LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[CharlieJ]

RC-
Thanks for the clarification...makes good sense to me!

[Jd1]

Originally Posted by Jd1

A couple or three questions re Winston batteries:
1) Is there any disadvantage to operating the cells in other than an upright configuration? I know they are supposed to be usable in any orientation but they do have a vent plug so I'd like to make certain.
2) Is there concensus if banding the bank is a requirement for boat applications or am I only concerned about imobilizing them in case the boat goes turtle?
3) Is there a rule of thumb about cell sizes and what size is best? Smaller cells have a lower energy density per volume but a dud cell is cheaper to replace. My choices are 100Ah, 200Ah or 400Ah.

Bump .... anybody willing to take a stab at these questions?

[deckofficer]

Jd1, I didn't take a stab the first time as others are better qualified than me, but here goes,

I think you can mount them any way you want and it is my assumption that banding may not be needed for < 1.0 C charge and discharge. As to cell size I always try to go for a single series string of the voltage and capacity that you need.

[diugo]

I agree with Deckofficer Bob. As to LFPs in "any orientation", note that upside down is generally prohibited.

[diugo]

Originally Posted by Maine Sail

I am actually operating to 20% SOC/80% DOD and back to 98-99% SOC. At 13.8 - 14.0V I have not found I have lost much capacity at all despite being well out of both knees

How is 98-99% SoC avoiding both knees? Cuz I thought the knees were the top and bottom 10%.

[Jd1]

Thanks Bob and diugo for taking a stab at my questions !

[Maine Sail]

Originally Posted by diugo

How is 98-99% SoC avoiding both knees? Cuz I thought the knees were the top and bottom 10%.

I don't discharge below 80% DOD so I never hit the low voltage cliff or knee (2.8 VPC) and I only charge to 13.8V which avoids the upper voltage spike or knee.. My current taper is slightly longer at 13.8V / 3.25 VPC but at 13.8V charged to 5A acceptance I can still get 400Ah out of our 400Ah bank while conducting a 25% of capacity load test. This leaves very little risk of hitting the upper knee/voltage spike unless the cells were to become grossly unbalanced. My cells still remain within a few thousandths of a volt even when pushed to .2A of acceptance at 13.8V after 204 cycles to approx 80% DOD.... As near as I can tell for my bank 13.8V to 5A of acceptance leaves a few Ah's of capacity on the table, about 7-10 Ah as near as I have been able to calculate. This means I am technically never pushing up into the danger zone where a cell could spike if the bank were to get out of balance. My bank has remained in near perfect balance over 204 cycles when charging mostly between 80% DOD & 13.8V to 5A accepted current.....

If I am missing something please let me know because the bank is on my bench for more winter experiments. I will not be top balancing again it until I see a need as I want to see how many cycles it takes to get my pack out of balance using the charging regime that I settled on after last winters testing..

[georgec]

I was looking for the flexibility of installing the cells on their side and asked Balqon that question just last week. Here is their answer:
"We do not recommend using cells laying on the side for long periods of time due to safety and reliability reasons."

[diugo]

Originally Posted by Maine Sail

If I am missing something please let me know because the bank is on my bench for more winter experiments.

It may just be a vestige from my lead acid days, where conventional wisdom says that bulk charging only brings you to 90% SoC. I did find one LFP charge curve which clearly shows the knee starting at 90%----but also lots of other curves where it was much closer to 100% (and some even above 100%).

With my older cells, higher Peukert, and low C/20 average solar charge current, I simply don't see the sharp spiky knee you are apparently seeing with your newer cells and lower Peukert. At C/20 my plateau is around 13.5V and once the voltage starts slowly rising, it takes two hours or more to reach 13.8V---which also supports my 90% figure.

But it may well follow that the lower the Peukert, the higher the SoC before reaching the knee. And it may be that the older the cell, the higher the Peukert. This may partially explain the widely held assertion that LFPs "have no Peukert effect"---when in fact these drain tests are simply performed on relatively young cells.

[T1 Terry]

Originally Posted by goboatingnow

Google is your friend. Info All over the place . Li batteries do have a fairly narrow usable recharge temp range

Ok say here
Charging Batteries at High and Low Temperatures
Dave

Google is only your friend when you work out how to sort the wheat from the chaff.
If Winston LYP cells really do have this problem I guess you should contact the manufacturer because they don't seem to have a clue.
Used as house batteries the cells are never likely to subjected to the discharge or charging currents an EV on take off or max regen can demand. Even then, a few mins light load or charging will warm the cell internals sufficient to accept quite solid charging and discharging currents.

T1 Terry

[T1 Terry]

Originally Posted by Jd1

A couple or three questions re Winston batteries:
1) Is there any disadvantage to operating the cells in other than an upright configuration? I know they are supposed to be usable in any orientation but they do have a vent plug so I'd like to make certain.
2) Is there concensus if banding the bank is a requirement for boat applications or am I only concerned about imobilizing them in case the boat goes turtle?
3) Is there a rule of thumb about cell sizes and what size is best? Smaller cells have a lower energy density per volume but a dud cell is cheaper to replace. My choices are 100Ah, 200Ah or 400Ah.

