Can you make your own lithium batteries?

Van Living Forum

Help Support Van Living Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Below is some pictures of the small ones I made. The largest I made was a 220ah lifepo4 battery from 160x 32650 cells. Pound for pound the 18650 cells are the best for any project, they take up less space then the 26650 lifepo4. 

The 9 ah lifepo4(26650) pack takes up the same space as the 27 ah li-ion(18650) pack. If I had used better 18650 cells (higher mah) I could have reached 40 ah instead of 27 ah, 40 ah is the same as a 700 dollar goal zero 400. The lifepo4 pack I'm building as a jump pack to jumpstart my van, those cells are rated 50 amp discharge, laptop cells max out at 2 amp discharge.

On ebay you can find under "lot of 18650" I only buy the ones that have been tested and have the mah on the side of the cells, I have good luck with those. Also people sell on craiglist laptop batteries, but you have to take apart and test. To test the cells you need an opus btc3100, it can test 4 cells at a time. For high discharge you need powertool 18650 or high drain 26650 lifepo4. They sell the spacers (to assemble the packs) on ebay. You can find BMS also on ebay as low as 4 dollars for 30 amps ones. 

For laptop cells I would stick with 3s, its all I ever use, I never had inverters shutoff on me from low voltage. They run everything I ran on a 12 volt lead acid battery.

For building lithium packs, I highly recommend a tab welder. I use the malectrics model (cost about 100 dollars) that runs off a 12 volt lead acid battery, I can build inside my van. Some people use solder irons to put together. Tab welder is alot quicker especially if you have hundreds of cells.

You can build a small 3s lithium pack to learn the basics on putting together and wiring up a bms, then move on to much larger packs. There is website "repackr", that will tell you where to place all the cells in your pack so it can stay in balance.

You can also build using the larger square prismatic lifepo4. You only need 4 of those large cells to get 100ah, but they are more expensive.


lithium.jpg
 

Attachments

  • lithium.jpg
    lithium.jpg
    97.2 KB
justawfulgamer said:
If you decide to do this, lmk. I might do it with you and we can compare notes. I live for this kind of stuff.

Sent from my Redmi Note 6 Pro using Tapatalk

Will do. I have 3 years of school and commercial electrical experience. This isn't out of my wheelhouse.
 
This is very interesting. It is also remarkable that I independently of this thread had an idea to make an 18650 14.8V bank just about 5 days ago. I have an old Chinese solar USB battery bank from Ebay that was DOA. The seller offered rather than replace it to refund like 50% of the purchase price. I accepted and had a roughly $3 phone charger that ran my bluetooth speaker almost exclusively off of the tiny solar panel and regular occasional charging for a year.

Turns out it was an issue with the switch. But now it's dead. And has 5 18650 batteries in parallel inside it. I looked on amazon and they have 18650 battery boxes just like you used to get at radio shack to hold AAs for small projects.

I'm thinking .... (and to be clear I have watched a total of zero videos) a simple project would be to buy the 4 X 18650 battery box, a lithium/FLA/AGM/etc compatible solar charge controller from amazon ($26 I think it was) and a smallish panel. I have no knowledge of BMS systems, their application, usage, or function so I'll need to bone up on that. That being said. I have some "garbage" sitting around that could very well be made into a power source for a 12v light by my grill at night.

I guess it's true what they say. There really is nothing new under the sun.... PUN INTENDED! :p

A sun pun? I'm a poet too. Ok I'll stop now. XD
 
jonyjoe303 said:
Below is some pictures of the small ones I made. 

