Subsea Electronics to 4500psi

I know ROV manufacturers submerge all circuit boards and motor drives in an oil bath, and then expose the electronics to the surrounding subsea pressure. At 10000 ft, this is about 4500 psi. What I don't know, is what specific components would have problems with the 4500 psi.

  1. ICs should manage the pressure OK as they are reasonably encapsulated without air trapped inside.
  2. I spoke with a circuit board manufacturer and they thought some surface mount component boards might have a problem, as ball grid arrays may have air trapped within the vias. Anything with trapped air would be subject to mechanical stress as the air is compressed. Any thoughts on circuit board design for high pressure applications?
  3. Clocks? I don't know much about what sort of clock packaging would survive OK under pressure. Help!
  4. Capacitors? What type of capacitors would be OK and which ones may have a problem? What if the packaging is surface mount?
  5. Connectors? The pressure may be OK for most connectors, but the insulating oil bath may cause a problem. I'm sure solder connections are better than pluggable headers, but I'd like to have as much off-the-shelf products as possible. COTS!
  6. The equipment I would like to package for a subsea application is PC104 based processors and I/O, and a 30hp VFD. Any thoughts on typical VFDs under pressure? Could be IGBT or back-to-back thyristors

- motor would probably be a brushless DC motor or 3 phase induction.

  1. What else?

Temperature should be above freezing, so temperature should be a managable problem and heat dissipation is definately a simplified issue.

Anyone who has any experience with such an application can email me at snipped-for-privacy@myway.com. Any info or past experience stories would be appreciated!

OJ

Reply to
OJ
Loading thread data ...

You might consider putting a 4500 psi pressure test in your specification for the circuit board manufacturer. And even the bare board could give you some problems. They too, get small amounts of moisture and gas entrapment. Make sure they put lots of test cupuons on the board frame and require them to provide serialized cross sections. That should help keep them honest! Good luck with your project....Ross

Reply to
Ross Mac

On Sun, 26 Oct 2003 17:46:57 GMT, "Ross Mac" Gave us:

Hahahha...

No, dipshit. PCB material is hygroscopic, which means that it can suck up a LOT of moisture, not "small amounts".

The solution is to full encapsulate the product in RTV or polyurethane encapsulant.

A slight fix would be to use a conformal coating, though it is not likely to help at that pressure.

You are old school.

Reply to
DarkMatter

On Mon, 27 Oct 2003 01:23:41 +0000, Guy Macon Gave us:

Solid forms are more likely. Liquid is not likely at all.

RTV is common. Hard epoxies are even used.

NASA uses polyurethane.

Reply to
DarkMatter

Silicone liquid is more likely.

Is this just curiosity, or do you have a project? It it's just curiosity, newsgroup answers are fine. If you are making something that has to work at 10000 ft. you need expert help. Your best bet would be someone who has made several 10000 ft. ROVs Your second best bet would be someone like me - see my resume at

formatting link
to see if I am a good fit for your project.

Not always. Some have cavities, many have bubbles. If I was in charge of the project, I would consider using QuickPak (

formatting link
) with a vacuum-based bubble remover.

Most surface mount components trap air in the attach epoxy. The problem with the boards isn't so much the vias but the blind vias. It's just that BGAs usually require blind vias.

Your fiberglass also traps air in it. I would look into Teflon PWBs

Exactly right. And it won't be easy to service, either!

Crystals don't work when encapsulated - they come in cans (or, in your case crushed cans!). I would have to research ceramic resonators. How accurate must the clock be?

You can go all ceramic.

Depends. I think the attach resin will be a problem.

Why do you think that soldering rather than using connectors isn't off the shelf? (I wonder if the mil-spec Radsok would work...)

BTW, have you thought about voids in the solder joints?

Oops! There goes your choices as far as connectors and ICs! I have never seen a PC104 CPU board without a can crystal. Have you?

Hard to tell. The power semiconductors shouldn't be a problem, but the rest? I don't know.

Have you solved any pressure related motor problems?

In an oil bath? Pretty much a non-issue.

Why you shouldn't ask for E-mail responses on Usenet:

formatting link

Reply to
Guy Macon

Having worked on the Space Shuttle, I can confirm that semi-solid is used in air and space. I have never done deep sea, but I can see some real problems with it if used that deep underwater. In that application, you need to fill the entire enclosure. A liquid seems like the best way to do that. A layer of semi-saolid (maybe 1/4 inch?) covering the boards would seem to be a good idea as well. These are just guesses, of course; I don't have the experience to know for sure.