(1) Best not to mount them upside down, it makes it very had to do the terminal bolts up, apart from that there is no restrictions... as long as you can protect the exposed terminals from anything conductive touching across them, not a good look when that happens and a change of clothing is required

(2) If you do not use some form of solid plating across the larger faces of the end cells and compress them together you will loose capacity as the cells will bulge. Has nothing to do with the rate you discharge or charge them, it is all to do with the cell voltage at any given time. Exceed the magic 2.8v and 3.4v limits for any period of time and the electrolyte will start to boil off the lighter vapours. Just like the coolant in the car, the higher the pressure with in the cooling system the higher the temp before vapour is released. Banding the cells with solid end plates stops the case from expanding and allowing the pressure to remain low. Even on the shelf cells in warm climate will start to bulge the cases just sitting there, that is why they are so well packaged in the boxes when they come from the factory.
If your cells have bulged, don't panic, in a warm environment simply use the solid end plates, 3mm aluminium will do, a using a "G" clamp or sash clamp, compress the pack back together, then strap it or make a battery box that the ends can be mechanically pressed against the cells larger face, the other end solid enough not to bulge under the pressure, and squeeze them up till all the cells sit flush with one another.

(3) If at all possible, use multiple cells in parallel to build capacity, connect these is series to build voltage. Only one 12v battery for a 12v system, far too hard to control cell balances between multiple 12v Li batteries in parallel, it makes herding cats look like child’s play.

T1 Terry

[T1 Terry]

The whole 80% SOC discharge thing requires some interpretation. It's another case of asuming things and read more into what is written than was written.
There still needs to be tested and proven using a number of test samples and a control sample to determine if the 80% discharge and 3,000 cycle life actually means the cells should only be discharged to 80% SOC if you want to get the 3,000 cycle life
or
After 3,000 cycles the cell will still have 80% of it's capacity remaining.

All the theory comes from the interpretation of the chart listed below.
Knowing the degree of additional information required to fully understand the manufacturers charging and discharging voltages, a number of different meanings can be read into the information shown in the chart. If the chart on the right does indicate the number of cycles one could expect if each line represented a depth of discharge, wouldn't all the lines start at the same voltage point?

[T1 Terry]

Originally Posted by goboatingnow

Egg heads will enjoy this recent article on LiFerrous , which has shown a " memory effect" exists in these cells


Memory effect now also found in lithium-ion batteries

Dave

I only had a quick read through the article but aren't they saying the memory effect is only in relation a small voltage "step" and not related to a reduction in capacity as such?
They refer to this being important when voltage is used for determining state of charge, but voltage related to state of charge doesn't have a linear path anyway so voltage can not be used as a depth of discharge indicator at the best of times, a columb count is the only truely accurate method.
If they are saying a voltage step would occur at 2.8v and 3.45v if this was the regular scale of the cells operation, I can't see that being an issue anyway.
So that leaves the question, of what value was the study?

T1 Terry

[goboatingnow]

Originally Posted by T1 Terry

Google is only your friend when you work out how to sort the wheat from the chaff.
If Winston LYP cells really do have this problem I guess you should contact the manufacturer because they don't seem to have a clue.
Used as house batteries the cells are never likely to subjected to the discharge or charging currents an EV on take off or max regen can demand. Even then, a few mins light load or charging will warm the cell internals sufficient to accept quite solid charging and discharging currents.

T1 Terry

Well terry , there's loads of articles that say Li should never be charged below freezing. I would be very suspect of that spec sheet. At the end of the day Li technology is the same across all Li -ion variants


Furthermore here's the CALB DATA

Technical Parameters
No
Item
Parameter Specification
1
Nominal Capacity
40Ah@0.3C Discharging
2
Minimum Capacity
40Ah@0.3C Discharging
3
Nominal Voltage
3.2 V
4
Internal Resistance
&le;1m&Omega;
5
Charging(CC-CV)
Maximum Charging Current
1C
Charging Upper Limit Voltage
3.65V
6
Discharging
Maximum Discharging Current
2C
Discharging Cut-off Voltage
2.5V
7
Charging Time
Standard Charging
4h
Quick-acting Charging
1h
8
Recommended SOC
Usage Window
SOC 􏰀10%~90%
9
Operation Thermal Ambient
Charging
0°C ~ 45°C
Discharging
-20°C ~ 55°C
10
Storage Thermal Ambient
Short-term (within 1 month)
-20°C ~ 45°C
Long-term (within 1 year)
-20°C ~ 20°C
11
Storage Humidity
􏰁70􏰂
12
Battery Weight
Around 1.4kg
13
Shell Material
Plastic




NOTE THE CHARGING TEMP RANGE which is consistent with Li tech
Dave

[goboatingnow]

And the Sinopoly cells


SP-LFP60AHA-S2 cell specification
Model name
SP-LFP60AHA-S2
Nominal voltage
3.2 V
Operating voltage under load is 3.0 V
Capacity
60 AH
+/- 5%
Internal impenetrableness
<2 mOhm
Operating voltage
min 2.8V - max 3,7 V
At 80% DOD
Discharging cut-off voltage
2.8 V
The cells is damaged if voltage drops below this level
Charging cut-off voltage
3.8 V
The cells is damaged if voltage exceeds this level
Recommended
charging - discharging Current
18 A
0.3 C
Maximum short-time discharging current
300 A
5 C period=10s
Life cycles
2000
0.3C, 80% DDC
Operating thermal ambient - charging
0°C ~ 70°C
The battery temperature should not increase this level
Operating thermal ambient - discharging
-20°C ~ 65°C
The battery temperature should not increase this level
Storage thermal Ambient
-20°C ~ 45°C
The battery temperature should not increase this level
Shell Material
Plastic
flame retardants
Dimensions
139 x 34 x 222 mm
Millimeters (tolerance +/- 1 mm)
Weight
1,9 kg
Kilograms (tolerance +/- 150g)



I think I can say QED. and the Winston spec is JUNK

Dave