Westfalia

Fridge Chill Booster

The Dometic ^TM 3-way fridge in Westfalias gets a lot of criticism for its efficiency, or lack of it. However, it must be remembered that the system works on boiling liquid ammonia, water and hydrogen under pressure, then separating each of them back into gases at the correct point in the circuit to transfer heat out of the fridge to the coils at the back, and then have that heat carried away by the surrounding air.
Despite such a complex process it is able to keep the interior of the fridge about 20^o Celsius below the ambient temperature. However, when ambient reaches 38^oC, then 18 degrees inside the fridge does not sound so good. Technically it could be called a "cooler" or "chiller" rather than a "refrigerator". As well, the heat whether from propane, 12V DC or 110V AC has to raise that chemical mixture to its boiling point before cooling can begin, so it is rather slow getting the chilling effect started.
Many modifications have been done to try to improve the efficiency, and here is one more.
The plan is to mount a Peltier unit (aka thermoelectric cooler or TEC) on the top of the Dometic fridge to assist or "boost" the regular ammonia cycle cooling.

Click on the image to see it full size.

The Testing Stage

Test Setup #1 To see whether it was worth cutting holes in my fridge, I first had to test the efficiency of the Peltier unit. For this I took the regular fridge door off, then cut a piece of 3/8" plywood big enough to cover the door opening in the fridge cabinet. I had a spare set of hinges, so mounted them on the plywood.
For insulation, I lined the inside of the door with Reflectix^TM insulation foil. Then I fired up the propane and compared the internal and ambient temperatures until the fridge reached its coolest point.

Propane Cooling Only

On propane, the fridge took about 1 hour to start cooling seriously and didn't reach its coolest for 3 hours.
This fridge is not my own, so does not have any of the other mods I have installed in mine, such as internal fan, external ducting and fan, and venting to the outside of the vehicle. Therefore it does not seem as efficient as mine. The temperature reached 4.8^oC, only 16.2 below ambient. Mine easily reaches 20 - 22^o below ambient. To be fair to this fridge though, it did not have an interior circulating fan for this test.

The Peltier Unit

Peltier units, or thermoelectric coolers, use DC current, usually 12 volts, passing through a solid state device to transfer heat from one side to the other. These are the devices used in 12V "plug in coolers".
This was scrounged from an office water cooler, so did not have cooling fins on the "cold" side.
This unit is rated at 3.8A, 12V, or about 45 watts but is being run on 13.26V here, which would make it about 50W, the average of those I have seen in camping coolers. On its own in a small insulated space it brought the "cool" side heat sink down to 25^oC below ambient temperature.

Test Setup #2 I took the door off and let the fridge warm up to room temperature again for the next test while I cut a hole in the door. The hole is to fit the heatsink stuck on the cold side of the Peltier unit. I have no idea what its efficiency is in transferring heat, but the inner heatsink will frost up while the outer heatsink barely feels warm.
The "hot" side heatsink of the Peltier unit is insulated from the outside of the door with another piece of Reflectix.
The circuitry below the fan is the elaborate controller that came with the unit. It converts 12V AC to DC, and has a thermal sensor which shuts the unit off when the cool side reaches 0^oC. I probably won't use the circuit board.
The small white board above it contains only LEDS to indicate "Power On" and "Cooling".

Results of Peltier Unit Only

Notice that it took about 1 1/2 hours for the temperature to start dropping, and even then it only reached 10.8^o, a difference of 11.2 below ambient.

The Inside Works

The cold side heatsink and fan were scrounged from an old computer. They will have to be changed to a lower profile type on the final version. The fan only needs to run at about half speed, so the voltage to it was cut down to 6.2V.
The RF remote thermometer sensor is sitting on the shelf in the middle of the fridge in all tests.

All in all, the insulation is probably not very efficient, but by using the same set up for all tests it will give a comparative result.

Propane & Peltier Unit Test
The third test was to see what the combined effect of propane and the Peltier unit would be.

