mould and machine afterwards, just watch the dust.

To get a decent quality moulding, you'll need to make a vacuum chamber. A 'Really useful Box' with a hole drilled in the side for a vacuum cleaner is generally sufficient unless you need the moulding to be aerospace quality. You just need to agitate and expand the air bubbles to get them to rise to the epoxy surface. If you do need aerospace quality, you'll need to degas the resin under high vacuum, but watch out if your PCB contains electrolytic capacitors or relays.

Cure the moulding overnight under a 150w halogen lamp. Before mixing the resin heat the bag to about 50C in an oven. Do not overheat! The bag should be warm enough to handle comfortably. This will make it much less viscous in the mixing and pouring process and gives a much better looking mould.

Epoxy resin will leak out of the smallest tiniest hole that even water wouldn't get through. Make sure ALL holes in the mould are sealed tight. Same with components, don't use non-waterproof relays etc.
I will not be pushed, filed, stamped, indexed, briefed, debriefed or numbered.
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6460
  • Country: nl
Re: potting compound choices
« Reply #2 on: April 23, 2014, 10:27:25 am »
if mass produced and cost is a major isseu: tar. Don't know the exact composition but it is used in outdoor lamp drivers for instance.

Sorrie this is for in a case. Why not use a case? Then it is no longer potting by the way because you won't have a pot  ;)
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #3 on: April 23, 2014, 10:55:56 am »
hm sounds like epoxy then.
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1384
Re: potting compound choices
« Reply #4 on: April 23, 2014, 11:41:22 am »
Why no potting box? Is the shape an oddball shape or something?  Potting boxes are dirt cheap compared to proper project boxes. If you need a custom shape, see if you can either find someone to do vacuforming or do your own vacuforming to make thin shells to put the electronics/potting in.   If you have the space for a thin shell, I'd highly recommend using one.  All the times IVe tried molding electronics in potting, Its always a ton of hassle compared to potting it in an enclosure. Plus you can do larger batches rather then however many the mold supports.   

Also, down to -45, I suspect epoxies may get a bit brittle that low, unless you find one specifically mentioning that.  A harder urethane rubber may be useable if its thin enough function as a potting well.  If you do find something hard like epoxy that will work over that temperature range, make sure to CTE match your board/components, and do thermal cycling to make sure its not going to rip components off the board. 

In the US I'd consider stuff from epoxies.com (dont know about their international service or UK epoxy vendors)  http://www.epoxies.com/products/potting-encapsulating-and-casting/   Ive used some optically clear epoxy from them before and it worked great.  Something like the 20-2523 urethane would be a starting point IMO. 

This page gives good examples of durometer numbers http://www.smooth-on.com/Durometer-Shore-Ha/c1370/index.html  Depending on the application, I'd think stuff in the 70A or 20D  range or higher would be pretty good.

« Last Edit: April 23, 2014, 11:45:00 am by ConKbot »
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #5 on: April 23, 2014, 11:47:45 am »
We started out wanting the PCB to go bare but now to try and protect it from shock and vibration I have to encapsulate it but still use the same mount holes - crazy I know - not my choice.
 

Offline Teledog

  • Regular Contributor
  • *
  • Posts: 202
  • Country: ca
Re: potting compound choices
« Reply #6 on: April 25, 2014, 04:46:05 am »
Conformal coating (spray on) & rubber standoffs?
« Last Edit: April 25, 2014, 04:48:43 am by Teledog »
 

Offline calexanian

  • Super Contributor
  • ***
  • Posts: 1881
  • Country: us
    • Alex-Tronix
Re: potting compound choices
« Reply #7 on: April 25, 2014, 04:51:08 am »
To pot or not to pot. Always the question. IF you can avoid potting, I would. If not only for servicing and fault investigation. Some of the soft conformal coatings have come a long way. We use Humiseal branded products and have had good luck. For just plain potting epoxy we use J Greer-Aero Marine. We find it to be a good value.
Charles Alexanian
Alex-Tronix Control Systems
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #8 on: April 25, 2014, 06:11:33 am »
I'd prefer not to pot and redesign the board to cope with the 500G requirement I have been given but because I work for a bunch of cheapskates that want to make life a misery over £50 tooling to do the damn thing again I am being asked to pot it and it must still be in a good enough condition at -46C to still withstand 500G of undefined shock and vibrations.
 

Offline mzzj

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: fi
Re: potting compound choices
« Reply #9 on: April 25, 2014, 07:59:27 am »
I'd prefer not to pot and redesign the board to cope with the 500G requirement I have been given but because I work for a bunch of cheapskates that want to make life a misery over £50 tooling to do the damn thing again I am being asked to pot it and it must still be in a good enough condition at -46C to still withstand 500G of undefined shock and vibrations.

 |O :-DD

What was your board size and mounting hole spacing?

You probably want extremely rigid potting compound if you don't have casing AND your board has to withstand 500G
Worst-case is that you add a lots of mass with your potting but not enough stiffness(ie silicone potting) to support increased forces from potting compound mass. -->resulting lot more board flex.

For your silly specifications  epoxy filled with hollow  micro glass balloons might be a good idea. More rigid than ordinary epoxy and lightweight. Or why not unobtanium-oxide reinforced epoxy,  goes well with rest of the spec..
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6460
  • Country: nl
Re: potting compound choices
« Reply #10 on: April 25, 2014, 08:17:43 am »
at -46C
No electrolytics on board or are you (pre)heating them?
 

Offline SArepairman

  • Frequent Contributor
  • **
  • Posts: 885
  • Country: 00
  • wannabee bit hunter
Re: potting compound choices
« Reply #11 on: April 25, 2014, 08:17:59 am »
it sounds like your employers have been up to a different kind of potting

this thread is like the typical engineering sad story which makes me not wanna be in this profession  :palm:

do they ever add fibers to epoxy to make it more rigid? or some kind of plastic honey comb to put over your board?
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1384
Re: potting compound choices
« Reply #12 on: April 25, 2014, 11:03:46 am »
Yeah... 500G with only a few mounting points for the PCB, inside the potting sounds like a recipe for fail.   That and the encapsulation/potting wont keep water out then (if thats an issue) as water can get right to the PCB along the screws/standoffs. 

If they are bitching about 50 quid for a re-spin, show them how much foam filled epoxy costs so the potting is stiff and lightweight. 

http://www.ellsworth.com/emerson-and-cuming-stycast-1090si-epoxy-black-1-gal-pail/

dont forget the hardener, sold separately. 


They bitch about nre for a respin, but want to get tooling made to cast encapsulated boards in, one batch at a time, instead of using potting boxes for a few cents/pence each? Why dont you see what other sort of one time expenses they will bend over backwards  to avoid at the expense of increased production costs.  You should make it your own personal project :p   "We get an FPGA if we buy software for it for $2k and then it will be $5/part, or we  can skip the cost of the program and build it on breadboard with 7400 series logic by hand for every unit"
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #13 on: April 25, 2014, 11:14:56 am »
I think it is partly not wanting to change the design of the mounting system, it's a 40mm board with 2 screw points. It needed to be reasonably small and i was not told about vibrations. There are electrolytics on board but they are not essential, a loss of capacitance won't affect the board as they are way over the minimum requirements.

So we can't get a box to mount without changing the mount plate so the idea is to just mount it with putting as one lump. I can design out the e-caps but I may have a problem with connectors.
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #14 on: April 25, 2014, 11:15:53 am »
apparently there was not time to redesign but I think things keep slipping with the main project as i could have done it by now.
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1384
Re: potting compound choices
« Reply #15 on: April 25, 2014, 12:35:31 pm »
I think it is partly not wanting to change the design of the mounting system, it's a 40mm board with 2 screw points. It needed to be reasonably small and i was not told about vibrations. There are electrolytics on board but they are not essential, a loss of capacitance won't affect the board as they are way over the minimum requirements.

So we can't get a box to mount without changing the mount plate so the idea is to just mount it with putting as one lump. I can design out the e-caps but I may have a problem with connectors.
A potting box (to make it easier to put together, and faster than molding) could still be usable.

Get a potting box of the right size (I.e. something like this just as an example http://www.digikey.com/product-detail/en/PB-1574-BF/377-1964-ND/439633 ) drill two holes though the bottom of it, and use 4 hollow /non threaded standoffs, and assemble the board/box so you have 2 standoffs on top and 2 on bottom of the board.  Temporarily bolt the whole thing in your potting box. Fill it with potting, but dont cover the stand-offs completely.  Bake/cure the potting, and take your nuts/bolts out.

