SOC reading incorrectly.

Dual Input, 120/240, stainless steel form factor design for unsurpassed surge capability
Post Reply
Rancho Partera
Forum Junior Member
Posts: 3
Joined: Wed Jun 22, 2022 11:08 am
My RE system: GS8040A, Flexmax 100 and Flexmax 80, Mate 3, communications hub, 20 405 W solar panels, 24 L16 400 A, FullRiver AGM, 48 volt system

SOC reading incorrectly.

Post by Rancho Partera »

I have an issue where intermittently my SOC percentage reading will just start dropping even during the middle of the day with full sun and the batteries fully charged showing 54.4 V. It will do this intermittently. Then go back to reading what seems to be more accurately. I have tried disconnecting the data cable from the back of the Mate 3 to reset the connection and that does not seem to help.

I have been told this % reading is unreliable and to focus on the V of the batteries but would like to see if there is a fix.

Thanks for your ideas.
raysun
Forum Emperor
Posts: 8024
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (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
Honda EU7000is gas fuel generator

Re: SOC reading incorrectly.

Post by raysun »

The Mate can report SoC, but does not track it. If there's an SoC reading in the system, there must be a FlexNET DC battery monitor.
FN-DC-2T.png
If one is in the system, it tracks DC charge and discharge to and from the battery. If the battery parameters are set correctly it is quite effective at deriving battery SoC.

The question becomes, why would there be unusual SoC readings that somehow self-correct? It may be something with the shunts or sense wires. Hard to tell without more to go on.

FYI: If the data cable is disconnected to reset, the FNDC will restart and show 100% SoC, irrespective of the actual battery state of charge.
Rancho Partera
Forum Junior Member
Posts: 3
Joined: Wed Jun 22, 2022 11:08 am
My RE system: GS8040A, Flexmax 100 and Flexmax 80, Mate 3, communications hub, 20 405 W solar panels, 24 L16 400 A, FullRiver AGM, 48 volt system

Re: SOC reading incorrectly.

Post by Rancho Partera »

When I disconnect the data cable then reconnect the SOC comes back exactly where was before the reset.
raysun
Forum Emperor
Posts: 8024
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (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
Honda EU7000is gas fuel generator

Re: SOC reading incorrectly.

Post by raysun »

Oh right, the FNDC needs to be power cycled 9n its battery + line to reset.
Rancho Partera
Forum Junior Member
Posts: 3
Joined: Wed Jun 22, 2022 11:08 am
My RE system: GS8040A, Flexmax 100 and Flexmax 80, Mate 3, communications hub, 20 405 W solar panels, 24 L16 400 A, FullRiver AGM, 48 volt system

Re: SOC reading incorrectly.

Post by Rancho Partera »

I power cycled the entire system and that solved the issue and it now shows 100% SOC. Trying to learn and not screw this up as I go learn.
raysun
Forum Emperor
Posts: 8024
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (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
Honda EU7000is gas fuel generator

Re: SOC reading incorrectly.

Post by raysun »

Rancho Partera wrote: Wed Jun 22, 2022 3:10 pm I power cycled the entire system and that solved the issue and it now shows 100% SOC. Trying to learn and not screw this up as I go learn.
Does 100% SoC reflect the actual battery state of charge? Was it truly 100% charged at the time the system was power cycled?
JRKsolar
Forum Whiz
Posts: 35
Joined: Thu Aug 20, 2020 11:35 am
My RE system: Outback GS8048A
4) Kilovault HAB 7.5
2) Outback Flexmax 100 AFCI
12) Mission Solar 375W
8) Canadian Solar 395W
Tigo TS4-A2F

Re: SOC reading incorrectly.

Post by JRKsolar »

I don't have fndc, but my batteries have a similar system built-in. Personally, I take the whole soc thing with a grain of salt. They tend to slowly drift and to keep them accurate you need to regularly charge the batteries to 100% based on.....reaching full voltage. I like to charge to a lower voltage in order to try to extend the lifespan, so my soc is often times not accurate or even the same on each battery. When it gets to be more than I can tolerate watching, I'll run a full charge cycle. As the mfg states, the soc is an approximation and as long as the voltages are the same, the batteries are working fine. It's certainly helpful to see, but not something to stress about. imho.
raysun
Forum Emperor
Posts: 8024
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (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
Honda EU7000is gas fuel generator

Re: SOC reading incorrectly.

