John and Cap'n Carter offer great points and advice to consider.
This year I had had enough with lead acid starting batteries on my Vanagon, which only gets driven really a few times a year. I looked at available lithium options that might work to replace the starting battery, but except for "maybe" one that checked at least most of the boxes, the price was in the higher hundred$ for a Lithium Ion option. Sure, LiFePo options are out there, but as pointed out earlier, there are BMS and other reasons they can't handle this application.
What did I do?
I made my own Lithium Ion starter battery. It was a lot of work and ended up not fitting, so I had to carve out some metal to get it to fit, but what an amazing starter battery I have now. Tested CCA (Battery Analyzer): 3000 CCA.
While it didn't cost me a whole lot to do, it was a lot of work.
In my case, I started with a known lithium ion cell stack I have used on my boat for 31-48v use (nom. 42v, 35ah, 12S2P config). I then disassembled all the interconnects and carefully made new ones to make a nom. 14v (4S6P) 105ah battery. I took the existing BMS card and made custom cabling, using just 4 of the monitoring channels. I already use these batteries and BMS cards on my boat pack (20P(12S(2P)) configuration for 700ah capacity) where I can monitor (and bypass if needed) any of the 240 cell pairs in the pack. For my Vanagon, there's just 4 cell hextets to monitor.
In either case, no active BMS is really needed given the operating conditions I have. In the boat's case, I almost never take the pack voltage below 36v and check and balance the pack about once a year (only maybe 4-7 charge cycles).
In the Vanagon's case, I measured the BMS readings at rest, during start and after charging and after a couple trips and have seen no change in differential cell voltages.
Performance: Rock solid starting (3000CCA vs maybe 250 or less with what I typically ended up with). Ability to disconnect and float the battery (and cells) for months or years at a time without lead-acid worry of "sulfation". Cost was under $300 total.
I still need to get the thing completely parked in place. I do have a new battery compartment lid I made, but still need to hammer out a replacement side wall. I also finally invested in a "house power" or auxiliary battery. In that case, I bought some of the lower cost nom. 12v LFP 25ah(?) boxes out there. The price was right, and they fit in the cavity nicely, but boy I really do hate that these battery manufacturers get away with embedding BMS electronics that cannot be disabled or won't self-disable during times of no discharge. Probably every one of these out there will "self-brick" within months if left alone. When these LFP batteries I got arrived, each one of them only read under 2volts!!! Fortunately, literature shows that most lithium chemistries will survive "self-discharge", just not reverse polarity (which is what you get when you serially discharge a series of cells). Still some of these LFP manufacturers (e.g. NEC) make batteries which, like laptop smart batteries, will blow an internal fuse or otherwise lock out the battery terminals from being charged or discharged if the voltage drops, say, below 10v. These may be recoverable as I found with the NEC batteries, but your mileage may vary…
One last thing: You might be thinking "Wow, you must have one heck of a beefy BMS if it can work as a starter battery!" Actually, no. The BMS cards I use is only good to bypass about 1amp around cell pairs. It's such a trickle that it can take days or longer to balance differentials of any significance. And that's okay. As configured, a 420amp starting current will draw just 70amps per cell with my battery. Now, that is not exactly insignificant as these are 17ah cells, so this is about a 4C draw. However, it's just for a very short time and the amount of amp-hours drawn during a start episode is miniscule---I measured drop in the low millivolts at the cell level during a start.
Anyway, long wind here, but for me, very happy with lithium ion for a starter battery.
-Myles
I checked out the post and the guy left out one simple yet important detail. You can't use a battery with a 100amp BMS for starting your car, which is what all of the cheaper prismatic cell lithium batteries come with. Thankfully most of them will point blank tell you not to use on your car but usually it's buried in the finer print that most people don't read or even follow half the time when they do.
As long as you shell out for the upgraded BMS and cells that will give you the CCA capacity you need then lithium is great but I don't think it's a better value because the batteries that can do that are much more expensive and that brings into questions the cost effectiveness of the extra investment, especially when you can pretty much count on a decent FLA to crank your car for 6 years or more. The lithium chemistry definitely outperforms Lead by leaps and bounds but it's the long-term reliability of the BMS I worry about. If you're paying 5 times more for a Lithium battery do you really think you'll get 5 times longer life? Do you even need 5 times longer life?
I'm a total Lithium convert and won't deny it's the near term future of energy storage. With the recent precipitous price drops, I finally converted the Arc from FLAs to LiFePo this year and am flabbergasted at how much better they are but I think it's safe to say lead is not quite dead. Yet.
Lurker here who enjoys all the informative info you all post.
This came down my Quora feed, I thought it might be of interest to you. It's by a Li-ion battery designer and he provides a lot of practical knowledge.
Founder, CEO
Act always as if the future of the universe depends on what you do, while laughing at yourself for thinking that anything you do makes any difference.
A Buddhist saying
