Hello all 

Any advice would be greatly appreciated. 

I need to buy a new battery for my home setup (to power the WiFi, one TV, one lamp and charge a cell phone during load shedding).

 

Before buying a new battery I would first like to check whether my inverter is compatible with LifePO4 lithium batteries. 

 

I have previously used a 12V 100ampH gel battery, but it doesn’t last very long (with current load shedding it only lasts about 6 / 7 months before the battery dies).  I therefore want to switch to lithium. 

 

I got my inverter in 2019. 

It is a Mecer Lobo 1200/IVR-1200LBKS

720w

Input: 230Vac 50/60Hz

Output: 230Vac 50/60Hz

10AMax 

Charge selector can be set to either 10A or 20A. 

 

Is this inverter compatible with lithium batteries? On the inverter, I am only able to change the charge setting (10A or 20A). I don’t think the inverter allows me change any other settings. 
 

I contacted Geewiz and they suggested I purchase this battery: 

 

https://www.geewiz.co.za/lithium-ion-batteries/209357-geewiz-12100l-12kwh-12v-100ah-lithium-ion-lifepo4-5000-cycle-battery-first-life-cells-3-year-unlimited-cycles-warranty-5000-cycles.html

 

Anyone know whether this battery (Geewiz) is suitable given my inverter type and usage needs? Or any other suggestions / recommendations?
 

Thank you in advance,

 

D 

13 hours ago, Dyl said:

Hello all 

Any advice would be greatly appreciated. 

I need to buy a new battery for my home setup (to power the WiFi, one TV, one lamp and charge a cell phone during load shedding).

 

Before buying a new battery I would first like to check whether my inverter is compatible with LifePO4 lithium batteries. 

 

I have previously used a 12V 100ampH gel battery, but it doesn’t last very long (with current load shedding it only lasts about 6 / 7 months before the battery dies).  I therefore want to switch to lithium. 

 

I got my inverter in 2019. 

It is a Mecer Lobo 1200/IVR-1200LBKS

720w

Input: 230Vac 50/60Hz

Output: 230Vac 50/60Hz

10AMax 

Charge selector can be set to either 10A or 20A. 

 

Is this inverter compatible with lithium batteries? On the inverter, I am only able to change the charge setting (10A or 20A). I don’t think the inverter allows me change any other settings. 
 

I contacted Geewiz and they suggested I purchase this battery: 

 

https://www.geewiz.co.za/lithium-ion-batteries/209357-geewiz-12100l-12kwh-12v-100ah-lithium-ion-lifepo4-5000-cycle-battery-first-life-cells-3-year-unlimited-cycles-warranty-5000-cycles.html

 

Anyone know whether this battery (Geewiz) is suitable given my inverter type and usage needs? Or any other suggestions / recommendations?
 

Thank you in advance,

 

D 

Geewiz normally know which units are compatible. Adhere to their advice to only charge at the 10A setting. 

 

On my Axpert inverter I've been using it in SUB mode during the day and switching it to USB at night by using software called "MultiSIBcontrol".
It allows me to set a timer to switch between different modes which is very convenient since on it's own the inverter stays on SUB mode until there is literally no sunlight before switching to USB mode automatically.

@Coulomb recommended me to lower the bulk voltage from 28.4V (as recommended by the battery manufacturer) to 27.6V and like he said I would, I haven't noticed any capacity loss.
Since my inverter is "dumb" I can't disable "float charge" so I decided to lower it to 27.4V. Everything seems to work perfectly fine and it's good to know that the lowered voltage will insure my battery doesn't degrade prematurely.

I am curious though, if I lower the "bulk" and "float" voltage even more would that be bad for the BMS in the battery?
The reason I'm asking is that I've noticed at night when the battery isn't charging it always settles on 26.6V and there it will stay. What if I lower my charge voltages to 26.6V. Is it safe? Will I mess up the BMS feeding it a too low voltage than what it recommends?

Edited February 22, 20242 yr by Gerrit84

3 hours ago, Gerrit84 said:

I am curious though, if I lower the "bulk" and "float" voltage even more would that be bad for the BMS in the battery?

Two problems:

1) The battery won't be fully charged, so you'll get reduced run time.

2) The battery's BMS won't see the battery voltage as full, so it doesn't regularly reset the state of charge counter to 100%. In time, the estimated state of charge will drift from its true value, and you won't be able to trust the reported state of charge.