good looking work. curious what your total price for that one on the left was, the 65 ah 18650 pack? including the tab material the plactic holders/spacers the bms (was this the 30 amp BMS you mentioned)  cells (are these the tested used cells you mention) what did it cost to build or buy a protective case to put it in to be able to use it as a normal battery with out worry of shorting out or damage (vs just a bench battery for messing with)

have you tested and confirmed the final usable capacity, did it match the specs of the cells you paid for? is this battery pack a 3s or 4s configuration (i saw you mention 3s but not in relation to this pack)
what cutoff voltage are you using when charging and what low volt cut off are you using when discharging, how many usable amp hours did that give you? (tested)


have you used this regularly for a while to see if the capacity is holding? (i ask since the used cells are already degraded in capacity and wonder how quickly they will lose more capacity)

thanks
 
I would also be curious to hear that answer. The charge controller I found was adjustable up to about 16v I believe. I had planned if I went thru with the project to set it to 14.8V but I needed more research to determine an appropriate low voltage cutoff. If I had found nothing I was going to model it after the low volt cutoff recommendations for lead acid batteries. I know lithium can withstand more depth of discharge than lead acid so that seemed a safe benchmark to me. But I'd be obliged to get the data from someone who has real world practical knowledge rather than my own adaptation based on the science for a different battery chemistry.
 
your upper and lower voltages will depend on if you have a 3s or 4s pack. 14.8 v would lead me to suspect a 4s pack. you want to find the per cell charge and discharge voltage recommendations for the cells you are using and then multiply by the number of cells you have in series.

just like a nominal 12 volt battery that gets a higher charge voltage so do lithium ion batteries.

very general standards for lithium ion as a chemistry (not lithium iron phosphate "lifepo4") is

charging
to 4.2 volts. stopping a tenth of a volt below the spec for your cells will increase life at the expense of capacity. you can find charts on line for specific cells showing volts to capacity on the high (charging) side and the low (discharge) side.

discharging
down to 3.0 - 2.8 volts for end of discharge. again, not going all that low will increase life at expense of capacity

for example, a 14.8 volt nominal lithium ion pack is going to be made of 4 3.7v nominal cells in series. referred to as 4s. these would be charged to 16.8v (4.2v per cell) or slightly less to increase life and then could be discharged to 11.2 volts (2.8v per cell) again, stop a little sooner to get longer life

so now we have a battery with a useful voltage range starting with a full battery of 16.8v and we can safely use it down to 11.2 volts going below that can really bugger things up.

if you want to run an inverter off this set up. check the specs. you will find the majority of 12v nominal inverters will turn off if the input volts gets above 15-16 volts (this can even be a problem just equalizing lead acid batteries at around 15.5v) you could still use an inverter like this, just find out what their upper voltage limit is and dont charge your batteries above that voltage. but if you only charge to say 15.4 when full charge is 16.8 you can see you will not be able to use much of the capacity of the battery. also you have to be sure your BMS is rated to handle the current your inverter can draw

if you plan to rely on the inverter shutting down as your low volt protection in this case. most inverters have the low volt shut off around 10.5 volts. your lithium ion 4s bank wont like that. so you might want to use a independant low volt cut off at what you feel is a good balance between long life and high capacity.

if you run the math on using just a 3s bank, you get similar problems on losing access to battery capacity below the 10.5 common cut off.

you can use buck/boost converters to compensate, but that adds complexity and a loss of efficiency due to the voltage conversion

also, when building a larger than 1 cell capacity bank. are you going to wire the cells in series then those in parallel. or put cells in parallel groups then wire those groups in series? and why?

lithium ion battery projects can be fun and entertaining and educational. but to build a simple cheap drop in replacement for the old rv battery. not so much

lifepo on the other hand, while still complicated and delicate. at least a 4s configuration gets you a battery with charge/discharge specs that aligns much better with standard 12v nominal lead acid
 
Thanks for that info. My first project would be with the existing 18650 batteries I have on hand. I junked them from a dead solar USB battery bank. I assume they used a booster in the Circuitry as they were wired in parallel and USB as we know is 5v. The phone charger was a cheap generic Chinese clone and as such the batteries in it are cheap generic knock offs. So technical specs would be useless even if I found them as I'm sure it was mfg with the cheapest available components. I wont be running an inverter on it as they're only 2600mAh batteries but a 3 watt 12v light bulb I think would do well on it. I suppose I would need to check the low volt cutoff on the bulb. I'd assume 10.5. As a simple project I was planning on just using the load and rigging a switch for the light so the controller would handle the cut out voltage.

Unless combining multiple banks is a BMS necessary? They would balance themselves in a series yes?

