question

Chris Wilberg avatar image
Chris Wilberg asked

Discrepancy between solar produced and power consumption

Some preliminary information:

I have several systems but only one with this issue.The affected system consists of a multiplus, bmv-712, two 100/30 charge controllers, and a cerbo gx. The battery is a custom 11kwh lithium iron phosphate battery. The BMS on that battery maintains normal charge level indications. There are no loads or chargers that do not go through the bmv. Peukert exponent is set to 1.

The issue:

In VRM the consumption kwh always shows about 10-20% more than the solar production. Day after day. If this were the case the battery would be slowly depleting, which is what the BMV shows, but that is not the case. The BMS on the battery shows that it is returning to 100% every morning. But the BMV will show it only returning to 99.8%. The next day will be 99.7%. The day after will be 99.6% and so on until I get irritated and reset the SOC in the bmv back to 100%

The system mostly sits idle and there is plenty of excess solar capacity that is not being used. I have tried disconnecting all charging and loads and resetting the current calibration on the bmv.

The more power I use, the higher the difference becomes. On idle days it's usually a difference of .5kwh. On days where I use 5-6 kwh the difference can be as much as 1.5kwh. If I use power after dark the discrepancy will show up the next day.

All of this leads me to believe that one or both charge controllers are actually delivering more power than they report. Is this possible? Is there a way to calibrate them? Shouldn't the bmv be counting the actual incomimg charge anyway? Or am I possibly missing something in the battery setup?

MPPT ControllersBMV Battery Monitor
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.

Chris Wilberg avatar image Chris Wilberg commented ·

Over the last year it shows 191 more kwh used than produced from solar or from the grid. That's almost 20%! As much as I would love a battery that can do this, I know it's not the case. screenshot-20230419-094122.png

0 Likes 0 ·
Matthias Lange - DE avatar image Matthias Lange - DE ♦ commented ·

What grid meter do you use?

Is your grid 1 phase or 3 phase?

Is the Victron system 1 phase or 3 phase?

0 Likes 0 ·
Chris Wilberg avatar image Chris Wilberg Matthias Lange - DE ♦ commented ·
Everything is single phase 120v. The grid charging is done by the multiplus. It's almost never connected to the grid. I grid charged the batteries only 3 or 4 times over the year and don't notice the problem when grid charging. The problem is that the solar shows less production every day than the daily power used when not grid tied.
0 Likes 0 ·
6 Answers
ejrossouw avatar image
ejrossouw answered ·

@Chris Wilberg What does the consumption tab show? The battery consumption is not accounted for on the overview page?

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.

Chris Wilberg avatar image Chris Wilberg commented ·

I guess a better description for the issue I'm having is that it shows more power coming out of the battery than going in to it day after day. Which couldn't possibly be. With losses there should be more power going in than coming out. So this makes the battery show a lower and lower percentage each day even though it is actually full. It also skews the consumption vs solar production numbers in VRM to look like I'm using more power than the solar produces day after day, which again couldn't possibly be the case as 99% of the time I have no power source other than solar.

0 Likes 0 ·
Chris Wilberg avatar image Chris Wilberg commented ·

Thought I put this up earlier but I guess it didn't go ...screenshot-20230419-173936.png

0 Likes 0 ·
ejrossouw avatar image ejrossouw Chris Wilberg commented ·

Below is a system charging the battery and the immersion heating the water from the grid. (top graph two peaks) Below is the grid consumption and charging the battery is not considered as you can see. This applies to charging from solar too. Understanding this, may help with interpreting the dashboard correctly.

1682406046487.png

0 Likes 0 ·
1682406046487.png (276.3 KiB)
astrayan avatar image
astrayan answered ·

Why is it nearly OK in March?

When you consume excess solar that does not go anywhere near the battery, does that show up on Consumption?

2 |3000

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

Chris Wilberg avatar image
Chris Wilberg answered ·

The way I understand it all power used shows up as consumption no matter where the power came from. Even if it's being directly used from solar production. It still sees power being produced but not showing up in the battery so I believe it calculates that as consumption as that's what it shows in real time on the vrm.

2 |3000

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

Chris Wilberg avatar image
Chris Wilberg answered ·

screenshot-20230422-094515.pngscreenshot-20230422-094405.pngscreenshot-20230422-093732.pngscreenshot-20230422-093709.pngThe problem isn't when it's grid connected. It'almost never connected to the grid anyway. Maybe 3-4 days a year. Either the solar chargers are reporting less than they actually produce, or the BMV is overestimating the loads. Day after day the consumption shows more than the solar has produced which is impossible. The battery would deplete. But the battery reaches charged voltage every day and then the solar chargers back down to only cover the loads and stop charging the battery. There is plenty of excess solar so l know it's being fully charged every day. The battery soc always shows higher at the end of the day than the morning so if anything the solar production should show MORE than the consumption due to inherent system losses! I know I could have the BMV set to reset the soc to 100 every time the charged voltage is reached but that would only fix half the problem. I still want the VRM to show a more realistic balance between the power produced and the power used as it is impossible to use more power than is produced, which is what it currently shows.screenshot-20230422-093659.png


2 |3000

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

astrayan avatar image
astrayan answered ·

I was messing around with my 12V system today, trying to figure out a voltage difference between what I set, and what was happening. I set 14.15 on all the MPPTs, but two of the MPPTs were doing 14.20, and the smaller solar panelled one was 14.10. But it occurred to me that if I was a perfectionist and put a battery monitor right on the battery, it would be measuring 13.8V for most of the bulk and absorption charge. The figures wouldnt add up for me, either. I know also, that when I run the 3000VA Phoenix inverter, it displays power consumption that isn't really explainable - displaying proudly VA, which is not really related to Watts - or at least that's their excuse. I would like it if it showed something accurate.

2 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.

astrayan avatar image astrayan commented ·

One thing you could do is get one of those cheap mains power meters and stick it on your output, and see if it's the solar that's out, or the AC.

0 Likes 0 ·
astrayan avatar image astrayan commented ·

Maybe it's a counter-current discrepancy, when consumption and solar occur simultaneously, the battery current gets unaccounted.

0 Likes 0 ·
bigbadbob76 avatar image
bigbadbob76 answered ·

I have found the multiplus power consumption measurement to be pretty inaccurate.

although in my case it seems to read low. maybe yours is reading high.


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.

Chris Wilberg avatar image Chris Wilberg commented ·
Most of the time the multiplus is off. But the discrepancy does seem higher when the multiplus is being used.
0 Likes 0 ·