Renogy lithium capacity

[dawsey]

I noticed an issue with our Renogy lithium batteries the other day, it’s no biggie and there’s nothing as such wrong with the batteries, but I though I’d post it up here in order that other folks can avoid getting caught out in the same way.

The batteries are Renogy Li Fe Po 100ah x 2 with bluetooth BMS. We use the Renogy app on the phone to check capacity. The batteries are charged by engine and solar via a 25/50A Renogy MPPT/B2B charge controller. The batteries are 16 months old and are discharged by about 30% every day then recharged as and when.

4 weeks ago when microwaving something one morning I noticed low voltage alarms. I checked capacity via the app - average 30%. But voltage had dropped below 12V. Strange I thought, I have gone down to 20% a time or two before with no issue.

2 weeks ago, again microwaving in the morning, low voltage alarms then a few seconds later total shutdown. The invertor shut down and even the good old elektroblok clicked off. I checked capacity and voltage - using the app - 40% capacity but under 11V.

Bugger, I thought, serious battery degredation. How is it possible with a relatively shallow charge discharge profile and only about 400days of continuous use? It’s about what you might expect from Lead Acids.

The lithium batteries were both genuinely fully discharged - that was self evident.

Turns out nothing was wrong with the actual batteries. It was the software the BMS used to estimate the remaining charge. It turns out that the remaining capacity indication is just an ‘estimation’ based on the measured discharge current, and it ‘drifts’ over time when the battery is repeatedly discharged without ever being fully recharged.

Over several weeks we had been re charging the batteries to 70 or 80% maximum then discharging to 40 or 50%. This had caused huge drift between what the battery remaining capacity actually was, and what was being calculated by the BMS and displayed. An indicated 40% had become a real 10%.

The advice is always to fully recharge any battery first… So that’s what I did. I ran the engine. Interestingly the capacity uptick got to 99.1% on both batteries and then stayed there for almost 2 hours despite 35A of charge being shared by both batteries.

After roughly 2 hours the charge current started to drop (as you might expect when you really get to 99% charge) and once it had dropped from 35A to 3A over about 20minutes both batteries went to 100% and it seems I had ‘reset’ the capacity calculator. I tested by discharging both batteries to 20% (just used the air fryer to make chips twice) and all was good, minimal loss of voltage.

There we go. A handy hint if you want accurate capacity indication on these batteries, periodically recharge fully to an indicated 100%

 

[peterc10]

Are you using some form of shunt to measure all the ins and outs of current? That is the only accurate way to calculate battery capacity.

We have 2 x 120Ah KS batteries. We also have a Victron BMV700 which uses a shunt to calculate more accurate figures than the KS app can. It also has the ability to synchronise the batteries and BMV to 100% SOC. If we want to know what the state of the batteries are we look at the BMV figures not those on the KS app.

 

[poppycamper]

The votronic shunt instructions say to calibrate shunt you need to charge battery for 24hrs..
Charged my 2x120 KS Energy battery's yesterday ready for the of tomorrow..
They were at 50% on Votronic display, Votronic display eventually read 100% but charger kept putting in 40a for another hour or so before it eventually decrease to minimul..

 

[bigtwin]

ran the engine. Interestingly the capacity uptick got to 99.1% on both batteries and then stayed there for almost 2 hours despite 35A of charge being shared by both batteries.

A good illustration that the %SoC isn’t a reliable indicator of true SoC.

There we go. A handy hint if you want accurate capacity indication on these batteries, periodically recharge fully to an indicated 100%

Yes, they require a full charge to properly reset the SoC indication.

Are you using some form of shunt to measure all the ins and outs of current? That is the only accurate way to calculate battery capacity.

A shunt will measure the current in/out but unless it’s charge algorithm is set to match the battery (i.e. the charge efficiency factor is set correctly) the shunt will not be accurate. As Raul has stated previously, he has adjusted both the Peukert exponent and the charge efficiency factor from the default LiFePO4 settings on his setup to get a reliable indication of charge status (perhaps he might remind us of the settings but IIRC he changed the Peukert setting from 1.05 to 1.02 and the CEF from something like 95% to 98%).

If we want to know what the state of the batteries are we look at the BMV figures not those on the KS app.

As above. The shunt is not necessarily more accurate than the KS App (with the proviso that the battery is properly charged on a frequent basis).

Ian

 

[Marie-Claire]

I have the same batteries and they show a difference of around 50% on the app, I have reinstalled the app charged each battery seperatly for 24 hrs and still have the difference although the voltage for each battery is the same, I put ehu to the van every 2 weeks for 24 hrs to top up the starter battery as solar isnt keeping up with the drain, will see what spring brings when the solar takes over again.

