This is part 1 of a four part project.
Click here for Part 2
Click here for Part 3
Click here for Part 4

I’ve been working on building my own Coolsculpting device as Zeltiq are offering.It is a machine which reduces fat cells in the body. The below video shows how it is done.

The way in which it does this is explained in the patent, found here.
The patent says a number of interesting things including:
An exemplary range for the cooling elements 130a-b can be from
about -20.degree. C. to about 20.degree. C., preferably from
about -20.degree. C. to about 10. degree. C., more preferably
from about -15.degree. C. toabout 5.degree. C., more preferably
from about -10.degree. C. to about 0.degree. C.

My machine is currently running at -3C and to avoid freezing the controller bounces it between -3C and 0C. I am still in testing stages in using liquids/gels which don’t freeze so that I can get to -5C and keep it there.  I am currently using salt water drenched cloth, but this freezes at -5C. If anyone knows what sort of liquid/gel I can obtain easily, mess free and doesn’t freeze at -5C, please let me know!!

Edit: I’m using a terry towel soaked in Propylene Glycol. This works fine for -5degC (freezing point of Propylene Glycol is -59degC).

From the patents, we can see there are a couple of mechanisms that make the Coolsculptor work:

1) The cooling effect keeps the fat cells at about -5 Degrees Celsius for up to an hour. If the cells get cooler you risk damage to non-fat cells. If it gets warmer the fat cells may not undergo apoptosis.

2) The vacuum ‘sucking’ of the fat also assist in limiting the amount of blood to the area being treated. This prevent the body from warming up the fat cells. The lack of blood is a double-edged sword in that when the device is switched off and the blood rushes back to the area, the cells are ‘shocked’ and this also kills the cell (there is a technical name for this which I have forgotten. Its in the patent if you’re interested).

So, knowing the above I created FrankenSculptor. The two aluminium plates serve to transfer heat (or in this case, remove heat) quickly while also having a clamping mechanism to ‘squeeze’ the blood away from the area (given that its not a vacuum I am hoping it is enough to cause apoptosis of the fat cells).
The two aluminium plates need to be cooled to below room temperature, so this means a normal radiator system wont be enough. An active cooling system is needed. Luckily we have such a device! The Peltier.

The Peltier (aka TEC, Thermo Electric Cooler) is a bunch of transistors
which when powered cause a temperature differential between each side.
The transistors are placed in an array and ceramic plates are glued to
either side to transfer the heat evenly.
TEC's are cheap. Around $20 each. Be careful when buying TEC's as you
may get cheap Chinese crap for much cheaper, but it may not work as well.

The TEC acts as a kind of ‘heat pump’ – it takes heat from one surface and places it on the other causing one plate to be cooler than the other. The more heat you can remove heat from the hot side, the cooler the cold side will get. The way the TEC works is quite inefficient and it generates its own heat, so you need to remove more heat than you would think.
To do this you need a very good cooling system. I used a computer CPU closed-loop water cooler. Essentially it is a block of copper which is liquid cooled using a pump, the liquid itself is cooled using a small radiator (like in your car). For what it is and does, this device is relatively cheap (around $150 each), the CPU coolers which you can build yourself are much more expensive but you get to choose how long the pipes are that connect the radiator to the cooling block (the ones I used had pipes too short and got in the way often).

The two aluminium plates on my FrankenSculptor required a TEC and a CPU cooler each. These devices are assembled using heat transfer paste and copper sheet (edit: copper sheet was removed, it was redundant) between the aluminium and TEC for heat transfer. The TEC’s run on DC power and so do the CPU coolers! Great! This means we can use a computer PSU to power everything.
For the prototype I am using a 800W PSU (make sure you get a decent one. A cheapo one may not power the TEC’s properly). So how can we control the temperature of the TEC? We don’t want it to go as cool as possible and stay there (for the record, the coldest I could get was just over -10C). The TEC’s can accept anywhere from 5v – 15v. If you vary the voltage you can control the temperature. The PSU’s only output 12v+/-, 5v+/- and 3.3v+. Using some clever electronics we could regulate the voltage, but I found this more challenging and convoluted than needed.
Instead, I opted for a controller which controlled the temperature using pulses (essentially, turning the TEC off when too cold and back on when too warm). The TEC’s don’t like to be controlled this way if you are switching too quickly, so be sure you don’t create a controller which switches the TEC too often.
My controller is Arduino-based. It runs off the 5v rail of the PSU. It monitors the temperature and adjusts accordingly. It also has a screen to let you know what is going on. I won’t go in to detail about this just yet as I still need to post up those pictures. Stay tuned!!

There are so many different facets of this project that needed to come together. Because of this the construction of this device took a while. To make things easier for me, I will be splitting the DIY in to multiple parts. At the moment it looks like it will be 3 parts but this may change later.

EDIT: Part 2 is now available. See main menu for link. Any questions or want to discuss the project? Post a comment!
Like my project? Follow and Subscribe!

Testing with multimeter
Initially the two plates would spin independently, which was bad. Later I installed a guide pin to prevent this.
Circuit board prototype!
Welding the nut to the plate for the threaded rod that keeps both plates together. This was needed later as explained below.
This is a drop of water that has frozen on the plate. It took about a minute to freeze. This is waaaaay to quick for human use. So the controller will be used to gradually decrease the temperature at a safe speed.
The plate is beginning to freeze the humidty in the air as you can see here.
Drilling and tapping the plates. Be careful here, the ally is not very strong and you can easily thread it if rushed.
Adding heat transfer paste. Be sure to wear gloves when doing this. Not so good if the paste gets on your skin.
Making thread
Making the guide pin for the clamp. This is to prevent the two sides of the clamp from spinning independantly
Some older pics of cleaning the copper. Should be put in Part 1
Marking out where the holes will go
Cleaning the plate with sandpaper
Cleaning the copper with sandpaper
Some tack welds to hold it in place
The origianl thread was directly in the aluminium, but this didnt work well as the ally is much weaker than the steel. So had to later weld this bracket and use the steel nuts thread.
The aluminium thread I created was beginning to bread down against the steel rod. So I made this bracket using steel nut which was bolted to the aluminium.

This is part 1 of a four part project.
Click here for Part 2
Click here for Part 3
Click here for Part 4