diy solar

diy solar

Xuba Electronics: DEAL - 280AH LiFePo4 cells. Purchase & Review

Doing a bit of rough math.

Assuming the following
4 cell stack
  • 0.5mm expansion per cell, so 2mm stack growth during charging from 30% SOC
  • clamped with a rigid end plate (thick aluminum etc)
  • 72mm nominal cell thickness for a total stack thickness of 11.3"
  • clamped with 4 1/4" steel threaded rods
When charging, assuming the cells expand 0.5mm, the rods will need to elastically stretch 2mm, or 0.078". Since each rods acts like a spring we can calculate the clamp load assuming zero clamp load at 30% SOC.

Stress area of 1/4 rod is 0.031 in2. Steels elastic modulus is 29,000,000psi. Calculating for Krod is (0.031*29,000,000)/11.3 = 79,557lbs/in.

So 4 * 79,557 * 0.078 = 24,821lbs.

Of course this assumes that the cells MUST expand 0.5mm each during charging, which I have not verified, and I doubt they will if constrained with a "300kgf fixture".

But it does pose the question if its best to clamp cells at low SOC. I would personally just clamp when fully charged, which is typically the cell state after a top balance. Then verify cell security after a discharge.
 
Luthj said:
But it does pose the question if its best to clamp cells at low SOC. I would personally just clamp when fully charged, which is typically the cell state after a top balance. Then verify cell security after a discharge.
Click to expand...

Someone did that and found his pack was loose after discharge. I don't know how many have noticed expanded cells after top balancing. But yeah I understand your point and it is another advantage of using threaded rods as they can be easily adjusted.
 
I think a plate or piece of plywood would apply pressure to the middle of the cell more effectively. That is what have done.
 
One option is to use a firm closed cell foam rubber (say 1/4" thick) under one or both end plates. They are available in 10-20psi for 25% compression.
 
Gents could I get your opinions please.
My cells arrived and I arranged in parallel and charged/top balanced up to 3.65v and down to .5amp using a 30v10a psu.

I numbered all the cells 1 through to 8. Using the 150w load tester @20a I began capacity testing all the cells down to 2.5v cut off. They all came home at around 282-4ah with the exception of cell number 7 it came in at 275ah.

The cells were charged/top balanced again then arranged in 2p4s configuration , bms (jbd) connected this time for testing. The 150w load tester reconnected @10a and the test begins.

Towards the end of the test, series #1 which contains cells #8 and #7 starts to drift away from the rest of the series around 320mV and cuts off the bms @2.5v as per attached pic.
The whole pack came home @564ah on bms shut down.

Could this be an indication of a crook cell (#7)? Would it be worth separating this cell and “exercising” it a few times to see if it gets a bit better?

I’m not to fussed about it because I can just adjust the bottom cut off up a bit and sacrifice a few ah because I’ll never use 100% of the ah anyway.

During the sales video Amy sent there was a cell "caught my eye" as being a little off from the rest of them, I wonder if this is that cell.
What are your thoughts lads?
 

Attachments

  • Screenshot_20200909-222403_xiaoxiang.jpg
    Screenshot_20200909-222403_xiaoxiang.jpg
    23.8 KB · Views: 32
  • 20200909_222537.jpg
    20200909_222537.jpg
    40.2 KB · Views: 33
Sharky722 said:
Gents could I get your opinions please.
My cells arrived and I arranged in parallel and charged/top balanced up to 3.65v and down to .5amp using a 30v10a psu.

I numbered all the cells 1 through to 8. Using the 150w load tester @20a I began capacity testing all the cells down to 2.5v cut off. They all came home at around 282-4ah with the exception of cell number 7 it came in at 275ah.

The cells were charged/top balanced again then arranged in 2p4s configuration , bms (jbd) connected this time for testing. The 150w load tester reconnected @10a and the test begins.

Towards the end of the test, series #1 which contains cells #8 and #7 starts to drift away from the rest of the series around 320mV and cuts off the bms @2.5v as per attached pic.
The whole pack came home @564ah on bms shut down.

