DIY LiFePO4 battery assembly from cells: Battery Management system (BMS) question.....

I'm curious about the possibility of assembling a 48V battery bank from purchased 3.2V lithiun iron phosphate cells and associated components.  Even though my goal is to use this bank in a golf cart farmyard vehicle as well as a project for learning more about the technology, I'm hoping those here who have done or tried the do-it-yourself approach can offer some insights.  It is now routine to purchase 36V or 48V LiFePO4 batteries as single units to replace the lead acid batteries traditionally used in these carts.  The typical size batteries for these carts were designated "GC2" and either the 6V or 8V lead acid batteries used in series to get the required 36V or 48V bank had a footprint of ~7.5 inches wide by 10.5 inches long.  For this reason, the new larger 36V or 48V stand-alone batteries require signifianct modification to to the battery bay in the cart in order to fit the larger battery case.  I'd rather not have to do this modification(s), which leads to my question.

A 48V single LiFePO4 battery typically will house 16 of the 3.2V cells wired in series and managed by a 48V BMS of the necessary amperage size.  Is there something about the close-packed wiring of the cells in such a battery that is required for proper operation of the BMS?  More to the point, would I be able to place 4 cells in each of 4 GC2-sized cases and wire ALL OF THEM in one mass series array, the positive and negative of which would feed into the BMS?  This would allow me to fit the battery array into the exisitng GC2-size trays already in the cart. [Yes, there are already 48V GC2-size LiFePO4 batteries out there for golf carts, but at rather low amp-hour ratings....and rather expensive.)  As long as the BMS is 16S and of an amperage to tolerate the surges necessitated by the motor, it seems that this should work.....???  Or what might I be missing with respect to cable length/size, etc. and let me know if something here needs greater explanation.  Thanks!

If anything, the close-packed arrangement of the cells is a detriment to performance and longevity, because the big enemy of LiFePO is heat. Cramming cells into a compact package reduces their ability to dissipate heat (and also the opposite, which can be an issue where you are, because they cannot be charged if they drop below freezing). I think this gives you an advantage if you're thinking about a DIY battery, because you could get creative with the layout for better thermal performance at both ends of the scale.

Hi John, I love this question.

So I am going to start off with a couple of assumptions that I didn’t see in the opening specs:  your golf cart originally came with a lead-acid battery, correct?  This fact right there is the probable reason that squeezing a Lifepo4 battery in the same place is so difficult and gives such poor performance.  All of this comes down to the battery’s power density and C rating.  C rating is a multiplier used to determine how fast a battery can discharge.  For simplicity’s sake, I will start off by using the C rating of a LiFePo4 battery which is 1.  The way this is used is that the C rating is multiplied by the battery capacity as measured in Amp Hours.  For instance, a LiFePo4 battery with a AH rating of 10 (10 amp hours) with a C rating of 1 can discharge a maximum of 10 amps for one hour.

Now for something like a power backup with lots of USB plugs, this is plenty, excessive.  But for a golf cart which has a far higher power need, a higher C rating is needed.  This is where lead-acid shines.  Lead acid batteries have C ratings that range from several C up to hundreds C (think car starter batteries).  Lead-acid doesn’t really store all that much energy, but it can give it up very, very quickly in a pinch.

In a golf cart, the lead acid batteries are kinda middling as far as storage goes, but they can give plenty of power.  To get the same power from a LiFePo4 battery, you will need a much larger battery, but it will run a very, very long distance.  This means that the LiFePo4 batteries simply can’t fit in the same physical space as the older lead-acid batteries.

Now you are right in that you can certainly create 24 or 48 volt batteries and they will be AWESOME batteries but they will take up some space.

John, I hope this helps and if you need more elaboration, just ask and I will see what I can do.


Eric

I've been there and done this so let me offer some advice. First yes you can split the cells into groups. The key is keeping all the balancing wires the same length which means adding to them to reach all the cells. This is more important than you can imagine!

As far as size and ratings that depends on what you are after. But I need to put this in here prismatic cells have to be compressed or they will expand and be destroyed. Also LifePO4 can NOT be charged below freezing or bye bye battery! They can't be used / discharged much below freezing either and do not like temperatures above 100f at all. Now having said all that 100-105AH cells make pretty good golf cart batteries! The reduction in weight is huge and quite noticeable! You will need an oversize BMS to cover starting amps but they are short bursts. Mostly unless you are hot rodding like me then you might start pushing 105s farther than wise.

