Random usage spikes in PVOutput

Below is the output from Solar.Web:


Looking at the PVOutput data over the same interval looks a little different (screen shot taken at 09:05 local time):

I can do a live csv upload to correct the issue:

But the next hourly log push ends up putting the spikes come back again.
The odd thing is that when I saw the random spikes at 07:10, there were only three, and I fixed then with a live csv upload. I checked again at 08:05 after the hourly push, and there were now four spikes. I fixed them again with a live csv upload. As can be seen from the above screenshot taken at 09:05, the spikes return again with the hourly log push.

Instant vs average maybe?

Definitely not as there is no usage above 180 watts during the period of the spikes. They are some random ghost event only pushed by the hourly log and not seen when viewed in Solar.Web.

So its the fronius push sending the spikes? Can you check the data its sending

The below is from the latest push update after 16:00:


There is now two additional spikes and the times of two of the spikes have changed as well. Where the data is appearing from I do not know but it is definitely not there in Solar.Web

Hi Decibel,

I’ve got the same problem this morning, random spikes around 35KWh at the same sort of time and missing consumption data from 2am to 2.55am. See below.

Weird eh?

ok Decibel i would rearly like to know how many watts you can ger out of the MPPT with the 14 panels on as i was told they only went 3000 watts each on the 5 KW inverter Thanks Jim

Wired as 7x2 in parallel in one MPPT and 6 into the other MPPT.
I can clean the spikes out tomorrow with a “live csv” upload for just the six points using just the time and actual consumption value and force a recalculate, but there is no issue with my Solar.Web data which is the strange thing. Un-noticed till now, just like “cammo” (well spotted, I missed that), I also show missing data between 01:55 and 03:00:

There is definitely no missing data in Solar.Web over the same interval:

Same for me Decibel, no missing data in that 01:55 to 03:00 period from Solar.Web

Did you swap over to Daylight Saving this morning? I am in WA so there is no daylight saving for my system.

I’m in Qld so no daylight saving here for my system either.

It could be a daylight saving thing on the PVOutput end. Will wait and see if others see the same, and see what tomorrow morning’s data looks like.

ok Decibel i would rearly like to know how many watts you can ger out of the MPPT with the 14 panels on as i was told they only went 3000 watts each on the 5 KW inverter and the most i have ever seen out of mine is 2940 watts i am also in W A Thanks Jim

It may be a fronius firmware bug, since log energy data reported is converted to power resulting in the spikes.

To capture the fronius data, you can add another log push to /service/r2/froniuslog.jsp instead of /service/r2/froniuspost.jsp

Thanks bankstown bloke. Could I check if you mean adding a second Push to PVOutput like below but changing the “Upload file name”?

The maximum wattage out of my 14 panel MPPT today was 3711 watts. Last month on a 37kWh day it peaked at 3803 watts, on a 40 kWh plus day it peaked at 3966 watts

It is all back to normal again today bankstownbloke. I have had no issue with the addition of the hourly log data push till now once the teething problems were sorted. It must be related to the “missing” one hour of data seen by both camo and myself in PVOutput, which is not seen when looking at the data in SolarWeb. I am in WA and cammo is in QLD. I wonder if it could be a daylight saving issue as this was implemented in a number of States yesterday, but not in WA or QLD

daylight savings.

is your time consistent with your actual time now since you don’t have daylight savings? or is pvoutput showing 1 hr ahead of what you are?

PVOutput is showing the correct time for me here in QLD.

Decibel, what’s the best method for fixing that missing hour of data and big spikes yesterday? Manual editing? Or is there a way to export a few hours of SolarWeb data to fix it?

This would be an additional “Push to PVOutput - Logdata” push.

Instead of “froniuspost.jsp” use “froniuslog.jsp” the rest of the settings should be the same. There will now be 4 pushes in total.

This will allow the raw data to be captured to investigate the cause of the spike. Please disable once data with issues has been pushed.