• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Does anyone else have an impression EVE MB31/MB31 seem to charge at higher voltages than other cells?

Luk88

Solar Addict
Joined
Apr 5, 2024
Messages
783
Location
Poland
I've done discharge tests and the curve looks normal. But it seems at least for the very first one or two charges the cells voltage raises a lot more than one would expect. (DC and AC IR are both good)

For example. My cells(MB31) were delivered at 3.33V. By any SOC/voltage table this is about 80% full. But the very first discharge test (from "as delivered") resulted in about 130ah (I have the curve elsewhere, I'll post it later if people are interested). I then charged it fully and a full discharge test showed 330AH and the curve is by the book.

Then I assembled a 16s pack. I charged it till 3.42 per cell on average (half of my cells at 3.45 half at 3.38 - there were two batches). I though surely they must be at 99% full by now.

So I set them up in parallel and I started charging with 40A. Many hours have passed and ~530AH later cells that were on 3.45 are on 3.53V cells that were on 3.38 are at 3.48. Almost no current flows to higher voltage cells (0.2A per cell) so this 530AH must have gone into 8 cells with lower SOC. So 66AH per cell above 3.38V !

Isn't that quite unusual for LFP? I never had EVE cells before and this post is not a complaint, the capacity and power delivery is fine. I'm just puzzled by these cells seemingly defying usual SOC/V tables during charging. Both my higee cells, old crappy CALB , as well as few types of pouch cells I had were almost fully charged at around 3.4V.

Perhaps this is only for the first few charges?
 
I believe when I check the specs on those they have a slightly higher internal resistance than say the LF 280k's I'll have to look back on that

But that would make them charge a slightly higher voltage their standing voltage should still be the same
 
I believe when I check the specs on those they have a slightly higher internal resistance than say the LF 280k's I'll have to look back on that

But that would make them charge a slightly higher voltage their standing voltage should still be the same
Indeed. I never had LF 280k so I can't compare like for like.

However, If it was only internal resistance wouldn't the difference in voltage go back to zero when there is no charging current? I suspect this is either just a feature of a very new cell, or maybe some new chemistry they are using.

I've had a look at the very first discharge curve attached. I'm pretty sure I measured 3.33 with a handheld meter (the AC IR meter everyone uses, I forgot the type). But the very first record shows 3.29V and when current is started (40A) it dropped only a couple mV, but it did drop 20mV within the first 2 minutes.

The "as new" discharge curve:
2024-8-1-11-36-47.jpg

Below is the same cell charged and discharged fully:
2024-8-1-11-35-44-full.jpg
 
I downloaded manufacturer specifications for both cells data is inconclusive because it is represented differently more less ballpark figures.

eve LF280k v3 <=0.25 mOhm
eve MB31 0.18 mOhm +-0.05 mOhm

Neither is specific and does not suggest that one is of higher or lower value than the other.

I will share my results once I have received and cataloged my cells for resistance matching though results can vary with between meters. I will preform capacity test and charge curve analysis on 4 liest fitting cells. Too many to do all of them.

If I find a live comparison of both cells I will link it.
 

Voltage rise like this?
Wow, yes, exactly like this. The only difference is that all the cells in my order are like this.

Is it a problem when they are all matched? Well... Probably not, but also maybe 🤷

I have a number of batteries in parallel. I went to great lengths (no pun intended) to ensure absolutely exact busbar and wire lengths between my batteries and I immediately see my new cells taking on most of the current load.

This will be a 33kW system in the end so I can't have a single battery bear all of this current. I'm hoping I can mitigate this by charging/discharging together few times.


I'm also a bit worried about a loss of capacity. As I'm normally charging only till 3.45 with a float of 3.35. This is well over 95% on my other cells. While these cells will be probably be at around 80%. Not ideal.


@upnorthandpersonal made a comment on the video that this may well be a feature of new cells during first few charge cycles. I hope so. Sadly I haven't saved my charging curves, just my discharge curves. I have to compare these two discharges (very first and second) and see if I'm imagining it, or is the second one about 7mV lower.
 
It is speculation on my part, but to reiterate what I said in the comment under that video: I have seen a curve like that in a cell that had zero (or just a few) cycles on them fresh from the manufacturer. It has to do with the SEI layer not formed yet/completed. After some cycles, the curves become more defined (steeper at the knees, flatter in the middle). I can only speculate that this is perhaps due to improved manufacturing and Q/A processes that either a) don't need cycling at the manufacturer to conclude capacity b) can determine defects before cycling so the cell can be discarded, c) effectively due to changes in cell composition (electrolyte/anode/cathode) that cause this effect to linger for more cycles.

I have not had time to test any of the new EVE cells. I'm in the middle of moving/upgrading the battery lab, and we're going to be busy with that for some time before we can start testing again.
 
It’ll be interesting to see how current crops of cells hold up, 280k v3 as well.

Knowing the factories are still prob running to maintain production numbers but demand is falling off (record low prices) I wonder how much tinkering is being done to flirt in the edge of proper design/chemistry, and this is a result.
 

diy solar

diy solar
Back
Top