If  you do this DO NOT FLIP THE CART OVER the FIRST TIME YOU TAKE OFF!!! Mine stood straight up! Granted it's lifted with oversize tires...

John,

I agree with everything that Larry just posted.  The only caveat I would throw in is that while it certainly is possible to have a LiFePo4 battery powered cart, it might be challenging to get that battery to fit in the same form factor.  I am not saying it can’t be done, but getting the right sized prismatic cells could be a real challenge.

Furthermore, everything he said about charging above around 100f or below freezing is dead on correct.  Especially the freezing.  Don’t do it.  LiFePo4 batteries don’t like it.  The upper limit is a bit fuzzy, but that lower limit is a pretty hard limit.  You can partially get around this by having self-heating batteries, but those require a constant, low-level discharge.

And the compression part is also a real thing, especially if that 1C limit is being pushed.

This might be a bit more of a project than you were expecting, but it is totally doable, and in my opinion is well worth the effort.

Good luck and keep the questions coming as you run into them!!



Eric

Thank you all for excellent and helpful responses!  I'm additionally on some other forums for golf carts and electric ATV/UTV users, but most are just purchasing LiFePO4s outright and are modifying their vehicles to accomodate the odd battery case sizings.

With regard to cold temperature use, indeed even with many pushing the conversion of lead acid to LiFePO4 for my situation, I realize the damages and issues that come with cold weather charging....and this has been the main reason I've held off moving into the swap.  Truthfully, I may do the swap with my golf cart sooner because LiFePO4 batteries, if done as linked smaller cases, are so much lighter weight than lead acids that I would be storing the cart for the winter and bringing the batteries inside the house for those winter months.  The other electric vehicle, a 4X4 Polaris Ranger EV, is my wife's main farmyard vehicle that she needs ~365 days a year, and some of these days will be in temps around -20 degrees F. In rhis case, it does appear that there are LiFePO4 with heating circuits....when they are plugged in for charging, no charging will commence until the batteries are heated to the proper temperature to do so.  As you might imagine, this adds additional cost to the product and additionally another level of technology that one hopes won't fail. So for that reason, that vehicle will get lead acids one last time this year with the anticipation that rapid advances in battery tech will solve the temperature-dependant charging issues down the road for LiFePO4 or other chemistry.

As Phil S noted, spacing the battery cells properly could aid in battery longevity by keepping them cooler, but as larry k. indicated prismatic cells, which I was leaning toward, appear to require some sort of grouped binding to reduce cell expansion from heat.  Thus, in some cases, the use of metal vs ABS plastic cases and supporting structure within the case.  I have seen videos with a fix for this as simple as a specialized tape that wraps around the cell group (???) but unless there is a specialized divider between each cell, would not this result in heat concentration within the cells? Is some sort of between-cell heat sink involved?  Also regarding C-rating, I've seen mention of "EV-quality cells", which I assume on some level is addressing this high-amp discharge quality that must be met for the amperage outflow to the motor when needed, is this possibly correct? Sizing of the BMS for this purpose is something I'm just learning about now along with adding bluetooth monitoring of the cell array for balancing, temperature, etc.  For charging, my ultimate goal would be to use a LiFePO4-approved charger as well as a solar panel on top of the cart which has been a successful highlight of dabbling in this solar tech to this point:  Even with near end-of-life lead acid batteries, the ability of the sun to recharge these between short short trips in the farmyard has been exceptional!  Worth noting as well that in this scenario for our use, depth of discharge is minimal since we are not using them for long excursions or deep power draw on the batteries, even as occasional surges will be needed (powering out of mudholes, use of 4X4 in mud/snow) and the effect of cold on state of charge must be considered.

Finally, yes larry .....I've seen those videos where addition of larger-power LiFePO4s to golf carts render them so much lighter that the fron wheels can rise off the ground during fast acceleration.  I'm not looking for that kind of acceleration in my cart....although I can't speak for the dogs when I'm not around to monitor them!  :-)

All of your comments have been cause for inspiration and I hope to update as I make my decision.  Although I may somewhat grudgingly opt for the more expensive GC2-sized 48V/27Ah offerings and parallel 2 or 3 of these to meeting my needs (they are sold as golf cart batteries and meet the C requirements for that task), I'm quite into the idea of DIY for this and future projects.  Other thoughts from this entry most welcomed.....and thanks again!

The big road block with the lithium cells you are looking  use is lithium cells of this chemistry cannot be charged at cold temps.          Most BMS have temp protection in them for when it is cold they will shut off charge to the batteries to prevent damage.

