The Refrigerator Group

Hello, we are the group working on refrigeration!

group_photo.JPGOur problem statement: Many women in the village of San Pablo expressed interest in keeping their fruits, vegetables, and other food items fresh longer. Our mission is to provide the people of San Pablo with simple refrigeration systems that can extend the life of their produce by a significant amount of time. In addition, the refrigeration system will provide a way to keep dairy and meat products fresh.

Our Group Members:
Erin Gibson: I am a third year animal science major, pre-vet. I have very limited experience building anything, but I am inspired and excited to learn! I look forward to helping make a difference for the people of San Pablo. Email:

Michelle Bewersdorff: Hi there! I’m a third year Graphic Communication student.I’m holding the “pneumatic shears” in the photo above. I am in Appropriate Technology because I want to push the boundaries of what I know about other cultures, as well as technology. I’m learning a lot about the people of San Pablo, and gaining valuable technical experience in the process of building our fridge. Email:

Robert Campbell: Hola, yo soy un estudiante de fisicas. Yo visite la cummunidad de San Pablo durante el invierno y estoy muy emocionado para ayudarlos. I am a third year physics major and I am really excited about the partnership that we have with San Pablo and what we are developing with them. Email:

Brecken Thomas: Hello I am a third year Microbiology major!! This class will give me an opportunity to “expand my horizons” and hopefully make a positive impact on the lives of the people in San Pablo! Email:

Current directions: (As of 1-22-11)
Solar Chimney: We are trying to design a cupboard type refrigerator that is very similar in appearance to standard refrigerators. However, it does not use electricity. We are going to use hot air currents generated from an exhaust pipe that comes out the top (Picture to come). As this air moves out the top cooler air will be sucked in from a pipe that runs through the ground. In principle, the cold air is drawn from its high pressure system to the low pressure system created by heating air at the opposite end of the set of piping. This would be a low maintenance system that would work well in the average climate of San Pablo, 10°C.
-Zeer pot: The zeer pot is used in hot desert country, and is a system that uses evaporative cooling. Two pots are set inside one another, and a layer of wet sand is laid between them. The products are placed inside the inner pot, and a damp rag is laid on top of it. The sand and rag must be rewetted 2-3 times a day, and via evaporation, the contents of the pots are kept cool. We decided against this option because we thought that the effort of maintaining this system was greater than the benefit. Additionally, the humidity of San Pablo would decrease the efficiency of an evaporative cooling system.
-Einstein: The Einstein refrigerator is one of the simplest refrigeration systems. It works by heating a circulation of water, which acts as a pump to essentially move butane through a set of tubing that removes heat via evaporative cooling. Butane has a very low boiling point, thus reacts at low temperatures and evaporates, then is re-condensed further along in the system. Though efficient, it requires a heating source and the use of chemicals, which are not ideal.
-Buried: This system utilizes the thermal-insulating properties of the earth in order to create a cooler environment in which to store food.
-Mini fridge: This would be the most ideal system, but the cost and use of electricity outweigh its efficiency and convenience when applied to the environment of San Pablo

Possible directions:
After countless hours of research based off the ideas we have listed above we decided to do a decision matrix to rate which type of refrigeration would be the best. Below are the ratings of the five different types of refrigeration that we considered trying to cocreate in/with San Pablo. Ultimately, the solar chimney received the highest rating.


Construction: 1/24, 1/26
-We began working on our first prototype.

Experiment 1:


To bury our “refrigerator”, a metal box we constructed, and compare the temperature in our box when buried underground to the outside temperature.
This will give us a baseline temperature that we will be attempting to decrease by using better insulation, and by converting it into a passive cooling system as described above.

Thus far, we have encountered problems with drawing a set of plans for sides that will match up when folded together. We discovered this when we were bending our metal box together, and had to unbend it and adjust for our mistakes.

We buried our refrigerator in the ground so that the face of the refrigerator is exposed to the outside environment. A thermometer was installed inside the box, and another thermometer was installed inside of the box. After a night, the outside temperature was seen to be 12*C and the temperature inside the box was 15* C. This result was hypothesized to be a result of the steady temperature of the earth, as our box was metal and thermoconductive. The earth did not reach the cold temperature of the outer air as the earth does not change temperature as easily as the outside air. Knowing this, we have decided to aim for achieving temperatures lower than 15* C, or the temperature that can be achieved by putting a box in the ground.

