Solar charging LiFePO4 without a charge controller?

[Chalo]

So I bought about 2kWh of LiFePO4 packs that aren't seriesable for 48V, so they don't pertain to any of my immediate needs. However, I have an offbeat 200W solar panel that has a peak open circuit voltage near 33V and a peak power voltage of 26.6V. nominally my packs are 26V with a top charging voltage of 29.2V.

My thinking is if I trust the BMS to cut charging when voltage exceeds the maximum, I could hard wire the panel to the packs without a charge controller. Is this correct?

Batteries will be used for LED lighting and small appliances on a lot that lacks utility power.

 

[amberwolf]

when the bms cuts charging, the load on the panel drops to nothing. what's the max voltage the bms fets can take? it has to be higher than the panel oc voltage.

also....do you really trust the bms to always cut charging? if it doesn't...can the cells handle the overcharge should the panel voltage rise that far under the minimal load of the already-full pack and the bms's balancers?

if it is a potential problem, there are some monitoring units poeple around here have used that open a relay if a voltage exceeds a limit (some are an lvc type, some an hvc, and some are either/or). i haven't tried any of htem, but it might be a useful safety net.


i considered one for my eig pack charging since my charger is built into the trike and i don't run a bms, but i never did buy any to try out. i'd meant to get one to try with my old "12v" panels so i could use them to charge the trike without an mppt or bms, etc., but ended up not doing it becasue i'm never home in the daytme to charge it except the couple days a week i'm not at work...and i only charge it once on one of those two days (to make up for the work commute the day prior), just about 5ah or so; probably low enough for the panels to do it or most of it, but unless i would be able to use the setup all the time or most of it, it wouldn't be worth spending the money on.

so i haven't actually tried direct panel charging yet. (other than whatever test i did way back when i got the panels from the thrift store, back in the sb cruiser thread).

 

[JackFlorey]

So I bought about 2kWh of LiFePO4 packs that aren't seriesable for 48V, so they don't pertain to any of my immediate needs. However, I have an offbeat 200W solar panel that has a peak open circuit voltage near 33V and a peak power voltage of 26.6V. nominally my packs are 26V with a top charging voltage of 29.2V.

A few reasons that is a bad idea.

1) Open circuit voltage can be much higher when it's cold out - say 39 volts. FETs have to handle that.
2) Vmpp will be much lower when it's hot out, so you may not get to 26 volts.
3) Often BMS switches cycle in odd ways i.e. never turn back on until voltage is below XX or something. That could lead to odd behavior.

It's not hard to find very cheap PWM controllers that will solve at least two of those problems.

 

[DingusMcGee]

If you have a modern programmable solar controller, e.g. Morningstar TriStar, you can program charging parameters. I have been charging 2 packs of 8p 4s 40ahr lifepo4 in conjunction with 4 150 ah agh lead acid cells to 14.35 volts for about 8 yrs. There is no BMS to manage the Lithium packs. The set up is in my van.

The LiFePo4 cells have such a low discharge rate that mine were getting frequent cell failures on edirtbikes. I switched to LiPo cells and have not had that problem. Apparently when paired with the lead acid batts, the LiFePo4’s do not reach their killing over discharge rate, not even when the system powers an inverter producing 1200 watts to the microwave plus lights.

 

[fechter]

I wouldn't use a BMS as the only control for limiting charge. It should be there only as a safety backup if the primary system fails.
Look for a MPPT solar charge controller. They are not that expensive.

 

[JackFlorey]

I wouldn't use a BMS as the only control for limiting charge. It should be there only as a safety backup if the primary system fails.
Look for a MPPT solar charge controller. They are not that expensive.

Just a note on this -

MPPT controllers have higher dropout voltages than PWM controllers, since the MPPT controller has to have a DC/DC converter in series with everything. And most will not boost voltage. So in the case where your Vmpp is very close to your needed voltage, MPPT can actually deliver more power by reducing your dropout.

(If the voltage in is more than a few volts higher than the desired voltage, though, they are far superior to PWMs.)