Reply to
Guy Macon

Since you are the expert....how do you handle moisture encapsulation in a plated through hole???? All it takes is a particle from the drilling process's backup material to be caught in the hole, then it goes to the electroless process where it gets wet, then to imaging and the plating process where it gets plated over.....That is why they use huge vibrators in the plating operation. To get the crumbs out from the drilling process! But!...it does not always work... It is well known that FR4 materials absorb moisture. That is usually controlled by keeping the material in an air conditioned, humidity controlled environment. So moisture from the material would not likely ever be a problem. Coupons old school ?....that is a requirement for Lucent, Tellabs, Cisco, Nortel Networks and many, many others and that is Today!....do your homework here....You are way off base! And you have ever seen a plated through hole blow?....well it does happen more than you know....I spent a number of years around that business...I don't think you want to argue with me on this one.

entrapment.

Reply to
Ross Mac

Out of curiousity, why do you believe that ICs don't have air trapped inside?

Have you cracked any open to see how they are constructed? If not, please do so, because you may be in for a large surprise.

Harry C.

Reply to
Harry Conover

Reply to
Ross Mac

That quote was from Guy Macon....check with him.....that is a bit out of my field of expertise.....take care, Ross

Reply to
Ross Mac

Actually it wasn't.

It was OJ who wrote:

OJ: ICs should manage the pressure OK as they are reasonably OJ: encapsulated without air trapped inside.

and I replied with:

GM: Not always. Some have cavities, many have bubbles. If I was GM: in charge of the project, I would consider using QuickPak GM: (

formatting link
) GM: with a vacuum-based bubble remover.

(BTW, I don't need to break open an IC to see what's inside. I just watch the top blow off when I am demonstrating it to some general...)

Here are some tips for making sure everyone gets quoted properly: How To Followup A Post On Netnews Properly

formatting link
Usenet 101: How to properly format a Usenet post
formatting link
Posting help for newbies
formatting link

Reply to
Guy Macon

Of couse he does. DarkMatter always wants to argue with you. That's why we all like him. And of course you always want to argue with DarkMatter. That's why we all like you.

May I make a modest suggestion, though? Could you fine fellows please take it elswhere - perhaps alt.flame - so that we can concentrate on electronics here? Thanks in advance!

***GROUP HUG***
Reply to
Guy Macon

You might want to try posting in sci.electronics.design. In particular, Win Hill has many years of experience designing oceanographic instrumentation and may be willing to offer some pointers.

This request is considered a bit rude on usenet.

Best regards, Spehro Pefhany

Reply to
Spehro Pefhany

That's correct....I missed the">".....later, Ross

Reply to
Ross Mac

Group Hug???? Yuck!!!!!!!!!!

Reply to
Ross Mac

Sorry Ross, my error. Harry C.

Reply to
Harry Conover

No problem at all....in fact I managed to make one myself in replying!.....take care Harry, Ross

Reply to
Ross Mac

One other thing, when you laminate a multilayer it is under high temperatures that removes moisture, then the board is again baked, for cure reasons prior to the drill process and again later in the process dependant on many other factors. By the time the board is complete, there should not be moisture in the material itself. You may also find desicant in the electrostatic sealed package that the board is shipped in. That leaves the plated through hole as the most likely culprit of moisture entrapment. That cannot be baked out but may show as a void. If the customer puts a requirement in their spec that the whole order is to be scrapped if more than a small percentage of boards have voids, it will go a long way toward preventing the problem.

specification

Reply to
Ross Mac

With such extreme pressures, even solids have a problem. The idea is to allow the pressure inside and outside the container to equalize. The forces on even a modest enclosure would squeeze pretty hard on epoxies or RTV, deforming them and the container.

Completely filling with oil (or silicone liquid) with a small 'bladder' or diaphram exposed to sea pressure works much better. As pressure rises, the diaphragm deflects a tiny amount, and a small amount of the *nearly* incompressible liquid is forced into the 'can', keeping pressure equal across the can walls. Of course if you leave an air pocket, more oil must be forced in to equalize, and if too much is needed, the diaphram deforms too much and ruptures. Now seawater is into you electronics :-(

Perhaps the best is to 'pot' the electronics so water won't affect them, then fill the free space in the enclosure with silicone liquid to keep the pressure equal. But the 'potting' has to be perfect. Any voids and it will add to the total voids the oil/diaphram have to accomodate.

daestrom

Reply to
daestrom

Reply to
**THE-RFI-EMI-GUY**

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.