This page describes a simple and practical design of a Peltier-cooled cloud chamber.

Considering the average blog-reader attention span (no offense!), I'll start from the end. Here is the finished thing, undergoing a cat scan:

Now, back to what the thing actually is.

The cloud chamber is a basic ionizing radiation particle detector. It's a container with supersaturated vapor. A charged particle passing through will produce ions that act as condensation nuclei, forming a visible trail of droplets. This virtual cloud chamber (click to see) gives you a rough idea of you what you can expect to see in a cloud chamber.

A typical home-made cloud chamber is often shown like this:

The top plate of the container is warm (or, at least, not-cooled) and has felt lining soaked in alcohol to keep the vapor close to the saturation point. The lower plate is typically cooled with dry ice and provides temperature gradient inside the container. The supersaturated vapor close to the lower plate is the zone sensitive to ionizing particles. Side illumination makes the condensation trails stand against the dark background of the lower plate.

Although this is a simple design, it has one disadvantage - you need to buy dry ice each time you use it. This gets old pretty fast. There is an alternative to dry ice - using thermoelectic (Peltier) coolers. This way one can experience the joys of ionizing radiation at any time by the click of a switch.

There are a number of Peltier-cooled cloud chamber designs online. Some work well, some don't, some are easy to build, others are (often unnecessarily) complicated. I tried a few different things, threw away ideas that did not work, and picked up something that is simple and convenient to use (because I'm lazy), easy to build (because I'm impatient), and is made mostly from scrap parts (because I'm stingy).

So, starting with...

The power supply

An old PC power supply is easy to find, has 12V and 5V, and works just fine for the purpose:

Keep in mind that the power supply will not turn on unless the PS_ON pin (see pinout) is grounded. If you have trouble with this, check this instructable.

The thermoelectric coolers

Nowadays those are quite cheap - I got a set of five TEC1-12706 from Amazon:

To test them I'd recommend plugging them to 5V (red and black wires of the cooler to red and black of the power supply). I tried 12V first and burned my hand on the hot side.

Cold plate

After playing with different combinations and number of coolers, I found that two Peltier coolers on top of one another work OK. Top cooler is connected to the 5V output of the power supply, the bottom cooler is running at 12V, stacked as shown below:

Make sure to stack the Peltier coolers with the cold side on top. To improve efficiency, use thermal paste at all junctions of the stack.

The "plate" is an endcap from a cylindrical Pillsbury dough box.

I was surprised by how much cold is lost (technically - "how much heat is gained", but who cares!) if the cold plate is not properly insulated. Putting some insulating foam around the edges of the cold plate makes for faster cooling and significantly lower temperatures:

Spray-painting the bottom of the dough box endcap a dark color provides a nice contrast for the vapor trails. I experimented with felt-covered endcap, but flat dark paint turned out to work better.

The assembled cold plate is shown below:

The cloud chamber container...

... is constructed of a piece of thin plastic (cut from a soda bottle) that fits into the groove of the aforementioned Pillsbury dough box endcap, a strip of foam, and a teacup glass plate: The foam ring soaked with 90% ethyl alcohol is the vapor supply. It also reduces the air currents that could sneak around the glass plate, and - most important - leaves a clear observation window in the center.

Assembled device

The assembled cloud chamber is shown below:

The best method I found for illuminating the cloud chamber was to use a powerful bicycle light. Make sure the light is set to illuminate the lower part of the chamber (where the super-saturated layer forms) and is shining parallel to the cold plate:

At this angle, the sides of the container do not provide the best view. Not only they are made of cheap plastic that blurs the view, but when the chamber is in operation, the sides will get fogged with condensation quite fast. This is the main reason for using transparent top. The view through the center of the foam ring is excellent:


Initially I was toying with the idea to have the heatsink air-cooled by a fan. This didn't work very well - the fan was effective only when the heatsink was at around 40-45°C, which was pulling the cold side up to about 0°C. The solution was to forget about the fan and dip the heatsink in water:

Ice cubes are optional - they just extend the length of time you can use the cloud chamber without changing the water. If you can pipe cold water continuously through the container, you don't need ice at all.

A few seconds after powering up the contraption, the temperature of the cold plate drops to respectable minus 23°C ...

... and, in a few minutes, ice forms on the cold plate:

The results

To use, drip some ethyl alcohol (90% works good) on the foam ring and close the chamber. Turn on the side light and darken the room. Power up and in a few seconds you should start to see condensation droplets forming in the air just above the cold plate. Then - wait for magic to happen:

Other projects

See my other projects here.