# Experiment 01 - Sensor Calibration

## Purpose

The purpose of experiment 1 is to map the analog output values of the hydration
sensor to different levels of water concentration in soil. This mapping will be
necessary to adjust the user experience of the final device.

## Components and equipment

- Arduino board to run code
- Three male to male jumper cables
- One moisture sensor
- A cup of soil
- Water

## Methodology

1. Prepare the sensor. The power input should come from the 5V output of the
   board. This is important since different voltage inputs to the sensor will
   alter the obtained measurements for an identical humidity.
2. Prepare the code. Three versions of the code will be used raw, slightly
   smoothed and heavily smoothed.
    - Raw version outputs the sensor's measurement to the serial port in
      real-time, unfiltered.
    - The slightly smoothed version will take 10 measurements with 50ms of
      delay between each and return the mean.
    - The heavily smoothe version will take 100 measurements with 50ms of delay
      between each and return the mean.
3. Prepare the soil
    - In a small recipient (weight: 80g), place 200g of soil.
    - Let it dry for X days (collection on 24/10/2022) to obtain the driest
      possible soil.
4. The next steps will be performed repeteadly by adding 10ml of water to the
   soil at each step. Start by first measuring the completely dry-soil. Ensure
   that the water is distributed uniformly across the soil. For each amount of
   water up to a max of 200ml, and for each version of the code:
    - Obtain measurements from the sensor. If the measurement is unstable,
      record the observed range and the most frequent value.
5. Finally, obtain measurements in a glass of pure water, with no soil at all.

## Results

See file `exp01_sensor_calibration.ods`.

## Conclusions

The results where a bit dissapointing. The sensor signal was clearly moving
when the dry soil received the first 10ml of water. But, from that point on,
the next additions of water barely modified the returned signal. Eventually,
there was no significant change in the sensor values. This is showing a
sensitivity way below what I was expecting.

I have been reading about some modifications in the circuit to improve the
accuracy of the sensor. I'll design another experiment to try them out.