On Oct 2, 2018, at 5:41 PM, 'Myles Twete' matwete@comcast.net [electricboats] <electricboats@yahoogroups.com> wrote:
Hey, who you callin' "odd"? J
Here's essentially what I use: https://enerdel.com/mp320-049-hc-bo-moxie-battery/ (12S * (2P) each)
Enerdel didn't consider it an odd voltage.
THINK didn't consider it an odd voltage when it used these to power its cars.
Two of these are in one metal case. I parallel these together thru a 15amp fuse: i.e. 2 * (12S * (2P) )
Then I have 10 of these cases, for a total of 20 * (12S * (2P)).
There are no exposed cells.
There are no joints to come loose---these same units are what powers 2011 THINK City cars---I have 51k miles on mine.
Each metal case has a pair of RLEC (remote lithium energy control) circuit cards that each manage one stack of cells.
In total, I have 20 RLECs that can be used to maintain every cell pair in the system.
Now, normally I do not power the RLECs as I believe the risk of always having them powered exceeds the risk of just occasionally using them, checking cells and balancing. Given probably less than 10 charge cycles per year, I think that is fair.
Anyway, sure, 39v could be considered nominal voltage if you cycle between 31 and 47v. I usually recharge before it gets below 38v and usually only charge it up to about 46v, so for me 42v is probably nominal.
My charger? It's effectively 2-stage, historically manual cutoff.
It's essentially just a 48v, current-limited power supply.
It runs in current limit until the CV is reached, then held there by the power supply.
If I have my balancer/control system powered up, I can set whatever target voltage I want (below 48v) and the charging will stop there. After, the cell balancers will do bypass control on each 12S2P stack until the cells are balanced.
At only 100ma of bypass current, this can take days.
If I did this with every charge cycle, it would not likely take that long.
In case this helps-
-MT
From: electricboats@yahoogroups.com [mailto:electricboats@yahoogroups.com]
Sent: Tuesday, October 2, 2018 1:45 PM
To: electricboats@yahoogroups.com
Subject: Re: [Electric Boats] Battery replacement
If you used LiFePo cells then 12s would be a nominal voltage of 39 volts, that's A pretty odd voltage, do you use a 48 volt charger or a 24 volt charger, this pack kind of lays in the middle. 12s1p would be 24 total cells 12s2p would be 36 total cells. What is the "20P" you list below?
On Oct 1, 2018, at 9:09 PM, 'Myles Twete' matwete@comcast.net [electricboats] <electricboats@yahoogroups..com> wrote:
Mine: 20P*(12S*(2P))
Delivering 700ah @ 31-48v since 2013
From: electricboats@yahoogroups.com [mailto:electricboats@yahoogroups.com]
Sent: Monday, October 1, 2018 6:19 PM
To: electricboats@yahoogroups.com
Subject: Re: [Electric Boats] Battery replacement
I would disagree with the recommendation to have more cells in parallel, I agree with the other James, that you should try and limit the the total number of connection. If you can not do 16S because of the battery box size, then use the next lowest "P" you can. So if 400 amp cells are a no go space wise, do a 16s2p. The chance of having thermal/arcing issues with loose terminal connection is many times greater then having internal resistance issues with cells all bought at the same time.
Or think of it this way, a good BMS is designed to over come slight internal resistance issues via balancing, but nothing automated is going to help you with arcing issues from loose terminal connections. Your batteries will be on a rocking moving boat, the less total connection you have the less chance you have of having issues, and the less time you will spend checking for loose connections.
As noted before in this thread once a year I check my battery under load with a thermal imager, loose connect is what I am looking for when I do that. That and a spot check of the voltages it the only maintenance I ever have to do on my batteries.
People that use a lot of dissimilar cylindrical (used laptop) batteries are the one that use lots of parallel strings to compensate for weak used cells, they generally use very accurate test equipment to sort the cells in to matched strings. This is not anything that really applies to building batteries with new prismatic cells. It is really a way to avoid throwing out marginal cells and getting the most bang from there used cells.
James Sizemore
On Oct 1, 2018, at 6:08 PM, clh5_98@yahoo.com [electricboats] <electricboats@yahoogroups.com> wrote:
James,
Thanks for the suggestion. A little more about my battery storage. I built my 2 battery boxes 6 years ago to fit 8 GC2 FLA batteries in each box.. With Lithium I can reduce to one box leaving the other box for storage, or bigger fuel tank for the generator. Without major cutting of the battery boxes, I'm limited to which cells I can use. 32 X CALB 180, or 200 won't fit without cutting the box.. 64 X CALB 100 wont fit either. I have found some taller cells that fit my space better. One is the FPT 220 Ah. I can easily fit a 16S2P pack using these cells. Another tall cell is the Frey New Energy 100 Ah. I can easily fit a 16S4P pack using these cells.
Does anyone have experience, or opinions about these cells?
I've read some that more cells in parallel can be beneficial in LiFePO4. Lower internal resistance is one with more in parallel, and that a weak cell can be "propped up" by the others and still be detected by the BMS as an issue. I don't know though.
Chris
Posted by: "james@deny.org" <james@deny.org>
| Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (95) |
No comments:
Post a Comment