Paper: The possibility for new battery chemistries in FTC/FRC- Evaluating batteries

#1

Hey y’all,

For my AP Research topic, I decided to solve one of the problems of FTC… there weren’t enough whitepapers!

Just kidding. But with how absolutely overpowered some robots are compared to the batteries supporting them, I decided to evaluate four common battery chemistries: lead-acid, NiMH, Li-Ion (Li-Po), and LiFePO4 for use in FTC scale robots, which will hopefully branch into FRC as the technology scales. Hope you like it!

Portability and Performance of Battery Chemistries in Applied Robotics- Named.pdf (1.0 MB)

If anyone has any questions, please do not hesitate to ask.

10 Likes
#2

silver-zinc!!

1 Like
#3

Wow. Very thorough and impressive.

1 Like
#4

You nailed it.

2 Likes
#5

Thanks for sharing.

If they switched us to LiFePO4 chemistry, would our batteries be any more resilient to damage from imperfect charging practices than NiMH? That’s what I dislike about NiMH. Our FTC batteries get trashed quickly when somebody forgets them on a charger, and I haven’t been able to identify an idiot proof NiMH charger. If somebody else has suggestions, please share.

#6

Oh yes… I actually have a funny story.

I normally supervised the tester when I was doing the discharge tests for the batteries, which took about 20 minutes. However, I got sidetracked when I got to the LiFePO4 battery and I left the room… only to realize about 20 minutes later. I ran back and the voltage read on the battery tester was 0.0v, and the cells of the battery were a little warm. I was expecting it to act kinda like a LiPo and blow up on me, but it actually charged back up and lost no capacity!

I didn’t have the time to do any cycle testing, but from manufacturer specifications, not only do the LiFePO4 cells have the advantage of over 4x the cycle count, but they also have a much better life characteristic where they do not self-discharge much higher towards the end of their lives.

In general, any battery chemistry is better than NiMH for self-discharge. We actually leave our batteries on the charger 24/7 during testing/comp because otherwise they’re completely useless if you let them sit for even 12 hours.

#7

No help here. Just an observation: to make anything robust, designers must consider foreseeable misuse. Many misuses are relatively easy to foresee; that is because most people are reasonably predictable.

The actions of geniuses and idiots are often difficult to foresee.

1 Like
#8

Yes. This is why it’s hard for me to figure out how to end up with consistently decent FTC batteries. People are guaranteed to occasionally do things like putting a mostly charged battery on the high amp charging setting and leaving it on there for multiple days. If there was a charger that would back off and trickle charge instead, that would track better with realistic expectations for human behavior. This is apparently more challenging do with NiMH chemistry than with lead acid. A charger like that would be worth a few hundred bucks, considering it costs $50 to buy one easily damaged FTC battery.

#9

That’s intriguing. It makes me think that maybe we should get some LiFePO4 batteries for practice and testing and bring out the NiMH batteries only when needed.

#10

First… You should only be charging your batteries at 0.9A. Full stop. Break the switch off your charger if you have it.

Second, the chargers should do exactly that, trickle charge the battery. Any “dumb” battery charger you buy should be CC/CV, meaning that it will do 0.9A until it reaches the set voltage (up to 15v), and then back off to as low as 0.05A when it is fully charged. The battery charger I used, the TB6B, had an interesting functionality where it would pulse charge all batteries- it would charge the battery, then stop charging, let the voltage normalize, and then resume charging, about every minute. The battery never got even warm during charging. Might want to give that a try?

Also, if you’re interested in a LiFePO4 battery, just know there aren’t any commercially available ones. Just buy some cells on eBay (the A123 ones are ridiculously awesome), buy a balance charger (TB6B), wire in a balance connector (don’t use a cheap eBay BMS…), and then put a low-voltage cutoff on the output. If you have a Batteries+ nearby, they might have a spot welder and will weld tabs to the batteries if they don’t have them already

Good luck with your battery adventures :slight_smile:

#11

Having previously used LiPo packs that were almost identical to those… I am not surprised at all. We killed 18/20 of them that we purchased within a year.

I would be interested to take a closer look at the Li batteries used in drills as a potential battery alternative. They’re much safer than “naked” batteries, are commonly available, and the chargers are fairly foolproof.

#12

The VEX V5 system has actually addressed many of these issues. There were a few issues with the roll out, but it seems to prove that lithium batteries can be used safely in mid size robots.

Check out the specs https://www.vexrobotics.com/v5-architecture.html
Jason

#13

Yeah, I referenced the new V5 system in my paper. I’d love to try it out since my last VEX experience was well over 6 years ago now- I’d especially love to take apart the battery and control circuitry and see how they back up the claims of 100% power at low battery. I think they’re going in the right direction.

Hit me up VEX… :grin: