Low voltage limit for LiFePO4 packs?

Hi y'all,

I have been building e-bikes since the '90s, but this is my first venture into lithium battery land. I am as baffled now as I was back then trying to decode Heinzmann motor specs.

Here's the context of my battery question: I'm electrifying a kick scooter for a dear friend of mine. I have salvaged a Wilderness Electric (Ri Chi) brushed hub motor from one of my early projects, and laced it into a 16" wheel. I have a couple of 36V brushed motor controllers in the 40A range that I can use. They are not programmable at all. I have lots of heavy cable and Powerpoles and solder, and plenty of experience putting these things to good use. But I want to get my friend set up with a reliable, not too expensive LiFePO4 battery that will play nicely with the rest of her system.

My friend went online and found various rechargeable tool batteries with predictably high prices, small capacities, and inflexible interfaces. I'm hoping I can do better than that.

I was looking at the LiFePO4 packs at HobbyKing, trying to figure out the simplest satisfactory arrangement for my friend, when it occurred to me that these controllers I have are intended for SLA batteries, and therefore they have a low voltage cutoff of about 29.5V or so. I don't know how deeply a LiFePO4 cell can be discharged, let alone a pack.

Ideally, I give my friend a system that does not require any close attention to a voltage readout, because she has other things to think about, and because she has adolescent kids who will use this thing. It would be ideal in this situation if the existing LVC could be matched to a LiFePo4 cell count that works for that value. I want to be as conservative with the discharge depth as necessary for long service life and reliability.

So here are my questions:

- What is the per-cell minimum voltage for LiFePO4? Does this number vary between different manufacturers' batteries?
- How much should I pad this value since I'll be using a pack of between 10 and 12 cells which must maintain balance?
- Does my existing 29.5V lo voltage cutoff correspond to an integral number of LiFePO4 cells?
- Can someone recommend a battery (or batteries) from HobbyKing or elsewhere? I'm looking for 4Ah or more.
- Can someone recommend a balancing charger from HobbyKing or elsewhere that would be a good match for this application?

Considering the objectives here-- long life, reliability, and brain-free operation-- is R/C LiFePO4 even a good idea? is there a better option?

Thanks for any insights from you knowledgeable folks.

Chalo

Can of worms this post is.

1) Figure out what the ACTUAL LVC is for the controllers you have. This will make things much simpler. (probably around 30v)
2) Try to see if you can hack the LVC of the controllers. It's generally not too hard. Post pictures of the controller, and maybe someone can help with this.
2) LVC for lifepo4 is.... well... the higher the better! By the time they get to 3v resting they only have 10% left - but many BMS are set as low as 2.1v - too low for my taste
3) If we assume controller LVC is 30v, then (30v/12s) = 2.55v lvc for lifepo4, which sounds about right.
4) Since the users will be non-technical and more importantly NON responsible (aka will drive it till the battery is dead every time) - a bms will be the only way to go for fool proof charging and balancing.
5) With out knowing how big the scooter is (or desired run time), I can't recommend a pack. But check out ping, headway, cell-man, and other common recommendations.

Sounds like a fun project. Off the top of my head, I'd say the easiest option would be to have cell man make you a custom a123 pack - 12s3p (36v 7ah) - it would be reliable and fool proof, but might be a bit costly.
Headways are freakin cheap, but probably too large.

Chalo;
From what I have gathered 2.2V is the absolute min. But, the life of your cells depends on your average DOD (depth of discharge). Closer you get to 2.2V each time you discharge the pack, the shorter your pack life will be. As an alternative, one guy I know takes them up to 4.2V just before he takes off, believing that it is better for the cells to get hit hard at the top than it is to get dragged down at the bottom. He has @ 18 months on his pack so far.

This thread will probably fill up with many expert opinions before very long.

Thanks, auraslip and Gordo, for your prompt replies.

In the FAQ here about lithium polymer batteries, someone says their depth of discharge should be limited to 80% for long cycle life. I have seen no such recommendation about LiFePO4 batteries, but it stands to reason that conservative depth of discharge would limit wear and tear and the opportunity for cell reversal. For simplicity's sake and to reduce potential points of failure, I'd like to use a balancing charger, but not a BMS that balances discharge. To me, this would be another reason to limit DOD.

On inspecting the controllers' specs, I find that one has an LVC rated 31.5V +/- 0.5V, and the other is rated 31V +/- 0.5V. The online retailer furnishing these manufacturer specs says the LVC "varies drastically from controller to controller" but doesn't say what that means voltage-wise. I don't have an instrumented power supply, so I can't easily test them.

It sounds to me like an 11 cell pack is the safe bet (corresponding to a rated LVC range of 2.77 to 2.91V per cell). I hope that does not complicate charging too much. I'm thinking two chargers, one for a five-cell pack and one for a six-cell pack?

Chalo

For simple charging on a low voltage LiFePo4, a BMS is really the way to go. No messing with multiple chargers, no messing with balance leads to the right charger. Dealing with all of that, especially for someone who has little to no clue (your friend) is a recipe for disaster. One wrong connection and ZAP, fried fingers, batteries, etc, etc. So with that said, a BMS doesn't really complicate things on the users end. It's one plug for charging, set it and forget it. Most BMS's use a separate charging plug so there is really no need to unplug the pack from the controller, unless your controller doesn't turn off for some reason. As long as you go with a reliable source for the BMS (cell man, ping) low voltage BMS's are pretty rock solid and trouble free. True, once in a blue moon someone on here comes in with one that goes out on them, but again, a quality vendor, like Cell man or Ping will take care of you promptly.

A lot of us are using Lipo ( cobalt ) from hobbyking but not very many are using the Lifepo4 packs from there, why ? , well, i have no first hand experience but i read that some of the specs are a bit exagerated...

That being said, the controllers being 40 amps max means you dont want to go with a Ping pack, or you need to get a 20ah pack minimum to cope with the load and this will likely not fit easily on a kick scooter.

About lvc on lifepo4, regardless of brand, the absolute lvc under load should be kept above 2.0v, the voltage profile of lifepo4 is very flat so expect to see a steady voltage reading during 80% of the capacity and a sharp drop at the end.. ( fully charged to 3.65v, drops to 3.4v overnight, and 3.2~2.8v during discharge is average )

Headway or PSI cells would do the trick, but cylindrical cells are bulky, those are pretty much the only cells at 10ah that can deliver 40 amps..

I"m always reluctant to recommend Lipo to newbs, but if you take care of the assembly and hook it up with some of gary's lvc boards and a hyperion charger ( pm " Ggoodrum on this forum ) , they are the smallest, cheapest, and most powerfull option.

cellman is certainly well known and recommended, he is in china, and does really good work, a pack of A123 cells ( 2.3ah per cell, so figure needed range and parallel as needed ) will cope with 40 amp controller easily.. so not trust the LVC on the controller imo.. either size the pack large enough that she will never drain it flat.. or get a BMS to monitor things.

Everything ypedal says

To elaborate on lifepo4 lvc; 2v is minimum. Lifepo4 is much more resistant to deep discharges. Sure. But a) if you're not going with cell level LVC protection like a BMS offers then you need to make damn sure they're in balance because with the LVC set at 2v per cell you have no room margin of safety and b) cycle life will suffer - not as drastically as with li-ion and lipo though

hobbyking lifepo4 packs are drastically over rated. They more like 2-3c IIRC from some threads here and on the rc forum. And they aren't as cheap as lipo either.

Yeah - it's a conundrum for sure.

Well, I think the lifepo4 you need for a 40 amp controller would either be A123's or headways, with a bms. That way the bms theoretically stops it in time. Have the bms set a bit high, like 36v. That's because teens are going to ride it till it trips, every single charge.

Another solution would be to tone down the power and go with a ping, vpower, or such lifepo4. Then you could go with a 10 ah battery, using a 15 amp controller.

Pure said:

For simple charging on a low voltage LiFePo4, a BMS is really the way to go. No messing with multiple chargers, no messing with balance leads to the right charger. Dealing with all of that, especially for someone who has little to no clue (your friend) is a recipe for disaster. One wrong connection and ZAP, fried fingers, batteries, etc, etc.

Click to expand...

I was under the impression that I would be able to securely stuff a bag or lidded case with batteries and balancing charger(s), and just leave the whole mess plugged or even hard-wired together all the time. A power cord would emerge from the bag when charging, and then stow for riding. Is this scenario only plausible with a pack-based BMS?

Chalo

most RC style chargers need the user to press buttons and confirm cell counts etc... :lol:

As others have said, Headways are a good choice because they can give you the 40amps you need without building a crazy high amp-hour pack for such a small ride. They are certainly bulkier than Lipo but nothing like as bulky as lead-acid. You should be able to hang them in a bag or box somewhere on your kick scooter. They are also really easy for DIYers to work with due to the screw terminal connections.

dogman said:

Well, I think the lifepo4 you need for a 40 amp controller would either be A123's or headways, with a bms.

Click to expand...

In this case, I think a 10Ah pack of any chemistry will be undesirably big and heavy and have more range than the vehicle's mission demands.

Am I stuck with A123s (which seem like a rather expensive option) or lithium polymer RC batteries to get 40A performance out of a sub-10 pound pack? (Sub-5 lbs would be much better).

This is the scooter: http://www.digglerstore.com/products/dirt-dawg.html#


The hub motor wheel weighs about 15 pounds and mounts in back. I was hoping to mount everything else, battery included, on the handlebar or the vertical frame member, or in the slim space between bottom frame rails.

Chalo

http://www.hobbyking.com/hobbyking/store/__16595__Turnigy_3300mAh_5S_30C_Lipo_Pack_USA_Warehouse_.html

2 of these would do.. 1 kg for 3.3ah and no problem with 40 amps.

Or you could get 5ah of 20C and put out the same 40 amps no prob..

http://www.hobbyking.com/hobbyking/...esAndProducts.asp?idCategory=239&LiPoConfig=5

If they fit anyhow.
> 20C batteries aren't worth the $ for our application. The cost per ah goes up quite a bit. Just add more AH.

And hand them to teens to manage...... right. Well, a teen aged Justin or such sure. But not a teen like I was.

No free lunch for 40 amps and foolproof operation by kids. I'd say lower amps or A123. With lower amps you could go with one of those water bottle batteries.

The 8 Ah headway cells can pull 40 amps, but that would be at their peak continuous Amp limit.

http://www.evassemble.com/index.php?main_page=product_info&cPath=2&products_id=20

I would personally go with the 10Ah cells for this, they will give you a little more wiggle room for amperage pull. Yeah they are going to be larger than than the A123s or LiPo, but as with anything, there's going to have to be a compromise. Cost, size, safety, it depends witch of the 3 are more valuable to your application.

That scooter at 40amps, ? Your friend my be spending some time on Her ass. Try a 25amp. controller with that 16in.

999zip999 said:

That scooter at 40amps, ? Your friend my be spending some time on Her ass. Try a 25amp. controller with that 16in.

Click to expand...

That might be a good idea if it opens up some more battery options. It certainly would make for more runtime per charge. And TNC Scooters offers a 36V x 25A brushed motor controller for only $18.

I could be wrong, but I doubt that 36V x 40A will result in a scooter that is too temperamental. I have run the same hub motor in a 26" wheel at 24V and a measured 50A, which was seriously underwhelming. 36V on a 16" wheel at 40A should result in a lower road speed, but a little more thrust on acceleration. This is a brushed motor that doesn't make as much "go" for any given electrical power as the equipment most of us are familiar with.

On the other hand, my friend is small and light, and so are her kids and the scooter itself. Realistically she'd get about twice as much zoom per watt as I did, on that basis alone.

Chalo