That should leave you with your board, potted into the box, with the same screw pattern, and the screws just run though the entire lump, but the aluminum sleeve provided by the standoffs keeps it from crushing the potting or box.  Plus  the box still provides better support then just trying to screw down a lump of potting. 
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #16 on: April 25, 2014, 01:00:11 pm »
thats a good idea, although we are still adding more and more weight
 

Offline lewis

  • Frequent Contributor
  • **
  • Posts: 704
  • Country: gb
  • Nullius in verba
Re: potting compound choices
« Reply #17 on: April 25, 2014, 01:38:08 pm »
Virtually no additional weight, the resin is much heavier than the box and standoffs.
I will not be pushed, filed, stamped, indexed, briefed, debriefed or numbered.
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #18 on: April 25, 2014, 01:47:25 pm »
I meant the box will be a little bigger than the board so more potting
 

Offline calexanian

  • Super Contributor
  • ***
  • Posts: 1881
  • Country: us
    • Alex-Tronix
Re: potting compound choices
« Reply #19 on: April 25, 2014, 02:28:10 pm »
Conformal dip.. Heavy bodies and multiple applications. That's the way I would go. I know they exist for the aerospace temp range. Also yes. They do have reinforced epoxies. The fill them with a number of different things for different effect. I have mixed fine cut fiberglass strands for strength before. Worked well.
Charles Alexanian
Alex-Tronix Control Systems
 

Offline lewis

  • Frequent Contributor
  • **
  • Posts: 704
  • Country: gb
  • Nullius in verba
Re: potting compound choices
« Reply #20 on: April 25, 2014, 02:28:54 pm »
I will not be pushed, filed, stamped, indexed, briefed, debriefed or numbered.
 

Online SeanB

  • Super Contributor
  • ***
  • Posts: 16284
  • Country: za
Re: potting compound choices
« Reply #21 on: April 25, 2014, 06:51:45 pm »
If it is mounted on a plate is it not possible to take the whole board, mount it on the plate then pot the whole lot after assembly. It is quite common for things like transformers in aircraft to have the plate as part of the unit. Solves a lot of problems. Not respinning a cheap board to fit in the same space, and to try to make a silk purse out of a pork ring sounds remarkably the same.

ps Do not google the pork ring, you will not like it.
 

Offline electronic_eel

  • Regular Contributor
  • *
  • Posts: 201
Re: potting compound choices
« Reply #22 on: April 25, 2014, 06:58:35 pm »
I have bad experience with epoxy potting over a larger temp range: the different tempco between board, components and epoxy can really rip the parts from your board.

I had a device pottet in some epoxy with good-sounding specs for permanent outdoor use. In winter (around -12 to -15°C) it stopped working. I analyzed the failure mode and carefully scrapped the epoxy away: several of the SMT components were ripped from their soldering joints, one ceramic cap broken in half.

Since this experience I have tried this: hard plastic case outside, inside filled with MS polymer. This stays very flexible but keeps condensation and water out. So far no problems.

If you don't want it that flexible, I'd look for a PU based material. That is a bit in the middle ground between the hard epoxy which rips your parts off, and the soft silicone/ms polymer.
 

Offline calexanian

  • Super Contributor
  • ***
  • Posts: 1881
  • Country: us
    • Alex-Tronix
Re: potting compound choices
« Reply #23 on: April 26, 2014, 12:45:46 am »
I have bad experience with epoxy potting over a larger temp range: the different tempco between board, components and epoxy can really rip the parts from your board.

I had a device pottet in some epoxy with good-sounding specs for permanent outdoor use. In winter (around -12 to -15°C) it stopped working. I analyzed the failure mode and carefully scrapped the epoxy away: several of the SMT components were ripped from their soldering joints, one ceramic cap broken in half.

Since this experience I have tried this: hard plastic case outside, inside filled with MS polymer. This stays very flexible but keeps condensation and water out. So far no problems.

If you don't want it that flexible, I'd look for a PU based material. That is a bit in the middle ground between the hard epoxy which rips your parts off, and the soft silicone/ms polymer.

Agreed. They have encapsulants that remain in a soft state to help with this. We used to buy a product, but I can't remember where from. May have been master bond.
Charles Alexanian
Alex-Tronix Control Systems
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: potting compound choices
« Reply #24 on: May 04, 2014, 12:54:29 pm »
the problem here is that it needs to be reasonably rigid as it's to stop stuff flying off the board so a movable mass of potting that will push stuff off the board under 500G of force is no god either. I also seem to remember that my sockets are not fully enclosed but have a slot up one side so it is looking like a nightmare.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf