Lithium Polymer Battery Packs?


#1

I'm an electric RC airplane nut so I'm already familiar with all the impellers and motor/esc combos you guys are using. I don't see any reason at all not to change your "C" battery tubes to be strings of lithium polymer batteries in series and parallel to give you much greater run time. The weight offset could be corrected using a little lead to make up for the lighter LiPo battery packs. Being rechargeable is a huge plus as well, but the longer run time would be huge in my intended use.

Have we discussed this before and i missed it, it can't really be considered cost-prohibitive since we are building ROV's. LOL The previous charging problems with LiPo's have been rectified by using balancing chargers that use a simple wiring harness to charge each cell individually. I am new to ROV's, but I can't really see an issue using them unless the PCBs are not compatible with the voltage combos of the 3.6v cells. Why aren't we already using LiPos for more run time and rechargeability?


Howto: Find the most popular posts of all time
#2

Jason,

That's a very valid point that people bring up quite a bit. In short- yes, I think that would work. The main reason we haven't done it so far really just relates to development time (we're swamped with other development tasks as it is). As you mentioned, buoyancy and balance would also need to be corrected for, as well as making packs with a form factor that accepts the new batteries. I liked using batteries that could be found almost anywhere since during many remote expeditions, charging is not an option and turn-around time is crucial (of course, just bringing several sets of charged battery packs would work too). Anyway, the bottom line is that I think is something worth perusing, but don't have any time to peruse it. We could use help!

Eric


#3

Thanks Eric, that does make sense to use a common, readily available power source. I'm spoiled in that I'm usually on a support vessel with a generator while hunting for shipwrecks. I have plenty of experience with LiPo's, so I'll be the test subject for them. Just have to get some approval from my superiors to order the ROV kit, but it shouldn't be an issue. Once here, I'll start working on a LiPo conversion and post my progress. I've already sourced cells that are the same width as the diameter of a C battery. Based on my calculations, we can get 4 amps of 11.1volts inside your existing battery tubes. Will do some weight calcs too, if I find another solution that will hold enough LiPos to acheive the same weight, we may be looking at over 10 amps...talk about some run time!

Maybe overkill for most users, but it sounds like fun to me. :-) Keep it up, this project is a great one.

Jason


#4

Jason, that searcher would be AWESOME! Please me know how it turns out and if there is anything we can do to help. This is the kind of thing I was hoping would happen by making the project open source and getting the community involved with it. Good luck!


#5

That was exactly my same question: lipo batteries would be much cheaper in the long run than normal C batteries. I haven’t done the math yet, but gut feeling says using a few lipos in parallel we might get even more than the 1hr you get with C batteries.
Or maybe a good option would be building some swappable battery tubes so that normally we use lipos and then, when needed we can swap the tube and use the C batteries.


#6

Has there been any progress on this search? After reviewing probable power sources I've deduced that Rechargeable Li-po batteries would be ideal. The problem that comes up with Li-Pos is that they come in 3.7v increments, which means we would either have to run at 11.1v or 14.8v. By my knowledge, running the ROV at 11.1v would not be a problem. The ESCs are optimized to run using "3S" batteries, which are 11.1v, the LED arrays run at 8-30v, and the beaglebone (from what I believe) is powered with a switching regulator with variable input (or could be changed to accommodate the change). The only real addition to the ROV, aside from larger/different sized battery compartments would be a voltage/current monitor to make sure the batteries don't get deep-discharged. For redundancy, an additional smaller 11.1v battery could be mounted in the electronics compartment (if there's room) to provide emergency backup power in the case of overly-discharged main batteries, switchable by software or hardware via a relay (with a capacitor to prevent power loss on the receiving end during the switch). Considering the very large variety of battery packs and sizes, power options using this method could be extremely scalable for both price and capacity.


#7

Hey SpringHalo,

The new OpenROV Cape we've been working on will be able to efficiently regulate down from up to 36v. This means lithium batteries in series that add up to anything less then that should be just fine. We've been very seriously looking at these batteries:

http://www.batteryspace.com/LiMnNi-Rechargeable-26650-Cell-3.7V-4000-mAh-4.0A-rated-14.8Wh.aspx

You're also quite right about a few of the other issues. We'll have to figure out what to do about adapting size, over-discharge protection, charging methods, mass (density) differences, etc.

If you're willing to do some experimentation, we'd love to get some ideas here!

Eric


#8

At the moment I'm planning on building one myself, as I have access to a laser cutter and quite a bit of tools. I also plan on using a raspberry pi with an arduino attached similarly. What interface does the OpenROV cape use to communicate with the beaglebone? Just an I2C connection and +3.3v and 5v lines, or does it use I/O pins as well? I'd love to use a prebuilt cape and just adapt it for Pi-use so we can save $$ on the microcontroller. Plus, if it IS I2C, there's essentially infinite scaling for sensors, one of which I plan to use (compass+accelerometer IC) for navigation.

As for the ESCs, I saw discussion about this one, and am wondering if that's viable, as it's a few bucks cheaper than the ones currently in the BOM.


#9

We've only tested using the 18A reversible from Turnigy. It might work, let us know if you decide to test it!


#10

At the moment the only difference I see between them is the burst current handling, with the one I linked only handling up to 25A burst compared to the 50A provided by the current choice. I think $2 extra is reasonable for double the burst handling, especially in the non-standard high-friction environment that we're using the motors.


