question

Kaj Lehtinen avatar image
Kaj Lehtinen asked

How much Kwh loss over a charge cycle is normal with pylontech batteries

Hi,

Started to look at my battery effeciency in VRM. I'm running a ESS system, last couple of months (November & December) there hasnt been any sun to talk about, and as such I'm running my system with charging the battery bank from grid during the 5 cheapest hours of the day and consuming the rest of the day.

My system is based on 3 phase Multiplus II 24/3000 with 8 pylontech UP2500 units for a total of approx 20 Kwh storage capacity. Coupled to this I have a small DC load constantly connected and a bunch of MPPTs. The DC load is connected over a SmartShunt so I know its draw. Everything connected to Lynx distribution boxes with its led indicators connected with the cable hack that circulates online.

When looking at VRM for last month it looks like this for Consumption

1703951697516.png And for GRID it looks like this

1703951753899.png

If I take the To Battery value and subtract the From Battery value its 597 - 369 = 228 kWh that has gone into the battery bank but not come out over the previous month. Divide 228 by 30 and you get approx 7.6 kWh/day. The SmartShunt load is approx 2.5 amps constantly, thats 70-80W, lets say 100W to have a slight margin. Thats 2.4 kWh per day or 72 kWh for the whole month. If I subtract 72 from the 'missing' 228 its 156 kWh still unaccounted for. Thats 5.2 kWh per day or a total of 25% if the batterybank would have been fully charged each day - which it isnt due to other grid loads and not overloading the grid and limiting charging to 5 hours per day.

The only things connected is the MPPTs, DC load, 3*Multiplus II's & the Cerbo - so where is the fault? In VRM calculations, in Pylontech BMS reporting (if that is what Venus OS uses as base for To/From Battery) or does the MPPTs draw that much power when not used? My understanding is that Lithium batteries perhaps should have less than 10% loss on the charge cycle, or have I misunderstood it?

Anything else that I dont see / think about that can explain the difference in what goes into the batteries and comes out of them?

The figures for December is the same approx - I will check to make sure on monday when we're in January.

/Kaj

ESSLithium BatteryPylontech
1703951697516.png (75.3 KiB)
1703951753899.png (70.2 KiB)
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

3 Answers
marekp avatar image
marekp answered ·

Why don't you use Advanced VRM?

You can see there "charged energy" and "discharged energy" calculated by SmartShunt.

Below you can see data about my LFP magazine.

screen-shot-2023-12-30-at-54053-pm.png

Battery efficiency 96.386%.


screen-shot-2023-12-30-at-54053-pm.png (48.8 KiB)
2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Kaj Lehtinen avatar image Kaj Lehtinen commented ·
Hi,


Pylontech BMS doesn't provide that info (for me at least) and the SmartShunt only measures the directly connected specific DC load (and configured as a DC load), not as a traditional DC Shunt.


/Kaj

0 Likes 0 ·
Show more comments
Alexandra avatar image
Alexandra answered ·

Alot of the charge efficiency depends on wiring and speed of charge ( and how hot wiring is getting.) And balancing. Then the batteries themselves consume some power for themselves.

The difference between the shunt and battery readings will probably be what is lost in cell/battery balancing battery self consumption.

And system self consumption would be part of your consumption too.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Kaj Lehtinen avatar image
Kaj Lehtinen answered ·

Thanks all,

So having thought about the issue for a while and reading the feedback the conclusion could either be 1) Pylontech BMS isn't accurate in reporting 2) VRM uses less optimal values to calculate what goes into the battery in regards to that it 'forgets' the self consumption of the victron equipment.

Considering that the 5 kWh that's 'missing' in the VRM figures equates down to 70W per multiplus unit (3 phase install) that's perhaps OK although a bit eye opening when one looks at the total amount drawn.

I will try to optimize the self consumption by programmatically switching the Multiplus over to 'charge only' when the ESS system knows it wont discharge the batteries for the coming hour (grid price < 10% above last charge price) - we'll see if that yields any measurable results.

/Kaj

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.