Anyway doing some figuring I should have using the 50% depth of discharge benchmark (i have read lithium can go to 80%) 2600mAh would be 1300mAh useable at 14.8V a 3 watt load would give 6.41 hours of light. Call it 6 hours flat to allow for losses. I'd say that's decent amount of usage for something I'd otherwise throw away. And it gives me a little experience with lithium. Then I can tackle other projects.

I know you said to charge to 16.8v but I'm a little uneasy going that high with used Chinese batteries. I'd like the extra capacity but I havent burned the woods down yet (although I came close burning a brush pile as a kid) and I dont wanna start now.
 
ya, i started out tearing apart a laptop battery i had that wasnt holding up almost 20 yrs ago. there is so much more info out there now i wish had been as easily accessed back then. just a note on that one. i found that the battery bms had failed and allowed the cells to get out of balance and that let a couple cell drop low enough that the cell level protection shut down the individual cells. now i am not recommending anyone try this at home. but i by-passed that cell level protection and got a little charge in those couple cells and charged and somewhat balanced the cells. then i shut down the laptop at like 25% charge after that so they wouldnt get to low. i got another year or so out of it

i think lithium ion are probably the worst for getting out of balance.
but hey, like you said getting some use out of scrap. who cares if it fails early. but for a 3 watt load on a small 4s1p bank you can get balancers and other protection circuits cheap on ebay and even amazon

i never said or encouraged 16.8 as a charge voltage. i just related that that was the general standard spec

in your experimental have fun and learn case. i would probably go to 16 volts. to start with. and i would stop discharging at around 11.5 - 12 volts. then play from there. increase or decrease a volt setting by a tenth per cell and see how that affects how long the light runs. play with it abit. just dont sleep with it under your pillow

i would also get some sort of battery monitor that records how many watt hours you are actually getting from the battery, even measure how many watt hours it then takes to recharge. good lithium chemistries are much more efficient when charging than lead acid. that is one of the biggest benefits i see in their use
 
Ok. So browsing thru amazon I find many 4S charger DIY kits. Some with BMS built into the PCB board. And they are very cheap. Much cheaper than a solar charge controller that is compatible to charge lithium.

All of them the charge as well as discharge are the same 2 solder points on the PCB board. So in my experimental scenario to charge a bank of 4s1P for the sole purpose of running a light at night could I bypass using a solar charge controller and just use the amazon charger/BMS device? The panels wouldnt be lit at night so I wouldnt over voltage the 12v light bulb when using it unless I forgot to switch it off before daylight.

That would make my project a lot cheaper and a smaller more compact unit to fit in a waterproof enclosure for permanent outdoor placement. Also in the eventuality of leaving the light switched on during the day the bulbs I looked at claim to be 12-24V so worst case scenario I'd be charging less efficiently in the day and have no light until I recover the charge on the battery.

The one I like best is $10 and has an over discharge protection of 2.5v per cell so 10v. And an over discharge of 4.25v so 17v.

Assuming the light has a low volt cutout of 10.5 or less, the panel does not exceed the volt or current rating of the charger, and the bulb can tolerate voltages from 12-24V then this is doable and can be put inside a small waterproof enclosure?
 
ya, i have seen those, but have not played with them. in general i agree with how youre reading the specs but cant comment on how it would turn out. if you do it sans solar controller, let me know how it goes. i've considered try similar

do you have a link to the one youre thinking of. id like to check it out?
 
Anmbest Enhanced 4S 16.8V 40A 18650 Charger PCB BMS Protection Board Li-ion Lithium Battery Charger Lipo Cell Module for Drill Motor with Balancer https://www.amazon.com/dp/B07RPW8MM6


According to the item description it works for 18650 (and other types) 4S and 4S2P configurations. Maximum input voltage is 18.1v so for a really heavy application it would not be ideal sans charge controller methinks if you're using a very high wattage panel or array of panels as I've seen a great deal of high wattage panels over 19 volts. But for my purposes the 5w and 10w panels I've found seem to be in the neighborhood of 17v so that would work. I'd imagine if you built a very very large bank of 4S batteries as previously pictured you'd need multiple units for every 2 banks or a different unit unit. But I dont know lithium. SLAs are what I have used for the most part.
 
