How to Make a Dehumidifier (Thermoelectric Cooling)





Introduction: How to Make a Dehumidifier (Thermoelectric Cooling)

About: My name is Simon Sörensen and I am the creator of RCLifeOn. I´m 19 years old and live in a town called Trollhättan, located in the southwestern part of Sweden. I´ve been in the RC hobby...

If you live in a humid area, or want to decrease the level moisture inside your house to prevent the growth of mold. This is the perfect Instructable for you, because I'm going to show you have to build a very easy to build dehumidifier from inexpensive parts you can find on Ebay. It will quickly remove moisture from the air by a phenomenon called condensation.

Basically we will build a device that extracts water from the air, maybe it also has a secondary use? Comment below!

Step 1: Materials

One of the main components is the peltier chip, also called a thermoelectric cooler. This is a 40x40mm chip that does something amazing! When powered from a 12V source (like a battery), one side will increase in temperature while the other side will decrease. The temperature difference is the required condition to make the water vapour in the air around us to transform into a liquid state, also called condensation.

You will also need an electric fan to pull the humid air through the device. The heat sinks will spread the heat and cold, making it more effective. Heat sink compound is necessary to achieve a good surface bond between the peltier and the heat sinks.

You can power the dehumidifier by either a 12V battery, or an adapter. In this Instructable I will be using a lithium polymer battery.

-Computer fan

-Large heat sink

-Small heat sink

-Peltier chip (thermoelectric cooler)

-Heat sink compound

-Scrap material, such as foam.

-12V battery

All these parts can be found on Ebay for under 30$! You will also need some tools:

-Hot glue
-Solder station
-Screwdriver + screws

Step 2: Assembly

Fasten the 80mm electric computer fan to the larger heat sink. This is the heat sink that will be increasing in temperature by the peltier chip. The fan will be pulling the air into the larger heat sink. Flip the heat sink over and add the compound to where the peltier will be placed.

Place the peltier chip in no specific orientation or upside/downside. If reversed, meaning the larger heat sink decreases in temperature, swap the positive and negative wires.

Place the peltier chip and press down firmly. Add compound to the smaller heat sink and push it agains the peltier chip. Screw it down if necessary.

Step 3: Stand

Take some scrap pieces of foam, or even cardboard, and cut 4 rectangular legs for the dehumidifier to stand on. Cut one of the short ends at a 45 degree angle and hot glue it to the fan as shown in the pictures.

Since the heat sink on the bottom will collect water it might start dripping if the air is humid enough. Make sure you have high enough legs to place a container underneath.

Step 4: Wiring

You will have a total of four wires, two from the peltier (black and red) and two from the electric fan (black and red). Solder the black wires to one of the pins on the electronic switch and solder another piece of wire to the middle pin from the switch.

Glue the switch to a convenient position on your dehumidifier, and solder the connector to the remaining two red wires and the black wire comming from the electronic switch.

I took the liberty to add LED lights, the main reason is because it makes it look better and lights it up during the dark hours. It also signals when the battery is connected to the dehumidifier, which can be very useful if you tend to forget things easily.

Step 5: Usage

Connect a fully charged battery, but remember to also purchase a small voltage alarm that will alarm you once the batteries is low.

After just 2 minutes drops of water is forming on the cold black heat sink indicating our dehumidifier is working properly. After 5 minutes the heat sink is covered with water extracted from the air. Proving that the device is very efficient and performs very well inside my garage where the humidity is low.

At just a 3 amps this setup will run for a little less than 1 hour. However, you can definitely get a much larger battery and achieve hours of run time on a single charge.

Don't buy does expensive dehumidifiers when you can build your own. Follow my instructions and you will have a device that runs efficiently, quietly and performs really well! Good luck, and comment below if you have any questions.

Please support me by subscribing to my YouTube-channel here:



  • Oil Contest

    Oil Contest
  • Creative Misuse Contest

    Creative Misuse Contest
  • Clocks Contest

    Clocks Contest

65 Discussions

Hi, I'm thinking of making this a means of making a controlled environment(a box,really) with controllable humidity. Do you think there's a way to do this by using an extension of this project? Thank you.

Also note you can buy a fully built dehumidifier with a bucket and autoshut off for under $30 and don't have to worry about building...

I am trying to build a miniature version of this. To make it small enough to be easily portable. I have a peltier module that has a max amp of 6 and a max volt requirement of 2.1.

My question is how to select the correct battery. If I choose a battery that has an output that exceeds the amp or the volts what happens?

1 reply

Gerdnaxela: You can use a voltage limiter to make sure you don't blow out the fan. Simply go to a RC shop and you can buy proper connections and get creative on how to make this work and look a bit better. You can also get a bucket to collect water and via gravity drain hav eit drain out an area of interest. If gravity drainage isn't an option, you can get a little pump that can turn on when the bucket fills up and dump it out to where you run a drain line...have fun tinkering...

use 2 peltiers.

one to cool the other, and use a big heat sink to cool the last one.

My brain instantly started wondering, looking at this design, what would it take to convert this to an outdoor solar powered water collector? (Or for that matter, build it so that it will pull distilled water.)

1) Build into a large mouth funnel. Likely add a "collar" to the top of the funnel. Cut vent holes all the way around the funnel above the area where the cool-sync will drip.
2) Add an "air filter" cloth around the vent slots.
3) Add a 12v solar panel lead and a battery charging circuit.
4) Put the funnel in a clean milk jug. Glass milk jug will have more weight.