To get around this some battery packs come with built in heaters that  will heat the batteries to a level that charging can start.

There are chemistries of Lithium batteries out there that can deal with the cold, but normally they are more expensive as I recall.

I would recommend  youtube channel  "Will Prowse"    as he tests batteries,  BMS  to see if they deliver what the label says they do,  many times they do not.

I also recommend looking at a site called "battery hookup"    they sell used batteries  at good prices.      David Poz on youtube has done conversions of golf carts with different types of  batteries.

Those are some places I would recommend checking out.

All issues have solutions. Far as the cold goes you have many options. I used 12V silicone heat pads wired in series and hooked to a thermostat. Far as overheating goes this never really became a problem for me and isn't likely unless you are doing long distance high speed runs in high temps. Also simply keeping the cart or UTV in a heated building in winter then using it for an hour or so in below freezing is NOT going to cause the cells to freeze! Also simply using the cells makes some small amounts of heat.

My GC also has a 300 watt solar roof! The added weight has been an ongoing issue and I will likely switch to a light weight flexible 200w panel that weighs 45lbs. less!

John, I have a question and a couple of thoughts:

1)  Are those solar panels on top of the golf cart?  If so, that's cool!

2)  Maybe, MAYBE, consider Lithium Ion batteries

Li-Ion batteries are very light, have spectacular energy density--far better than LiFePo4, and have an absolutely staggering C rating.  I don't remember it off the top of my head, but it rivals lead-acid and might even supersede.  It also has a very high charging efficiency.

The downside is that they are slightly prone to fires.  Now I have plenty of Li-Ion tool batteries and have no fear of them, but they are tiny by comparison.  But being so light and so powerful, they might be the right option given that you are driving a vehicle which is sensitive to weight.

This is just a consideration, so if you consider it, just be cautious.



Eric

Eric Hanson wrote:John, I have a question and a couple of thoughts:

1)  Are those solar panels on top of the golf cart?  If so, that's cool!

2)  Maybe, MAYBE, consider Lithium Ion batteries

Li-Ion batteries are very light, have spectacular energy density--far better than LiFePo4, and have an absolutely staggering C rating.  I don't remember it off the top of my head, but it rivals lead-acid and might even supersede.  It also has a very high charging efficiency.

The downside is that they are slightly prone to fires.  Now I have plenty of Li-Ion tool batteries and have no fear of them, but they are tiny by comparison.  But being so light and so powerful, they might be the right option given that you are driving a vehicle which is sensitive to weight.

Thanks for the questions and thoughts, Eric.    Yes, that is a single 327W panel on top of the golf cart and I;ve added a photo of the spec sheet below. At 41 lb, it was relatively easy to replace the existing golf cart roof with this panel and guide the wires down under the seat to the charge controller and from there to the main battery (flooded lead acid) terminals for charging.  On good days the panel is pushing ~ 5 amps and around 260 - 300 W into the battery bank.  A quick side digression to note perhaps as before that this battery bank is severly worn out.  My experimental interest here has a lot to do with how the vehicle is used.  We use it solely as a 'go-fer' unit, shuttling feed buckets, tools, firewood, (dogs!... :-) ), and occasional light towing around the property.  For this reason, I have neglected a LOT of maintenance on the cart....wheel bearings, brakes, motor/controller efficiency, etc.....partly just to see how durable the mechanics of an off-the-shelf, used, older standard golf cart would be.  I've added a bit larger tires for clearance, but no 'lift kit', the latter of which raises the body, but not the axles.  Oddly the weight of the 6V batteries (~70 lb each!) while being a hassle to maintain, pays back somewhat in weight to the cart which can provide good traction when needed.  Also, without having played much with LiFePO4, I lament a bit the added BMS requirement over lead acids the latter of which need some babying, but no need for a BMS.  (It's unclear at this point in battery tech if Na-ion batteries will require a BMS for all applications although they certainly are being sold with BMS modification at this time.)  But my aging body is a good reason among others to go LiFePO4 when it comes to removing the batteries from the cart.  I don't anticipate heating the garage in the near future for winter storage and would rather just move the batteries indoors for that time.  But a main point I wanted to make with the digression is the combination of small trips, large gaps (hours) of time between use, and the presence of the solar panel on top means that I'm rarely discharging the bank to any great extent -AND- it's constantly being topped up by the sun.  That alone is helping to maintain battery health, even with end-of-life batteries.  I can only guess that with new LiFePO4's, this effect would be even greater given their high cycle lifespan.