Notice the interior temperture started dropping immediately after switch on. At about 2 hours it reached its lowest point, 2.2^oC, for a difference of 19.8 degrees, over 3 degrees better than on propane only and in 2/3 the time.

Heat Gain or "Losing its Cool"
As a test was to see how good the insulation in this experiment was, this graph shows how rapidly the fridge "lost its cool".
The temperature started rising immediately after propane and power were switched off and rose 8.7^o in the first hour and 6.1 in the second hour. The internal temperature was already up to 16^o, only 4 below ambient when the test was stopped at 2 hours.
How much of this loss is due to the plywood door and Reflectix insulation, and how much is due to the overall Dometic construction might be be proven below.

Dometic Fridge Temperture Loss As a comparison with the above test, I cooled my own fridge to its maximum over night on 110V AC. I left the small interior fan circulating air inside the fridge. Notice that this fridge cooled to 21 or 23^oC below ambient (depending which thermometer you believe. I used two thermometers for comparison.)
Again, the temperature started rising immediately the power was switched off. It rose by 4.2^oC by one thermometer, and 7.5 degrees by the other (average 5.8) within the first hour. In the second hour it rose 7.6 on one and 2.6 on the other (average 5.1). Only two or three degrees better, on average, than in the test fridge, above!
At 6 hours the temperature was up to 15^oC and by the end of 11 1/2 hours, it was at 20, a change of 24 and 21 degrees by the two thermometers. So, it would seem the insulation in a stock Dometic van fridge fridge is not very efficient.
As a side note: the greater the difference between ambient and internal temperature, the more rapid will be the heat transfer between the outside and inside. As the difference diminishes, the rate of change of transfer appears to decrease. Its just one of those laws of physics - but I don't know which one!
In all these tests, the fridge was completely empty with just a thermometer sitting on the middle shelf. Of course, a fully packed fridge would take longer to cool down, and longer to lose its heat as well.

So far it looks like installing the Peltier unit will be worthwhile. Cooling time is cut to 2/3rd and overall temperature is lowered by 3 degrees.
The aim of this mod is to run the Peltier unit along with another heat source (12V, propane or 110V) on start up, then with the normal 12V heat system while driving during hot days. At night when the ambient temperature usually drops to a reasonable level, the fridge can run on propane only. The Peltier unit draws too much current to run it at night on the single battery I have in my van.

The owner takes no responsibility for anyone else making these modifications.
Photos provided by owner. Contact the postmaster for permission for use.

Back to
Home Page. Back to
Spitfire Hangar.

Copyright ^© 2017
F. Griffiths

Last updated December 5, 2016

webmaster: F. Griffiths,
To contact us, please type the following into your email address
griffco AT griffco DOT ca
to avoid automated email spammers.

