Battery EQ and BMS discussion.

[australsolarier]

kurt,
thanks for all your pictures and explanations. would you mind telling us how you are manually "equalizing" the cells of your lifepo4 battery?
i have a similar setup like yours bought from the same seller, although the blue evpower box has different disconnect relays at the bottom. i prefer not going past 56volts for the equalizing to take effect on the little circuit boards.

thanks in advance
greetings urs

[offgridQLD]

To avoid having to spend long periods at 3.6v and above (could be as high as 3.8v) using the built in shunts on the boards. I just charge to 3.6v x the number of cells 57.6v for me (I set it up as a manual EQ in the classic) then I stick by the bank with a good acurate digital MM and keep tabs on the cells as the rise in voltage . If I see any cells that are lagging behind the others. For example 15 of them are at 3.58v and one is at 3.52v. I will boost charge the low cell with a 10A single cell charger. (just a little hobby charge designed for RC cars) I set it for one cell 3.6v and about 5A and within 10 - 15 min I can have all the cells within a few MV of each other. (within 15mv at 3.6v I call it a day) No need to be pedantic about a few MV.

I haven't had it yet ( usually just one or two low) but if I had a high cell that got to 3.6v well befor the others the shunt on the boards would start to bleed off the current by its self. Just using the single cell top up charger speeds up the presses rather than it taking hrs for a low cell to catch up the the others.

The chargers can be had for about $30 - $50 from places like hobby king online. (handy for other jobs to as they charge lead acid, NMH,Lipo,Lifepo4 fully programmable. They also discharge and balance up to 8 cells at a time.Handy tool to have in the shed. You power them from another 12v battery or a 12v power supply (just make sure the power supply  used is isolated if your charging the very same  battery that is powering the external charger) As in its charging its self

I have scheduled the balance to be done every 3 months on my calendar I'm sure you could stretch it out longer but nice to keep tabs on it 4 times a year.

Kurt

[australsolarier]

thanks kurt,
i appreciate your help and experience. i will give your instructions a try.

the 12v system about 1mV max difference per cell at 13.5V

however the 48v system is a different kettle of fish. there is about 10mV at 53.4V.  but much higher at the 58v level: 0,2V. probably the circuit boards might have influenced that voltage reading. there seems to be 2 kinds of cells, low ones and high ones and not much in between.
this was the first time i looked at the voltages of the 48v battery bank as it is fairly new and only slightly used yet. 600w panels hooked up. once that NSW 60 cents gross feed in tariff ends i will switch over to fully use the 48v battery bank and probably go off the grid altogether

using a separate power source to do the top up charging is an important point.

thanks again
greetings urs

[offgridQLD]

i appreciate your help and experience. i will give your instructions a try.

the 12v system about 1mV max difference per cell at 13.5V

however the 48v system is a different kettle of fish. there is about 10mV at 53.4V.  but much higher at the 58v level: 0,2V. probably the circuit boards might have influenced that voltage reading. there seems to be 2 kinds of cells, low ones and high ones and not much in between.

Just want to make sure your reading your MM correctly
one MV is 0.001v

you will never get things that tight (particularly at the top of charge) and there is no point in doing so even if you could.

Ignore the imbalance in the middle of the SOC as this will never show anything  they always read the same voltage out of the upper and lower knee.

0.2v is  200mv that's worth leveling out that's starting to  represent some capacity imbalance between cells. You will get a feel for what it represents in AH or mah as you top up or discharge cells to level it all out.

Only the discrepancy at the the 58v mark count. within 10 - 20 mv  you my as well call it quits until next time rather than chasing your tail as your  only balancing to try and keep tabs on it befor it gets out of hand....enough to reduce your usable capacity.

[ClassicCrazy]

I don't understand what kind of lithium balancing you are doing ?  It does not sound like bottom balancing .  Top balancing ?

Larry

[offgridQLD]

Ballancing at the top.

Ballancing at the bottom is hardly practical in a offgrid system that's powering live loads at the time.

Ballancing = ironing out any small SOC differences between individual cells. The unbalance only becomes a problem when ...if it becomes so great that one cell reaches a low SOC or high SOC well befor the others. Effectively reducing  the usable capacity scope of the bank.

