Lithium battery SOC drop with peak solar?

A place where users can meetup with other FlexPower One users and can find support for one another. A great place for people who have not purchased the FlexPower One yet to look up information.
Post Reply
Dan-norcal
Forum Junior Member
Posts: 3
Joined: Mon Jun 17, 2024 11:52 am
My RE system: FP1: VFXR3648A-01, MATE3s, HUB10.3, FM80, FLEXnet DC, Battery: EG4-LL 48V 100AH Solar: 3.2kw, 9 x REC365NP2

Lithium battery SOC drop with peak solar?

Post by Dan-norcal »

After running this system for a few weeks, we have questions on the SOC of the EG4 battery and whether we should be concerned. The state of charge drops throughout the day from 100% to around 80% while the battery voltage stays at around 53.6. Towards the end of the day, the state of charge suddenly shoots up to 100%.

Is this what we should expect to see?

(Currently have the Absorb set for 54.2 V (0.5 hours) and Float 54.2 (11 hours).)
optics-re-screenshot.png
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

Dan-norcal wrote: Mon Jun 17, 2024 12:24 pm After running this system for a few weeks, we have questions on the SOC of the EG4 battery and whether we should be concerned. The state of charge drops throughout the day from 100% to around 80% while the battery voltage stays at around 53.6. Towards the end of the day, the state of charge suddenly shoots up to 100%.

Is this what we should expect to see?
No, that sounds atypical.
Flat voltage discharge curve is likely normal.
100% SoC -> 80% SoC -> 100% SoC might an artifact of the FNDC Battery Charging parameter settings.
Alternatively, it could simply be the 54.2V Float Voltage is acting like a "long, slow Absorb" cycle.
(Currently have the Absorb set for 54.2 V (0.5 hours) and Float 54.2 (11 hours).)

optics-re-screenshot.png
The technical specs from the EG4 LL manual suggest a charge (absorb) voltage of 56.2V (at 0.1H) and Float of 54V.
Last edited by raysun on Mon Jun 17, 2024 1:34 pm, edited 3 times in total.
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

Just a note about "Solar" and "To Battery".
I asked one of the Outback engineers about the disparity between the two, and the response was "To Battery is more accurate."
chrisski
Forum Guru
Posts: 103
Joined: Fri May 17, 2024 7:23 pm
My RE system: 8048 Radian w/ Load Center
Mate 3
Flex max 100 afci X 2
Fortress eFlex 5.1 kWh Battery X 4
TP1000 for Internet Connectivity
Location: Central AZ, lowland desert

Re: Lithium battery SOC drop with peak solar?

Post by chrisski »

Dan-norcal wrote: Mon Jun 17, 2024 12:24 pm After running this system for a few weeks, we have questions on the SOC of the EG4 battery and whether we should be concerned. The state of charge drops throughout the day from 100% to around 80% while the battery voltage stays at around 53.6. Towards the end of the day, the state of charge suddenly shoots up to 100%.

Is this what we should expect to see?

(Currently have the Absorb set for 54.2 V (0.5 hours) and Float 54.2 (11 hours).)

optics-re-screenshot.png
What do your logs say?

I noticed the same issue starting last month in May. I have an Outback Radian. Really need an alternate method to measure SOC for the battery whether its an EG4 app or even just voltage. Do you have a way to monitor the EG4 batteries to see the true state of charge?

The Batteries of course are not reading correctly from the SOC% in the Mate3S display or Optics.

I have 20 kWh of Fortress Power EFlex batteries, 5.1 kWh each. I have noticed like you say, at some point during the early night or just as the sun is rising, the battery SOC will rest to zero.

I have looked in my logs and it seems to me that the one time I was present at the OUtback when it happened, sunrise this weekend at 5:30 am, my MPPTs were cycing on and off as they were just at wakeup amperage. I think this cycling on and off several times may do something to the battery SOC internal in the Outback that resets the SOC to 100%.

Please check your logs to see if you have something similar. My cutoff amperage is .5 amps, so perhaps that is too low and it should be a little higher.

AFAIK, the outback does not use SOC% as trips for different modes, but uses voltage, so that this inaccurate SOC won't prevent us from coming out of different off grid modes. I consider using voltage a good thing for the situation we're in.

=====================

I'm not sure how much the rest of this applies.

This inaccurate reading from the Outback Radian got me into a situation where my batteries died. The SOC on the Outback said 100% which I monitored for several days not knowing its inaccurate, but the batteries died from undercharge. Each of my four batteries has 4 LEDs for a SOC indicator that I had not been monitoring. So when the inverter shut off, one battery was completely depleted with a red SOC trip and three others were at 25% with one bar.