^{_{Subscribers  do not see these advertisements}}

 

[peterc10]

A good illustration that the %SoC isn’t a reliable indicator of true SoC.


Yes, they require a full charge to properly reset the SoC indication.


A shunt will measure the current in/out but unless it’s charge algorithm is set to match the battery (i.e. the charge efficiency factor is set correctly) the shunt will not be accurate. As Raul has stated previously, he has adjusted both the Peukert exponent and the charge efficiency factor from the default LiFePO4 settings on his setup to get a reliable indication of charge status (perhaps he might remind us of the settings but IIRC he changed the Peukert setting from 1.05 to 1.02 and the CEF from something like 95% to 98%).


As above. The shunt is not necessarily more accurate than the KS App (with the proviso that the battery is properly charged on a frequent basis).

Ian

When I fitted the Lifepo batteries I adjusted both of those settings on the BMV to the figures recommended by Victron for LiFePO4 batteries. From memory the Peukert setting was set at 1.01 and the CE to 99%.

 

[Marie-Claire]

The soc values were identical for about 9 months , a previous post suggested they will re align when charged and used more often

 

[tinkertaylor]

When I first got my lithium battery it was charged upto to 14.6v with a constant voltage set at 14.6 until the current draw was 0A. I used a regulated power supply, you can use a victron dc to dc or solar controller to do this, this is a vital step, 0 current drain at 14.6 volts.

I set the victron shunt at 100% state of charge.

I then set a test rig that had a current draw of 100 amps ( inverter with an oil filled radiator) which I ran for 2 hours, 200 amp hours.

While doing this I monitored the voltage of the battery to ensure it never went under 10 volts, the minimum my battery should operate at, it never, got to about 10.2 volts. The state of charge on the shunt was 0%, I knew the battery could deliver its stated capacity and the SOC was correct.

The main thing with lithium is max and min voltages, 14.6v max and 10v min on a 12v 4 cell battery. Each cell has a max of 3.65v and min of 2.5v but the bms should ensure these values aren't encroached. It's useful to be able to see the cell voltage values as balance between these cells, or lack of it, can alter the total battery capacity once the cell values start to drift apart. This is what causes the failure of lithium batteries and if the bms isn't a decent one can cause over charging of the individual cells and safety issues. A good bms is vital.

 

[dawsey]

I have the same batteries and they show a difference of around 50% on the app, I have reinstalled the app charged each battery seperatly for 24 hrs and still have the difference although the voltage for each battery is the same, I put ehu to the van every 2 weeks for 24 hrs to top up the starter battery as solar isnt keeping up with the drain, will see what spring brings when the solar takes over again.

Are they installed correctly ? Main positive and negative leads to different batteries and link wires ? If not they will charge/discharge unevenly.

 

[Raul]

When I first got my lithium battery it was charged upto to 14.6v with a constant voltage set at 14.6 until the current draw was 0A. I used a regulated power supply, you can use a victron dc to dc or solar controller to do this, this is a vital step, 0 current drain at 14.6 volts.

I set the victron shunt at 100% state of charge.

I then set a test rig that had a current draw of 100 amps ( inverter with an oil filled radiator) which I ran for 2 hours, 200 amp hours.

While doing this I monitored the voltage of the battery to ensure it never went under 10 volts, the minimum my battery should operate at, it never, got to about 10.2 volts. The state of charge on the shunt was 0%, I knew the battery could deliver its stated capacity and the SOC was correct.

The main thing with lithium is max and min voltages, 14.6v max and 10v min on a 12v 4 cell battery. Each cell has a max of 3.65v and min of 2.5v but the bms should ensure these values aren't encroached. It's useful to be able to see the cell voltage values as balance between these cells, or lack of it, can alter the total battery capacity once the cell values start to drift apart. This is what causes the failure of lithium batteries and if the bms isn't a decent one can cause over charging of the individual cells and safety issues. A good bms is vital.

The 14.6v is only valid if the cells are balanced within 0mv. Shunt in its spec sheet,has enough large voltage calibration tolerance, to take you outside 14.6v spec. The only reliable way of doing this, is with calibrated equipment, and balanced to a delta 0mv. A safer way is in parallel cells.But, I ask myself why you have this need of 14.6v? What do you benefit of it? Why you keep the cell at 3.65v until current tappers to 0? Cell degradation happens fast at elevated voltage, hence charger off once hits 3.65v.
Normal charging is, hit it with 0.2C and charger off when current tappers off to 0.05C. So when you charge with less than 0.2C you can absorb a little, same with CV lower than 3.65v, it gives you a chance for absorb. This absorb its beneficial for a working pack. A daily cycled pack it will drift, so the absorb time it’s spent for balancing. If you charge at the parameters you described, you will not leave any room for absorb.
Even with new battery, the cells have been charged to 3.65v under compression, before they ended up in a pack, so you don’t need to do that again. That’s a one time event to form the SEI layer, and gas out any air trapped, but under compression. Any subsequent charging, does not need to reach 3.65v to get full, just absorb a little, at a lower voltage.

