Large Capacity Testing - advice

E

Eric Koshinsky

Member
Joined
Jul 22, 2017
Messages
75
Hello all,

I recently acquired two 100AH lithium-ion (LiFePO4 based) 12 v drop-in replacement 12v batteries. All of my testing gear is designed for single cells, so I'm at a bit of a loss as to how to test these for capacity. Even if they have fairly reduced capacity, I would prefer to keep them intact so I can use them to power my trolling motor and save about 50 pounds.


image_hhwlki.jpg


https://www.lithiumbatterycompany.com/product/12v-100ah-lithium-ion-battery/


image_paxlrd.jpg


https://www.lithiumion-batteries.com/products/product/12v-100ah-lithium-ion-battery.php


It would be nice to test them at something in the range of 15 - 30amps (the max my motor draws is 30 amps). This is well within their continuous draw capacity of either battery. I would just use my motor, but it is winter here and the motor needs to run in water, so that won't work for another 2.5 months.

Any suggestions would be very much appreciated.

Thanks.
Eric
 
The crude way, take an automotive headlight bulb and after charging the battery up, hook it up and see how long it takes for the voltage to drop to 10V.

So, if the bulb pulls 5A, that'd be about 60W. If the battery takes 2 hours to discharge, then you 5A * 2H = 10Ah

I think I got my math right there. I'm sure someone else will chime in and correct it ;)
 
You could use any appropriate load and a "power meter" of some kind, different ones available on eBay and Hobbyking and so on. They basically accumulate the current flown and do the math for your. More convenient and more accurate than doing it manually.

Another way would be an iCharger, but they are much more expensive. The bigger ones have a rather decent discharge capability on their own and can dissipate 200W. But they also offer external discharging (for example with a array of resistors) or regenerative discharge (back into in battery they are running on, obviously not when running on a DC power supply) and in these modes you can put up to 3kW of load on any battery to test it.
 
Just hook up some 12 Volt halogen spots up i prarralel and messure how much current they draw with an amp meter. Old car headligt would also do. The register time and volts as you go.
Let it rund until you reach the low end voltage, and calculate how much capacity there was in the battery.
Not the most presice, but will give a good idear og how much the can deliver.
 
One of these sorts of power meters can measure watt hours accumulated in discharge (or charge if you reverse the current shunt), while you use a large load like a 12V headlight to discharge the battery. You would still need to monitor it for cutoff voltage when to stop the test. Handy thing is to set up a camera to video the whole discharge so you can go back and figure out the watt hours at any part of the SOC curve.

https://www.amazon.com/dp/B01MF8S82C/ref=cm_sw_r_cp_apa_1w3KAb0KRERHM
 
rev0 said:
One of these sorts of power meters can measure watt hours accumulated in discharge (or charge if you reverse the current shunt), while you use a large load like a 12V headlight to discharge the battery. You would still need to monitor it for cutoff voltage when to stop the test. Handy thing is to set up a camera to video the whole discharge so you can go back and figure out the watt hours at any part of the SOC curve.

https://www.amazon.com/dp/B01MF8S82C/ref=cm_sw_r_cp_apa_1w3KAb0KRERHM
Click to expand...

These batteries have their own BMS. They are LiFePO4 chemistry. Designed and built as drop in replacements for lead acid batteries. They have their own over discharge and over charge protection.

However, if the current sensing device was not powered by another source, whenthe BMS shut off, you would loose your data.

It depends how accurate you wish to be. I would personally hook up a ZB2L3 with a 1a load in parallel with a large load of a known current - ie a headlight. The ZB2L3 would act as a 'timer'. So if you combine the loads together and multiply them by the AH that the ZB2L3 reads, you should get a reasonably accurate number.

I depends I use the above method with my RC charger parallel with large loads. Its discharge capability is very limited, so I simply set it at 100ma and use whatever load suits my purpose.
 
