Pressure compensation using mineral oil?


has anyone tried using mineral oil in the electrical compartments? using this clear liquid would eliminate the pressure problem and mineral oil is a non-conductive fluid. there are a few examples on the web of computers operating in a tank filled with mineral oil.

Manufacturing a shell for deep dives at 1000m+

Is this what you are refering to??

This does seam very relevent.


yup that is it. a good example and explanation.


Good call. We've talked about this a lot. Would love to see/hear about your results if you try it!

We're going to be really excited when getting deeper than 100m is our biggest design challenge - what a great problem to have!!


Probably adding oil to the electrinics chassis might also change the current buoyancy as will weight more and will not have air inside. So something else must be thought to keep it from sinking :)


Right. We've had situations where the ROV was slightly negatively buoyant. Adding small blocks of syntactic foam can work.


Have a look at this

(photo electric protected panel)

This can go down.. up 4000m... so its a nice solution to use mineral oil

Buoyancy will be affected but... you can go deeper


Cool Idea, I also often wondered about this. Got the idea from watching the "Abyss" . LOL, Has anyone ever tried this? I read about a hobbiest enclosing his prop motor in Oil. Though he did run into a problem with the motors over heating due to the small closed space of the Oil. The motor was well lubricated being suspened in the Oil, but the motor still over heated. He original thought being enlosed and submersed in water would dissipate some of the heat, perhaps if he had used a metal enclosure it would have acted as a heat sink to cool the motor. But PVC Pipe didn't seem to do that for him. He also used a good idea with the motor as the motor was attached to a maget, and the prop was attached to another magnet out side of the motor enclosure, the magnetic field connected the prop so he could completely seal the motor assembly and didn't need to weaken the assembly by drilling holes in it. The other nice thing is if something got caught in the prop, its not directly attached to the motor, so the prop was less likely to get damaged. Depends on the strength of the magnets used and how they are mounted. For motors he was using R/C car motors which are fast, but get hot quickly, he thought with the motors being in oil it would help with the heat, it did a little but not enough. Burned out the motor. I was hopeing he would try a metal enclosure and see if that made a difference, never saw anything further mentioned about it.


I've used mineral oil, silicone oil, and flourinert as pressure compensation fluids for high pressure electrical compartments. Mineral oil works well but is really messy and not fun to clean off for maintainence. Flourinert is nice, but really expensive and is about 1.8x the density of water (so its really heavy). Currently i'm using silicone oil, its cheap, not as messy as mineral oil, and flows better (so you can get air bubbles out easily)

This liquid compensation technique works because of the fact that all liquids are pretty much incompressible. Thus by filling a compartment with a non-polar liquid (so that it isn't conductive, and it doesn't mix with water) you essentially eliminate the pressure differential between the internal compartment and the external environment. Alot of commerical deepsea thrusters use this method; its typically called things like 'oil-filled' and 'pressure-compensated'

I used this technique in the tailsection of this underwater robot, and we pressure tested it to 1.5km of depth without any issues. Since there isn't any pressure differential, the electrical compartment doesn't have to be terrible structural! (in our case we machined the tailsection out of marine grade delrin).

But there is one design consideration that must be added; you have to have some sort of flexible diaphram to allow for the liquid in the compartment to change in volume. All liquids are 'essentially' incompressible, but they still slightly compress at high pressures. So you accomodate for that volume difference, and ensure that there is no pressure differential, you need to have some sort of flexible diaphram for the compartment.


Filling electrical boxes with mineral oil is pretty standard in the oceanographic field. Some prefer peanut oil and some use kerosene for its low viscosity.

Colin's comment about a small flexible diaphragm is important. Often a couple of inches of Tygon tubing will work for this. Either the Tygon can be filled with oil and stuck out into the water, or the tubing can be filled with water and stick into the oil filled housing.


As mentioned by others the use of various oils inside electronics housings is very common in the world of underwater robotics. All the work class ROV's I have been associated with use oil compensating systems to allow smaller housings for the required electronics - i.e. Schilling UHD and HD, and the Oceaneering systems.

The choice of oil depends on a couple factors. I am a fan of carnation light mineral oil due to the minimal enviornmental effects when it leaks. I have used plenty of Tellus and Royal Purple oils as well which are standard in the off-shore oil and gas industry.


Would the oil have any effect on the camera's clarity?

It must have an impact on the light refraction at the very least.


Oil in contact with lenses is going to totally mess up the optics. The only way you can do this it to have a lens with a flat side, or a flat window pane between the exterior fluid and the interior air. Either that or do the optical math to calculate the effect based on the curvature of the lens and the index of refraction of the particular fluid you are using.

The only underwater camera that I have ever worked on that did not have a flat window to the sea was a military one with a thick quartz dome. There were pages of math describing how the focus of the camera had to be corrected for salinity and temperature of the seawater (both affect index of refraction) and the flexing of the quartz dome due to pressure.


Also, having a flat lens makes it so you will be focused both in and out of the water (however your field of view will change).

I assume that a gas-less optic system would look something like this:

I've been thinking a lot about this recently, and have come up with a few criteria for what the ideal imaging system for an ROV would be like:

1. Very wide angle (~120deg??) field of view
2. Sensitive even in low light conditions (may conflict with #1 due to aperture size requirements)
3. Low cost (this probably means not using anything custom)
4. Works in air and in water (difference in field of view okay)
5. Physically fairly small
As you all probably know, lenses work by exploiting the difference in index of refraction between different mediums. The greater the index of refraction difference, the more powerful a lens of a given size will be. As a reference, the difference in index of refraction between air and glass (which is how most lens systems we use every day work) is about 0.5.
Here is a list of (approximate) index of refraction values for materials that may have a part in this design:
Water (sea or fresh) 1.33
Mineral oil: 1.47
Acrylic: 1.49
Glass: 1.52
Sapphire: 1.77
N-SF11 Glass: 1.78
Note the the index of refraction for some materials is observably different at different wavelengths, which can cause what's called "achromatic abberation" - that is, certain sides of the light spectrum not being focused the same as others. For underwater purposes, I've generally been pretty comfortable not worrying about achromatic aberration issues, as the range of wavelengths we'd be looking at is fairly narrow. Also, having a long depth of field might be nice so the camera does not need to be dynamically focused during a deployment, and having that long DOF will reduce achromatic aberration. All the same, these concerns can be addressed in exchange for increased cost by adding additional lenses to correct for the phenomena if needed.


Great video, thanks for posting. That explains the issues and the concerns (vent holes to completely fill whatever it is you're working with and get ALL of the air out).


That's great stuff right there. And thanks for the clarification of 'essentially' non-compressible.


Thanks for this post.

It brings up the subject of having different compartments for different components. We need to have 'eyes' in the water and that part (for all intents and purposes) needs to have air or at least a clear gas in it. The rest though: Not so much. Maybe in the case of a motor it needs to have some built-in cooling (and possibly the electronics as well, who knows?) but if you only have to worry about having a gas in the camera compartment it means the other part where the electronics are just has to be sealed, it doesn't need to withstand tremendous pressure differences.

From a design standpoint that changes the requirements of an ROV quite a bit if you're willing to dunk all of your electrical stuff in some gooey liquid. :)