Even ballancing every 3 months or so I only see very small discrepancies. In the order of 800mah or so on a 400 ah bank. So perhaps 3.2ah over a year.I still like to trim them all up regularly as it's a quick job and potentially catches out something odd that could have hapend.

Kurt

[russ_drinkwater]

Hi Kurt,
Sorry for off topic , lol.
Easter is all good. Mate we have spare beds in the the house, lol.
Not cold at that time so whatever floats your boat
I will have a fugitsu (split system) 3.5 kw inverter A/c from bundy by then.
Bring your bow as Doug will appreciate dragging you around after rabbits.
Give me a ring or I will ring you closer to time.
Russ
Just Having a gander at the the WA ev power site.

[ClassicCrazy]

Ballancing at the top.

Ballancing at the bottom is hardly practical in a offgrid system that's powering live loads at the time.

Kurt

I don't understand why bottom balancing would be any different when running live loads ?

Larry

[australsolarier]

because the battery is empty and you won't do much running of "live loads" then.

+ we try the batteries to be charged as much as possible.

[offgridQLD]

I don't want my cells at 2.5v volts what ever voltage your taking them down to when bottom balancing.  Particualy when a 5kw+ load comes on from the house when they are down there. For one the inverter would most likely drop out on LVD . Then I a have a bank thats most likely at 10% or less SOC and I have a house , workshop and electric car to run and a bank at 10% SOC ...great!

In a stationary storage situation...particularly one where you have a big PV array ample battery capacity and located in a area with lots of sun all year. The likelihood of a High voltage ...top of the SOc curve imbalance event is much more likely than a low SOC situation. As in. I have never seen my 400ah bank below 300AH remaining capacity . I never see the bottom area so why make the cells balanced at the bottom. A area they will never see.

In a electric car where you want to suck every last Ah you can from the bank to get the most usable range its not uncommon to be lurking  in the lower 10 - 20% of the SOC you might be concerned with how the cells are balanced at the bottom.....and really only if you don't have a LVD on each cell.

Again having a bank that's not balanced isn't a big issue if you have LVD and HVD at cell level. The worst that can happen is you will trigger the LVD or HVD  befor you were expecting it due to the weekest link or cell that triggers it 1st. Effective reducing your usable capacity scope.

In my case I would probably need a cell to be a massive 100  150ah out of balance befor I would notice it triggering a LVD or HVD. I have had  reports from other offgrid systems where they are working there pack very hard over the full capacity and went for a very long period without any balancing at all. The result they had to trim about 30ah back into a 400ah bank when they did get around to doing a balance.The outcome. They decided to perhaps balance a tad more regularly.

As mentioned I'm trimming under 1Ah pr 3 months on a 400ah bank. Hardly even worth bothering with really but i do it anyhow to keep  trimming job quick and regular chance to glance over the bank for other issues. More or less a reason to hang out in the Air conditioned power room and play with everything

Some of it is luck with how well matched the cells are that you ended up with also there can my very small differences in the tiny load that the cell level boards apply to each cell. How hard your working the bank will have some effect on how often and how much balancing you might want to do.

Kurt

[australsolarier]

i balanced the 48V/400A/h lifepo4 battery:

as mentioned earlier the last time i balanced with the small circuit boards from evpower, i left off at 200mV max difference between the lowest and highest cell.

this time heeding Kurt's advice i helped bringing up the  4 lower cells with an external isolated power source.  they came up relatively quickly. but it is a bit like chasing bits of paper in the wind. you bring them 4 up, some others go down. (the overall voltage of 57.4 volts stays the same of course, so other cells have to go down, not evenly though.)  the little balancing circuit boards probably do their bit in addition as well.  and like Kurt mentioned there is not that much energy in the lift up. anyway the max difference is brought down to 45mV. so this is much better.
as for the 12V/300A/h small lifepo4 battery, the max difference is 16mV at 14.4 Volts. this has been running for a year and so 50 to 60% SOC every day.
so i am much happier with the 45mV than the 200mV. makes for easier sleeping at night. heheheh

greetings and thanks for that professional advice, Kurt

urs

[ClassicCrazy]

I thought  bottom balancing was just where you start them off initially - so after that you use the Whizbang to tell when they are full enough at about 90 % or so and let that terminate absorb  and go into float .  All the other loads during the daytime when the sun is out would just be available to the loads as usual.