The Testing Stage
Test Setup #1	To see whether it was worth cutting holes in my fridge, I first had to test the efficiency of the Peltier unit. For this I took the regular fridge door off, then cut a piece of 3/8" plywood big enough to cover the door opening in the fridge cabinet. I had a spare set of hinges, so mounted them on the plywood. For insulation, I lined the inside of the door with Reflectix^TM insulation foil. Then I fired up the propane and compared the internal and ambient temperatures until the fridge reached its coolest point.
Propane Cooling Only	On propane, the fridge took about 1 hour to start cooling seriously and didn't reach its coolest for 3 hours. This fridge is not my own, so does not have any of the other mods I have installed in mine, such as internal fan, external ducting and fan, and venting to the outside of the vehicle. Therefore it does not seem as efficient as mine. The temperature reached 4.8^oC, only 16.2 below ambient. Mine easily reaches 20 - 22^o below ambient. To be fair to this fridge though, it did not have an interior circulating fan for this test.
The Peltier Unit	Peltier units, or thermoelectric coolers, use DC current, usually 12 volts, passing through a solid state device to transfer heat from one side to the other. These are the devices used in 12V "plug in coolers". This was scrounged from an office water cooler, so did not have cooling fins on the "cold" side. This unit is rated at 3.8A, 12V, or about 45 watts but is being run on 13.26V here, which would make it about 50W, the average of those I have seen in camping coolers. On its own in a small insulated space it brought the "cool" side heat sink down to 25^oC below ambient temperature.
Test Setup #2	I took the door off and let the fridge warm up to room temperature again for the next test while I cut a hole in the door. The hole is to fit the heatsink stuck on the cold side of the Peltier unit. I have no idea what its efficiency is in transferring heat, but the inner heatsink will frost up while the outer heatsink barely feels warm. The "hot" side heatsink of the Peltier unit is insulated from the outside of the door with another piece of Reflectix. The circuitry below the fan is the elaborate controller that came with the unit. It converts 12V AC to DC, and has a thermal sensor which shuts the unit off when the cool side reaches 0^oC. I probably won't use the circuit board. The small white board above it contains only LEDS to indicate "Power On" and "Cooling".
Results of Peltier Unit Only	Notice that it took about 1 1/2 hours for the temperature to start dropping, and even then it only reached 10.8^o, a difference of 11.2 below ambient.
The Inside Works	The cold side heatsink and fan were scrounged from an old computer. They will have to be changed to a lower profile type on the final version. The fan only needs to run at about half speed, so the voltage to it was cut down to 6.2V. The RF remote thermometer sensor is sitting on the shelf in the middle of the fridge in all tests. All in all, the insulation is probably not very efficient, but by using the same set up for all tests it will give a comparative result.

Propane & Peltier Unit Test	The third test was to see what the combined effect of propane and the Peltier unit would be. Notice the interior temperture started dropping immediately after switch on. At about 2 hours it reached its lowest point, 2.2^oC, for a difference of 19.8 degrees, over 3 degrees better than on propane only and in 2/3 the time.
Heat Gain or "Losing its Cool"	As a test was to see how good the insulation in this experiment was, this graph shows how rapidly the fridge "lost its cool". The temperature started rising immediately after propane and power were switched off and rose 8.7^o in the first hour and 6.1 in the second hour. The internal temperature was already up to 16^o, only 4 below ambient when the test was stopped at 2 hours. How much of this loss is due to the plywood door and Reflectix insulation, and how much is due to the overall Dometic construction might be be proven below.
Dometic Fridge Temperture Loss	As a comparison with the above test, I cooled my own fridge to its maximum over night on 110V AC. I left the small interior fan circulating air inside the fridge. Notice that this fridge cooled to 21 or 23^oC below ambient (depending which thermometer you believe. I used two thermometers for comparison.) Again, the temperature started rising immediately the power was switched off. It rose by 4.2^oC by one thermometer, and 7.5 degrees by the other (average 5.8) within the first hour. In the second hour it rose 7.6 on one and 2.6 on the other (average 5.1). Only two or three degrees better, on average, than in the test fridge, above! At 6 hours the temperature was up to 15^oC and by the end of 11 1/2 hours, it was at 20, a change of 24 and 21 degrees by the two thermometers. So, it would seem the insulation in a stock Dometic van fridge fridge is not very efficient. As a side note: the greater the difference between ambient and internal temperature, the more rapid will be the heat transfer between the outside and inside. As the difference diminishes, the rate of change of transfer appears to decrease. Its just one of those laws of physics - but I don't know which one! In all these tests, the fridge was completely empty with just a thermometer sitting on the middle shelf. Of course, a fully packed fridge would take longer to cool down, and longer to lose its heat as well.
So far it looks like installing the Peltier unit will be worthwhile. Cooling time is cut to 2/3rd and overall temperature is lowered by 3 degrees. The aim of this mod is to run the Peltier unit along with another heat source (12V, propane or 110V) on start up, then with the normal 12V heat system while driving during hot days. At night when the ambient temperature usually drops to a reasonable level, the fridge can run on propane only. The Peltier unit draws too much current to run it at night on the single battery I have in my van.