Automated Irrigation Control by Dimension Change of Wood

Wood will expand, when it gets wet and shrink, when it dries. This will be known by nearly everyone.

I had been wondering, if this effect could be used to control irrigation of plants in pots. I am showing here a version which has been working for more than one summer. Earlier versions had been made of metal and Pine wood, see last picture. These also worked for at least one summer but Pine wood is not so stable and steel bolts had corrosion effects.

So here we use a block of Douglas Fir and printed parts of ABS together with a Silicone rubber hose

What is the effect?

A dry block 60x18x25 of Douglas Fir of length 60mm (parallel to the circumference of the tree) will grow by 0.6mm within about 20 minutes, reaching 0.8mm after 40 minutes, when it gets wet. To dry and shrink will take several hours.

0.6mm travel is enough to pinch a small rubber hose to close it tightly when wet and open after drying.

The diagram shows data from a block of Pine wood. Travel here was larger than measured with Douglas Fir. Fast growth within minutes and slow shrinking over 30h can be seen. Duration of drying will depend from air temperature and humidity.

The time needed to grow and to shrink can enable some sort of oscillation, which is desirable to prevent midges to breed. In other cases the system acts as level controller.

Experiences

As said the system has been working. Of course I would recommend to try it (and any other automatic irrigation method) some weeks before you go into your holidays to be able to adjust and learn to know.

I must say that I am a little bit surprised, that the growing and shrinking of the wood works so often even with our water here which leads to limestone scale. I have some replacement blocks, but did not need them during one summer. The travel will be reduced though: A 60mm block of Douglas Fir, that had been used during the last summer did now grow after wetting by 0.47mm after 50min.

The photograph above shows the item in the bottom plate of a large flat pot. You also can see one of several wicks, which are used here to have more space for the water and for the controller. The bottom plate is dry in the picture, but this system is actively in use these days. The wood is still dark, we are in the drying phase of the cycle. In this case the water supply comes from the bottom, which is possible, if the plants have roots in this region. It worked like this with Physalis, Tomatoes, small Sunflowers and Herbes like Thyme.

I also have tried to use microcontrollers to measure soil humidity, but reliability was rather low, because varying contact between soil and sensor. The Blumat drip system however seems to have similar reliability as the method shown here.

If you are trying this, it would be nice to leave some note about your experiences!

Have Fun! Christof

1. Wood Block Douglas Fir 60x18x25 mm 60mm in direction of tree rings
2. Bolt ABS, printed
3. Nut ABS, printed
4. Washer metal 10x18 mm
5. Female Part ABS, printed
6. Hose Silicone rubber 5x3mm

The bolt is printed from ABS lying horizontally to have good strength. The thread of the bolt is a 10x2.0mm special coarse metric one. At the bottom the thread might need some sanding.

The thread of the nut is 10.4x2.0mm.

The female part (blue in the 3D picture) has a rectangular opening for the head of the bolt for easy movement and acts to squeeze the hose. Also it is useful to prevent rotation of the bolt, when adjusting.

The wooden block is cut from a plank of Douglas Fir of width 18mm. This type of wood can withstand to get wet better than other types of wood and will rot less easily. My experience is, that it will work for at least one summer. As the dimensional change is much dependant of direction, it is important, that the long side is in parallel to the tree rings.

There is one bore 12mm along the length for the bolt with plenty of play and two bores 9mm to help to dry.

The parts are assembled according to the picture.

The hose must be filled with water. A few cm of height difference from the water supply are sufficient.

The wooden block is wetted, wait 30 minutes, then the nut is tightened until the hose is just blocked completely.

The photograph shows an item, which has been in use for the last summer.

Of course the basic principle of the "system" can be varied. The picture shows older metal parts, which also worked. Length was a bit less with 55mm. Diameter of the bolt 6mm. The nut was made of aluminium.

The block shown is fresh and made of pine wood.

The mix of of metals lead to corrosion and a screwdriver was needed for adjustment. The female part is a metal ring, which tends to shear the rubber hose.