#11

I would say definitely worth the extra few bucks...I have seen measurable current draw changes when flying (electric RC aircraft) in sea level air as compared to flying in thinner air at higher altitudes. Pushing water will definitely create some very high burst scenarios.

Sorry I dropped the ball on my lithium polymer tests, we made a significant shipwreck discovery in the Dominican Repubic which has had me running like crazy since we found it. I likely won't get back to my ROV project until well after the new year (provided the world doesn't end). :-)

www.arsdr.com has details of the shipwreck if you guys are interested. Starting deep water stuff next year with AUV and ROV equipment...can't wait!!!!

Jason


#12

That's some neat stuff you've got yourself into. I hope you have fun with it. Thanks for the insight on ESCs. I think that if I go with a LiPo design (highly probable because of the ridiculous C battery cost) I would use two 5000mAh 3S 20C batteries in parallel. By my calculations they should be able to provide 400A of burst current (I think; limited knowledge of RC terms). Does that design look probable to you? For voltage monitoring, a simple voltage divider on the main power line would scale the voltage down to <5v levels for the arduino's analog inputs to measure. ESCs have programmable limiters for when voltage gets too low as a secondary measure (3.0v per cell would cut power, 3.3v per cell would limit speed), and 3.4v per cell (10.2v on the mains) would trigger a warning on the control station to return to base or something similar. We could also get a 'run time remaining' algorithm set up for better knowledge of the ROV's status.

Sam


#13

You may all know this but I'd like to add that I just found out that the batteryspace c-form batteries that Eric linked to could play very nicely with the stock battery tubes:

3 of those li-ion batteries is nearly the exact same length as 4 alkaline c-cells, just a few mm shorter. Even better is that 3 of them weigh about only about 5 grams or so less than 4 alkalines. I am also worried like you about the overdischarge protection. Batteryspace sells some pretty slender overdischarge protection chips that could conceivable fit into the main electronics tube, I am not sure. They could also be incorporated into a longer battery tube, again not sure though. They are hooked up to and balance each 3.7V cell of the system... I really like what you are saying about the esc's cutting out when the voltage gets too low. Then we wouldnt need that chip. But I dont have any experience with esc's but it sounds like each 3.7V cell must be attached independently. So it doesnt look like there is any way around of needing a bunch more wires coming out of the battery tubes than the stock setup?


#14

The only problem I see with those is that they're rated for 4A continuous and 8A peak discharge, which pales in comparison to common RC batteries (100A+ continuous discharge), so they might not be adequate for the application that you're looking at. In addition, they don't have built-in overcharge protection like most battery packs sold for RC purposes, and you'd have to somehow add that separately.

Since I don't have much experience at all in this subject though, I think it would be wise to go to a RC forum and inquire there to get the opinions of people more knowledgeable in that field. I might also do that, then report back with any results I find.


#15

Actually, I just do all the individual wiring at the connector coming out of the battery bank. If I'm using 3 cells, I use an 8 port balancing connector. The two outside pins are pos & neg out in series, the inside 6, 2 go to each cell for charging each one individually.

That is a very basic description, but I'm sure you get the point. I like the E-Flite system, look for them, they have batteries already wired up this way with balancing chargers to match. You can buy the connectors and charger from them to simplify...Turnigy also has a lot of this stuff too. Search for a "balance charging wiring diagram" and you can see a diagram for how to do this. Nice find on the C-form batts....those may be the answer we have been looking for.

Jason


#16

Started a thread on RCgroups.com, already have some good suggestions and ideas. Feel free to contribute or add your own questions because I'm not entirely sure of our goals here.


#17

Hi guys i am from the RC Community too...!!! I was thinking about the ROV batteries and i think that i found something that might do a good job..!!!

What about these batteries.... A123

These are A123 batteries..!!! I know that they are a bit pricey for example you will need 4 for an ROV in order to supply 12 volts..!! But they are cheap in the long run.

Moreover they have a really really high discharge rate of 30C...!!!!!

I think that i will give them a try..!!

Also if you connect them in series to get the 12V and then the two packs in parallel you get 9200 mAMPS..!!


#18

I suggest you look at the thread I linked, where discussion about the topic has already begun. As for the cells that you linked, they're only rated at 2300mAh so you would actually need 3 or 4 sets of 4 in order to supply the 7-10Ah's that the default battery combination (C batteries) provides.


#19

What I've gathered from the thread over in RC forums is that going with LiFePO4 battery cells is going to be the ticket. They're supposedly much more hardy than regular LiPo cells, albiet 20-50% more expensive from the get-go. These proposed cells are going to end up being extremely easy to implement into new ROVs with a simple change in tube length and diameter (1.5" ID, Unsure of length). Charging is as simple as designing a 4-battery enclosure and popping the batteries in when need be, connecting that to a low cost charger. No soldering cells together or messing around with funky connectors. They act just like alkaline batteries, except for the small fact that they're twice as long, a bit heavier, and cost $23 a pop!


In addition to larger battery casings, there will have to be additional positively buoyant cells to compensate for the additional weight (~1.4kg), but the ROV runtime should be more than 25% longer than with standard C batteries (estimated using the maximum alkaline capacity and minimum LiFe capacity)


#20

HI

I know nearly nothing about batteries. But whati noticed with my new car, that the battery broke down at plus 3 degrees celcius. What about the temperature tolerance of conventional batteries compared with LiPos??

Regards

Tobi