From this experiment and in order to optimize the Planned Solar Chimney we decided to take some data of the ground temperature over the course of a day. We put temperature probes in the surface of the earth and then one at 1′, one at 2′, and one at 3′. Below is the data that we collected over the course of 48 hours.


  • From this data for the ground temperature, we decided that the temperature was well maintained at depths as shallow as one foot and that even if we went three feet down it did not give us a significantly cooler medium to work with.

Experiment 2:

To construct an experiment to test Pete’s thermal siphon hypothesis.

In order to build a functioning thermal siphon, we must seal a section of tubing that is full of water and no air. This presents an issue of how to do that. We managed to bend aluminum tubing in a loop, hold it under water, and when it was full of water we attached rubber tubing to both ends and clamped them down with pipe clamps. This has proved to be a satisfactory solution

We built two insulated boxes out of styrofoam. One we decided to be a control, and we installed the therm-syphon in the second identical box. The control will act to provide a gauge of temperature change that the thermal siphon may or may not provide to a sealed, insulated environment. The focus of this experiment was on the function of the thermal siphon system, not on the materials we used to construct our “refrigerator” box.

The thermal siphon results were inaccurate due to poor testing conditions, and the damaged experiment. The temperatures taken across several days of both boxes were almost exactly the same between the test box and the control box. There were holes made in the test box on construction that were patched, but it may have effected the insulation capabilities of the box. Additionally, the size of the thermal siphon tubing was likely too small of a surface area to have a significant effect on the temperature of the box. Because of these results, we decided to move in the direction we intended to at the beginning rather than expand on this concept further.


Experiment 3:

To construct and test the solar chimney concept. The general concept can be seen in the sketch to the left. Cool air is drawn into the bottom of the cupboard through a pipe running under the house, with an opening to the surface outside. A second pipe at the top of the cupboard vents to the outside of the house. The external portion of this vent has access to full sun and is painted black. This causes the air inside it to heat and rise, which in turn causes cool air to be sucked in through the bottom pipe. Our idea for our solar chimney was based on the design by permaculture company EcoEdge Design Ltd.

With the lessons learned from the importance of sealing our test box, we used significant amounts of caulking to seal our construction, as well as epoxy to secure our foam box together.

We used the control box from our first experiment as the material for our new test box. We installed ABS black 2.5″ pipe as the chimney and sealed it with caulking. We installed the black rubber plastic pipe in the side of the box and sealed it with caulking. Below is a picture of our model during the construction phase.


Experiment 3 Results, Test 1:

The first test we ran was in a synthesized environment. The ground tubes were placed in a bucket of ice and a set of lights was set on the chimney to warm it. We took temperatures of various sections of the solar chimney over the course of the two hours. It was conclusively shown that the inner temperature of the refrigerator was maintained at a lower temperature than the outside environment. Below are some pictures of us conducting the experiment and below the pictures is a graph of the temperatures that we recorded for the experiment that made us want to move forward with this prototype.



  • From this data you can see that the temperature inside of the refrigerator was consistently lower than the temperature of the outside air temperature even despite the very high temperatures of the chimney. This made us want to move forward and actually test it in the ground.

Experiment 3 Results, Test 2:

We planted our ground pipes at a depth of one foot and set the solar chimney box outside where it could be exposed to sunlight, which would heat the chimney and activate air-flow. Data was taken over the course of three days. We placed one temperature probe in the box, one probe in the chimney, one probe at the depth of the ground pipes, and one probe in open air. The results indicated that our box was maintained at a temperature colder than open air, but warmer than the ground temperature.


Discussion/Future Directions:

The final results of our solar chimney indicate that the mechanism of the solar chimney does work. Many steps can be taken to improve the design, including:

  • Increasing the length of the ground pipes to allow for more cooling time
  • Increase the separation of the refrigerator box from the heated component, the chimney
  • Enhance the insulation
  • Maintain the refrigerator in an environment not exposed to direct sunlight
  • Experiment with different materials of insulation, such as cork or fiberglass.