48v 100ah LiFePo4 Pack Build with BMS

Power wheelchair board for REAL info!

POWERCHAIR MENU! www.wheelchairdriver.com/powerchair-stuff.htm

48v 100ah LiFePo4 Pack Build with BMS

Postby Williamclark77 » 21 Feb 2023, 23:25

This is not a tutorial or how to guide. It’s just what I’ve had success with and learned over the past few years building these packs.

I needed to replace the Headway pack in W2. I didn’t design it for battery replacement. It’s a pain. It needed to be replaced because I was overseas in Asia for a month. Someone came to my house to feed the dog while I was gone. They moved the chair and left it turned on. Not only turned on, but put some things on top of it, which were on the joystick, holding it slightly applying throttle against a wall for three weeks until the Roboteq shut off at 8 volts.

I recharged, balanced, and capacity tested. No bueno.

Sooooo…..

I think this is the third pack I’ve built using RJ Lithium cells. Not the cheapest but best I’ve tried. Grade A virgins. Buy once. Cry once. FYI – Shipping was $588 plus $55 Paypal fee. Took about two weeks.

Very well packaged
Image

All connected in parallel to charge/balance to 3.6v prior to assembly. That’s a piece of Romex 12/2 household wire I stripped. This took three days. That’s 1400ah worth!
Image

Image

Making the bus bars from .0625 x 1 inch oxygen free 101 copper. Overkill but what I got.

https://photos.smugmug.com/photos/i-ZKCQskf/0/5ed0d57c/1920/i-ZKCQskf-1920.mp4

How you have to pick them up when you’re a gimp that can’t reach the floor. I need a copper magnet.
Image

Short video machining them
https://photos.smugmug.com/photos/i-gJJQBkv/0/7bd8efce/1920/i-gJJQBkv-1920.mp4

All wire ends properly crimped then lightly soldered to reduce resistance. All bus bars were thoroughly sanded to remove oxidation, cleaned with acetone, and coated with heavy duty red Lucas grease as they were installed. Copper will oxidize in minutes, causing major issues later on. We use this same grease on our 5th wheel hitches at work. It comes factory on the big tractor truck battery bus bars. It’ll still be there airtight on the bus bars years from now.
Care must be taken though. This grease will destroy many plastics over time and is electrically conductive. It will cause shorts! It’s also messy.
Image

Image

Balance wiring. All equal length. There’s a sheet of .031 inch thick buna n rubber between each cell.

Image

Image

Image

These two bronze studs on the end will be the positive/negative connections.
Image

Almost ready for install
Image

Image

That’s the BMS on the front. More on it later.
Image

Image

Image

The thoroughly beaten Headway pack awaiting a viking funeral. These chairs are used hard. Actually, what to do with these old cells?
Image

The Headway pack was 60ah. It's being replaced with 100ah.

Image

Snug fit! Hard to see but the pack is surrounded by 1/8” thick buna n rubber sheet.
Image

Image

And a layer on top. An 1/8” plastic layer went on top of this. Only the battery connections are exposed. They’ll be covered.
Image

The BMS I used is a JK B2A24S. This is not an advertisement for them. It’s the only BMS I’ve even remotely considered since building my first pack in 2012 or so. I probably still wouldn’t use one for 8s or less. Charging 14s is a bit more challenging though.

It balances/charges up to a 24s pack, balances at 2 amps, and is good for up to 200 amp continuous discharge. It also measures ah used and so far seems to give a pretty accurate battery percentage indicator. I’ve gotten a display working with it. Haven’t quite figured out how or if I’ll use it. The phone app is capable.

You must set everything up correctly. All of the necessary settings are there.
Image

Image

You can turn the charging, discharging, and balancing off/on
Image

So far, it balances fast while charging and properly does no balancing below whatever volt you choose. I have it set to turn the balancing on above 3.45v.
Balancing at 2 amps
Image

Pulling out 2 amps. From testing, it appears to pull energy out at up to 2 amps to fill its capacitors, then dumps that energy into the lowest cell. So, on paper it’s moving two amps, but that works out to 1 amp over time I guess when you consider both directions. Still much better and much more efficient than dumping the extra energy into a resistor as heat to balance.
Image

I tried testing how quickly it responded by connecting the charger at 8 amps with the high voltage protection at 3.6v after the pack was fully charged/balanced t0 3.55v. It reacted withing ten seconds and cut the charge current down. No cells got over 3.62v. Very good.
Image

Within one minute
Image

Within two minutes
Image

Image

Within five minutes all cells were back down to the 3.55v target. Supply voltage isn’t super critical. Set it 2 or 3 volts over the max pack voltage. I was worried it couldn’t resist overcharging if the supply was over the max full pack voltage. It cuts it back fine. It cannot add voltage. The pack won’t charge if the volts aren’t slightly over.

