5 years of CALB LiFePO4 battery operation

Getting into the nuts and bolts of wind and solar power components, installation and performance issues, ask technical questions, answers and tips

5 years of CALB LiFePO4 battery operation

Postby Gordon-Loomberah » Thu Jan 18, 2018 6:01 pm

My nominal 21kWh capacity CALB battery has now been in use for over 5 years (since Dec 2012), and it gets a pretty hard workout in high temperatures over summer, but is still performing quite well.

One cell was taken out of service a couple of years ago as it had reduced capacity relative to the others, and was replaced with a 400AH Wintson, which was an inconveniently slightly different size. Initially configured as 800AH @26V, it was changed to 400AH @52V a few years ago.

Typical daily cycling in cooler weather is to around 50% DOD, but frequently in summer and sometimes in cloudy weather at other times of the year, it starts the day down around 20%SOC (80%DOD). Temperatures in the power shed over summer often peak in the range of 40-48C.

They appear to have perhaps 90% of usable capacity, based on recharging on Sunday from a 17.5kWh (83%) discharge, graphs to follow.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5762
Joined: Thu Sep 24, 2009 9:41 pm
Location: Loomberah NSW Australia

Re: 5 years of CALB LiFePO4 battery operation

Postby Gordon-Loomberah » Fri Jan 19, 2018 2:48 pm

The cell balancers vary in performance, here is a comparison of the 2 halves of the battery voltage for 14/1/2018, from as low as 17% SOC to fully charged.
20180114half-voltages.gif
voltages of 2 halves (8 cells) of the battery.
Eastern Australian Standard Time
CLICK ON IMAGE TO ENLARGE


Sometimes the half voltages diverge significantly near the top, and at others they are quite close, varying on time scales of weeks to months. I've given up worrying about it long ago.

20180114netkW.gif
Gross and Net charging and discharging
Eastern Australian Standard Time
CLICK ON IMAGE TO ENLARGE


There was a bit of cloud around, so solar charging is up and down a lot on short time scales for some of the day.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5762
Joined: Thu Sep 24, 2009 9:41 pm
Location: Loomberah NSW Australia

Re: 5 years of CALB LiFePO4 battery operation

Postby australsolarier » Sun Jan 21, 2018 3:21 pm

thanks Gordon,
for all your pioneering work and inspiration. it has certainly made the work for us followers that much easier. and also proved the the lithium sceptics wrong. and renewable energy sceptics as well.

thanks again Gordon, all the best
urs
australsolarier
Solar Crusader
Solar Crusader
 
Posts: 518
Joined: Sun Mar 28, 2010 9:27 am

Re: 5 years of CALB LiFePO4 battery operation

Postby Gordon-Loomberah » Sun Jan 21, 2018 4:52 pm

Thanks Australsolarier :)

My (actually mostly my rainbow trouts') energy demands in this current heatwave are certainly working the system hard- 50kWh used yesterday, and down to 14%SOC this morning.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5762
Joined: Thu Sep 24, 2009 9:41 pm
Location: Loomberah NSW Australia

Re: 5 years of CALB LiFePO4 battery operation

Postby jaahn » Mon Jan 22, 2018 7:45 am

Thanks Gordon. :)
You'r an inspiration to us all who might want to follow. And graphs etc too in great detail. :geek:

Would you care to generally outline your loads in this time of high useage. What do those 'freeloading' trouts do to use all the power. Martinis on ice all day or ?? :shock:

Do you think the high temperatures in the battery shed cause some loss of battery life at all.
Jaahn
jaahn
Solar Crusader
Solar Crusader
 
Posts: 220
Joined: Sun Apr 25, 2010 8:42 am

Re: 5 years of CALB LiFePO4 battery operation

Postby Gordon-Loomberah » Mon Jan 22, 2018 10:55 am

Thanks Jaahn :)
Yes, I'm sure the high temperature isn't ideal, and have plans to shade the entire shed, which will help a bit, but when the outside air is near 40C and I have a couple of charge controllers outputting ~5kW and an inverter running 3-4kW loads, there is a fair bit of heat being produced inside the shed, pushing the temps up.

A couple of graphs from yesterday, when ~50kWh was used again. I was a bit over 4kWh short of fully charged at about 16:00, so ran the generator and battery charger (1.3kW) for a few hours in the evening. Very smoky air, mostly from a large bushfire 40km SE of here and high temps (38.4C yesterday) meant 7kW of PV wasn't producing as much as I'd like.