the above linked "battery" is just 1 cell, it would take 4 to get to a 12v nominal battery. plus bus bars and BMS.

as far as price and quality, who knows. do you trust them. sell as used (they are really "new" and we are sure to tell you that, but we call them used to cover our butt) with no claims to performance could leave you hanging
 
Yeah I know I'd need 4. I just don't know what exactly to be looking for. Luckily with eBay you have buyer protection, and it's actually really good.... Almost too good

Sent from my Redmi Note 6 Pro using Tapatalk
 
yes ebay buyer prtection can be real good. if they said it had full capacity as new. they could certainly be held to that. but they just say came out of function units sold as used. so if it powers up and runs a bit but not a full 40 amp hours worth. i dont know if buyer protection goes that far

there are some known good quality lifepo4 cells out there. names like caleb, sinopoly, winston come to mind. i always get suspicious when they dont mention the name/brand. there is just as much or more chineese knock offs that are horible. and some middle of the road

but yes 4 of those would make a 12v nominal battery. highly recomended to have some sort of cell balancing capability, high volt cut off, low volt cut off, low temp cut off and maybe even over current protection. and most critically, read enough to know why you need each of those and how they interact with each other. it could be a great learning project

but i have no idea if those particular cells are a good deal or quality.
 
A Goal Zero YETI 1000 open box is $714.97.   It has 1045 watt-hr of energy, a 1500 watt pure sine inverter, a small pwm solar charger and is 40 pounds.   

Just as a comparison when you are thinking of buying individual cells and such.

I couldn't make the math work on building my own but as a fun science project to tinker around with it is ok.

Sort of like building your own solar panel...doesn't make much sense nowadays.
 
XERTYX said:
Anmbest Enhanced 4S 16.8V 40A 18650 Charger PCB BMS Protection Board Li-ion Lithium Battery Charger Lipo Cell Module for Drill Motor with Balancer https://www.amazon.com/dp/B07RPW8MM6
looks possible. i would probably want something configurable on the charge voltage. so i could set it lower like 16.4v or even 16. also, i was not convinced from what i read there that it would regulate and step down the input voltage toanything in particular. it gives a "charge volt" range from 16.8 to 18.1  most buck converters (step down) need to see around 1.5v higher than the output. ii dont see min and max input voltages i dont see an "output" voltage. i am wondering if this is just going to shut down any cell that gets over 4.2? if so, it might protect the batteries but just doesnt seam like it is meant to take an unregulated voltage source.
hard to say, the documentation on these things can be a little vague. might work, might not even do anything with direct pv connection, might let the smoke out. take video and report back... lol
i feel like i would put a small  adjustable buck boost converter between the panel and that protection board. make sure it is rated a bit higher than your panels max out put. then i could feed the protection board the charge voltage i wanted, say 16.4 and the batteries would charge, but each cell would be individually protected if they were out of balance. like if 1 cell was getting up to 4.2 while the others were still at 4.0 your total volts for the bank would be below the input chrge volts of 16.4 but if you kept going the one cell could over volt.
 
I thought about a buck converter between the PV and BMS but I didnt know if that would be wise. Or feasible i.e. too many amps as you mentioned. If I can find the 4 5w panels I used when I had my boat I'll definitely give it a go with just the board. If I have to buy a 5w panel I may not go thru with it. The 5w panels seem to be $3 a watt at best. I dont wanna waste that one just a project. But $10 for a board on scrap material with existing useless panels I'll definitely chance it.
 
Here's a link that has DIY kits, no spot welder needed.

https://vruzend.com/

I was looking into this for ebike motor kit. In the end it will prob be cheaper to just buy a 48v 20ah battery sourced from China for under $500 that comes with a 1 year warranty(US support).
 
IGBT said:
A Goal Zero YETI 1000 open box is $714.97.   It has 1045 watt-hr of energy, a 1500 watt pure sine inverter, a small pwm solar charger and is 40 pounds.   
I almost bought this on eBay earlier this summer.  Often these systems get little respect from the DIY crowd out of hand...
 
Top