To actually charge my batteries closer to 100%, what I do is I am Grid TIed and use "Drop" for a couple hours a day when my batteries are 1 or 2 LEDs, 25%/50%, to push the 5 kWh - 10 kWh into the battery to get the true SOC closer to 4 bars, 100%. This "Drop" method keeps me at a good SOC through the day that I don't worry about batteries duiong at night. If I did have a power outage, this also leaves me enough power to run for a few hours.
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

While SoC via the FNDC must be taken with a grain of (lithium) salt, I have found it quite predictable when synched to true 100% battery charge. The only way to achieve that is to charge the battery to the "full charge" voltage and amperage. The latter is typically 2% of the C20 Amp Hour capacity for LFP.
chrisski
Forum Guru
Posts: 103
Joined: Fri May 17, 2024 7:23 pm
My RE system: 8048 Radian w/ Load Center
Mate 3
Flex max 100 afci X 2
Fortress eFlex 5.1 kWh Battery X 4
TP1000 for Internet Connectivity
Location: Central AZ, lowland desert

Re: Lithium battery SOC drop with peak solar?

Post by chrisski »

raysun wrote: Mon Jun 17, 2024 1:54 pm While SoC via the FNDC must be taken with a grain of (lithium) salt, I have found it quite predictable when synched to true 100% battery charge. The only way to achieve that is to charge the battery to the "full charge" voltage and amperage. The latter is typically 2% of the C20 Amp Hour capacity for LFP.
In grid tied mode, I am having trouble getting the full charge. Either the Battery sell voltage gets set too high and the battery gets charged but I lose about 20% - 30% sellback, or the battery sell voltage is set too low and I sell back a lot, but don't fully charge automatically.

For now, the installer is tweaking those settings.
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

chrisski wrote: Mon Jun 17, 2024 2:21 pm
raysun wrote: Mon Jun 17, 2024 1:54 pm While SoC via the FNDC must be taken with a grain of (lithium) salt, I have found it quite predictable when synched to true 100% battery charge. The only way to achieve that is to charge the battery to the "full charge" voltage and amperage. The latter is typically 2% of the C20 Amp Hour capacity for LFP.
In grid tied mode, I am having trouble getting the full charge. Either the Battery sell voltage gets set too high and the battery gets charged but I lose about 20% - 30% sellback, or the battery sell voltage is set too low and I sell back a lot, but don't fully charge automatically.

For now, the installer is tweaking those settings.
My simple mind is greatful for off-grid living. 😉
Dan-norcal
Forum Junior Member
Posts: 3
Joined: Mon Jun 17, 2024 11:52 am
My RE system: FP1: VFXR3648A-01, MATE3s, HUB10.3, FM80, FLEXnet DC, Battery: EG4-LL 48V 100AH Solar: 3.2kw, 9 x REC365NP2

Re: Lithium battery SOC drop with peak solar?

Post by Dan-norcal »

Having had a few Lithium batteries in an electric car as well as a few tablets and smart phones it has been my reasoning that it is not a good idea to fully charge or fully discharge Lithium technology and I have read a few opinions that agree with this assessment. For this reason, I don't charge my EG4 batteries up to 56 volts. I don't think EG4 has any technology that makes their batteries any different as they are using the same technology as other battery manufacturers. Considering it is really a 48 volt battery, I think 54 volts is a good number.

Also, when I did try to charge the EG4 up to 56 volts it took a long time to reach that voltage and I feel it is too high for Lithium cells.

I talked to EG4 and apparently they are working on getting the ability to interface in their batteries with Outback systems.

One think I would like to ask is, we have set "Float" with a time of 11 hours and I am not sure if we should just set the Float voltage without putting in any time?

Thanks for the replies and would welcome your thoughts.
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

Other lithium chemistries are more susceptible to degradation at full charge than LFP, especially as anode materials and separators in LFP have evolved. That being said, all lithium is happy at PSoC. A researcher equipped that an LFP kept between 70% and 30% SoC would last 1000 years. Nobody alive will prove him wrong.
The trick is managing battery capacity. Fairly easy in a Tesla, more of an issue in your power shed.
"Full charge" is a broad definition in LFP when it comes to voltage. I can reach full capacity by a constant current charge to 56.4V. I can also reach full capacity with a CC/CV charge at 54.6V. It's simply a matter of the time it takes to shuffle enough ions.
Your "full charge" excursion to 56V taking a long time is a sure indication there was a lot of shuffling to do.
Float is another factor. My battery specs 54V float. That's fine, but I like reducing Float so there is a very small (100mA) discharge from my 450Ah battery. Why? Because it's fun to play with the physics.
11H Float on the inverter is a waste of AC, and not aligned with your PSoC goals. Set Float Time: 0
chrisski
Forum Guru
Posts: 103
Joined: Fri May 17, 2024 7:23 pm
My RE system: 8048 Radian w/ Load Center
Mate 3
Flex max 100 afci X 2
Fortress eFlex 5.1 kWh Battery X 4
TP1000 for Internet Connectivity
Location: Central AZ, lowland desert

Re: Lithium battery SOC drop with peak solar?

