Merged LiFeYPO4 1000Ah Winston prismatic cells and all electric galley...

[goboatingnow]

Originally Posted by CaptainRivet

Just saw this.
You are sure thats the issue?
Active or passive balancing?
At voltage level did you stop charging daily, 3,55V?
It depends how and when balancing starts with the REC, if done from 3,45 volt on and you go to 3,55 then it might not be enough time but if it done from 3,30V or continous then it should be.
There is a ongoing discussion if a active balance can work properly and eg Dacian, the developer of the electrodacus BMS states it cannot and thats why he implemented a passive one.
If thats right I have not enough experience about that to judge, still learning here....

Interesting article from Eric Bertscher from Nordkyn design when terminating the charge:
https://nordkyndesign.com/charging-m...battery-banks/

I see no argument that active balancing can’t be used.

Firstly cell Imbalance is it the end of the world. If you implement cell level monitoring ( and every lithium system should be cell monitored then all that happens is the charge terminates on the first cell to reach the designated charge cutoff parameters.

Now in most cases cell imbalance where fractional C usage is deployed ( typical of boats ) you will see imbalances of 5-10 % all this means is you have slightly less energy available then the theocratical Ah number suggests. So your 500Ah system might now be a 475Ah. Given you should have built in some capacity headroom anyway , big deal , so what.

In my experience cell imbalance tends to remain fairly static over time. So it’s not rust things get progressively worse.

Balancing is sometimes presented as somehow “ drawing off charge “ to reduce the voltage to a balance , this is not right. What ( top) balancing is in fact is diverting charge current away from the cell that has reached charge stop so as to allow the other cells to continue to charge to their charge stop points.

With active or passive top balancing , therefore you need to have two conditions , (a) the charge current at the start of balance must be within the allowable range of the balance load. And (b) there has to be enough time to achieve the balance.

So if we have say 25aH imbalance and a 1A balance load , then it will take 25 hours at 1A charging so the charging process must have this time available. This can be problematic on boats.

Secondly most top balancing is made nonsense of by the battery continuing to supply loads while charging , because in many cases ( low capacity mains chargers or falling sunlight ) the load currents exceed the available charge currents and a discharge cycle begins , top balancing is useless here

Active and passive balancers ( in terms of terminology , these terms are used in the electronics industry to simply mean the load type ) do the same thing they provide a load which diverts charge current away from the out of balance cell either lost as heat for passive balancers or converted into charge current for active balancers.

The other issue is of course that the battery charging current at the time of a cell reaching its charge stop point may be to high to engage the balance load. If your charge current is 2A and you balance load can only handle 1A , and you reach a cell charge stop point , you are now forced to stop charging the whole bank irrespective of whether you can balance it or not.


I’m coming to the conclusion that cell based charging is actually the best way to handle lithium recharging. Ie a dedicated isolated boost converter dedicated to each cell. All 12v nominal charges sources are paralleled into the cell charger and the BMS has one charger per cell to control and NO balancing is every necessary as each cell is dealt with individually

I really think the issues we have are largely brought about by trying to shoehorn lithium into a overall charging system designed for a different battery technology entirely.

In my opinion in process passive balancing , ie diverting current from cells charging slower is a travesty of engineering

[newhaul]

My question is why does almost everyone want to complicate a simple job ?

[goboatingnow]

Originally Posted by newhaul

My question is why does almost everyone want to complicate a simple job ?

How is it simple !

[newhaul]

Originally Posted by goboatingnow

How is it simple !

My system is 250ah calb cells 4p1s 12 v nom . 200 amp DALY BMS passive top balance at 3.6v .
300 amp shunt battery monitor. Tracer 30 amp Mppt solar controller. ( fully programmable via Bluetooth)
On my vessel all normal loads go through the load circuit on the controller except the inverter charger. ( trace 2512 ) fully programmable charging )
3 years liveaboard. With no (0) issues. It does its job so I don't have to think about it .

[CatNewBee]

Originally Posted by Catapulte

CatNewBee....


Would it be possible to show the integration of the Mastervolt inverter on the original Cerbo GX Touchscreen when running your software ?
Or asked the other way ... Could your software be run in the original Cerbo GX instead of the original software from Victron Energy or is this only possible when it runs on the Raspberry Pi ?


BTW: Thank you for the great work you are doing here and all the inspiration.

It is the modified Victron software, so yes, it would run on Victron hardware too (most of it, the part that uses USB Interfaces). The temperature sensors not sure. Cerbos have GPIO ports, maybe configurable for 1Wire or I2C, maybe not. But in general it is just python / qml script software.

Well, the changes I made were not meant to be used on other systems, some are hard coded based on my setup.

