[speculawyer]

Dave Kois has done a wonderful job of establishing a good pipeline to Chinese-made LiFePO4 batteries. See here:
http://www.evcomponents.com/SearchResults.asp?Cat=28

With LiFePO4s now costing less than $350/KWH, is there really any point to getting lead-acid batteries anymore? Granted, the LiFePO4s are still more expensive than lead-acid, but the fact that they are supposed to be good for some 2000 cycles means that they should easily last more than twice as long as a lead-acid set.

I think the only advantages of lead-acid at this point are:
1) Upfront price . . . they are less up front, but in the long run they will cost more.
2) Durability - Lead acid batteries can be abused and over-charged and they'll still work fine. LiFePO4s are much more sensitive. If you over-charge them or over-discharge them, they may die.

But (2) can be addressed with a good Battery management system. And this is the only area where I see real weakness in lithium-ion batteries. The available Battery Management Systems are crappy, expensive, flaky, and/or not easy to use. If someone produces a good inexpensive mass-produced easy-to-use Battery management system, I really think lead-acid is on its last legs.


This idea came to me when I was looking at helping someone deal with an NEV having battery issues. To replace a full set of 6 12-Volt lead acid batteries for an NEV can be pretty expensive. If you get good deep cycle marine batteries, they can cost close to $200 a piece. And since they are so damn heavy, you know the shipping will push them to that $200 price such that a full set of six batteries for a 72Volt based NEV is $1200.

Well, if one picks up 24 of these 60 Amp-hour lithium iron phophate batteries, the total price is just $1584:
http://www.evcomponents.com/ProductD...de=TS-LFP60AHA
$1584 is just a little more expensive than the $1200 price for a full new set of lead-acids. But that is more than made up for by the longevity, lighter weight, and improved performance of the LiFePO4s.

But the problem is that you really should have a BMS system. You can probably get away without one for a while (Just hook up the 24 cells in series and charge them up . . . maybe even with the existing charger.). But if you do that, you are taking a risk. If one cell shorts out, you may end up over-charging the others thus destroying the whole pack.

If someone made an inexpensive BMS, I think lead-acid would start going away.

[JRP3]

Actually I'm seeing pretty good evidence that a cell level BMS is not necessary if you keep well within the discharge/charge parameters. Jack Rickard has made a pretty good argument to split your pack in half and monitor each half. As long as they are equal you know they are in balance, and as long as you don't pull high C rates they should stay in balance.

[jorhyne]

Another trick I have heard of is occasionally (or at least initially) is to equalize all of your batteries by connecting them in parallel and charging them as one big 3.2 volt battery. Once they are fully charged rearrange them back into your series alignment and supposedly they will stay close to balanced for a good while.

[speculawyer]

Actually I'm seeing pretty good evidence that a cell level BMS is not necessary if you keep well within the discharge/charge parameters. Jack Rickard has made a pretty good argument to split your pack in half and monitor each half. As long as they are equal you know they are in balance, and as long as you don't pull high C rates they should stay in balance.

I figure that an NEV that ran on lead-acid batteries won't really stress a LiFePO4 cells. And if I periodically test the cells and manually balance them somehow, I would think it would work fine.

Originally Posted by jorhyne

Another trick I have heard of is occasionally (or at least initially) is to equalize all of your batteries by connecting them in parallel and charging them as one big 3.2 volt battery. Once they are fully charged rearrange them back into your series alignment and supposedly they will stay close to balanced for a good while.

Is there some kind of special charger for this?

[JRP3]

An adjustable 5 volt power supply should work, though it might take a loooong time to get it done unless it's a high current unit. I suppose you could charge the pack in series to maybe 3.5V per cell or so then finish off in parallel for the final charge to 3.6. You would need to monitor closely during the series charge because if one or more cells is at a significantly higher SOC they could start overcharging before the others are done.

[sunworksco]

A BMS system is so easy to install on your individual batteries.Why is everyone here getting their shorts in a bunch?