Solar Assistant battery charged vs discharged totals seem incorrect

[wayne530]

I’ve been running my system now for almost 4 months and noticed that in the aggregated data, Solar Assistant is reporting that battery charged is often more than 2.5x battery discharged, which just doesn’t make sense to me. Has anyone else seen this? I know there are inefficiencies in converting between AC and DC but would not expect this much of a difference. Are there any potential issues I should investigate? Thanks in advance for any suggestions!

I’m running 2 LV6548s for split phase 120/240V output and the battery bank consists of 6 EG4 LiFePower4 48V packs in an EG4 cabinet with built-in bus bars. Each pack is connected to the bus bars with #4 cu welding cable (about 12” per terminal) and inverters are directly connected to the same bus in a diagonal configuration with about 12’ (total loop) of #2 cu welding cable. There is a 200A class T fuse between the + bus bar and + battery terminal of each inverter and a DC disconnect switch between the - bus bar and the - battery terminal of each inverter.

 

[Mattb4]

Don't know if the reported totals are accurate but you have loss while charging a battery compared to how much you get out of it. That is not in regards to inverter losses.

 

[Lt.Dan]

@Mattb4 is correct, and I believe it states somewhere that the numbers are not exact because solar assistant cannot account for inverter and mppt efficiency losses. I think it also states that some inverters are terrible at reporting the correct numbers, which will skew numbers even further.

 

[wayne530]

@Mattb4 @Lt.Dan Thanks, fully understand there are charging losses as well, but I don't think they should result in a 2.5x+ difference. What I'd like to better understand is if other folks with a similar setup see similar differences between charged and discharged in Solar Assistant. I have only one other data point (same inverters, same battery configuration) that suggests this is not normal. Do you both run Solar Assistant? Please share what approximate ratio you see between charged:discharged.

 

[Mattb4]

Sorry I don't run SA, or for that matter "2 LV6548s for split phase 120/240V output and the battery bank consists of 6 EG4 LiFePower4 48V packs".

 

[Lt.Dan]

I run SA. I'll check my stats when I get home

 

[houseofancients]

@Mattb4 @Lt.Dan Thanks, fully understand there are charging losses as well, but I don't think they should result in a 2.5x+ difference. What I'd like to better understand is if other folks with a similar setup see similar differences between charged and discharged in Solar Assistant. I have only one other data point (same inverters, same battery configuration) that suggests this is not normal. Do you both run Solar Assistant? Please share what approximate ratio you see between charged:discharged.

did you set your inverter's self consumption correctly ?

 

[100FUEGOS]

Turbo Energy (Deye) 5kW inverted DIY 18650 Li-Ion battery.

 

[wayne530]

did you set your inverter's self consumption correctly ?

I set mine to 70W. I don't know about the correctness. It's not listed in the inverter documentation so I just googled and found a few references to this value. Would that account for a 2.5x difference though?

 

[TomC4306]

Add 7 days charging and discharging data together and then divide by 7 to come up with an average. You're forgetting that this issue crosses midnight. Unless you are charging to the exact same state of charge and discharging to the exact same state of charge each day then there will be differences from day to day

 

[TomC4306]

I've had SA for 9 months and don't find those two columns particularly useful.
Edit . Except as a spot check from time to time to check capacity

 

[100FUEGOS]

did you set your inverter's self consumption correctly ?

I do not think that value can be set for Deye / Sol Ark inverters, at least it is not available under Battery / Advanced like it is for Voltronic inverters.

 

[wayne530]

Add 7 days charging and discharging data together and then divide by 7 to come up with an average. You're forgetting that this issue crosses midnight. Unless you are charging to the exact same state of charge and discharging to the exact same state of charge each day then there will be differences from day to day

I'm not forgetting that charging today replenishes the battery from discharging the day before. The values I showed are monthly aggregates and every month shows > 2.5x delta between charging and discharging.

 

[TomC4306]

I'm not forgetting that charging today replenishes the battery from discharging the day before. The values I showed are monthly aggregates and every month shows > 2.5x delta between charging and discharging.

Hadn't looked at the screen grab. What I see shows relatively close numbers on a month to month basis for charge and discharge 393 versus 391.8 pretty close

 

[wayne530]

Hadn't looked at the screen grab. What I see shows relatively close numbers on a month to month basis for charge and discharge 393 versus 391.8 pretty close

Sorry, I reposted those from someone else on the forum. See my first post in this thread for my numbers.

 

[TomC4306]

No pv. So, grid - load = net battery.
How much did you put into the initial charge?

 

[wayne530]

No pv. So, grid - load = net battery.
How much did you put into the initial charge?

The batteries were at 100% SoC when the system went online.

 

[timselectric]

Solar Assistant is just reporting the information it receives from the inverters. (Rarely accurate)
This is why I use a Victron smart shunt for battery monitoring and reporting to Solar Assistant.

 

[wattmatters]

Solar Assistant is just reporting the information it receives from the inverters. (Rarely accurate)

↑ This ↑

Solar Assistant is reporting that battery charged is often more than 2.5x battery discharged, which just doesn’t make sense to me.

Check the source of data. How is SA getting its data on battery charge and discharge?

With EG4 batteries I'd have thought you could have those feed data directly to SA:

EG4 battery setup in SolarAssistant

How to connect an EG4 battery to SolarAssistant.

solar-assistant.io

 

[wayne530]

Solar Assistant is just reporting the information it receives from the inverters. (Rarely accurate)
This is why I use a Victron smart shunt for battery monitoring and reporting to Solar Assistant.

100% understand about the inaccuracies but to see such a large disparity in values with other forum members that use the same inverters seems anomalous to me, so I was just trying to better understand whether I have a potential issue in my setup.