Building a prototype Clapp-Gouriet Oscillator

Building a prototype Clapp-Gouriet Oscillator

Soil Moisture Sensor Project

Making a first prototype Clapp-Gouriet Oscillator

It wasn’t too hard to build a working Clapp oscillator. I had hoped that I could use it to measure soil moisture levels.

Clapp Oscillator on breadboard
Clapp Oscillator on breadboard

I didn’t explain the principles behind a Clapp Oscillator in the post where I explained why I chose this design. In short, the Clapp Oscillator places an extra capacitor in the inductor-capacitor tank circuit that LC oscillators use. If you change the value of this extra capacitor which is connected in series with the inductor, you change the total capacitance of the resonant tank. This changes the frequency that the oscillator produces. When the capacitance increases, the frequency decreases. The value of the extra capacitor in the Clapp Oscillator can be varied, changing the output frequency.

You may be wondering why you can’t just change the value of one of the other capacitors instead of having to add another one (and I did wonder about this). If you have looked at a Clapp Oscillator circuit, and understand complex schematics) you will have noticed that some of the output of the transistor is being fed back into the tank circuit in between the capacitors. This feedback is what keeps the oscillations going. If you change the values of either of the capacitors connected to the transistor, you change the amount of feedback to the circuit. This could be bad.

The buildup begins…

Breadboards are a great way to quickly develop a prototype circuit, but they won’t work at high frequencies. That site will tell you that they don’t work, but not why they don’t work. Breadboards are made up of parallel strips of metal that conduct the electricity along each row. Each parallel strip of metal is effectively a capacitor. This is not a problem when you are following Dick Smith’s Fun Way into Electronics Volumes 1 to 3. It quickly becomes a problem for a low voltage circuit that is creating pulses that switch on and off again millions of times per second. The capacitance created by the parallel strips of metal interferes with the high frequency signals. I don’t really know how, it just does. As an example, the simulation of this circuit gave an output of about 150MHz. On the breadboard I got less than half that figure.

Nevertheless, it was time to test how the circuit reacted to water.

I used a design by Acolomitchi for the sensor (their write up and links to the design files are here).

Acolomitchi sensor design
Acolomitchi sensor design

Printed circuit boards are not waterproof, but in doing some background research I came across this…. warning: “SEARCHING FOR THE BEST INSULATION FOR THE FOOT PART, It took many weeks.”

Luckily, the author of that foreboding statement had written about all the methods they had tried and rejected. And they had written about the method they used to make the circuit boards waterproof – epoxy glue. To find epoxy glue you’ll need to head to your local hardware store (e.g. Bunnings – like Spotlight, just for building stuff). You may be able to find it at your local art and craft store (e.g. Spotlight – like Bunnings, but for art and craft).

I also wanted to test smaller sensors to see how how small I could make the tuning capacitor. Background: the smaller the capacitance, the faster the oscillator would run.

Cut down Acolomitchi sensors
Cut down Acolomitchi sensors

After cutting the sensors down to size and coating them in epoxy, it was finally time to test.

And now the testing begins….

Using a full-size sensor (Acolomitchi’s design) connected to the breadboard:

  • In air: 60MHz
  • In water: 45MHz

This seems strangely… successful.

Brendan’s thoughts on designing this project (art vs science):

Dark Arts ———-(*)—————————————————————– Science

Not so dark, but definitely precious.


Go back to the main page.

One Reply to “Building a prototype Clapp-Gouriet Oscillator”

  1. Some remarks about the insulation of the sensor board. Electronic boards usually are made from a material like FR-4, which are glass-reinforced epoxy laminate sheets. This material is almost perfectly water proof (also the solder mask and copper). The real problem are all the bacteria and organism in the soil. This organism seem to like the material of the solder mask. After some months, the bare copper comes into contact with the soil which also dissolves and in “large” amounts it can have a negative effect on the plants.

    Enclosing the whole electronic (including the wires) into epoxy is a very good solution, but you have to choose the right epoxy/hardener combination. In many epoxy glues, there are additives to keep the final material flexible (to improve the adhesive effect and make it resistant against micro fissures). This is definitely not something you want into the sealing layer for the sensor. Best is to choose an epoxy/hardener combination which is used for building ships. Like Resin L with Hardener S or L from R&G, or any other epoxy made for ship building.

    Be very careful with epoxy, it is a very dangerous material. Most hardeners go easily through any gloves (even nitrile ones) and directly though the skin. Also mixed epoxy produces a lot of heat, too large amounts can self ignite and start to burn.

Leave a Reply

Your email address will not be published. Required fields are marked *