SoC jumps upwards when it shouldn’t be.

Question

question

thanar asked Mar 16, '23

SoC jumps upwards when it shouldn’t be.

ESS GX system, 5kW MP-II, 3kW SmartSolar, just 5kWh LFP battery (no networked BMS).

The battery has a high internal resistance and as soon as charging at over 20A, it’s voltage peaks to the bulk set voltage, hence percentage jumps from whatever calculated value to the set percentage as set on “State of charge when Bulk finished” by VEconfigure on the MP-II.

Is there a way to disable this adjustment of percentage?

I’ve adjusted “State of charge when Bulk finished” to 60% (because that’s roughly where it really is), but now SoC jumps all the way to 100% for some reason half an hour later, while the battery still sips 1kW of power for at least two hours.

Only thing I can think of regarding the 100% jump, is that the BMS kicks in and cuts off charging for a blink, so Victron translates this as full. Depending on your perspective, that could make sense, so I will be trying to cap charge current a bit and see how it goes.

However, SoC with no adjustments could also be an option. Is it really that difficult to leave SoC alone and only adjust it according to the measurements made by the interal MP-II coulomb meter?

battery charging SOC float

e7af822c-42ca-413f-80bd-c2f63ddd127a.jpeg (110.0 KiB)

2 |3000

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

Modifications Space

Node Red Space

Answer 1 · 2023-03-16T09:08:48Z

nickdb answered · Mar 16, 2023 at 09:08 AM

Best practice is to install a shunt which you can tune. The internal monitor is not particularly accurate nor configurable, especially with external charge sources.

1

2 |3000

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

thanar commented · Mar 16, 2023 at 09:40 AM

It’s not the lack of accuracy that’s the problem, it’s the fact that SOC jumps all the way from 20% to whatever % because “Bulk is complete” and then to 100% for no apparent reason. I’m looking for a way to completely disable this SOC adjustment.

0 ·

Answer 2 · 2023-03-16T11:37:35Z

tom w answered · Mar 16, 2023 at 11:37 AM

As long as all the chargers are from the Victron universe, the reading should be good enough assuming the DVCC settings are correct. Without an SOC meter the ESS does not know when to stop discharging.

On my system I had the Shared current sensing set to on. That resulted in using only the Multiplus current sensor for the SOC. The charge current from the MPPT was ignored. After setting the shared sensor to off, the readings are correct.

Additionally the ESS needs some charge cycles to get the SOC right. The battery needs to go to at least one Absorption phase for top balancing all the cells.

Jumping from 20% to 95% indicates to me that the MPPT charge current is not added to the SOC. I guess you have no AC coupled PV.

2

2 |3000

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

thanar commented · Mar 16, 2023 at 11:49 AM

The jump occurs as soon as the charging voltage reaches 54.5V, which is the maximum voltage for my battery. I have set both Bulk and Float to that same value. Problem is, that charging voltage depends on the current, and due to the fact that this specific battery has a high internal resistance, voltage spikes without the battery being full, of course. I would prefer to ditch bulk/float altogether and go plain CV/CC, which is the right thing to do for lithium batteries, but I don’t see such interest from the team at Victron.

I will give DVCC another try, although it’s supposedly not suit for the specific battery.

1 ·

nickdb ♦♦ commented · Mar 16, 2023 at 11:50 AM

Your answer is nonsensical. He is referring to the native configuration of the internal battery monitor.

10.1.7. State of charge when bulk finished

Use this setting to specify what the "state of charge" is set to when the Bulk phase is finished. This helps to calibrate "state of charge" value due to inevitable measurement errors that accumulate over several charge/discharge cycles.

1 ·

Answer 3 · 2023-03-17T06:02:28Z

kevgermany answered · Mar 17, 2023 at 06:02 AM

LFP and high resistance? Something wrong.

Might be the BMS causing your problems.

1

2 |3000

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

thanar commented · Mar 17, 2023 at 07:50 AM

I couldn't agree more... This specific battery indeed has a capacity of 5kWh (tested at 5A discharge, sure that's just 0.05C), however the resistance adds up to around 100mΩ, when calculated using data off of the VRM.

Here's the installation, in case someone wants to have a look: https://vrm.victronenergy.com/installation/278343/share/f53e777c

0 ·

question