MultiPlus SOC dropping when charger in Storage mode and inverter is off

Question

question

moonlightrider asked Nov 19, '18

MultiPlus SOC dropping when charger in Storage mode and inverter is off

Hi

I recently set up a small UPS using a MultiPlus 12/800 and a 105AhC20 AGM battery. Everything works as expected and I'm also running Venus OS on a Rpi 3B to be able to monitor the system.

The MultiPlus is working as standby UPS (no ignore AC settings), so it only uses the battery/inverter if AC goes down or misbehaves.

Battery Monitor on Venus is set to MultiPlus on VE Bus.

I noticed that SOC keeps dropping about 0.5% every day. This should not happen as the battery voltage is 13.4V (temp. comp.) and no load is applied. I also see a strange random load of 1A being drawn from battery. This is probably related.

I was able to catch one, see below...

I been thinking...

Is this maybe caused by temperature compensation? (temp raising so charger lowers the voltage maybe causing false current readings?) If so, this can also confuse SOC calculation?

Any thoughts and help would be very much appreciated.

UPDATE:

So, if we had solar/external chargers or loads connected we shouldn't use the MultiPlus to calculate SOC, but if we have a basic setup like mine, with no external charger/loads, I think that would be a good idea that internal calculation could hold SOC at 100% after entering storage mode and if inverter was not being used, so it would ignore 'noise' readings like the ones I get that I think are caused by temperature compensation adjusting voltage.

MultiPlus Quattro Inverter Charger SOC feature request

multiplus.jpg (23.1 KiB)

soc.jpg (27.8 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.

MultiPlus Product page / Quattro Product page

MultiPlus 230V Manual / Quattro 230V Manual

MultiPlus 230V Datasheet / Quattro 230V Datasheet

VE.Bus Error codes / VE.Bus configuration guide

Additional resources still need to be added for this topic

The feature requests topic tag can help Victron R&D find feature suggestions for products.

There are many factors that go into the Victron product roadmap, and feedback from end-users can help.

Modifications Space

Node Red Space

Answer 1 · 2018-11-19T09:47:54Z

ben answered · Nov 19, 2018 at 09:47 AM

Probably measurement error. The state estimation done inside the inverters is, in particular, known to be a rough approximation.

Does the battery voltage ever actually drop, like over the course of weeks?

4

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.

moonlightrider commented · Nov 19, 2018 at 11:32 AM

Battery voltage stays mostly the same, it only changes according to battery (room) temperature compensation from charger.

Today SOC reports 98%, 13.4V, still on storage mode = so no battery load for at least 24H... Pretty weird issue.

0 ·

ben ♦ moonlightrider commented · Nov 19, 2018 at 04:35 PM

A few days of data is not enough to say for sure, but I still suspect you are seeing small metering inaccuracies accumulated over time.

0 ·

moonlightrider ben ♦ commented · Nov 19, 2018 at 07:21 PM

The 1 Amp 1 second random load I'm seeing is what intrigues me most and possibly causes this.

0 ·

ben ♦ moonlightrider commented · Nov 19, 2018 at 07:41 PM

I think Victron is going to tell you that the MP metering is just not that accurate. Even with a BMV, which can be calibrated to 0A, there is still SOC drift over time. It's necessary to "resync" the SOC every so often to get back to a known-good baseline.

I'm not sure, though, why the MP algorithm won't do that automatically while you are in storage mode, since I think that's where it is supposed to go when it knows the battery is full. I would have thought it would just ignore slight current readings and hold the SOC at 100%.

1 ·

Answer 2 · 2020-11-22T16:18:25Z

richardp63 answered · Nov 22, 2020 at 04:18 PM

Same situation here:

MultiPlus 12/1600 with 4x 150Ah C20. Used for USV, no cycling.

After installation and configuration, before power on the supply (grid), I saw that approx. 13 ... 15 W were taken out of the battery - would be the consumption of the MultiPlus itself, I think (no load on MultiPlus output),

After powering on the supply an putting load on the output, there were approx. 2 ... 3 days for Bulk/Absorption/Float.

After this, storage mode was entered. I see nearly constantly that 10 ... 15 W were taken out of the batteries and, that the power from grid is always slightly lower than power delivered to load.

After approx. 5 days in storage mode, SOC falls below 80% - and the relay assistant shut down my server cabinet :-(

The Ah consumed from batteries in this 5 days confirm the SOC calculation to be ok.

Questions arising are:

Why is the Multiplus taking power (its self consumption plus more) from batteries?
Why does it ignore that grid power is below load power?
Why doesn't it fill up batteries? I set the check mark at "USV-Mode"
Storage mode should not accept discharging - why isn't the voltage set to get the battery current to zero or little current into batteries?

I can go to the MultiPlus in a little bit more than two weeks - I think, I have to disable storage mode.

And Im hoping, MultiPlus will not inactively watch the SOC going below the switch off level before I can cange configuration...

Any Ideas to get storage mode useable?

1606060536985.png (24.2 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.

Answer 3 · 2022-03-18T07:30:33Z

martin-van-biljon answered · Mar 18, 2022 at 07:30 AM

Good day guys. I have the same issue here. Any one here to help solve it?

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.

Answer 4 · 2023-02-06T14:46:42Z

vet answered · Feb 06, 2023 at 02:46 PM

Unbelievable: four years have passed since problem was reported, and nothing done by manufacturer. Is it really normal behaver for the world’s leader?

I’ve just bought MultiPlus 12/1600/70-16 and got the same issue: after battery is fully charged, SOC drops by 1% per hour. So I’m actually can’t use it safely discharging the battery to only 40% SOC from 100%.

It is obvious, that problem could be solved by software update. Also it’s obvious, that it’s manufacturer bug (SOC should not be calculated down from the moment when the voltage is at the level of Float or Store mode)! Is where any hope, that manufacturer will finaly hear the customers?

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.

Answer 5 · 2023-03-03T12:39:34Z

vet answered · Mar 03, 2023 at 12:39 PM

Recently I have got an answer from service department of Victron. They admitted, that it is production defect (bad “calibration or something”) and MultiPlus should be sent to dealer to be serviced. So if you have the same problem, best solution is to get service from Victron’s dealer.

In my case I couldn’t stay for a long time without Multi, so the only way for me was to find a possibility to fix calibration issue by myself. And I’ve made it! If you are in the same situation as me here is the procedure:

1)Find current measurement transformer of “charger/inverter” chain inside of you Multi.

2)Find +12v from internal psu, which is used to power all smart functions (microcontroller, operational amplifiers etc.). I used VCC pin number 4 of 324A operational amplifier.

3)Using 68 kOm resister connect +12v you have find in step 2 to one of secondary coil pins of transformer you determined in step one. In my case it is right pin of coil if you are looking from the pcb side (see picture attached). If wrong coil pin will be used, probably no positive or negative effect will be seen. So it’s easy to find out write pin. BUT UNDER ANY SIRCUMSTANSES DON’T USE HIGHT VOLTAGE LOOP PINS WICH IS ON THE OTHER SIDE OF THE TRANSFORMER!!!

That's all. Now my MultiPlus works perfectly! After battery once been charged, at any given time I have all the battery capacity that can be safely used up to a certain SOC level (I set 40%)!

P.S. Resistor nominal in your case could be different. Mine was experimentally determined in range of 15 to 450 kOm. Best stable affect without decreasing of SOC parameter in storage mode I have got with 68 kOm. Any negative impact on the system charging or inverting was not found.

1.jpg (2.5 MiB)

2.jpg (2.6 MiB)

3.jpg (7.1 MiB)

4.jpg (6.9 MiB)

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.

question