I just got some 100 AH Calb but haven't started using them yet. So that is why I am trying to understand the different lithium charging philosophies
Larry 

[australsolarier]

there seem to be two ideologies: the bottom and the top balancers.

like kurt already said, the bottom balancing is mostly used in car batteries. you do it once at the beginning of installation. the reason behind that is that when you deplete the battery to almost zero, all the cells are still delivering current. and not one or some going under the dangerous destructive voltage, whilst the overall voltage is still ok. (they have many more cells than house batteries)

however batteries running the electricity in houses spend a lot of time at the top voltage. many systems go into float in the morning. so you do not want to just charge them only 90%. i don't think there is a provision for the whizzbang jr to go into float at 90%. you would have to change absorption and float voltages to get an approximation of 90%. (but the way we charge them might be only 90% for all i know)
so here in australia the company evpower sells battery packs including all the bms (battery management system)  and bcu (battery control unit). this disconnects the battery when one cell goes over- or under voltage, low or high voltage disconnect. over current. etc.  the little circuit boards on the battery also do balancing at the high end of voltage. unfortunately this is at a voltage we try to avoid for reasons of longevity of our investment. so in my case absorb is 3.5V, but the bms kicks in at 3.6v . so what i have done with the small 12v system, charge it once a month with 3.65v absorb. and measure the cells with the voltmeter. to see they are reasonably close together. (the multimeter i use measures down to 1mV)
my 48v system is fairly new, has for the time being only light use and it was the first time i balanced the cells. so that's why i had 200mV difference between the highest and lowest cell. that why i asked kurt how he does his cell balancing, because i did not like the battery  being in that 3.65v area for a longish time.
a cell unbalance seems to work accumulative. every cycle it gets a bit worse. but we think it might take a long time to get into danger zone.
basically we take care of hour investment and want it to last  for the longest possible time. that is the reason we check them periodically and install air conditioners. most of the time we have the free electricity to run the air conditioner anyway. the weather is usually hot when the sun shines.
and all with the lifepo4's is a bit new. basically we learn of other people"s mistake. the chinese manufacturers are not really interested in experimenting how to charg them. (they leave that to us the users) and they do not greatly appreciate advice. i think it has something to do with their ideology of loosing face.

so that is the reason why there are so many threads of how to balance and charge lifepo4 batteries. also the manufacturers specification might not necessarily be the best.
ok, hopes helps a bit for future lifepo4 users
urs

[mike90045]

Ballancing at the top.

Ballancing at the bottom is hardly practical in a offgrid system that's powering live loads at the time.......

I'll disagree.   
Top Balance  =  Charging cells to the top 90% of allowable long term float voltage, and generally relies on "Balance Boards" to bleed off the lower capacity cells that charge up quicker to full.   But a 2A balance board really will do nothing with a 40A charger, still 38A going into the full cell.
So the lower capacity cells are bumping the Full limit often.
Then at discharge, those low cap cells reach the 80% discharged point and start sliding down the voltage knee way before the better cells hit the knee.    If you don't have generous LVD set, you risk reverse charging a cell and destroying it.

Bottom Balance = bleeding each cell down to it's 80% discharged point and then while recharging, the Controller can sense the top end at 90% full, before the knee starts and terminate charge.

I guess it's the devil you trust vs the unknown.   Top:  I distrust 15+ little balance boards to work properly for as long as the batteries are expected (hoped) to last. if one fails, or a single cell hits the knee and reverses, it's spending time.
Bottom: The fail points are the inverter's LVD setting and the charge controllers Hi Volt disconnect.  But you save 15+ balance boards and their wires from being added to the failure route.

[australsolarier]

in god we trust

do you trust the electronics of the solar charger?

do you trusts the electronics of an inverter?

do you trust the electronics of a smoke detector?  (or not having one because it might fail?)

during absorb the voltage is fixed. so as your 40amp charger approaches full charge the current bleeds off to the programmed end amps. depending on the size of your bank (400ah = 8amps)

so is it possible your mathematics are flawed?