I keep reading that oil compensation would allow for most housings, even PVC pipes, be (theoretically) rated for full-ocean depth. But I’ve seen notes that some components would not function at these pressures inside of an oil-filled housing. Naturally, I could see how some capacitor types, relays, and batteries, components that are hollow but sealed, and cannot be oil-filled, would fail. And that eventually, some integrated circuits would crack, or that the solder joints on the PCBs would shrink and snap. What I’ve yet to see is a “straight” answer on when any of this would happen.
To those who have used oil-compensated housings in the past, what has your millage been? How deep could they go, what components were housed in them, and what components had to be housed in non-compensated 1-atm housings?
This topic isn’t one I have much experience with, so am unfortunately not able to go into depth about specific details and expectations. That said, from a theoretical standpoint component failures in an oil-filled component would be dependent on the crushing strength of the materials and structures involved.
I imagine many of those who have used oil-filled components won’t have had reason (or the necessary equipment) to test them to failure, so you may find limited information as to what doesn’t work. I would at least expect integrated circuits and solder joints to withstand much higher pressures than “can” capacitors, or other hollow components. That kind of thinking can perhaps guide you as to which information is most important/relevant to what you want to achieve (i.e. only the weakest component matters, and validation beyond a particular point stops providing useful results).
Do you have a use-case or some limitations in mind (e.g. can X components withstand Y pressure), or are you mostly putting out feelers for gotchas/limitations that others have found that you’ll need to look out for if pursuing an oil compensation approach?
My experience is with work class ROVs and oil filled electronic boxes is very common with PCB and control electronics immersed in the oil for intelligent hydraulic valve operation. Have a look at this link:
I know that crystal oscillators can fail under pressure. I have built some oil-filled servos (commercial R/C servos in my own waterproof case) that had crystal oscillators that failed at about 400 meters. You may be able to fix that by putting a blob of hard epoxy over the oscillator to protect it from the pressure. Of course, that technique requires careful surface preparation and a sufficiently hard epoxy.
All the other components in these servos did fine up to 600 meters. If your device needs a motor, try to use a brushless motor instead of a brushed motor.
Hi everyone,
I know this is an old post, but I´m looking into oil compensation aswell.
I was wondering if anyone has experience with some sort of “hybrid” system. What I would like to do is to use a 5" enclousure and fill it with Shell Diola or silikon oil, without a hydraulic compensator. This would not necessary increase the depth rating, but also not increase the pressure on single components like in conmpensated systems. The enclosure would still be absorbing the pressure.
Now why fill it with oil in the first place? I had alot of struggle with condensation, and want a different solution then silika gel.
The oil would prevent condensation, increase heat transfer from electronics to the alluminium enclousure and also prevent leaks.
Now I am wondering what you guys think about that!
Hi @michaelsloschek -
While that approach would help heat transfer, it may not make too much of a difference when it comes to leaks. If you’re able to fill the tube entirely with the oil, with no air present after sealing things up, the enclosure will transfer the external pressure to the internal fluid, by deforming the enclosure tube. As the oil is not very compressible, this is likely to help your depth rating, but will expose internal components to potentially significant external pressure…
Condensation is quite easy to manage - to me, keeping desiccant recharged with a toaster or microwave is vastly easier than dealing with large quantities of oil - this is quite messy!
Hi @tony-white
I have been doing some thinking and kinda don´t want to give up on this topic yet. I do alot of deep and long dives with my ROV and as mentioned always run into condensation problems. I usually use silica gel and fill the enclosure with Argon or Nitrogen.
Now that I am using a 5" Alluminium enclosure there is no way to monitor the amount of condensation and water accumulation inside. That´s why I am thinking about the Oil filling approach. Even tho it´s a mess. Once closed, it should be fine. Unless you need to open it of coure.
What do you think about filling the enclosure only partially with oil? Maybe 3/4 way. All the electronics would be covered in oil, no humidity should be able to affect any components, there is only a small amount of air left to condensate, and the pressure on the components should not be a problem, because the deforming of the tube will only compress the air.
Also if there is a leak, the water would displace the air, the enclosure compleatly filled and the electronics should still be safe in the oil. assuming the water is heavier and sings to the botom 1/4 of the enclosure.
I’m surprised, even shocked, that with dessicant and nitrogen filling you’re still running into condensation! I would have thought those measures, done properly, should prevent this, as I’ve never run into condensations despite typically operating in very humid (surface) environments.
As for partially filling with air, your assertion
This is incorrect - both the internal air and oil pressure would increase to the same level! So in this partial fill case, you wouldn’t really see any depth rating increase benefit without adding an external, oil-filled container that is compressed with depth, increasing the oil and air pressure, and minimizing the difference between external and internal pressure across the seals. This of course can damage components that are not pressure rated.
I’d suggest trying an oil-filled enclosure with a smaller system and seeing just what a mega mess you’d be dealing with!
A highly absorbent cloth or paper towel should collect any condensation without issue. If you’re seeing more water than that, I’d start to suspect you have a very small leak - are any of your seals using potted penetrators?!
Thanks for getting back to me @tony-white
aperrently I have a slightly wrong understanding of the pressure acting on air vs oil. The reason why I wanted to use the oil is to avoid condensation and condensation only. I don´t need to increase the depth. 1000m is more then enought. I won´t be going deeper than 400ish m.
The thought about having a very small leak never came up, because the amounts of water where too small for me to suspect a leak. On my last dive I had about 5-7 ml of water in the enclosure. Again I suspected condensation, because I yust reasembled the enclosure after changing to a Navigator board on my ROV.
Now thinkimg about it, it seems like a lot of water compared to my first ever dive and the condensation on those first dives.
So maybe the problem is somewhere completly else.
Yes, all my penetrator are potted on the current build. The new build will have Wetlink or Blue Trail Engineering connectors.
Are you thinking the leak comes from the penetrators?
Hi @michaelsloschek -
In an enclosed volume with a liquid and gas, if the volume is decreased, the pressure goes up in both fluids. One may compress more than the other though!
More than a drop or two of water is definitely a leak, and not condensation. Potted penetrators are notoriously vulnerable to leaks - this is definitely what is plaguing you! This video demonstrates how you can find the leak with some soapy water and your vacuum pump set to pressurize - I would bet one or more of the penetrators will have bubbles and need to be re-terminated. It is possible to upgrade to WetLink penetrators - see this post!
