Diy LiFoPo4 Battery & Charger Help

JuicyApple

1 µW
Joined
Sep 23, 2020
Messages
2
The Ebike i currently own:
https://www.voltbike.com/voltbike-nitro.html
Its a 750w 48v 17.5ah 80lbs ebike

My Situation:
I avg 40-50km range with it since i mainly use the throttle and rarely pedal. This would be fine if it didnt take 4 hours to charge 0-100%. I know i could probably charge it at 480w without stressing the battery that much however...

My Solution?:
I plan to build a 48v 30ah using 90 (3.2v 5000mah) cells in a 15s 6p configuration.This custom battery will have almost twice the capacity, twice the amount of charge cycles, a safer chemistry, 2-1 hour charge time (but i would need to have enough nickel between series connections to handle the amps), and it will be around 30-40Ibs.

Im fine having it be a big battery since i plan to upgrade my ebike to 1500w and uprade my battery to a 16s and replace the bms with a higher amp smart bms.

Cells: https://batteryhookup.com/products/100-brand-new-lifep04-32650-3-2v-6000mah-cells

My Problem:
I know that i want to charge at about 720w since that gives me a little over a 2 hour charge time and my bms has a charge current limit of 15a so it would be perfect. Sadly this is where my completely theoretical knowledge ends, I don't know all that much about making a homemade charger and I would very much appreciate any and all help/criticisms.
 
I don't recommend making your own charger. If you know how to design and build current limited constant current constant voltage power supplies, you can build a safe enough one, but otherwise you should buy a charger designed to do what you need it to do. It would be cheaper, anyway (during design and test you'll probably blow up enough stuff to have paid for at least one charger, and if you don't have test equipment then that will also cost at least as much as a good charger).

If you build one and it's not designed or built right, it could cause a fire in a number of ways, including by overcharging the battery (too high a voltage) or forcing too high a current into it. You'd want to test the possible failure modes of different components, until you find which ones fail which way, to design failsafe things into to prevent bad things from happening.

People *have* built bare-bones "bad boy" chargers, but these are not failsafe, either.

You can "build" a charger using a power supply of sufficient voltage and then a current regulator/limiter, there are at least a couple of threads around here for that.

Some of us use LED PSUs, which work *almost* the same way a charger does, in place of a charger. They have no safety shutoff features to prevent overcharging, however, so you'd need a BMS on the battery to ensure that doesnt' happen.

I use the Meanwell HLG-600H-54A for built-in charging on my SB Cruiser trike, a 11A for 57v, about 700w or so max power. This would work by itself for your purposes, albeit a little under your 720w goal.


I have a parallel pair of them that can do twice that charge rate when I really need quick charge at home, but that's very rare. They can also be wired in series to charge my lawnmower pack (which is two of my spare packs for the trike wired in series), if I need to do that in a hurry without disconnecting the series connections.

I also have a Cycle Satiator battery charger to charge various batteries, as it's programmable with multiple profiles...but it is only a 300w charger.


BTW; you should ask BH for the various "curve charts" (discharge, charge, etc) for those cells, which will show you how much voltage sag there is for various discharge currents over the full capacity range of the cell, etc. See https://lygte-info.dk/ for a bunch of such info on various cells for examples of such testing and charts. It's very helpful to determine if a cell will *actually* do what you want it to (since the way most companies state their ratings is not realistic usage/results, just the max numbers they can put out without lying (well, usually, anyway)).
 
Firstly thank your reply it is greatly appreciated :)

In regards to my "homemade charger" I should have specified I do wish to go the route of "building" a charger utilizing an LED PSU, bms and maybe a boost converter? As I said I don't know much in the way of building a charger, but I would love to be directed to any resources on how to "build" a charger using the same or similar parts.

amberwolf said:
BTW; you should ask BH for the various "curve charts" (discharge, charge, etc) for those cells, which will show you how much voltage sag there is for various discharge currents over the full capacity range of the cell, etc. See https://lygte-info.dk/ for a bunch of such info on various cells for examples of such testing and charts. It's very helpful to determine if a cell will *actually* do what you want it to (since the way most companies state their ratings is not realistic usage/results, just the max numbers they can put out without lying (well, usually, anyway)).

Thank you very much I will get right to that. If I read/understand this correctly my ebike controller has a peak draw of 20a with my 6p battery that means each cell has to provide just ~3.3a per cell at full throttle which is substantially less than its stated continuous discharge. When attempting upgrading it to 1500w however, that will most definitely be a concern. Hopefully upgrading my battery to 16s will alleviate some of the potential voltage sag?

Lastly I know I probably just talkin outta my a*s since and I don't really know all that much, so thank you for your understanding.
 
JuicyApple said:
In regards to my "homemade charger" I should have specified I do wish to go the route of "building" a charger utilizing an LED PSU, bms and maybe a boost converter? As I said I don't know much in the way of building a charger, but I would love to be directed to any resources on how to "build" a charger using the same or similar parts.
If you're using an adjustable-voltage/current LED PSU, you don't need any boost/etc converters; the LED PSU does everything you need it to for the charger part. (setting the voltage and limiting the current).

If you use a boost (or buck) converter, then you don't need to use an LED PSU; any PSU that can supply what the converter requires will work. Then the converter itself must be adjustable voltage and current, and must do current limiting in a CC (constant current) manner, rather than the "crowbar" type.

As noted, depending on the final charge voltage you need, the Meanwell HLG-600H-54A should do what you want, just shy of the total power you want to charge at. If you need more power you could get two of them and put them in parallel, then turn the current limit on both down until it's just where you want it. (you don't have to get them both as 600H versions, one of those and then a lower-power version of the HLG series (but still with the A on the end for Adjustable, and 54 for the same voltage range)

If you need a higher charge voltage, you can put more in series and adjsut the voltage down. IN your case you wouldn't use the 54 version then, you'd want to divide the final pack voltage in half and then see which HLG version would provide the adjustment range that lets you divide the pack charging between them. (or use dissimilar ones, one higher voltage and one lower, as long as the adjustment range on each allows the total you need).

If you want to carry the charger with you, or build it into the scooter, I would definitely recommend the Meanwell HLG series (potted and sealed and simple).


The BMS would be a separate thing, that stays in the battery and monitors the cells themselves.




Thank you very much I will get right to that. If I read/understand this correctly my ebike controller has a peak draw of 20a with my 6p battery that means each cell has to provide just ~3.3a per cell at full throttle which is substantially less than its stated continuous discharge. When attempting upgrading it to 1500w however, that will most definitely be a concern. Hopefully upgrading my battery to 16s will alleviate some of the potential voltage sag?
More cells in series doesn't change the voltage sag under load; it just increases the total voltage.

More cells in parallel means less load on each cell, and so less sag for the same total load all other conditions being the same.

The less load placed on each cell, the less sag, and so the more total power you get out of the pack with less heat generated inside it (wasted power).

Some cells are better than others, the charts for them will show you how they perform under various loads under specific conditions, making it possible to predict how a pack made from them will behave under your conditions and compare them to see which will do the job best. :)



Batteries are complicated; I'm not an expert and I've been learning about them for years. :lol:
 
Back
Top