Post by chrisski »

FWIW, my Outback measures my batteries are having SOC erroneously change after dark to 100% with no charging every night.

The time I mentioned in an earlier post correlated with an MPPT turning on in the morning, but no log events for any of the others.
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

chrisski wrote: Tue Jun 18, 2024 12:18 pm FWIW, my Outback measures my batteries are having SOC erroneously change after dark to 100% with no charging every night.

The time I mentioned in an earlier post correlated with an MPPT turning on in the morning, but no log events for any of the others.
I take it there's an FNDC in the system?
What are the Battery Charging parameter values?
Dan-norcal
Forum Junior Member
Posts: 3
Joined: Mon Jun 17, 2024 11:52 am
My RE system: FP1: VFXR3648A-01, MATE3s, HUB10.3, FM80, FLEXnet DC, Battery: EG4-LL 48V 100AH Solar: 3.2kw, 9 x REC365NP2

Re: Lithium battery SOC drop with peak solar?

Post by Dan-norcal »

raysun wrote: Tue Jun 18, 2024 12:01 pm Other lithium chemistries are more susceptible to degradation at full charge than LFP, especially as anode materials and separators in LFP have evolved. That being said, all lithium is happy at PSoC. A researcher equipped that an LFP kept between 70% and 30% SoC would last 1000 years. Nobody alive will prove him wrong.
The trick is managing battery capacity. Fairly easy in a Tesla, more of an issue in your power shed.
"Full charge" is a broad definition in LFP when it comes to voltage. I can reach full capacity by a constant current charge to 56.4V. I can also reach full capacity with a CC/CV charge at 54.6V. It's simply a matter of the time it takes to shuffle enough ions.
Your "full charge" excursion to 56V taking a long time is a sure indication there was a lot of shuffling to do.
Float is another factor. My battery specs 54V float. That's fine, but I like reducing Float so there is a very small (100mA) discharge from my 450Ah battery. Why? Because it's fun to play with the physics.
11H Float on the inverter is a waste of AC, and not aligned with your PSoC goals. Set Float Time: 0
Thanks, I agree. Changing the float to zero time and we'll see what happens! :grin:
chrisski
Forum Guru
Posts: 103
Joined: Fri May 17, 2024 7:23 pm
My RE system: 8048 Radian w/ Load Center
Mate 3
Flex max 100 afci X 2
Fortress eFlex 5.1 kWh Battery X 4
TP1000 for Internet Connectivity
Location: Central AZ, lowland desert

Re: Lithium battery SOC drop with peak solar?

Post by chrisski »

raysun wrote: Tue Jun 18, 2024 12:20 pm

I take it there's an FNDC in the system?
What are the Battery Charging parameter values?
Screenshot 2024-06-18 164149.jpg
I just put two and two together that these settings will return the SOC to 100%. Since voltage is jumping to 100% when its lower than these settings and between 50% and 70%, I'm not sure its the FNDC settings. These are my Radian Battery Charger Settings:
Radian Battery Charging settings.jpg
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

Is this system connected to the grid?
chrisski
Forum Guru
Posts: 103
Joined: Fri May 17, 2024 7:23 pm
My RE system: 8048 Radian w/ Load Center
Mate 3
Flex max 100 afci X 2
Fortress eFlex 5.1 kWh Battery X 4
TP1000 for Internet Connectivity
Location: Central AZ, lowland desert

Re: Lithium battery SOC drop with peak solar?

Post by chrisski »

raysun wrote: Tue Jun 18, 2024 5:48 pm Is this system connected to the grid?
Mine is, but I wonder about the op @dan
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

Rebulk Voltage:51.2
Is battery getting there?
raysun
Forum Overlord
Posts: 11402
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01 - Series Stacked, (2) FM80, MATE3s, FlexNetDC
6 SimpliPhi 3.8-48 (48v @ 75AH. 450AH total)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
2 Midnite Solar MNPV6 combiners w/20A DC disconnects.
Honda EU7000is gas fuel generator
Off-Grid

Re: Lithium battery SOC drop with peak solar?

Post by raysun »

Towards the end of the day, the state of charge suddenly shoots up to 100%.

Is this what we should expect to see?
If it's what I suspect, it is by design. "Force Float" is an artifact of lead acid charging. When the battery loiters in Bulk long enough, the charger is switched to Float (as a battery protection measure) by the Mate/FNDC. The Float trigger is 100% SoC.
Currently have the Absorb set for 54.2 V (0.5 hours)
The low Absorb Voltage is right on the charge "plateau" and will stay there from roughly 70% to 100% SoC. 30 minutes likely times out at an indeterminate State of Charge.

The FNDC Charged Return Amps and Charged Time values are unorthodox. What's the purpose?
Post Reply