18
$\begingroup$

I'm looking to build an underwater glider robot that will need to remain autonomous for long periods of time, perhaps even months. Power draw should be minimal, and I'm thinking of including some form of charging device (such as a solar charger) however I'd also like the battery capacity to be large enough so I don't hugely need to worry about this. Large current draw isn't really needed, but the battery does need to hold its charge effectively for long periods of time. Considering this is an underwater vehicle, weight and size are also a concern.

Cost isn't too much of an issue, as long as it's within reason of a hobbyist project.

I am looking to understand the pros and cons of each technology (Lead acid, LiPo, NiCad, fuel cell?), so I can decide what type of battery would be best suited to my purpose. As such, I'm looking at battery technology rather than looking for a specific shopping recommendation.

$\endgroup$
  • 2
    $\begingroup$ I think this is at its essence a shopping recommendation. It would be better to ask about what characteristics of a battery make it a good choice. $\endgroup$ – jonsca Oct 23 '12 at 22:08
  • 6
    $\begingroup$ I disagree, he is asking which type of battery, not trademark etc. Those types do have quite different characteristics, but unfortunately I'm no expert at this and not able to give you a good answer. $\endgroup$ – bit-pirate Oct 23 '12 at 22:52
  • $\begingroup$ @bit In the context of SE, the word "shopping recommendation" is not necessarily for commercial questions. This one is borderline shopping, IMO, since it specifies some okay-ish constraints. $\endgroup$ – Manishearth Oct 24 '12 at 11:59
  • $\begingroup$ @berry keep in mind that LiPo explodes/catches fire/etc if it's shorted, so be extremely careful near water... $\endgroup$ – Manishearth Oct 24 '12 at 12:01
  • $\begingroup$ @MarkBooth Thanks for that, yup I agree that does make it clearer! $\endgroup$ – berry120 Nov 6 '12 at 21:41
14
$\begingroup$

The best commonly available battery technology when it comes to power density versus weight, and at reasonable cost, is LiPo.

I would prbably recommend LiFePO4 batteries. LiFePO4 are somewhat more expensive than LiPo for the same watt-hours, and weigh somewhat more, but they hold the charge better, they age much better, and they can be re-charged five to ten times more than LiPo batteries before they lose their capacity.

For both of those kinds of batteries, if weight matters, you can design the "battery enclosure" as part of your robot, and buy cells that are just the chemistry enclosed in thin insulating plastic, and build your own battery pack from those cells to fit a custom-shaped cavity in your robot. Make sure to protect the battery against over-charge / over-discharge, though, because it will ruin the battery if you don't. LiPo may even catch fire when you do, whereas LiFePO4 will not.

$\endgroup$
  • $\begingroup$ Awesome, thanks for that - especially LiFePO4, not something I'd heard of but now off to take a look! $\endgroup$ – berry120 Oct 24 '12 at 0:43
  • 3
    $\begingroup$ Cell overcharge and discharge protection, and particularly cell balancing are VERY important considerations if you are trying to build your own battery pack. Especially with LiFePO4 which are potentially dangerous is mistreated. Day job confidentiality considerations restrict my answer further! $\endgroup$ – Andrew Oct 24 '12 at 10:15
  • $\begingroup$ I have been working on AUVs for the last couple of years and have been using LiPO As far weight and charging time is concerned they are superstars. But their charge lifetime is not really great. LiFePO4 seems like a great alternative $\endgroup$ – user1101791 Oct 24 '12 at 10:20
  • $\begingroup$ LiPo may even catch fire when you do, whereas LiFePO4 will not. I beg to differ! $\endgroup$ – Andrew Oct 24 '12 at 11:40
  • $\begingroup$ I would appreciate links to credible resources showing LiFePO4 catching fire under "normal" abuse conditions. Yes, if you over-force way too much energy into it past charging end, any battery will heat up, and do it enough, and the enclosure will catch fire if nothing else. But so will a hot dog under those conditions :-) $\endgroup$ – Jon Watte Oct 30 '12 at 3:37
6
$\begingroup$

When taking into account size / weight / Ah, fuel cells and micro power turbines have an energy per kilogram and (with liquid fuels) an energy per liter that is an order of magnitude better than any battery.

In particular, I keep hearing that liquid methanol fuel cells are "relatively practical". a b c d e. And I've heard that Hyuk Kim and Seungdoo Park built a prototype diesel fuel cell.

Prototype microturbines also show promise.a b

Alas, fuel cells and micro power turbines (as well as some high-energy density batteries such as zinc-air) are not really designed for underwater vehicles -- they are designed to breathe air.

$\endgroup$
5
$\begingroup$

Lithium Ion Polymer batteries are probably your best bet.

  1. They will present the least amount of magnetic interference (followed by Alkaline... NiCD interferes the most).
  2. Unlike the regular lithium ion cells (which are usually round), these ones will stack nicely.
  3. I hope you'll check the manual before trying this -- many of them are pressure tolerant. I know of some AUV labs that just put stacks of cellphone-sized batteries into rectangular boxes filled with oil (for pressure compensation) and go with it. For example, Autosub 6000's battery packs -- depth-rated to 6000 meters, without the aid of a cylindrical or spherical pressure housing: Pressure Compensated Battery Pack (http://www.thesearethevoyages.net/jc44/autosub6000_page1.html). For other battery technologies, you'll have to decide how to pack them into your pressure housing, which may be a bit more complicated.
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.