[goboatingnow]

Originally Posted by newhaul

My system is 250ah calb cells 4p1s 12 v nom . 200 amp DALY BMS passive top balance at 3.6v .

300 amp shunt battery monitor. Tracer 30 amp Mppt solar controller. ( fully programmable via Bluetooth)

On my vessel all normal loads go through the load circuit on the controller except the inverter charger. ( trace 2512 ) fully programmable charging )

3 years liveaboard. With no (0) issues. It does its job so I don't have to think about it .

I didn’t think Daly did balancing ?

[newhaul]

Originally Posted by goboatingnow

I didn’t think Daly did balancing ?

Yes they actually do top balance. Here is a shot of the spec sheet

[more]

Originally Posted by BetterDays 2020

I have been searching for a 500 amp Marine Rated Circuit Breaker, and can't find one. Who makes them and where can I find one?

Thanks,

https://www.eti.hr/proizvodi-i-uslug...mounting-plate

https://www.megashop.hr/product/1859...DC-690-VAC-1-S

[CaptainRivet]

Originally Posted by more

https://www.eti.hr/proizvodi-i-uslug...mounting-plate

https://www.megashop.hr/product/1859...DC-690-VAC-1-S

Ok the first 500A breaker that’s up to the spec, like NH fuses high amp public grid equipment that’s also DC rated. Can‘t find the resistance though as with primarily AC fuse this is not a critical parameter,especially in the 690 and 1000V range.
The 1000V one will have a very high resistance for sure, same with NH3 1000V spec which is too much drop for 12 or 24V. 690V can or cannot work well.
Best is 400V spec equipment for low voltage dc use as here resistance is a critical parameter for purchase decisions and therefor resistance is as low as possible.

[more]

Originally Posted by CaptainRivet

Ok the first 500A breaker that’s up to the spec, like NH fuses high amp public grid equipment that’s also DC rated. Can‘t find the resistance though as with primarily AC fuse this is not a critical parameter,especially in the 690 and 1000V range.
The 1000V one will have a very high resistance for sure, same with NH3 1000V spec which is too much drop for 12 or 24V. 690V can or cannot work well.
Best is 400V spec equipment for low voltage dc use as here resistance is a critical parameter for purchase decisions and therefor resistance is as low as possible.

pick Size 3 / 630A Resistance (R) = 8.4E-5 ohm (Ω)

[CaptainRivet]

Originally Posted by more

pick Size 3 / 630A Resistance (R) = 8.4E-5 ohm (Ω)

Good
Thank you, learned something new

[Ulstue]

The problem with balancing not only on top is that each cell has its own individual charging curve. The one that bottoms out (without balancing) at around 3.4V may be the same one that tops out at around 3.55V (without balancing). If this cell has already been topped up at 3.4V, it will end up much further above the others at 3.55. Therefore balancing below the top range is counterproductive. It forces debalancing at the end of charge.

The charge curves of the individual cells can be very irregular with LiFePO4, sometimes in the range of 3.3V even at higher state of charge the voltage drops by mV to rise again later. If this is then rebalanced, the result is debalancing.

[CaptainRivet]

Originally Posted by Ulstue

The problem with balancing not only on top is that each cell has its own individual charging curve. The one that bottoms out (without balancing) at around 3.4V may be the same one that tops out at around 3.55V (without balancing). If this cell has already been topped up at 3.4V, it will end up much further above the others at 3.55. Therefore balancing below the top range is counterproductive. It forces debalancing at the end of charge.

The charge curves of the individual cells can be very irregular with LiFePO4, sometimes in the range of 3.3V even at higher state of charge the voltage drops by mV to rise again later. If this is then rebalanced, the result is debalancing.

Thats why your BMS rhould also track charge current, nos only Volt. Then it can balance earlier without getting into debalancing.
If you are 4S =12V and charge or load stays under 50%of its max cont C-Rate then no till little balancing needed which also takes a small amount of time to do so.

[CatNewBee]

It doesnt work like this, cells change their resistance when being charged or discharged, sometimes lower cells during bulk outrun the higher cell when hitting absorption and may even cause disconnects. You can only judge imbalances when the cells are near full.

[newhaul]

Actually did capacity test on Tuesday at full charge 3.6v on 3 cells 3.42v on cell #2 drew down to 12v to test and on full recharge got 265.8ah out of the cells that were sold to me as 240ah units . Now all 4 cells are within .1v of each Other at 14.4v cells were as follows 3.6v, 3.59v, 3.6v , and 3.61v
Reset my meter to read as a 250ah bank all is good. It was reading at 265.1ah 3 years ago at start of service life