5 replies

This is a terribly inefficient method to obtain water. You shouldn't drink the water you collect as it is either as the heatsink fins are a great breeding ground for bacteria, even more so when dust builds up. You could filter it and boil the water if you really wanted, but this is consumption we're talking about let alone the inefficiency of using a peltier.
"While peltiers are useful, they're not very efficient. Only around 5% of the electrical energy used to power them gets used for cooling."
In addition, photovoltaic (solar) panels only make electrical energy out of ~22% of sunlight.
In conclusion, yes it would work but you're better off taking fabric soaked in sea water, dirty water or even leaves from tree branches and having them sit in the sun inside a mini glass house. The suns direct thermal energy (~55% of sunlight) will evaporate the water and it will condense on the glass/plastic.

Awesome!!! Glad to see someone is working on it! Exactly what I envisioned - particularly the bike model. While this wouldn't work in the desert, it could be amazing somewhere humid!

Unless your airborne humidity is very high (ie, high dew point temperatures), the likely hood of a peltier module getting it's cold junction plate below the dew point of free air and making water is low and production would likely be very limited. A mechanical refrigeration system would be more likely to be effective on a cost per gallon basis IMHO.


There are commercially available peltier cooler dehumidifiers. You could dry your cellar and get distilled water. However, the efficiency ratings tend to be vague.

They should last a long time if made well, and the parts should be easily repaired (power supply, fan, cooler module) compared to a refrigerant based machine. Then again, I have a 30 year old dehumidifier that still runs like new.

How did you connect the Lipo Battery to the Computer Fan and the Peltier Chip wires?

Does a large or small heat sink matter?

Nice idea but have a look at the efficiency of the thermo-electric
device. You'll find that the number of 'input watts' is much higher
than the number of 'output watts' of cooling. Another way to look at
this is the 'hot side' has to dissipate both the 'input watts' from the
12V supply and the 'cooling watts' pulled out of the 'load' (air/water
vapor) so will get much hotter than the cold side gets cold. Coolers use these because they 'have to' to make a mobile
device. Its far more efficient to use a refrigeration cycle if you
don't need portability. If this wasn't the case, a lot of refrigerators
would use the same technology.

6 replies

@toddbatzler, reading through all of the comments it seems yours is the most honest and reasonable comment regarding the actual efficiency of this device. Many people seem to have no idea just how much energy is needed to dehumidify basements, etc. in humid climates. Peltier devices have been available for several decades, but never replaced traditional cooling systems for good reason! They are ONLY a better choice than mechanical refrigeration pumps when portability, small size, or quietness are absolutely required features and efficiency (and cost, in utility bills) can be sacrificed in exchange for these special cases. It is laughable to call any Peltier-based cooling or dehumidification device "efficient", unless perhaps comparing it to some even worse method - like repeatedly baking out dessicant so that it can be used to "efficiently" remove humidity from a dank closet.

I wouldn't say it is "laughable" to say it is efficient... if you have an unused 12V solar panel laying around, use that instead of, or in conjuntion with a battery. Use the sun to do the work. THAT'S efficient.

OK, if you need only a small number of Watts of cooling power to effectively dehumidify a space and you can get your power free from the sun. So perhaps to control the humidity in a drawer, or small cabinet. Anything more than that, it still doesn't make sense.

Many commenters here seem to think that could be enough to dehumidify a room, or even a whole basement. An average "room" dehumidifer uses about 500W, and the adjacent air and soil provides essentially unlimited fresh humidity into a space that's being actively dehumidified, so a duty cycle of 100% is not unusual. In the Northeastern US, it can easily cost hundreds of dollars worth of electricity to run a dehumidifer half of the year:

A Peltier cooler is 1/3 as efficient, roughly speaking (yes, it really is TERRIBLE efficiency in comparison to mechanical refrigeration). So a comparable Peltier-based dehumidifer would need about 1500 Watts to do the same work.

So if someone needs enough cooling power to justify even a very small commercial dehumidifier, and they would like to make the most efficient use of their spare solar panels it would be better by far to convert the DC output of the solar panels to AC using an inexpensive inverter, and use that to power their mechanical dehumidifer. A single Peltier device being used to draw moisture from the air is a nice demonstrator of the concept, but it is basically a toy for most of the applications where dehumidification is needed.

True. It is inefficient compared to a refrigeration unit. And obviously solar would be better being used to power a conventional dehumidifier.
However, since i began using Peltiers around a quarter century ago, their prices are now maybe 2% what they used to cost. "laughable" was what i took exception to.
Who knows... tomorrow all refrigerant gases may be banned as environmentally unacceptable. And Peltier devices could be mass produced to fill the gap, making them cost efficient.

There are good possibilities for dehumidification in a basement, though. One could run a cooling-water loop to the outside, and during the coolest part of the morning, turn the thermoelectric condenser on for an hour or three. This way, most basement moisture condensation happens over your drain, rather than on the concrete walls...

Despite a lot of hopeful comments here (maybe based on the false impression that Peltier devices offer high efficiency for cooling), any basement-scale dehumidification really would make more sense using a traditional compressor-based refrigeration system if it is to be affordable to operate, or for it to have sufficient capacity (as noted by JerryS42) without spending massive amounts of money on Peltier devices (and generating 3x the amount of waste heat in the process).

How the cooling loops are arranged to collect condensation, or how waste heat is removed from the cooler, is more adaptable to personal choice and experimentation though.