Hi there Weather Fans!
I have successfully assembled and downloaded the SParkFun sketch for my SparkFun Arduino IoT Weather Station
KIT-22636 with ESP32 processor:
From the Spkfn_Weather_Station.ino file:
/*
Sketch generated by the Arduino IoT Cloud Thing “Spkfn_Weather_Station”
I have an Arduino Cloud Entry account.
I added a clear front enclosure and a weather resistant vent to the enclosure.
I have made a dashboard and included the parameters from the Thing successfully. All appears to be working wirelessly and I have it powered through the USB connector.
Most of the functions appear to work correctly on the dashboard I created.
However there are a few discrepancies:
- Rain gauge in the dashboard is wildly off compared to an analog rain collector placed nearby:
Analog rain collected: 4.75"
Dashboard rain display for the same period: 64.8"
Error factor: = 64.8/4.75 = 13.6x
I did some research and verified that the sketch code does allow for switch de-bounce at 100mS, which was sited as a possible error source in a previous post: (0.2794mm or 0.011in per tip as seen elsewhere as well):
// The rainfall detector contains a small cup that collects rain water. When
// the cup fills, the water is dumped and the total rainfall is incremented
// by some value. This value defaults to 0.2794mm of rain per count, as
// specified by the datasheet
calibrationParams.mmPerRainfallCount = 0.2794;
// The rainfall detector switch can sometimes bounce, causing multiple extra
// triggers. This input is debounced by ignoring extra triggers within a
// time window, which defaults to 100ms
calibrationParams.minMillisPerRainfall = 100;
So what gives?
Do I need to increase the de-bounce time from 100ms to something else?
It seems like a lot to attribute to this error…
Do I need to adjust the calibration factor in the sketch?
Seems unlikely that it would be off by 13x…
Is there some other SW or Sketch error to be aware of?
Second Issue:
- The humidity sensor never shows more that 64% even in a downpour when it should be close to 100% as verified by an analog gauge
Is there a calibration factor to adjust for this?
the enclosure is vented, but could this be an error factor?
Third issue:
- The UV index always reads 0
dashboard uses uvIndex Thing variable
Is there something I need to add to the sketch or setup to make this work correctly?
Thanks very much for any help you can provide on these 3 issues.