question

saskak avatar image
saskak asked

smart shunt discharged energy 95% of charged energy

Hi all,

My first post, so be gentle. I have a victron smart shunt 500A (and temp sensor) with victron smartsolar mppt 100/30, 24V battery bank (2x FLA Banner 95751, C20 100Ah) and 2x 340W panels connected in parallel (shading on one of them). Both the shunt and mppt are in a bluetooth network and connected through ve.direct cable to raspberry pi for logging purposes.

I have had the system for ~6-7 months, and the shunt has had 19 automatic synchronisations with total Amps drawn -481.60Ah. The shunt shows total discharged energy: 11.92KWh and charged energy: 11.32KWh. So, somehow the batteries get undercharged by ~5% or 11.32/11.92 =~95%.

This does not seem normal to me. Should it not be the other way around, it shoud have put more in the batteries, accounting for the inneficiency of the FLA. The charge efficiency factor is 95% with Peukert 1.25. In all my thinking there should have been more charged than discharged to account for the 95% efficiency. It is like the brand new Banner FLA batteries are way more than 95% efficient in accepting charge or the shunt is undercounting. Nothing is bypassing the shunt in terms of energy, i.e. only one negative cable comes out of the battery and goes to the shunt, everything comes out of the other end. Maybe I need to calibrate the shunt's 0 again. I only have a non-victron inverter and no other power coming in other than the solar.

Every time I draw certain amount of Amps, then the mppt charges to ~95% of that. Consistent undercharging. I suspect that this is better than overcharging, but this will kill the batteries eventually.

Any help is more than appreciated!

screenshot-2023-02-13-at-161432.png

SmartShunt
screenshot-2023-02-13-at-161432.png (123.1 KiB)
4 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.

Alexandra avatar image Alexandra ♦ commented ·
@saskak

FLAs are not usually 95% efficiency. That more lithium territory.

Does the bank reach target voltage programmed? How long it is kept there? Do you have adaptive charge on?

1 Like 1 ·
saskak avatar image saskak Alexandra ♦ commented ·

I completely agree with the 95% efficiency. I would have thought more like 85-90%. Yes, the victron 100/30 reaches target volts with temp correction, that is for absorb and float (28.6 & 27.6 respectively). A bit higher now as the temp is ~10C (temp correction -32mV/C). The algorithm is adaptive of 6h with 1amp tail current. That was given to me by Banner, so it stays longer in absorb until it reaches the 1amp.

My normal cycle is to draw whatever amps (<30% or so) and then let it fill up to until it reaches float and amps under full sun are something like 0.1-0.2A. This will be bulk/absorb and float ~2-3h. Then I draw again. When it reaches float every time, it is ~5% less, as measured on the shunt. I have a dehumidifier that I run every 2-3 days depending on when the batteries are full.

0 Likes 0 ·
kevgermany avatar image kevgermany ♦♦ saskak commented ·
Just for info, my lead carbons have total discharge of 111.5kWh, total charge of 133.9kWh.

Charge efficiency/Peukert values/settings shouldn't interfere with recorded power in/out of the batteries. They're there for calculating SOC and time to go.

Might be that you have long periods of very low charge rates, small currents can be missed by the shunt, there's a setting for this, current threshold. Anything below this is ignored by the shunt. Default is 0.1A. If this was increased, it might explain your differences.


1 Like 1 ·
saskak avatar image saskak kevgermany ♦♦ commented ·

Yes, thank you. Yours seem what one should expect, more charged than discharged to account for the efficacy of the batteries. Yours is even ~83%, which seems about right for lead carbon.

I thought about this as I have set it at 0.12A and as you have mentioned the default is 0.1 with the difference 0.02A. 0.02A even multiplied by 24h is nothing to account for the 5% difference. If that was the case it would be constant loss within every 24h or whatever time it takes to charge the batteries (say 48h), but mine seems relative to the discharge rate or roughly 5%. I never really have very long periods of charge rate, they get charged to full every about 2-3 days, sometimes every day, so not very long of the 0.02A to accumulate with time.

A discharge of say 10A one day I get ~0.5A not recovered, but on a discharge of say 20A I get ~1A not back. It is weird and not too sure where it is coming from. It is either the mppt undercharges, the batteries get knackered every day or the shunt is undercounting. Everything works as expected, volts get to where they are supposed to be, the shunt synchronises, the voltage of the batteries does not sag under constant load (relatively constant voltage/SOC decline). I generally draw about 7A, so a bit more than the C20 (5A).


0 Likes 0 ·
1 Answer
saskak avatar image
saskak answered ·

I think I managed to solve it, but will have to monitor in the next few days to see. The Smartshunt was undercounting by ~0.08A, so either I have forgotten to zero it out at the beginning (zero calibrate) or it has drifted a bit with time. I simply removed the negative system cable from the shunt, so nothing coming out, reduced the current threshold to 0 in the victron connect app and saw -0.08 current coming out. Once I zero calibrated it, the current was 0.

So, in every 24h it was counting ~1.92Ah that have come out of the battery. So, my 0.10A current threshold was just above that and I never saw it, but it was accumulating with time. I will report back if that has fixed the issue, but I am nearly 100% that it will have fixed it.

Thank you all for the advice.


UPDATE:

Yes, that was the problem. The smartshunt was not zero calibrated or it has drifted with time. Probably I forgot to do it.

Last few days I was experimenting with small/large discharges and the difference between the discharged and charged energy is slowly reducing. Eventually will go to the normal ~85% for lead acid more charged than discharged.

Date Charge Discharged_Energy[KWh] ChargedEnergy[KWh] Difference[KWh]
15/02/2023 FullCharge[float at ~0.3A] 12.06 11.47 -0.59
16/02/2023 FullCharge[float at ~0.3A] 12.11 11.53 -0.58
17/02/2023 FullCharge[float at ~0.3A] 12.15 11.60 -0.55
19/02/2023 FullCharge[float at ~0.3A] 12.67 12.15 -0.52

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.