Small data gaps

Support
1

Hello IoTaWatt community
I have a small issue. I changed from EmonCMS.org to our own server and since that, not all data is recorded/uploaded. See the following picture on our Server: (The data holes are the same if directly connected to the IoTa)

image
image747×419 50.4 KB

These are 3 different IoTaWatts and all 3 have a backup power bank to measure all the time and are connected to the same Wifi. The server runs v10.5.5.
Any help is much appreciated.
Thanks, Daniel

2

Did you forget the picture?

3

My mistake. Updated the question :slight_smile: Thanks

4

I see. So to recap the situation:

You were previously uploading to Emoncms.org and were not experiencing these gaps. The IoTaWatt were using UPS both before and after this issue developed.

The plot that you posted is from Emoncms, not the IoTaWatt.

To explore this further, can you clarify some things:

Two of the voltages appear to be the same source, but about 7 Volts different. Do you know if this is a VT calibration issue or if there is another reason why they would be different?

Why is TESTOPS:1 so consistent?

Can you show a plot for KITCHEN:1 for the same one hour from both the IoTaWatt Graph+ and Emoncms?

Can you post an IoTaWatt message log for that same one hour period?

Are the VTs also connected through the UPS?

5

Thank you very much for this fast reply. I gathered the information below. Please let me know if you need additional information

  • Yes, the IoTaWatts were also using a UPS (these are power banks able to work as UPS)

  • Yes, the plots are from Emoncms on our server.

  • The 7 Volts difference is intentional for now. It’s easier for me to see if data is missing on one of them if they are not the same. That’s why one of the VTs is not yet calibrated.

  • The VT for TESTOPS is on the output of a solar system which is very consistent in comparison to the Ugandan grid on the other VTs

  • The Graph+ doesn’t show anything but here is the graph from the IoTaWatt

    image
    image869×510 51.1 KB

  • And here is the graph from Emoncms for the same time:

    image
    image739×417 40.7 KB

There is no specific message for the period shown in the graph (25/06/2021 9:55 to 10:25). I made a txt document here with the messages from the 24/06/2021 to 26/06/2021

6

My initial impression is that this is being caused by the poor WiFi strength. Your RSSI is -87. Good WiFi is typically in the -60s or better. RSSI alone is not the determining factor. Other WiFi APs on the same channel, as well as other 2.4MHz devices, can cause problems.

The IP firmware in the IoTaWatt is an open lwip layer and is not anything I can modify or improve, even if I knew how to do so. My experience is that when RSSI is weak or there is other heavy interference, the lwip seems to block and lose heap memory. I see that in your system as it is restarting at about 10 minute intervals because of degraded heap storage.

When WiFi is strong, this does not happen and the IoTaWatt will run for up to six weeks when it does a routine restart.

Is there anything you can do to improve your WiFi?

7

Thanks for the fast reply. Indeed this Wifi has many devices connected.
Let me test another Wifi network and
see if it improves.