I started to get the power consumption values in Venus OS by using a Shelly 3EM. But with this grid meter I had to rely on the wifi connection. In addition, I was looking for the fastest reaction time to power changes in my home grid. So I changed to the Victron EM540. It is a nice smart meter. The power value refreshing in Venus OS is very fast, now.
But I have one issue. After the change I have 0,1kw/h power import on some days. It is exactly 0,1 never more or less. I can measure the import with my smart meter, as well.
It is strange behavior. Sometimes there is one week without an import. Next week I have 3 days in one week with import. My grid setpoint is currently set to -100w. A change of the setpoint shows the same behavior. Venus OS is up to date.
I’m not afraid of the costs. But I’d like to understand why it is worse after the installation of the EM540
many loads are pulsing, e.g. switching on/off for short periods of time knows as pulse wide modulation. E.g. an oven at 50% might be 1 minute on, followed by one minute of.
With each change it switches in zero time from 0W to maybe 2000W.
The German grid code (your screenshot is German) allows a grid tied generator to change power at a rate of 400W/s. So the MP2 will need at least 5 seconds to reach the 2000W.
thanks for the quick reply. I know the behavior of the MP2 and the 400W/s. Strange for me is, that it worked better with Shelly grid meter. Additionally, it is always exactly 0,1kW/h. Why is it never more or less? With the Shelly I had different figures.
I have the same issue. 1-2 times per day I feed 0,1kWh into the grid, even without high loads switching. I figured out, that when this happens, the multiplus switches a short time into passthrough mode. My EM540 is connected via Victron USB/RS485 interface. I didn‘t inspect this further yet, since it isn‘t a huge issue for me. My guess is, that there is sometimes a short communication error between the GX and the EM540 or an USB reconnect or something like that.
I don’t think a connection problem is the root cause here.
As Björn already mentioned, this is a great result. The 0,1kWh is the smallest resolution that is displayed. This might be just 0,051kWh that is then rounded up to 0,1kWh. In the month summary you see then the “real” consumption, that is even better.
I have had a similar situation but after playing around with some parameters this changed a little bit.
What I did, - I changed the battery charging do DC only. So the AC power is exported to the grid, as soon as available, what also reduces the grid consumption.
And the already mentioned set grid point to e.g. -10W.
This is something you can test and I’m sure, the grid consumption will change from daily 0,1kWh.
