Updated 22nd April 2015
In 1968 my first job was in the Ministry of Works, Water and Soil Division. I worked on hydrological instrumentation. There was a problem with data collection from the field at the time. Some of it was manual and some of it comprised long records with not a lot happening. Rainfall was a good example. By measuring a sample of rainfall with a tipping bucket rain gauge and logging a digital record of the time when it happened, a very short record was obtained. A patent was obtained. This mode of recording is now standard.
Tipping Bucket Rain Gauge
I have always wanted to have a tipping bucket rain gauge at home. At right you can see my first effort. The bucket is made from a vitamin C tablet container by slitting it and folding the two halves back.
The pivot is spring steel wire. The pivot supports are simply the slotted heads of the two screws. One end of each slot is blocked with epoxy to limit end movement.
The funnels are part of a set of three from the Warehouse for a few dollars. The rims have been cut off and sanded. A wire mesh filter has been fitted to the cut-off base of the larger funnel. A teflon dripper with a 1.5mm hole restricts the maximum flow rate and makes for easy calibration. Either of two funnels made from a plastic report cover catches the water from the bucket, directing it outside.
A magnet is mounted on one side of the bucket with epoxy. I used a part of a magnet from an old hard drive. A magnetically operated reed switch is soldered onto two screw-terminals. The terminals are mounted through the base of the aluminium box.
Some materials used are not corrosion resistant. These parts will be updated if necessary.
An updated version was built recently complete with operating instructions.
The counter used for recording the number of bucket tips is a pedometer obtained for the grand total of $2.50. The mechanism is removed and the wire from the rain gauge is connected in the same place. A 0.1 microfarad capacitor is connected across the terminals and a 220 ohm resistor is connected in series with the wire from the rain gauge. These two components will limit any stray signals from the long antenna now connected to the pedometer.
Recently I have reduced the diameter of the funnel and recalibrated so that each bucket tip now represents rainfall of 1mm.
Thermistors are somewhat non-linear but they can be made nearly linear by connecting a resistor in series. The value of this resistor is made the same as the thermistor at a mid-range temperature. The other end of the thermistor is connected to the positive 3.3V supply and the remaining end of the resistor is connected to ground. The thermistor - resistor junction goes to an analog input.
Calibration is done at two temperatures to get the gradient and at one temperature to get the offset. The calibration temperatures should be near the range of interest.
The thermistor is housed outside, under a screen made from some bent-over white plastic strip. This is covered with a cut-down white medicine bottle. Some ventilation holes are drilled in the bottle to complete the screen.
A second thermistor is painted black and is placed in a small glass bottle. This is exposed at an angle of 45 degrees so that the sun illuminates the bottle from the side. This thermistor heats up more when the sun shines. On a fine sunny day the reading will be approximately double that of the shade thermistor. On a clear night the reading will be lower than the shade thermistor due to heat loss by radiation.