Geek said:
rev0 said:
One of these sorts of power meters can measure watt hours accumulated in discharge (or charge if you reverse the current shunt), while you use a large load like a 12V headlight to discharge the battery. You would still need to monitor it for cutoff voltage when to stop the test. Handy thing is to set up a camera to video the whole discharge so you can go back and figure out the watt hours at any part of the SOC curve.

https://www.amazon.com/dp/B01MF8S82C/ref=cm_sw_r_cp_apa_1w3KAb0KRERHM
Click to expand...

These batteries have their own BMS. They are LiFePO4 chemistry. Designed and built as drop in replacements for lead acid batteries. They have their own over discharge and over charge protection.

However, if the current sensing device was not powered by another source, whenthe BMS shut off, you would loose your data.

It depends how accurate you wish to be. I would personally hook up a ZB2L3 with a 1a load in parallel with a large load of a known current - ie a headlight. The ZB2L3 would act as a 'timer'. So if you combine the loads together and multiply them by the AH that the ZB2L3 reads, you should get a reasonably accurate number.

I depends I use the above method with my RC charger parallel with large loads. Its discharge capability is very limited, so I simply set it at 100ma and use whatever load suits my purpose.
Click to expand...

This sounds like a good idea (the others before too). I'm a very visual learner, so it's gonna take me a bit to put it all together, but it makes sense. (got a picture?)

Thanks everyone for your advice and ideas.

Eric
 
I'm assuming the construction of these batteries are 4 cells in a pack so given the discharge curve of LiPo4 batteries I would suggest the cut-off voltage should be much higher to avoid damaging the battery. !2 Volts in a 12Volt 4 cell type system.

As can be seen it all happens in these cells between 3.35 Volts and 3.00 Volts .Going outside of these values would seem pointless to me on discharge . Slightly higher settings to recharge just enables the current to flow.


image_snlpob.jpg
 
Not all LiFePO cells are the same. Especially not after they have been used. Depending on the load the usable capacity might be zero if you set the cutoff to 12V. 10V is much more reasonable if you have matched cells or 2.5V for the lowest cell if you haven't and are using a BMS or some sort of monitoring.
 
I'm a bit sceptical on this post above DarkRaven .
Don't be offended but I think its bad advice to discharge one of these cells to 2.5Volts.
I don't have much experience with LiPo4 but I have not seen one used or otherwise change its voltage as per the curve above to less than 3.2V and 3V is as low as I would take it.2.5V is just too far down.

Do you have any reference material on this that you can cite?
 
tytower its correct to test down to 2.5 or 2.6V. At 3.2V those cells can have 80% of its SOC left... or 30%.

I guess you are mixing LiFe with LiIon? Even LiIon can or should be discharged to 2.8 or below for full range for most cells.

And just for ref: Victron set their low point to save cycles at 11V. Thats 2.75V. https://www.victronenergy.com/uploa...-Volt-lithium-iron-phosphate-batteries-EN.pdf
And to take a high C Life cell the low Voltage is 2V:
http://docs-europe.electrocomponents.com/webdocs/12fd/0900766b812fdd0e.pdf

And here you have Valence that is another HUGE brand and same as I use. They set their low cut off at 20V aka 2.5V.
https://www.valence.com/resources/ds-download/
Note that prebuilt cells always have higher to save on cycles meanwhile proper datasheets show the full performance.

With that said if you want full or atleast 80% capacity out of a life you are between 3-3.35V in general

I have my low cut off set to 2.6V and normal operation voltage low cut of at 2.9V
 
The cells I have posted the charge and discharge curves for are LiPo4 cells. No other cells have i mentioned . In your post
normal operation voltage low cut of at 2.9V
Click to expand...
That says it all doesn't it?
The "charge to" figure on mine is 3.55 max. Never over 3.6V So 3V flat 3.55 V full is how I work mine .

If you look at the graphs it fits in nicely.
I'll let you know how long they last and what voltages are being shown in 20 years or so.


I thought it interesting because I have noticed with Lead Acid cells . You buy a battery ,use it and the minute you let it go flat all the charge seems to come out of it and never goes back fully .
They are on a downhill slope from the day you buy them and one silly mistake which flattens them once really starts the degradation flaming.

