question

derik-alwyn avatar image
derik-alwyn asked

Mutiplus 2 configured parallel single phase ESS setings for feed back

The system seems to work fine except for the following issue:

When the batteries are fully charged, the PV input is throttle down to zero Watt. Then it seems that the battery starts feeding the load, until the battery drops a few milli volt. Then the MPPT,s are throttle up again to say about 4KW that charges the batteries and the MPPT' are throttle down again. This cycle repeats throughout the whole day until sunset.

2 x Multiplus 48/5000/70 in parallel

2 x 250/85 MPPT's

2 x Solar MD 7.4Kwh batteries

1 x Solar MD V2 logger (bms)

1 x Color control

8Kw solar panels.

battery charging
2 |3000

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

3 Answers
Daniël Boekel (Victron Energy Staff) avatar image
Daniël Boekel (Victron Energy Staff) answered ·

Hi @Derik,

Your battery's bms is constantly changing the CVL (Charge Voltage Limit), and the Victron system is reacting on this, even discharging the battery to get at the lower CVL, the battery then changes the limit (higher) again, causing an oscilation.

Please ask your battery manufacturer to fix this issue, or if allowed with this battery set a manual (lower) charge voltage.
If you set the MPPT's at 53.5 V, and the Multi's at 53 V or even lower, feed-back to grid should work perfectly.


3 comments
2 |3000

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

Kaloyan Dimov avatar image Kaloyan Dimov commented ·

@Daniël Boekel (Victron Energy Staff) Hi There, I am from the battery manufacturer. How do you think the best should be to set the charging voltage. We have been dynamically changing the charging current which worked better, and Victron stuff recommended to change dynamically also the voltage. It works well for throttling the MPPT when the battery is charged (without export). At the moment few clients complain about the above-mentioned issue when in export. We are happy to solve it quick over the firmware update if we know the best way for it.

0 Likes 0 ·
Daniël Boekel (Victron Energy Staff) avatar image Daniël Boekel (Victron Energy Staff) ♦♦ Kaloyan Dimov commented ·

Hi @Kaloyan

Look at the screenshot, and see what is going wrong, adjust your algorithm until it works well.

0 Likes 0 ·
Kaloyan Dimov avatar image Kaloyan Dimov Daniël Boekel (Victron Energy Staff) ♦♦ commented ·

@Daniël Boekel (Victron Energy Staff) Thanks for this. Our R&D worked 3 days on a solution and it has released a remote software update on our batteries. Works like a dream now, but we had to be very creative. It does a machine learning to find the Victron sweat spot for export via the MPPT.

@Andre du Rand thanks for your training last week


@Derik Alwyn, happy?


0 Likes 0 ·
platar86 avatar image
platar86 answered ·

@Daniël Boekel (Victron Energy Staff)

Hi Daniel, main difference from Lead Acid and Li-Ion is Voltage-Capacity curve is not linear like in Lead Acid:

In order for Lithium battery to keep itself at specific SoC normally require to control the current from charging/discharging system connected to it. As fast the control is SoC can be pinpoint with very small margins. Voltage level controls work Fine for lead acid to point specific SoC. Changing frequently Voltage and Current Does work with most inverter. Please keep in mind that maximum reaction time for frequency shift control is less than second. The messages from normal LiIon Management system are send 1 a second depending of the system performance. When you said: "causing an oscilation " what is the minimum time that voltages must be change?


1562910519783.png (238.1 KiB)
3 comments
2 |3000

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

Hi @platar86

The graph you show there is not a LiFe(Y)PO4 lithium battery, as is the battery in the system.

Also this system is DC-coupled, with MPPT's not AC-PV inverters.

A battery / BMS manufacturer will have to test the algorithms in many different situations,
All this to prevent unwanted effects in all situations (ESS / off-grid / battery full / empty / high load / low load / etc. etc.)

0 Likes 0 ·
platar86 avatar image platar86 Daniël Boekel (Victron Energy Staff) ♦♦ commented ·

So if lithium bms change only the current , will this be enough for victron system to control charge discharge base on maximum current allow send over can? btw this forum keep scrolling to top at mobile at back space pressing, please forward this to the web department.

0 Likes 0 ·
Daniël Boekel (Victron Energy Staff) avatar image Daniël Boekel (Victron Energy Staff) ♦♦ platar86 commented ·

You will have to test what works best in what circumstances.

using voltage and current can work fine, but not in the way that this system was programmed.

0 Likes 0 ·
derik-alwyn avatar image
derik-alwyn answered ·

Thanks for the response.

I set the following values in response to the feed back, but no change in the situation. Hope I understood the suggestion correctly.

Multi absorption 54.00v - Float 53.00V

MPPT absorption 54.40v - Float 54.00V

The status now oscilate between absorption and float every few seconds and it still powers down the mppt to zero watt every few seconds. As a result the grid feed back also oscilates. The problem also exists if I switch of the feedback option to feed the load only. Then the load gets powered between the grid and PV in oscilation.


1 comment
2 |3000

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

That's because the batteries bms is telling the system what voltage to charge, not the Multi settings.


You really have to ask your installer to fix things, the battery is not behaving correctly.

0 Likes 0 ·