Li-Ion instead of LiFePO4....  I had not really considered that even as I sometimes pondered the idea of having power tool batteries running my cart!  We are invested in Dewalt tools and the 20V batteries even as I know they sell 60V sizes for beefier needs.  I have wondered how many 60V/15Ah batteries would be needed to run my cart....with much less weight!  As a side note, hats off to EZ-GO and like golf carts that for years have built in amazing power flexibility into their motors....many report running the 36V motor with 48 and 72V battery feeds with no problems as long as the motor controller and solenoid can handle the surges.  But I also have to admit to shying away from anything too concerning with regard to fires.  That is why both LiFePO4 and Na-Ion technology are my two favorite considerations at this point after lead acid.

My wife's Polaris Ranger EV is a bit of a compromise vehicle as she wants minimum "messing around" with a unit that she needs to function for feeding her animals daily.  I had wanted to add a solar panel here as well to the vehicle, but she ends up driving through brush on the property and did not want to wreck the panel doing so.  So I've added the solar panel (380W) to the carport roof with wires feeding a wall-mounted charge controller and it plugs into the port on the side of the vehicle where I've added a plug (and implored her to NOT accidentally plug this into a wall outlet!.....there is a separate wall-outlet, onboard charger for 120V grid-based charging).  This vehicle currently operates on 8 X 12V lead acid batteries (HEAVY!!) that provide 2 banks at 48V each bank.  The motor is AC instead of DC which apparently are more efficient at power conversion and the electronics overall are more sophisticated in this vehicle than the golf cart. That said, it's been operating pretty well and the added 4X4 capability pretty nice.  So I'm chomping at the bit to add LiFePO4 to this vehicle, but I need to guarantee to wife that it will not come with winter hassles when she needs it to function in the sub-zero F weather.

Keep the questions coming and I hope to be adding my own as I continue with the research.....  Thanks!

John,

I wanted to reply to your posts earlier, but I was delayed by preparation for and attendance of a funeral.  At any rate,  here are a few of my thoughts:

1). To start off, somewhere I read that when a person first discovers/encounters/attempts a new project, hobby, endeavor, etc., they do so with wild, complex ideas about how something should turn out.  As the person then becomes more experienced in said endeavor, the person looks for simplicity.  This project is reminding me of my new hobby of making battery boxes.  I am glad that you appreciate simplicity where I still get fascinated by the complex.

2). Dead on correct that plain old flooded lead acid batteries, though needing a bit of maintenance, are pretty rock solid, cheap, and simple to utilize.  LiFePo4 is much more complex.  Technically better, LiFePo4 needs additional components precisely because of its flat discharge profile—voltage does not indicate start charge except in extreme and dangerous conditions.

3). Also as has been stated, given your usage patterns, those LiFePo4 batteries are unlikely to get below freezing as per your usage.  Heat *MIGHT* be more likely, but it is still manageable.

4). Ahh, the Sodium Ion battery!!  Although they are commercially available now, I don’t know if they should be treated like a LiFePo4 battery, a lead acid battery or their own unique battery.  And thus far I have seen no associated electronics.  I want to know more, but I also want someone else to go first.  I wish I could offer more.

5) Lithium Ion:  that’s the stuff!! Its technical specs are hands down the best!  And the fire danger is mostly a thigh of the past if obtained from a reliable vendor and utilized properly.  But I also get your concerns and understand your reservations.  As a bit of a compromise though, I have found that third-party batteries for brand-name tools are both perfectly reliable, powerful and far, far more affordable than the name-branded ones.  I bought a pair of 18v, 6 AH batteries two years ago, used them literally every single day till empty and had my first one fail this summer which is exactly on the predicted lifespan.  I simply bought another pair, but by now they were 7AH for the same price.  Perhaps an 80, 5+ AH battery (or two, just saying) could be worth something.  It’s just an idea, do with what you want.


Overall I find this a fun project to watch, and I will help where I can and learn when I can’t.  Please keep this updated!!




Eric

Eric Hanson wrote:

5) Lithium Ion:  that’s the stuff!! Its technical specs are hands down the best!  And the fire danger is mostly a thigh of the past if obtained from a reliable vendor and utilized properly.  But I also get your concerns and understand your reservations.  As a bit of a compromise though, I have found that third-party batteries for brand-name tools are both perfectly reliable, powerful and far, far more affordable than the name-branded ones.  I bought a pair of 18v, 6 AH batteries two years ago, used them literally every single day till empty and had my first one fail this summer which is exactly on the predicted lifespan.  I simply boiled another pair, but by now they were 7AH for the same price.  Perhaps an 80, 5+ AH battery (or two, just saying) could be worth something.  It’s just an idea, do with what you want.

