Fuzzy logic

One common approach is to use soft silicone under the harder potting compound. That tends to make things like radial-lead capacitors survive a lot better. The pressure inside the silicone would be pretty nearly hydrostatic, I should think.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Hi thanks, Yes that might work, but how will silicone behave with the surrounding polyurethane potting? will they bond and be stable, or will they react with each other and maybe dissolve, or other nastiness?

I also remember from somewhere that silicone is quite compressible so I might have to calculate a rather large volume change...

Br, Rune

Reasons not to go for a 1-atm container: There are 5 AWG 10 coaxial cables and 4 AWG 24 cables connected to this PCB. These cables will require penetrators, preferably coaxial type, which is not easy to find. The penetrators will be large and bulky, and occupy space I don't have available.

Dunno. It works well under epoxy.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

--
So use connectors instead: 

http://seaconworldwide.com/wp-content/uploads/CM2000.pdf 

More data at: 

http://seaconworldwide.com/products/electrical-underwater-mateable/cm2000/

50

ls

ew

r n a R f e

As far as chemical interaction, no way will the two polymers dissolve each other, etc. Phil's idea is mechanically sound, but I question whether you n eed a bond between the silicone surface and the urethane. I don't think you do, if the PUR surrounds the Silicone completely. Silicone potting was wid ely used in the IBM 360. Of course, you shouldn't use RTV, as it contains acetic acid. Another fscto r is that epoxies at least will increase leakage current if the circuit is sensitive to pico or femto amps.

Ther are primers for potting compounds that will make them stick to other materials.

Cheers

On Tue, 28 Jan 2014 11:18:56 +0100, Rune Gave us:

Cytek's "Conap" is NASA "Space Approved"

It is VERY tough stuff. So is the price.

Thanks, I will keep those connectors in mind if I should end up using a

1-atm container. As the requirements now stands they are still to space consuming. And I would really like to fulfill those requirements. So unless I somehow can build a 1-atm container, rated for 5000psi, with penetrators/connectors that adds less than 2" to the diameter of the 20" pipe the 1-atm container is a no go.

Thanks, No I probably don't need them to bond since it is all covered in the outer polyurethane potting. I have also been considering using a softer PUR potting (ShA 45) around the PCB, Is there any reason to choose silicone or other materials over PUR?

One, probably minimal, concern, but needs to be considered is: The lower bulk modulus of the silicone (1.5 GPa or so, PUR is about 2.3GPa I think, found little data on this) will led to a difference in volume change between the two Potting materials. this again will lead to some degree of stress and deformation on the surrounding PUR potting to fill inn the volume reduction of the silicone.. The deformation of the surrounding PUR needs to be controlled so that it occurs where it does not lead to stress on the PCB trough its rather stiff cable connections.. The deformation must probably occur normal to the PCB plane facing the sea water so that the PUR potting between the silicone and the sea water will act as a flexible pressure compensation membrane between the silicone and sea water.

This would not need to be a concern if the Bulk modulus of the two material where equal, so maybe a softer inner PUR potting is the way to go anyway?

Circuit is not sensitive to leakage below mA. Its a 1:4 MUX for a short

200A current pulse. Switching only occurs at zero current so ~no heat dissipation to consider.

Br, Rune

--
Hmmm... things are getting interesting! :-) 

What can you tell us about the cables and their connections to the 
PCB, the PCB physicals, and the PCB schematic?

The very low shear strength of soft silicone is a plus in this application, because that helps to make sure the pressure inside is hydrostatic. Otherwise you can get all sorts of shear stress on the solder joints. I don't think you care if the silicone delaminates, because its only real job is to keep the polyurethane away from the board.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Well, I can see a couple concerns. One is that if you use a soft inner potting material, the outer potting will squeeze inward, and you have to make it robust enough to not crack under the pressure.

5000 PSI is a LOT of pressure for plastic materials.

Second, I think there will be some sort of pressure gradient from outside to inside, and the wire feedthroughs need to remain hermetic or the water gets into the PCB. So, you most certainly can't just pass PVC-insulated stranded wires through the outer potting, you might need bare, solid conductors prepared for a good bond to the potting, and then have a bit of a snake to the conductors so they stay put and don't push inward axially.

The people who do this undersea stuff and oil well borehole gear must already have worked this all out. Maybe there are papers available describing what does, and DOESN'T, work.

I don't know the size of your PCB, but if it is small, then it seems a stainless cylinder with a hermetic flange with copper O-ring, and a Kovar seal for the electrical terminals would do it. I do more with vacuum feedthroughs than high pressure, but I know this stuff is commercially available.

Jon

Do you have any example of such a primer for a PUR silicone bond?

Thanks

pot the circuit board in a polyurethane box using silicone, pot the box in polyurethane. stuff doesn't like sticking to silicone, but silicone sticks to stuff.

--
For a good time: install ntp 

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

Example for Sylgard..

Cheers

You convince me. So thank you guys for all contributions so far. I will set up a cyclic pressure test on this silicone inside polyurethane potting concept, and we'll see how it does.

Kind regards,

--
In order to make your test(s) more ralistic, don't forget that sea 

temperature of a bit less than 2C.

Beware: My recollection from the late 1960s is that silicone and polyurethane interfere with each other's cure. Don't know if there were compatibility problems once both are well cured, but it's a possibility. Things may have changed since then, but I would suggest that you discuss your problem with plastics experts from the resin manufacturers.

We did coat glass components in urethane rubber as a buffer coat before potting in hard epoxy, precisely to prevent component failures due to cracks due to stress in the glass. (This was for the space program.)

I do like the idea of a soft potting material surrounded by a hard but still flexible potting agent, to ensure isostatic pressure on the components. At that pressure, everything is compressible, so some modulus matching is in your future.

Joe Gwinn