Can you post an example of what the entry should look like in the Push service?
1 Like
Confirming the screenshot by @blackdesk is correct.
Note this is an additional push. There should be three pushes -
- SolarAPI v1 - CurrentData - Inverter
- SolarAPI v1 - CurrentData - Meter
- SolarAPI v1 - LogData - Data
Thanks Blackdesk. Just what I needed. It was not clear that this was a selectable item from the drop down list of “Data Format”. I have now added the extra push and will wait and see how it goes over the next few days. Unfortunately the day I added the extra push was the worst day of power generation for over 15 months (1.2kW)
There might be a bug with the logdata though, see https://www.pvoutput.org/intraday.jsp?id=66252&sid=58912&dt=20180702
Around 5pm you can see a phantom jump in generation. That’s about when I enabled the logdata push so maybe it’s double counting? It also affects the consumption.
Having a look at the energy jump in the 5-minutes, it would be not possible for that high value to be calculated from power unless the energy value was sent by the inverter push (CurrentData - Inverter).
It could either be a coincidence that this happened the moment LogData push was added or the Logdata push caused issues with the DAY_ENERGY energy reported by Fronius.
Can you add &nodayenergy=1 to the inverter push config? This tells PVOutput to ignore the energy sent by Fronius and calculate it from power.
This issue is described in the thread below -
It’s odd that it coincided with the settings change on my side and I’ve not had any problems before. The “power used” also had a large spike that slid along the chart every hour finally stopping at 11pm.
Today’s metrics seem fine so far, so I might hold off on changing the nodayenergy flag as I’ve not had any problems before.
Is it possible through the UI to edit out those false reading?
Actually, no, the data for today looks weird as well compared to solarweb. I’ll add nodayenergy=1 when I get a chance at home.
Is it recommended to add the logdata push if the other inverter and meter push are mostly reliable? Right now it seems like it’s causing more issues for me.
If you’re not having any issues with the current setup (no gaps or 0W) then the logdata push isn’t necessary.
2018-07-02 has been fixed, the Live Loader can be used to correct the energy values.
https://pvoutput.org/liveload.jsp
The following sample data resets the energy back to 16920Wh from 17:10 to 17:20
17:10,16920
17:15,16920
17:20,16920
Load Configuration
Thanks for the tip to look at the Live Loader. I’ve exported the data from Solar Web and hopefully mapped the values correctly.
However, if the power goes off or your WLAN link fails for (less than?) an hour, I assume this Push edit will fill in the gap?
I added the new Push instruction this afternoon before reading your suggestion and it’s working fine. No spikes like those mentioned above, although the Fronius Solarweb server did fail early this evening (for me) but all the PVO Push data continued uninterrupted.
The log data contains the cumulative entire’s day log, as long as the connection is back up by the last hour of the day (23:00 for meters) and sundown for inverter only, all gaps will be filled.
This is similar to solar web receives the data, although they seem to get additional pushes if you happen to be offline for a few days.
The log data format was tested with a gross and net meter push and working fine. It maybe the firmware version that’s producing some undesired results for @blackdesk.
But if yours is working then its best to use it since it will help with the many connection issues.
1 Like
I’m fairly certain I’m running the latest firmware on both the data logger (3.10.2-1) and the inverter (fro28500.upd). Is there anything else I can do to help try and diagnose the issue if there is one?
Please send the log data to froniuslog.jsp instead of froniuspush.jsp this will capture the raw data for analysis.
Thanks for helping and the inconvenience this is causing.
My frequent connection issues are all in the past with a LAN cable installed a few months ago, but this will be welcome for any other outages … thanks!
(Which reminds me; my system installer came back 2 years after the install and wired in the LAN, but I’ve yet to see a bill. I think I’d better call and remind him … best kept on-side 
1 Like
Done. Happy to help.
This was quite useful, the problem was the system has two inverters, and the service only took into account one inverter.
The post on the 02 July where the value jumped to 30kWh as reported by current inverter push was correct. Energy values before 5pm only used one inverter from the log data push.
The problem is now fixed and all inverters are added from log data. Today’s log data 04 July should now produce the following graph.
Ah, yes, sorry should have mentioned the setup I had with one data logger and two inverters. The chart you’ve got there matches what was recorded without the logdata so it looks good.
I’ve flipped the settings back to froniuspush.php and we’ll see how it goes. Thanks.
Seems there might a problem with time zones? After the logdata push the power generated drops to 0 at 10am then starts again at 16:50 (about 10 hours after it should). Then there’s data for tomorrow loaded somehow too.