Could this be an indication of a crook cell (#7)? Would it be worth separating this cell and “exercising” it a few times to see if it gets a bit better?

I’m not to fussed about it because I can just adjust the bottom cut off up a bit and sacrifice a few ah because I’ll never use 100% of the ah anyway.

During the sales video Amy sent there was a cell "caught my eye" as being a little off from the rest of them, I wonder if this is that cell.
What are your thoughts lads?
Click to expand...
I have my BMS set to cut off at 2.875V and 3.5V per cell.
 
Your capacity seems fine. There will always be one cell the hits bottom first.
 
Sharky722 said:
During the sales video Amy sent there was a cell "caught my eye" as being a little off from the rest of them, I wonder if this is that cell.
Click to expand...

What was different? The voltage, internal resistance, or both?

Sharky722 said:
Could this be an indication of a crook cell (#7)? Would it be worth separating this cell and “exercising” it a few times to see if it gets a bit better?
Click to expand...

I would exercise it in the pack like it is. No need to separate it.

Sharky722 said:
I’m not to fussed about it because I can just adjust the bottom cut off up a bit and sacrifice a few ah because I’ll never use 100% of the ah anyway.
Click to expand...

I wouldn't worry about it either. I have 8 cells coming and will never get to use the full capacity unless we have a major power outage.

Sgt Raven said:
I have my BMS set to cut off at 2.875V and 3.5V per cell.
Click to expand...

Those are good settings and I am going to do the same.
 
Gazoo said:
Those are good settings and I am going to do the same.
Click to expand...
It's not like they are written in stone. If I felt I needed to change them because of a situation I was in, then it's only a few clicks away.. That is the beauty of building my own battery and the smart Bluetooth BMS, I have complete control over it.
 
Gazoo said:
What was different? The voltage, internal resistance, or both?



I would exercise it in the pack like it is. No need to separate it.



I wouldn't worry about it either. I have 8 cells coming and will never get to use the full capacity unless we have a major power outage.



Those are good settings and I am going to do the same.
Click to expand...
The resistance on 7 cells was .11-.12 then 1 cell was .15
 
Sgt Raven said:
It's not like they are written in stone. If I felt I needed to change them because of a situation I was in, then it's only a few clicks away.. That is the beauty of building my own battery and the smart Bluetooth BMS, I have complete control over it.
Click to expand...

I totally agree. But I jumped the gun anyways. Those are the settings I am going to use for my inverter. The BMS will have higher and lower cut off settings.
 
Sharky722 said:
The resistance on 7 cells was .11-.12 then 1 cell was .15
Click to expand...

That's still close. It would be interesting to know if cell 7 was the one that measured .15. I don't know that would make any difference. Perhaps someone else would know.
 
Gazoo said:
That's still close. It would be interesting to know if cell 7 was the one that measured .15. I don't know that would make any difference. Perhaps someone else would know.
Click to expand...
Thats what im keen to find out.
I understand 1 cell will always hit bottom first but there was just such a big gap at the end of 320mv which just happened to be #7
 
OK, let's step back a second.
320mv is not significant... please, too many get hung up on such MINUTIA, really, it is almost nitpicking and really serves little purpose. Even perfect matched cells wobble voltage wise at the top. All LFP will settle from 3.65 down a bit while they rest, often unevenly making one or two cells standout, not unusual. A reminder, that these cells are Within the Grade-A Spec of the manufacturer but thee are not Voltage & IR Matched and batched cells and so such deviations is not unusual. This is one reason the Top & Bottom Balancing is mentioned so often.

Assume a generalized rule of 1mv per amp hour capacity is not unusual. Often a Passive balancer (which burns off hi volts only from hi cells) can address such small deviation but Passive is very slow. Larger Capacity cells are often too big for passive to be efficient due to it's slow process.

Longevity practices dictate that we stop the cells at 10% from bottom & 10% from top providing 80% of their capacity for use. 2.75 & 3.28 respectively per cell.

