PROJECT 82 – Machining a Pressure-Resistant Camera Viewport

Build/Make
1

Hello everyone,

I am currently building PROJECT 82, a tracked underwater crawler designed for underwater inspection and exploration.

Today I finished machining a pressure-resistant viewport for the front camera housing. The viewport consists of steel flanges and an acrylic window and is intended for operation at depths of approximately 20–30 meters.

I would appreciate any feedback regarding viewport sealing, acrylic thickness, or pressure testing methods.

Note: The linked build video is in Russian. However, I can provide additional information and answer questions in English here on the forum.

Thank you.

2

Hi @Proekt82, welcome to the forum, and thanks for sharing your project! :smiley:

  1. Both acrylic and steel are quite stiff, so getting a leak-free seal with a direct interface between them is likely very difficult
    • Have you considered something squishy in between, like an o-ring or gasket?
    • This guide may give a sense of some possible options, and how they can go together
  2. It looks like you’re using substantially thicker metal parts than is likely required by the depths you’re targeting, though that is a bit hard to estimate without provided dimensions (e.g. in a drawing or model)
    • The Under Pressure software may help you to get a more streamlined estimate of required thicknesses
    • Extra steel means extra weight, which makes for less convenient transport, but may also be desirable to make the vehicle negatively buoyant (so it stays “crawling” on the sea floor)
  3. There are various ways to pressure test, and which one(s) make sense depends on your robustness requirements, and what information you’re actually looking for
    • Destructive pressure testing is ideal for determining maximum depth ratings, but can be quite expensive, and may require substantial equipment
    • Submersion and holding at or a bit beyond the intended operating depth can determine depth capacity sufficiency, but doesn’t give a sense of safety factors / over-engineering
    • Vacuum testing can be a useful check for whether an enclosure is well-sealed
    • Dunking a sealed enclosure in water and looking for bubbles can also test seal quality, and help find leaks, but is risky if you already have electronics inside
    • Coating all seal points in a dishwashing liquid solution, then pumping in air to achieve a slight positive pressure can help to find leaks without requiring submersion