question

nick-a avatar image
nick-a asked

double power usage - problem with system

Hi,


I just upgraded my raspberry with the newest developer version.

something is not right:


I use the following:

bluesolar charger MPPT 75/15

Phoenix inverter 12V 500 VA 230V

BMV-700

21 AH 12V LiFePO4 battery

raspberry firmware: v2.60~35

all Victron devices are attached to the raspi using way too expensive Victron usb dongles.

all firmwares are the latest of today (22 july 2020)


I attached a 200 Watt non-inductive load (Infrared philips lamp) to the inverter.

According to my AC load on the display of the raspi it uses an ac load of 240W. That's seems ok- altough a bit high (perhaps the Philips ir lamp uses more than its rated value).


Problem:

The soc goes down a lot faster than anticipated. In addition, the output of the VRM portal shows a load of 496W.


Something is not correct.


Anyone any idea what to do?


MPPT ControllersBMV Battery MonitorPhoenix InverterRaspberry Pi
victron1.jpg (130.7 KiB)
2 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.

Izak (Victron Energy Staff) avatar image Izak (Victron Energy Staff) ♦♦ commented ·

Hi @Nick_A,

You mention three concerns, 1) SOC drops faster than expected, 2) Power appears to be higher than expected, 3) The consumption on VRM is double what it should be.

Regarding the SOC, this is measured by a BMV and if it is wrong it is likely that the BMV is configured incorrectly. I noticed that you had a DC load of approximately 20A for the hour between 21:45 and 22:45, and during that time the SOC dropped by almost 50%. That seems to indicate that the BMV is configured to expect a 40Ah battery, which is in accordance with your answer below.

Regarding the power use: The lamp most likely doesn't use exactly 200W, and the inverter is also not a precision instrument, so I m not concerned by the slight discrepancy.

Regarding the 496W consumption: You're right. This is a bug. You will note on the Overview that the DC Power block shows a load of 244W. The DC Power is an estimate that is calculated by subtracting the current used by the inverter from the total battery current, and multiplying by the voltage. The difference between the two must logically be used by other DC loads (if any). This is only done if you turn on "Has DC system" under Settings -> System Setup. One way to work around it is to disable that option (unless you have actual DC loads).

The bug is that we don't correctly subtract the current of VE.Direct based Phoenix inverters, so the DC Power is wrong (it should be closer to zero). Once the total consumption is calculated by adding the two, the number ends up double.

We'll fix this problem in a future release. Thank you for reporting.

0 Likes 0 ·
nick-a avatar image nick-a Izak (Victron Energy Staff) ♦♦ commented ·

Hi Izak,


Many thanks for you reply. I really apprecitate it. Yes, I do have a small dc load - a raspberry and a usb power delivery hub.


I'll wait unit the next update arrives. Oh, about updating to the next beta. It is a pain in the ass, since I have to reconfigure cq install other stuff like display, touchscreen, brightness of the display. I've seen scripts here and there to alleviate these problems, but could it not be easier?


Cheers,


Nick

0 Likes 0 ·
5 Answers
Kevin Windrem avatar image
Kevin Windrem answered ·

The numbers in the main page shapshot don't add up

~-240 watts out of the battery

~250 watts of AC load

AND

~240 watts of DC load

The first two roughly cancel but the AC and DC loads add up to about what VRM says is the consumption.

If you don't have any DC loads, I'd turn off the "Has DC System" to see of power numbers fall in line.

Not at all sure about the SOC numbers. They seem very low. A 200 watt load should deplete a 21 AH 12 volt battery in well under an hour. The SOC curve looks more like a 100 AH 12 V battery given a 200 watt load.

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.

nick-a avatar image
nick-a answered ·

Hi Kevin,


I made a mistake. The battery is a 42 AH lifepo4.

Yes, there is a dc load of about 4 Watts (raspi).

It is a bug then. The system adds the dc and the ac load together.


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.

muddy1 avatar image
muddy1 answered ·

@nick-a

Presumably this problem still remains to be sorted. I have a Multiplus II 48v 5KVA. When running but with the output isolated it reports 53.53v and -0.7a, giving 37.47W standby power.

Clearly greater than the specified 21w standby power. The reported current is incorrect, meter gives 397mA which would give the correct 21W standby power. It is disappointing to find such a large discrepancy on a quality product.

This causes problems with negative DC currents which have been discussed on numerous posts.

Is there any way to zero the current meter in this unit ?

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.

amirb avatar image
amirb answered ·

Hello ,

My setup is : smart shunt , mppt smart solar 100/30 phonix 24/1200va 230v ve direct and rpi runing venus 2.66

I disabled the "have dc" and disconnected all dc loads .

The only dc load is just the inverter !

I have similar promblem that is described in this post .

The ac power reported at the output of the inverter is much greater then the input power to the inverter wich is reported by the smartshunt .

This is impossible !

Even if the inverter was ideal with 100% efficiency then the Dc power that is drawen from the battery was equal to the power consumed by the ac load .

But the power consumed by load is way higer .


screenshot-20210820-215144-chrome.jpg

Here we can see 31w is drawn from the baterry this is correct it is 10w self consumption of the invetter + 21w small ac loads like laptop . How ever thes ac loads are not reported and it shows 0W . When a larger ac load is added e.g. the refrigerator kicks in then as you can see here the power from the battery is 128 w which is correct ( 31w+ small refrigerator) while the ac load is way wrong and shows 184w
screenshot-20210820-222600-chrome.jpg

screenshot-20210820-222600-chrome.jpgon top of that according to vrm the inverter is generating power ! It shows negative current ! I hope that these bugs will be addressed and solved , i consider victoron products as premium products and this is really dissaponting .



screenshot-20210820-215144-chrome.jpg (443.9 KiB)
screenshot-20210820-222819-chrome.jpg (176.8 KiB)
screenshot-20210820-222600-chrome.jpg (451.1 KiB)
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.

hut43 avatar image
hut43 answered ·

Hi,

I'm seeing the same issue (DC consumption reporting much too high, goes away when inverter is disconnected). Any prospect of a fix?


my off-grid setup is:

-MPPT 100/50

-Phoenix 800

-BMV-702

-170AH LifPo4 battery

-Venus on RPI 3+ connected with 3 VEtoUSB cables


edit: the root cause seems not so simple…I’m now seeing more sensible figures reported.

1 comment
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.

Al avatar image Al commented ·
I'm seeing strange reporting after Venus update yesterday to 2.73.


Seemingly roughly double the consumption in Kwh on System Overview Historical data chart when it's dark and then almost zero consumption when the sun is charging the batteries. But the live System overview shows the expected data, is this an old bug resurfaced?

Happy for Victron remote if it helps.

Al

0 Likes 0 ·