Here is some Excellent Information from Orion BMS Folks (a pretty darn credible BMS Company who knows their stuff. The BMS specific stuff is moot as it depends on what your BMS is, but the battery info is what we want anyway.

High Resistance Cells : Again Orion hs some helpful info.

Something from Battery U:

BU-803a: Cell Matching and Balancing – Battery University

Learn about battery technology, battery maintenance and battery life at Battery University
batteryuniversity.com

Hope the info helps.
 
Steve_S said:
OK, let's step back a second.
320mv is not significant... please, too many get hung up on such MINUTIA, really, it is almost nitpicking and really serves little purpose. Even perfect matched cells wobble voltage wise at the top. All LFP will settle from 3.65 down a bit while they rest, often unevenly making one or two cells standout, not unusual. A reminder, that these cells are Within the Grade-A Spec of the manufacturer but thee are not Voltage & IR Matched and batched cells and so such deviations is not unusual. This is one reason the Top & Bottom Balancing is mentioned so often.

Assume a generalized rule of 1mv per amp hour capacity is not unusual. Often a Passive balancer (which burns off hi volts only from hi cells) can address such small deviation but Passive is very slow. Larger Capacity cells are often too big for passive to be efficient due to it's slow process.

Longevity practices dictate that we stop the cells at 10% from bottom & 10% from top providing 80% of their capacity for use. 2.75 & 3.28 respectively per cell.

Here is some Excellent Information from Orion BMS Folks (a pretty darn credible BMS Company who knows their stuff. The BMS specific stuff is moot as it depends on what your BMS is, but the battery info is what we want anyway.

High Resistance Cells : Again Orion hs some helpful info.

Something from Battery U:

BU-803a: Cell Matching and Balancing – Battery University

Learn about battery technology, battery maintenance and battery life at Battery University
batteryuniversity.com

Hope the info helps.
Click to expand...
Thank you for that info Steve cheers.
 
FYI GENERAL re XUBA & LUYUAN (their sister company)

They are presently trying to work out how to offer Fully Matched & Batched Cells and they have run into a learning curve about the process and it's costs. Bottom line, this requires serious Test Rigs and Equipment and it is not cheap. Then there is the time factor as it is Time Consuming running cells through complete cycles. The realization that this is a long & tedious process which adds cost per cell hit home. This would add cost per cell / bundle but now that would provide for better pack builds and more performance.

Here is a USA Vendor that offers a Cell Matching Service and they outline (very basically) the extra cost & process. I think it helps explain it just a little.
www.batteryspace.com

Custom Job for matching cells

8818
www.batteryspace.com

Here is a concern which will arise from such a service being offered as well, POSSIBLY. The cells must be connected/mounted to the test rig, meaning the terminals will get scratches and I know that some Derwood would run of claiming "the are used cells". It DOES happen, there always has to be a few that just have to go "Chicken Little - The Sky is Falling" mode.

This Service may be offered starting in October or so, depending on what they sort out and if they can make arrangements with a test facility.|
 
Last edited:
Steve_S said:
FYI GENERAL re XUBA & LUYUAN (their sister company)

They are presently trying to work out how to offer Fully Matched & Batched Cells and they have run into a learning curve about the process and it's costs. Bottom line, this requires serious Test Rigs and E
Click to expand...
I would imagine that to a be substantial outlay for the company not only in the set-up costs but also the operational and maintenance cost also.
There must be a demand for the service though.
 
Testing equipment to quickly and efficiently cycle big batteries is custom, large, and expensive. Especially when data logging is required. Typically 6-7 figures USD.
 
I keep seeing references to "300 kgf" as if it's something odd, but this is a relatively typical way to refer to forces in the metric system although it is not the standard unit.

300 kgf is the force exerted by 300 kilograms of mass at standard earth gravity (9.806 m/s^2 acceleration), or 2942 Newtons. This is the proper SI unit of force.

For us used to degrees Foreignheit and length based on a monarch's thumb, naturally this is equivalent to 20.56 slugs and by multiplying by the 'Merican acceleration due to gravity (a perfect 32.17 ft/s^2) we come to this being the force exerted by about 640 pounds of mass. Naturally this is not the exact answer since we rounded several times in these calculations.