Real time view of pack amps being pulled out. -20.2 amps. It refreshes about once every second. So, not really easy to see spikes.
Image

So far I’m pleased. Time will tell if it’s good. It’s nice being able to disable the entire electric system with one tap. That alone will prevent this pack from being over discharged.
It can also be set to disconnect at a certain voltage x time. Ie – if the pack is below 42v for 60 seconds it will disable.
WcMade.com - Get nearly anything you need made
Williamclark77
 
Posts: 1059
Joined: 21 Mar 2013, 01:18
Location: South Mississippi, United States

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby Burgerman » 22 Feb 2023, 04:10

Hope it works out. And the special smoke stays in. How does it end the charge. Meaning what tell the charger when to stop?
User avatar
Burgerman
Site Admin
 
Posts: 65282
Joined: 27 May 2008, 21:24
Location: United Kingdom

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby 11ydy » 22 Feb 2023, 05:02

Burgerman wrote:Hope it works out. And the special smoke stays in. How does it end the charge. Meaning what tell the charger when to stop?


The MOS of BMS will turn off the charging
11ydy
 
Posts: 21
Joined: 04 Oct 2020, 01:06

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby Burgerman » 22 Feb 2023, 10:44

Either the charger will do so at some random period. Or at some unknown point when the current drops low enough. That may be too long, or before its actually balanced the cells or completed the charge. Is the charger asimple power supply? That doesent switch off, If so then:

The correct time would be:
When all cells are balanced and the balance current has dropped to a couple of mA. And then after the voltage as been held steady at 3.550V for another 15 to 30 mins. And when the CV current, not including any balancing, drops to around 300th to 400th of theAh capacity. Is that correctly configurable? You should not just hold them high, or stop before completion. The question was how does it decide? If you hold it at charge voltage it will deteriorate the cells. They dont like that. They must be allowed to relax after charge and drop to their nominal full voltage of 3.4x volts per cell. A bit like lead batteries need to be allowed to drop to a float voltage. Although lithium needs no float. Just off.
User avatar
Burgerman
Site Admin
 
Posts: 65282
Joined: 27 May 2008, 21:24
Location: United Kingdom

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby AgGuy » 22 Feb 2023, 16:08

Very nice work and attention to detail! Great write-up about your BMS and results. Much appreciated.
AgGuy
 
Posts: 42
Joined: 27 Jun 2020, 16:38
Location: Illinois USA

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby Williamclark77 » 22 Feb 2023, 18:57

It's just a dumb adjustable power supply like https://www.amazon.com/dp/B0B18MM1YZ?re ... tails&th=1

The bms is full of mosfets. I ASSume it uses one to manage input. It's able to throttle the amps, not just on/off.

It turns off balancing after all cells are at 3.55v and within .005v (what I have it set to). I haven't left it on charge long enough to see what it does after the cells naturally drop to 3.4 or so. I ASSume it will turn back on once they all drop below the balance start voltage (I have it set to 3.45v).

Burgerman wrote:Hope it works out. And the special smoke stays in.


I don't trust it yet. It can be bypassed with one jumper wire to get home should it fail.

I'm more concerned with the Roboteq and finding an alternative. It has started acting sketchy and Roboteq no longer shows any HBL controllers available. Curtis Instruments has released a brushless controller intended for forklifts, floor sweepers, pallet trucks, etc that looks promising. I would know if I could get my hands on the programming software.

AgGuy wrote:Very nice work and attention to detail! Great write-up about your BMS and results. Much appreciated.


Thank you very much!
WcMade.com - Get nearly anything you need made
Williamclark77
 
Posts: 1059
Joined: 21 Mar 2013, 01:18
Location: South Mississippi, United States

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby woodygb » 23 Feb 2023, 00:40

Curtis programming software is available.
An expert is a person who has made all the mistakes that can be made in a very narrow field.
Niels Bohr
User avatar
woodygb
 
Posts: 7072
Joined: 12 Mar 2011, 18:45
Location: Bedford UK

Re: 48v 100ah LiFePo4 Pack Build with BMS

Postby woodygb » 23 Feb 2023, 00:54

woodygb wrote:Curtis programming software is available.


It would seem that the CURTIS INTEGRATED TOOLKIT (CIT) is required rather than the standard 1313 software.... So ignore my previous statement.
An expert is a person who has made all the mistakes that can be made in a very narrow field.
Niels Bohr
User avatar
woodygb
 
Posts: 7072
Joined: 12 Mar 2011, 18:45
Location: Bedford UK


Return to Everything Powerchair

Who is online

Users browsing this forum: swalker, UnicornBurner and 49 guests

 

  eXTReMe Tracker