Post by raysun »

The great thing about LiFePo4 is its perfectly happy at partial state of charge. The hassle with LiFePo4 is figuring out what capacity is available at partial state of charge. The easy way to solve the dilemma is to simply have more capacity than is ever needed. That only takes $$$$$$.

Cell voltage is an important metric, and can be an accurate determinant of capacity, as long as the battery isn't actually being used. Not too practical in a working system. In the latter case, cell voltage creates special challenges all its own. The vaunted "flat voltage curve" of LiFePo4 is a blessing and a curse. The blessing comes with consistent energy output. The curse is knowing how much energy remains. Also, while the voltage curve is flat across a broad range of capacity, the absolute voltage varies with discharge rate. The voltage at 50% capacity @ C5 is very different than the voltage at 50% capacity @ C20. Voltage as an indicator of available capacity in an active system is an interpolation exercise that is prone to error. Again, lots of extra capacity allows for a wide margin of error with fewer consequences.

Pop Quiz: Johnny has a 450AH LiFePo4 (16 series cell) battery that reads 52.8V. How long can Johnny keep the lights on?
(I hate story problems.)
Screenshot_20220623-080152_Firefox.jpg
Visual clues:
19H - Voltage 52.8, SoC 85%
23H - Voltage 52.9, SoC 75%

Though fraught with its own imperfections borne of the complexities of battery charge and discharge (curse you Peukert!), "coulomb counting" is still the best method for dynamically determining available capacity. Since State of Charge is a derived metric, it is prone to metering errors and drift, however it can be made to be useful and if properly done, is far less prone to interpretation errors than cell voltage under load. Syncing SoC is a simple and unambiguous process, accomished by fully charging the battery.

BTW, reaching "full charge" voltage, in and of itself, is no guarantee of an actual full charge. Another factor is required. For example, in my LiFePo4 reaching the manufacturer's termination voltage at a charge rate of 225A leaves the battery about 40AH short of maximum capacity. A typical (and more rational) charge current of 75A leaves the battery about 5AH short of full capacity at termination voltage. How to know when full charge is reached? When the termination voltage is held constant, and the net charging amps decline to a specified rate, typically 2% of the C20 AH value for LiFePo4. So termination voltage and end amps are the two factors that unambiguously identify full charge.

There's some merit to not fully charging a lithium battery. The degradation effect on the Anode is most pronounced in some of the other lithium chemistries, but LiFePo4 is not immune, however a great deal of materials science goes into the typical quality LiFePo4 cell these days to negate the degradation to a good degree. Its not clear that most folks would benefit from a PSoC program, especially if it meant Depth of Discharge excursions below 20% of battery capacity. PSoC is a luxury for those battery owners who have enough surplus capacity to, ironically, not need worry about extended battery life. End of Life is considered loss of 20% of overall capacity. Someone with a battery that starts out with 20% or more surplus capacity won't know their battery is statistically "dead" when EOL comes. That, and by the time the battery needs replacing, bigger, faster, better, and cheaper batteries will dominate the landscape.

FWIW, I use the FNDC and Victron BMV700 series meters to track battery SoC and voltage. Both track each other (and presumedly the battery state) to a remarkable degree of precision. Since I have a fairly large battery and cycle daily, neither metric garners more than a glance, as I have a good intuitive understanding of our system and loads. If I was running closer to the limits, I'd pay more attention. Its good to know that I can rely on the readings.
JRKsolar
Forum Whiz
Posts: 35
Joined: Thu Aug 20, 2020 11:35 am
My RE system: Outback GS8048A
4) Kilovault HAB 7.5
2) Outback Flexmax 100 AFCI
12) Mission Solar 375W
8) Canadian Solar 395W
Tigo TS4-A2F

Re: SOC reading incorrectly.

Post by JRKsolar »

I agree with what you're saying.
My situation is a little different in that I run in Mini-Grid: off-grid, but with grid as an automatic back-up source. My system automatically reconnects to the grid when the battery voltage drops to 51.6 for at least 10 minutes. That's about 25% soc at the loads that I run at that time of night. I've connected twice this year.
Admittedly, if I didn't have soc built in to the batteries, I wouldn't have a way of knowing the soc at those levels. And yes, I watch soc obsessively anyway :grin:
Post Reply