Or you could say 300 kg is 661 lbs, and do that. For us used to one relatively stable gravitational field the line between "force" and "mass" is very blurry...
 
chicagoted said:
I keep seeing references to "300 kgf" as if it's something odd, but this is a relatively typical way to refer to forces in the metric system although it is not the standard unit.
...
Or you could say 300 kg is 661 lbs, and do that. For us used to one relatively stable gravitational field the line between "force" and "mass" is very blurry...
Click to expand...

Though I was unfamiliar with the unit, it wasn't too hard to grasp once I looked into it, and it started being discussed.

Where (I perceive and feel) the uncertainty to be, is how to implement that spec in a design, and when and why it is beneficial.
 
Dzl said:
why it is beneficial.
Click to expand...

High charge and discharge currents can create voids in the roll/sandwich. Compressive force reduces or prevents these voids. In prismatics it also provides support for the large, but very thin layers. Physical support is also seen in other chemistries with thin layers. For example pure lead AGM batteries make use of a higher density AGM mat between the cells. If memory series, lifeline uses the case to provide some compression when they assemble, which provides some of their legendary lead acid vibration resistance.

Since most of use are not exceeding 0.5C, and 90% of the time we are under 0.2C, its much less of a concern. But with 1C rates? It obviously makes enough difference that the cycle life chart calls it out.
 
  • Like
Reactions: Dzl
Luthj said:
Since most of use are not exceeding 0.5C, and 90% of the time we are under 0.2C, its much less of a concern. But with 1C rates? It obviously makes enough difference that the cycle life chart calls it out.
Click to expand...

This was one of the things I have been pondering with no way to answer based on the datasheet alone. Under what conditions is this compression beneficial, I'm sure it won't hurt no matter what, but it would be useful to know whether (like active or passive cooling) it is beneficial at high 1+ C rates or if it is beneficial regardless of C-rate.

It would also be useful to know how they measure it (100% SOC? 50% SOC? etc).
 
chicagoted said:
I keep seeing references to "300 kgf" as if it's something odd, but this is a relatively typical way to refer to forces in the metric system although it is not the standard unit.

300 kgf is the force exerted by 300 kilograms of mass at standard earth gravity (9.806 m/s^2 acceleration), or 2942 Newtons. This is the proper SI unit of force.

For us used to degrees Foreignheit and length based on a monarch's thumb, naturally this is equivalent to 20.56 slugs and by multiplying by the 'Merican acceleration due to gravity (a perfect 32.17 ft/s^2) we come to this being the force exerted by about 640 pounds of mass. Naturally this is not the exact answer since we rounded several times in these calculations.

Or you could say 300 kg is 661 lbs, and do that. For us used to one relatively stable gravitational field the line between "force" and "mass" is very blurry...
Click to expand...
Working in metric units.
Building a compression frame with 2 end plates and 4 threaded rods with nuts, one in each corner.
It would be handy if someone could calculate the torque in newton/meters that would have to be applied on each nut to create a force of 2942 newtons on the cells.
The newton/meter torque on the nuts would have to be converted to tensile force applied to the rod and therefore compression force on the cells.
Other than doing this one wouldn't know what compression weight/force the frame was applying to the cells without using some sort of force gauge.
Any statics engineers here that could help
 
You must log in or register to reply here.

Similar threads

W
Flexible or rigid bus bars?
Replies
21
Views
769
tinyt
T
H
Charging and Discharging Current for Parallel Cells
Replies
4
Views
246
Helles
H
F
Connecting JKBMS in parallel
2 3
Replies
58
Views
2K
Cogito44
C
S
Best vendor for ~300Ah cells? And some related questions
Replies
2
Views
381
hwy17
hwy17
B
Battero Tech BT280 3.2V 280Ah LiFePO4 Cells - US$55.00 _ GobelPower.com
Replies
8
Views
1K
Skypower
S

diy solar

diy solar
Back
Top