Neither of these is particularly bad for the battery or the BMS, but they are really annoying.

Lowering the charge voltages to 26.6 V will severely undercharge the battery. 26.6 V is an average of 3.325 V per cell; LFP has a rested "plateau" at about 3.330 VDC; 3.325 under load is perfectly normal for that plateau. The plateau means that the battery voltage will hardly change with SoC, from some 95% down to about 75%, where it transitions to another slightly lower plateau, all the way down to about 30% SoC. That's just the characteristics of the LFP chemistry and cell dynamics.

All batteries (not just LFP) need a higher charge voltage than where they are resting to be able to overcome internal barriers, and "push" power into the battery. The difference between "voltage in" and "voltage out" causes a less than 100% energy efficiency. LFP batteries have one of the highest energy efficiencies there is, and their current efficiency (total current in versus total current out) is nearly 100%. That's why the BMS is "counting coulombs (amp·seconds) to update its SoC estimate, and why it generally gets it pretty right, if it resets to 100% at the end of every charge, and has a good estimate of the battery's State of Health (its real-life Ah capacity as a percentage of its nominal, near-new capacity).

21 hours ago, Coulomb said:

Two problems:

1) The battery won't be fully charged, so you'll get reduced run time.

2) The battery's BMS won't see the battery voltage as full, so it doesn't regularly reset the state of charge counter to 100%. In time, the estimated state of charge will drift from its true value, and you won't be able to trust the reported state of charge.

Neither of these is particularly bad for the battery or the BMS, but they are really annoying.

Lowering the charge voltages to 26.6 V will severely undercharge the battery. 26.6 V is an average of 3.325 V per cell; LFP has a rested "plateau" at about 3.330 VDC; 3.325 under load is perfectly normal for that plateau. The plateau means that the battery voltage will hardly change with SoC, from some 95% down to about 75%, where it transitions to another slightly lower plateau, all the way down to about 30% SoC. That's just the characteristics of the LFP chemistry and cell dynamics.

All batteries (not just LFP) need a higher charge voltage than where they are resting to be able to overcome internal barriers, and "push" power into the battery. The difference between "voltage in" and "voltage out" causes a less than 100% energy efficiency. LFP batteries have one of the highest energy efficiencies there is, and their current efficiency (total current in versus total current out) is nearly 100%. That's why the BMS is "counting coulombs (amp·seconds) to update its SoC estimate, and why it generally gets it pretty right, if it resets to 100% at the end of every charge, and has a good estimate of the battery's State of Health (its real-life Ah capacity as a percentage of its nominal, near-new capacity).

My idea was mainly to keep the battery at about 70-80% SOC when we don't experience any load shedding, instead of keeping it full all the time.
I understand from what you explained that it is basically impossible to accurately determine SOC just form voltage alone since the voltage curve is so flat.
Unfortunately my setup isn't able to accurately determine what the actual SOC is, therefore I will stick to the charge voltage that you recommended.

Thank you (again) for explaining everything so clearly!

17 hours ago, Gerrit84 said:

My idea was mainly to keep the battery at about 70-80% SOC when we don't experience any load shedding, instead of keeping it full all the time.
I understand from what you explained that it is basically impossible to accurately determine SOC just form voltage alone since the voltage curve is so flat.
Unfortunately my setup isn't able to accurately determine what the actual SOC is, therefore I will stick to the charge voltage that you recommended.

Thank you (again) for explaining everything so clearly!

I run my Axpert with lithium in the 26.2 to 26.6V region under no load. As I have a low power requirement run time is thus not a problem. 

I use 3 different brands of lithium in my 3 banks of which 1 is fitted with a BMS that shows each cell voltage and an active balancer. 

Below is a picture of my normal SOC after charging. Balancer is set to start at 8mV delta and starting at 3.3V per cell. Current charge is 7.1A which is what this 1 of 3 bank is receiving. Currently having the washing machine running and powered from PV while having LS. 

When under no load or charging volts would drop to about 26.4V.

Edited February 24, 20242 yr by Scorp007
Added BMS graphic

Thanks for the feedback.
Seems that you have your setup tuned perfectly. When I get to upgrade my system I will definilty invest in getting a setup that allows me to monitor everyhting more accurately.
For now it's really a hit and miss at what SOC my battery is and because of my modest setup I can't afford having the battery too low. I will HATE myself if we have 4 hours of LS and the battery dies.