On the gross graph the 10 min ~800W loads every hour are a water pump, and there's another smaller pump coming on nearly every hour too, for shorter periods in the day and longer at night. There are 3 ~40W air pumps running continuously, a battery charger than comes on every 2 hours at a quarter past the hour overnight (PV panel supplies daytime power), to reduce DOD on a couple of 150AH Pb-acid batteries that run a small inverter powering one of the air pumps, as a back-up. There's another circulation pump running continuously on a water filter too.
In the house there are 2 chest freezers and a fridge, plus various other small loads, lights, TV computers etc, and washing machine from 11-12:00.
The big ~2.5kW load that is on much of the day is the water chiller, keeping the water at a suitable temp for the trout. I lost quite a few of them over a few days recently after the water got to near 20C overnight, trout prefer <18C, so have been running the chiller longer hours lately plus some overnight time to minimise temp rise (which continues through the night in really hot weather). That has meant a bit of generator use is needed. I've located some 2nd hand panels that I will add into the system soon, to reduce genny usage under these conditions.
In hot weather I usually start the chiller an hour or so before sunrise, to take advantage of the coolest air, which improves the EER/COP of the chiller. I run a water mister near the air intake when it is hot, which also improves performance.

20180121gross.gif


20180121net-SOC.gif
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5762
Joined: Thu Sep 24, 2009 9:41 pm
Location: Loomberah NSW Australia

Re: 5 years of CALB LiFePO4 battery operation

Postby tom rickard » Mon Jan 22, 2018 5:12 pm

Well done Gordon, and thanks again for all your postings over the years. You have helped a lot of people.
tom rickard
Solar Crusader
Solar Crusader
 
Posts: 111
Joined: Tue Nov 26, 2013 2:32 pm

Re: 5 years of CALB LiFePO4 battery operation

Postby Gordon-Loomberah » Mon Jan 22, 2018 6:36 pm

Thanks Tom :)
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5762
Joined: Thu Sep 24, 2009 9:41 pm
Location: Loomberah NSW Australia

Re: 5 years of CALB LiFePO4 battery operation

Postby karrak » Wed Jan 24, 2018 11:33 pm

Thanks Gordon for posting the update and thanks for all your experimenting and being prepared to try new things. It was your posts on charging at 3.45V/cell and floating at 3.35V/cell that convinced me to go the same way even though my understanding of the accepted wisdom was not to charge LFP batteries to nearly 100% and float them at such a high SOC.

My LFP battery is nearly five years old now. The lowest SOC my battery got down to last year was 8% measured using a current shunt. From the cell voltages at that time I calculated the SOC to be ~6%-7% of the rated capacity so I haven't lost very much capacity compared to the name plate capacity. My battery has had a far gentler life, average daily cycling is 21% of battery capacity and maximum charge is ~0.15C and discharge current is ~0.6C. The voltage drop under load hasn't changed over the five years. I calculated the impedance of my battery including the interconnects as 5mOhms. I was wondering if your battery impedance had changed much over the five years?

I see you still aren't monitoring the individual cell voltages. There are now available some reasonably priced 16S BMSs from China like this https://www.aliexpress.com/item/16S-60V ... 97854.html where you can access the individual cell voltages and battery current if you wanted to. I have made an interface between my battery monitoring software and these units. You can pay more money for a unit with a higher charge/discharge limit or bypass the battery disconnect circuitry if you wanted to just monitor the voltages.

Another thought is that if you are using reasonably high power via the inverter for long periods of time that you will be subjecting the battery to high ripple currents which will decrease its lifespan. Someone on another forum has made a ripple filter. This thread gives some more background information.

Simon
Off grid 24V system, 6x190W Solar Panels, 32x90ah Winston LiFeYPO4 batteries installed April 2013
BMS - Homemade Battery logger https://github.com/simat/BatteryMonitor
Latronics 4kW Inverter, homemade MPPT controller
karrak
Solar Evangelist
Solar Evangelist
 
Posts: 84
Joined: Wed Jan 21, 2015 1:14 am
Location: Cookernup (near Harvey) SW Western Australia

Re: 5 years of CALB LiFePO4 battery operation

Postby australsolarier » Thu Jan 25, 2018 5:35 am

karrak,
the link you added is totally unsuited to lifepo4 batteries. i mean totally useless.
for one thing it has a 50A limit. and the balance function starts way over the lifepo4's.

that bms is for bikes, for a 60V lithium ion battery. they have different voltages.

but even the biggest bms for lifepo4 batteries they have on their website, is totally unsuited. the shunt current is tiny. 50mA.

if i remember correctly Gordon has the ev-power bms installed.

there is a thread somewhere on this forum that deals with bms's
australsolarier
Solar Crusader
Solar Crusader
 
Posts: 518
Joined: Sun Mar 28, 2010 9:27 am

Next

Return to Components, Installation And Performance

Who is online

Users browsing this forum: Bing [Bot] and 3 guests

new solar power specials