I treat my LiPo4 's the same and don't deep discharge below 3V.

I looked at your links. The Valence one was on Lithium Iron Magnesium Phosphate ?
The "High C" was on Lithium Iron Phosphate in 18650 form
And the first was again on Lithium Iron Phosphate batteries .
Did I get it wrong ? I thought the OP was talking about LiPo4?

Ha Ha I see I was wrong . They are Lithium Iron phosphate .....my error sorry
 
Yeah as I said, you can't really NOT discharge LiFePO cells below 3V. If you do you might get nothing out of them because of the voltage drop under load.
2.5V loaded is totally fine for all LiFePO cells, some are even specced a bit lower and state 2.0V as the end of discharge.
 
As Darkraven said you need to distinguish frequent use and what you can do without hurting them for testing.
Most LiFe batteries can go down to 2.5 and as shown down to 2.0V without any damage for testing. If you want 80+% DOD you need to go below 3V on many of them.

Lead acid cannot be compared to ANY lithium based battery :p They should always stay 100%SOC compare to Lithium that will last longer at 40-60% and such ;) LA will degrade ALOT every cycle that goes to 0.

I had to read your small text 2 times before i saw it :p Yeah they are mentioned a bit different all but the common part is that they almost all can take low votlages but for longevity you should of course not go to low.
 
OK. So to bring this back a bit closer to my original post, I have now got a charger that can charge the 100Ah batteries. It has both a specific lithium mode, and several others for normal/AGM lead acid (Noco genius g7200).

From absolutely dead flat, it should charge the battery in less than 24 hours. Didn't happen! Having put the battery on the charger for 2.5 days, the voltage has basically levelled off at 3.45 V and only goes up about .01v/hour. Considering it should charge up to 14.6, clearly something is wrong. Time to do more digging for info.
 
What is the AMP rating on that charger? and is it putting out it's rated Current?
 
You won't get it up to 14.6V for two reasons:
1. The Noco charger puts up to 14.2V across the battery with 7.2A maximum. Charging a 100Ah battery with it should take 15 hours, more or less. The specs of the battery say 14.6V is the maximum charging voltage, which is well within what should be expected, and that means 3.65V end of charge per cell. Anything between 3.6 and 4.0V is normal for a LiFePO battery. The charger can't to that though, 14.2V is 3.55V per cell.
2. LiFePO cells don't always keep their end of charge voltage like other lithium chemistries do. Given the fact that you can't charge it to 100% SoC with this charger I'd say that 3.45V is a decent idle voltage for a LiFePO cell.

So your battery is probably just as charged as it can get. Do you have a clamp meter? You could easily check whether the charging process runs according to the specifications. On the website of the battery manufacturer there is a picture with the battery case opened. If that's easily done then you could also check the single cells inside and check current flow through the BMS while charging to make sure that the BMS is ok.
 
Thanks for the input DarkRaven.

In the manual I got with the charger it says it charges 12v Lithium to 14.6 v at 7.2A. The specs seem to show that 14.6v @ 7.2A should be possible (even at 85% efficiency)


image_pkaugb.jpg

image_laojrb.jpg


I do realize that just because it is in the manual doesn't make it reality. Regardless, your math makes good sense, and other points are quite helpful. I have next to zero experience with LiFePO4, so I'm glad for all information.

Before I crack the sucker open, I'm going to try one or two more things with some standalone power supplies I have (D3806 or BST900) and set the voltage to 14.6 and see what happens.
 
Interesting, I looked it up on their website and it says 14.2V there. They also provide a datasheet and it says 14.2V as well so I just assumed that this is correct. It's a bit on the low side but I thought that might be a specific safety decision or just for longevity of the cells.

Since you have the battery at a decent SOC, or at least that's what we think, you could also attach a load now and see for how long it runs. You don't necessarily need to get it to 14.6V. If the manual is right then you've done it anyway. As I said, you're probably not even going to see or be able to measure a difference as the cells won't keep their end of charge voltage.
 
You must log in or register to reply here.
Back
Top