^{_{Subscribers  do not see these advertisements}}

 

[Raul]

I noticed an issue with our Renogy lithium batteries the other day, it’s no biggie and there’s nothing as such wrong with the batteries, but I though I’d post it up here in order that other folks can avoid getting caught out in the same way.

The batteries are Renogy Li Fe Po 100ah x 2 with bluetooth BMS. We use the Renogy app on the phone to check capacity. The batteries are charged by engine and solar via a 25/50A Renogy MPPT/B2B charge controller. The batteries are 16 months old and are discharged by about 30% every day then recharged as and when.

4 weeks ago when microwaving something one morning I noticed low voltage alarms. I checked capacity via the app - average 30%. But voltage had dropped below 12V. Strange I thought, I have gone down to 20% a time or two before with no issue.

2 weeks ago, again microwaving in the morning, low voltage alarms then a few seconds later total shutdown. The invertor shut down and even the good old elektroblok clicked off. I checked capacity and voltage - using the app - 40% capacity but under 11V.

Bugger, I thought, serious battery degredation. How is it possible with a relatively shallow charge discharge profile and only about 400days of continuous use? It’s about what you might expect from Lead Acids.

The lithium batteries were both genuinely fully discharged - that was self evident.

Turns out nothing was wrong with the actual batteries. It was the software the BMS used to estimate the remaining charge. It turns out that the remaining capacity indication is just an ‘estimation’ based on the measured discharge current, and it ‘drifts’ over time when the battery is repeatedly discharged without ever being fully recharged.

Over several weeks we had been re charging the batteries to 70 or 80% maximum then discharging to 40 or 50%. This had caused huge drift between what the battery remaining capacity actually was, and what was being calculated by the BMS and displayed. An indicated 40% had become a real 10%.

The advice is always to fully recharge any battery first… So that’s what I did. I ran the engine. Interestingly the capacity uptick got to 99.1% on both batteries and then stayed there for almost 2 hours despite 35A of charge being shared by both batteries.

After roughly 2 hours the charge current started to drop (as you might expect when you really get to 99% charge) and once it had dropped from 35A to 3A over about 20minutes both batteries went to 100% and it seems I had ‘reset’ the capacity calculator. I tested by discharging both batteries to 20% (just used the air fryer to make chips twice) and all was good, minimal loss of voltage.

There we go. A handy hint if you want accurate capacity indication on these batteries, periodically recharge fully to an indicated 100%

You just discovered the limitations of a bms hardware, nothing to do with the firmware, no mater how many times you update it. The bms accuracy in current detection it’s not so fine, so small draws goes undetected, leaving you with a much discharged battery than claimed. Renogy bms, under 2A are really bad for keeping track of what’s going on.
Only 2-3 bms’s in the world are using a shunt and measure all the current, they cost more than 2 renogy batteries, and are manly used in large storage.

 

[tinkertaylor]

The 14.6v is only valid if the cells are balanced within 0mv. Shunt in its spec sheet,has enough large voltage calibration tolerance, to take you outside 14.6v spec. The only reliable way of doing this, is with calibrated equipment, and balanced to a delta 0mv. A safer way is in parallel cells.But, I ask myself why you have this need of 14.6v? What do you benefit of it? Why you keep the cell at 3.65v until current tappers to 0? Cell degradation happens fast at elevated voltage, hence charger off once hits 3.65v.
Normal charging is, hit it with 0.2C and charger off when current tappers off to 0.05C. So when you charge with less than 0.2C you can absorb a little, same with CV lower than 3.65v, it gives you a chance for absorb. This absorb its beneficial for a working pack. A daily cycled pack it will drift, so the absorb time it’s spent for balancing. If you charge at the parameters you described, you will not leave any room for absorb.
Even with new battery, the cells have been charged to 3.65v under compression, before they ended up in a pack, so you don’t need to do that again. That’s a one time event to form the SEI layer, and gas out any air trapped, but under compression. Any subsequent charging, does not need to reach 3.65v to get full, just absorb a little, at a lower voltage.

At 3.65v the cells still aren't fully charged until they pull 0A , this takes about 30 minutes after they hit 3.65v, this is all done on full calibrated lab equipment.

My batteries are lithium, they don't emit any gas unless things are about to go tits up.