please post the findings of your research.

So, I have ordered 4 TECs (cheaper, worse performance than TEG) 40x40 mm 12706 modules and a voltage converter 1.5V to 12V. Later I'm gonna get two arctic cooling alpine M1 heatsinks with fans (0.36 W) - 80x80 mm.

One heatsink will cool the peltiers on the compressor and the other heatsink will cool condenser coils (I have the passive ones)

The compressor warms up to 45°C (113°F), ambient temp is around 24°C (75° F).

The fan with heatsink arrived (it was from local seller), I just ran it of 6 AA (actually 4AA and 2 AAA) batteries, so 9 volts are enough. I'm guessing it will run at 0.2 W since max is 0.36W. With 2% efficiency of peltiers, I need just 10W of thermal energy from compressor. This seems doable, since it consumes 80-90.

Another update: Voltage regulator just arrived from China and I was able to run the fan from just one AA battery. It's starting to look hella good :-)

I think your concern is valid, and I think that your interest is a good one, despite the analysis (?) that appeared below your original post.

I think that every BTU should be used if possible. But even more importantly, I think that every BTU of 'wasted' energy should be scrutinized and eliminated if possible.

The best approach is to start drastically reducing the waste energy.

If you have a refrigerator that is putting out enough waste BTUs to appreciably heat a house in the winter, and to be a significant cooling problem in the summer, it indicates to me that you are dealing with a very inefficient refrigerator, probably an older one.

Many of the newer refrigerators, 'Energy Star' refrigerators, use far less energy and produce far less wasted heat than conventional units... maybe half of the energy use compared to older units, half of the energy wasted compared to older units.

The next step in the Refrigerator Warrior's attack would be to re-purpose a modest-sized freezer, either chest type or cabinet type, to become a refrigerator. The chest type will ultimately be the most efficient. Some argue that the chest type is less convinent.

I have re-purposed a cabinet type and it is extremely efficient., and convenient to use.

With both types, there will be condensation issues. Chest types have a drain hole in the bottom which is perfect for eliminating condensation. A carefully chosen chest type should have a drain hole, too. I've been using mine for a few years now, and I don't think that condensation is a deal-breaker.

My project is fully explained in the link below. (Please note that I purposely chose an inefficient freezer, to see if even a bad freezer could make a good refrigerator)

http://ecorenovator.org/forum/applia...onversion.html


MEMPHIS91 did a chest type conversion. He has been super good about including photographs and detailed explanations.

http://ecorenovator.org/forum/geothe...eigerator.html

So, my advice is NOT to abandon your project, but to attack the problem at the root, which would be the drastic reduction of wasted energy.

I wish you great success.

Sincerely,

-AC_Hacker

So first test conclusion is just as jeff5may predicted - low dT doesn't produce enough "juice" to run the fan. With hot pot underneath and cold pot sitting on top of peltier plates I got around 1V directly from modules and slightly above 3V from step-up booster. I will continue tests and will probably try more than 4 modules (I guess I can fit 12 on the compressor)

Another test (in the first one I didn't have surface that would be flat enough) shown 2,65V on modules and around 12V on step-up booster, but when I connected the fan, the voltage sharply dropped. I guess the modules are very soft power source. I have found some RC helicopter engine on aliexpress and propellers (which together acts just like fan), so I'll order those and try it.

I finally got around to install TEGs on the fridge compressor and results are very dissapointing. At most I am getting 0,4V without load and between 4 and 5 mA short current. Which is too low to run small electric motor I salvaged from solar toy.

I ran out of thermal paste (surface of compressor isn't flat, so I need something to fill the gaps), adding more should help a little. Also my heatsinks are too small - voltage rises from 170 mV to 320 mV fairly quickly, but than flats out, because heat sinks are "saturated".

I have ordered two more TEC 12706 (there is still room on the compressor) and bigger heat sinks (height 30mm instead of currently used 11mm). That should be enough to run the motor with propeller.

When the motor is running (from a battery) - and cooling the heat sinks - then the voltage rises to 0,7V (with spikes around 0,8V) and short current rises to 10 mA.