Visualization in emoncms

[UCLteam]

We have issues visualizing the datapoints in emoncms. We know data has been recorded as they appear in inputs and are refreshed every 5s.

However, when we try to add a feed and visualize the data, scales are strange and we don't manage to change them correctly or to see anything.

@Ashure @jmatsushita

[jmatsushita]

Please make sure you're using the very latest committed code. https://github.com/opendroplet/opendroplet-avr/commits/master

This is what a working configuration looks like:

Then use the Vis link to test the realtime feed:

You can then create a dashboard:

[UCLteam]

We didn't know you changed the code again. We will try the new one.

[UCLteam]

For a constant noise:

here is what we see for node 8:2:

and here is what we see for node 8:3:

According to the code, it seems that node 8:3 represents the peak to peak value and node 8:2 represents the spl value. But we have difficulties understanding:

• the physica meanings of these values:
  why don't we just use the amplitude of the sound with "analogRead(MICPIN)"
• how the spl value is calculated:
  from the code, it seems it is an average over a periode of 50 ms of the peak to peak values/10
• these visualizations:
  shouldn't the spl value be constant for a constant sound? Maybe 8:3 is the spl vaue and 8:2 the peak to peak?

[jmatsushita]

Dear @UCLteam,

You are still not using the latest version of the code. You need to git pull and upload the code to the microcontroller again.

In the latest version of the code .2 is the approximative sound level (over 2 secs), and .3 is the latest peak to peak value in the 2 second interval (mainly for debugging).

analogRead(MICPIN) is the instantaneous value of one sample at the sampling frequency (probably around 8KHz). It is the tyinest part of what makes up the sound and is not meaningful for measuring sound. See http://en.wikipedia.org/wiki/Sampling_(signal_processing)

Yes the SPL is not any approximation of actual Sound Pressure Level, it is a basic integral of the signal over time and should be more or less proportional to the sound volume.

The fluctuation is noise (amplified with the integral). There is no such thing as signal without noise when doing electronics.

You can modify the code if you want to measure other values, it only requires a very basic understanding of C and of Sound Sampling.