I will try to post more as developments arise, but fall has arrived, we almost frosted last night, and now pre-winter chores are upon us.  Thanks for your interest, comments, and advice, Eric.  For sure the LiION angle is one that I had dismissed but I suspect that was done too early without propoer prior investigation.  Indeed, not sure why eBikes use LiION and not LiFePO4, but that implementation of LiION has sent me down a rabbit hole :-)    Using the chart below as an example, it appears that I might be able to use higher-power eBike LiION batteries to meet my needs, at least in a cost-effective experimental way..(??)  There appears to be pre-made 48V LiION packs for eBikes that can power up to a 1500W motor....exactly the rated power in my golf cart.  As these come in 30 Ah configurations and BMS upper limits of 50 A, then by itself it would be straining the system....but two or 3 of them in parallel should be able to serve my needs -AND- avoid the issues with series connection, something other cart owners have steadfastly advised against.  Hmmmm.... the plot thickens!  If I were to test a set of LiION batteries in my cart, would I use a LiFePO4 charging profile in my solar charge controller?  The batteries appear to come with a charger as you would expect for use in an eBike....but would I be able to charge 2-3 batteries wired in parallel with this charger or require a larger charger?....and would one of the many 48V LiFePO4 chargers out there sold for golf cart charging suffice for his purpose?  From what I can see from the specs, such a configuration of batteries would once again be even less weight than LiFePO4s for roughlly the same power capability, perhaps?  Much food for thought here.... Thanks!

Hi John;
I am still too early in my plans to have made decisions about the products I will ultimately choose, but this is the Battery system that was suggested by my installer.
https://discoverenergysys.com/products/lithium-batteries/helios-ess
Very large and not cheap, but could be the best buy for my money.
It will be truly incredible to be able to weld RMH parts from solar power!

John Weiland wrote:

Eric Hanson wrote:

5) Lithium Ion:  that’s the stuff!! Its technical specs are hands down the best!  And the fire danger is mostly a thigh of the past if obtained from a reliable vendor and utilized properly.  But I also get your concerns and understand your reservations.  As a bit of a compromise though, I have found that third-party batteries for brand-name tools are both perfectly reliable, powerful and far, far more affordable than the name-branded ones.  I bought a pair of 18v, 6 AH batteries two years ago, used them literally every single day till empty and had my first one fail this summer which is exactly on the predicted lifespan.  I simply boiled another pair, but by now they were 7AH for the same price.  Perhaps an 80, 5+ AH battery (or two, just saying) could be worth something.  It’s just an idea, do with what you want.

I will try to post more as developments arise, but fall has arrived, we almost frosted last night, and now pre-winter chores are upon us.  Thanks for your interest, comments, and advice, Eric.  For sure the LiION angle is one that I had dismissed but I suspect that was done too early without propoer prior investigation.  Indeed, not sure why eBikes use LiION and not LiFePO4, but that implementation of LiION has sent me down a rabbit hole :-)    Using the chart below as an example, it appears that I might be able to use higher-power eBike LiION batteries to meet my needs, at least in a cost-effective experimental way..(??)  There appears to be pre-made 48V LiION packs for eBikes that can power up to a 1500W motor....exactly the rated power in my golf cart.  As these come in 30 Ah configurations and BMS upper limits of 50 A, then by itself it would be straining the system....but two or 3 of them in parallel should be able to serve my needs -AND- avoid the issues with series connection, something other cart owners have steadfastly advised against.  Hmmmm.... the plot thickens!  If I were to test a set of LiION batteries in my cart, would I use a LiFePO4 charging profile in my solar charge controller?  The batteries appear to come with a charger as you would expect for use in an eBike....but would I be able to charge 2-3 batteries wired in parallel with this charger or require a larger charger?....and would one of the many 48V LiFePO4 chargers out there sold for golf cart charging suffice for his purpose?  From what I can see from the specs, such a configuration of batteries would once again be even less weight than LiFePO4s for roughlly the same power capability, perhaps?  Much food for thought here.... Thanks!

E-bikes use L-ion instead of LifePO4 mostly because of weight and secondly so they can sell replacement batteries. L-ion lasts about 1/3-1/8 of the life of LifePO4. They (L-ion) also cost more per AH if I remember correctly.