First test results Winston vs LiTime

[CaptainRivet]

Originally Posted by wholybee

You are good for balancing purposes. You are likely overcharging the batteries slightly unless you are able to achieve a high C rate. Whether this will dramatically reduce the life, or only slightly reduce the life, I don't know.

How many amps is the charger you are using? Do you have a Victron or similar meter that you can set to alarm at 14.2V so you can manually stop it? Or you can use a Victron battery protect, and set that to a lower voltage.

no you are not overcharging the battery with 3,65V. As you can see all their own charger are 14.6V so 3,65V per cell is what they are designed and tested. The BMS in them has 70mV passive balancing and if you drive/cycle them heavly you defintily need to go into 3,65V, especially if you have LI time with paralleled cells (jedi doesn't but most are). my experince with these chinese surprise boxes are they are tested and designed with 3,65V per cell charge (or even higher) in a system with standard values in MPPT and DC2DC, just look at LI-Time own ones and you can see what they are designed for. So by trying to pamper them for longer life and go to 3,55V or worse to 3,45V you reach the opposite as the battery needs the 3,65V to do proper balancing. And as i wrote its the only indicator (besides a capacity test where you actually do the same, charge and discharge till that cut offs) where you can see if a problem is approaching=imbalnce or not as you have no external communication. and with 70mV passive balancer if you get a 150mV runner in a 200AH battery (that approx when you charge to 3,55V and then get cutoff at 3,5V) then most likely its a take out a multitool and start cutting open operation to rescue it or a painfull long process keeping them forced at 3,5V or higher for a long time to get them balanced which stresses and damages the cells more then you initially charge them up to 3,65V and don't get into that situation.



If you have external communication and see cell values then its a different cup of tea as you can see when balancing is not enough and you start to get a runner.


and yes that they share and charge quite evenly is also normal and no surprise if you eg watch the tests that Andy from Offgrid garage channel did, eg a 280AH and a 10AH in parallel or even his Frankenstein battery in parallel with different manufacturer, different capacities and even significantly bulged cells=even the worst chinese surprise box is better then that.

[Chotu]

thanks! i have a pair of 200AH LiTime batteries

[sailorboy1]

Rivet why do some feel the need to add thread drift? Aren't there lots of lithium threads to post your expertise on? Can we not just stay focused on the topic?

[s/v Jedi]

Let’s be honest: if your battery needs balancing every single charge cycle then maybe you need better batteries. I have no experience with LiTime batteries that have parallel cells so who knows…

[s/v Jedi]

I just double checked our Winston cells. It has been three years plus two weeks since these were balanced. At that time, max. deviation was 8mv.

This picture is from now with the cells at 75% SOC so I will check again at fully charged for fair comparison. But this is the balance you can expect with good quality cells.

[barcoMeCasa]

Originally Posted by s/v Jedi

I just double checked our Winston cells. It has been three years plus two weeks since these were balanced. At that time, max. deviation was 8mv.

This picture is from now with the cells at 75% SOC so I will check again at fully charged for fair comparison. But this is the balance you can expect with good quality cells.

Do you let the cells sit idle for a while before checking for balance voltages?

My individual cell voltages vary considerably with each other depending on if charge is going in/out. Never sure what's the most useful data.
Do you have the data to graph what's actually going on over time? Pretty sure I have one cell a bit low. But overnight an outside cell will gradually drop less voltage compared to an inner cell when the fridge runs.
Don't plot the data, you have no idea what's going on inside each cell over time.

[s/v Jedi]

Originally Posted by barcoMeCasa

Do you let the cells sit idle for a while before checking for balance voltages?

My individual cell voltages vary considerably with each other depending on if charge is going in/out. Never sure what's the most useful data.
Do you have the data to graph what's actually going on over time? Pretty sure I have one cell a bit low. But overnight an outside cell will gradually drop less voltage compared to an inner cell when the fridge runs.
Don't plot the data, you have no idea what's going on inside each cell over time.

I only see an 8mV deviation on the mean voltages… that is not a considerable variation. You are zoomed in too much, you’re looking at cabling resistance as much as cell performance.

[barcoMeCasa]

Originally Posted by s/v Jedi

I only see an 8mV deviation on the mean voltages… that is not a considerable variation. You are zoomed in too much, you’re looking at cabling resistance as much as cell performance.

How long is the mean? & how do you know what you are looking at is accurate out of balance voltage? These cell differential voltages can vary a lot between idle, charging and discharging.
Hard to see how to get anything reliable without letting the cells sit idle for a good few minutes first of all.

Can you really not graph any of this data? Opens up a whole new world of information often unexpected

[s/v Jedi]

Originally Posted by barcoMeCasa

How long is the mean? & how do you know what you are looking at is accurate out of balance voltage? These cell differential voltages can vary a lot between idle, charging and discharging.
Hard to see how to get anything reliable without letting the cells sit idle for a good few minutes first of all.

Can you really not graph any of this data? Opens up a whole new world of information often unexpected

The only way I can move cell voltage up/down is by doing a lot of amps. My refrigeration turning on/off doesn’t change it even 1mV.

The reason you see variation changes between idle, charge and discharge is because the cells have different internal resistance. In that case you need to measure voltage while idle. When the cells are matched, like mine and afaik all Winston deliveries, then the deviation doesn’t differ much.

I am much more interested in the graphs like attached.

[barcoMeCasa]

Originally Posted by s/v Jedi

The only way I can move cell voltage up/down is by doing a lot of amps. My refrigeration turning on/off doesn’t change it even 1mV.

The reason you see variation changes between idle, charge and discharge is because the cells have different internal resistance. In that case you need to measure voltage while idle. When the cells are matched, like mine and afaik all Winston deliveries, then the deviation doesn’t differ much.

I am much more interested in the graphs like attached.

OK, so in one way your bank is so big that the relatively low loads don't show up any voltage drop either due to cell inbalance or connection resistance. Bet if you accurately plotted individual cell voltages each second you'd see much more going on

[s/v Jedi]

Originally Posted by barcoMeCasa

OK, so in one way your bank is so big that the relatively low loads don't show up any voltage drop either due to cell inbalance or connection resistance. Bet if you accurately plotted individual cell voltages each second you'd see much more going on

Yes, at sub mV level. At some point you can see big peaks that change as you move your hand closer to the cell.

Really, you are zoomed in too much. Remember I have 16mV maximum deviation after more than three years not balancing. Tiny mV changes due to load are of no consequence.

[barcoMeCasa]

Originally Posted by s/v Jedi

Really, you are zoomed in too much. Remember I have 16mV maximum deviation after more than three years not balancing. Tiny mV changes due to load are of no consequence.

Much more interested in what is actually going on inside each cell second by second, unexpected patterns can be hidden deep in the data but you need to go looking to find them. simple to zoom out if you have the data to begin with.
, Really, you are zoomed out too much...

[s/v Jedi]

Originally Posted by barcoMeCasa

Much more interested in what is actually going on inside each cell second by second, unexpected patterns can be hidden deep in the data but you need to go looking to find them. simple to zoom out if you have the data to begin with.
, Really, you are zoomed out too much...

Molecular workings inside the cell are interesting but not for this thread. You could add live measurements of internal resistance and electrolyte pressure etc. for even more data, but plenty of white papers are available already.