I’m building a new battery pack for our BlueROV2 using 21700 cells I found that that have some really nice characteristics for the price. (I’ve used these with great success in 48v ebike packs.)
A 4s6p rack of these cells gives 28 Ah for $120. That’s an all-day battery for us. (With a longer tube, we could even fit 4 more cells and bring it over 32 Ah.)
Problem is, the 21700 cells won’t fit in the stock 3 inch enclosure in an efficient way. To keep costs down, I’m thinking about permenantly sealing the pack in a piece of 3.5 or 4 inch schedule 80 PVC.
I’d run the power cables through one of the end caps, secure the battery pack inside the tube. Then glue the end caps on with PVC cement.
I’ll use a waterproof connector to attach the battery tube to the electronics enclosure or charger. (I’ve got several spare Subconn connectors laying around.)
This site puts the collapse pressure of 4" schedule 80 higher than 300 psi. Water pressure at 100m is about 160 psi. I could install a stainless service port for vacuum tests.
The permanently sealed battery enclosure appeals to me. The end caps are far too easy to knock out in the stock configuration, especially when lowering it over the side of the boat. Also, saltwater always drips into the tube during battery swaps offshore.
Am i overlooking any glaring reason why this wouldn’t work?
I see no reason why this wouldn’t work, providing you pot the wires exiting the tube correctly (strip insulation, pot in epoxy/silicone) so water can’t ingress along under the insulation.
Theoretically, you could pressurize the tube for deeper dives, offsetting the pressure pushing in. I have 4 used 5-gallon freon tanks that i use for static buoyancy in my ROV, pressurized to 200psi.
EDIT: Looking at the data on that PVC pipe site, i think they got their burst/collapse data fields reversed. I cannot imagine that PVC pipe would be able to withstand higher internal pressure than external pressure. 325psi burst pressure and 1,100psi crush pressure would make alot more sense.
Curious…what BMS are you using in your 4s6p pack builds?
I have used 4" Schedule 80 PVC as a battery housing in a towed camera sled and it works great. I have sent it to 1400’ deep ( >600psi) with no issues, and it has been well over 1000’ many times. I made a cap that allowed the housing to be opened to swap battery packs, but a sealed enclosure is probably even more robust.
A nice thing about PVC is that it is non-conductive, so you can make very low budget (single wire) penetrators by tapping a small hole (6-32 for example) and threading in a small screw with wires connected at both sides. Epoxy that up really good and you have a very reliable penetration for your low voltage battery power.
That should work just fine as long as you seal everything well. If you want a more oem look, I make/sell 1" taller side frame panels that allow fitment of a 4" BR bottle. You have to match drill the holes yourself, but it’s easy to do and the panels are fairly cheap at $80 + shipping ($18 for USPS Flat Rate).
Blockquote EDIT: Looking at the data on that PVC pipe site, i think they got their burst/collapse data fields reversed. I cannot imagine that PVC pipe would be able to withstand higher internal pressure than external pressure. 325psi burst pressure and 1,100psi crush pressure would make alot more sense.
I can’t vouch for that website’s numbers, but my guess is that it is a different failure mode. PVC is possibly a lot stronger in tension than compression?
Also, I usually skip the BMS. Occasionally I’ll check the voltages and balance manually if necessary.
Since this battery will be sealed up, I’m using one of these 100 A eBay BMSes, and hoping for the best.
I expect you are thinking about it but feel it needs to be said if others aren’t - any failures with this kind of configuration could easily result in an explosion and shrapnel. PVC is nasty when it does fail, too.
Make sure to design in failsafe wherever possible and develop processes for minimizing risk.