Instructables
Picture of Chillanator
13, 11:23 PM.jpg
Introducing the Chillanator, it uses a Peltier chip I purchased from deal extreme.
If you have ever set watching TV or reading and looked over to see your drink is all "warm'" well no more with Chillanator.
The total cost of this project was about 10.00 dollars US.

The Peltier chip was 7.00
The fan was one dollar and the LEDs just a few
Bucks.
 
Remove these adsRemove these ads by Signing Up

Step 1: Order The Stuff

Picture of Order The Stuff
13, 11:23 PM.jpg
13, 11:23 PM.jpg
13, 11:23 PM.jpg
Unless you want to pay out the nose I suggest you shop around, I have seen people say they have spent as much as 30.00 bucks for one.

I chose to make my own heat sink from scrap copper and brass.

If you are astute you may have noticed the Peltier mounted to 1/2 inch copper, this is not required or important, a friend offered a chunk and go thought I'd give it a try.

I plan to make a new instructable using two chips and will discuss heat removal then.
The fan was from electronic gold mine it's a 12 v squirrel cage design, I think this is far more efficient than a muffin fan because it blows air at a higher rate of flow

Step 2: Build A Box

Picture of Build A Box
13, 11:23 PM.jpg
13, 11:23 PM.jpg
13, 11:23 PM.jpg
I chose some left over Ash plywood, it seemed the easiest to lay out the base plate first, 1/2 inch birch ply was my choice, I suggest what you have in the scrap bin is probably the best choice.
To make the leg cut outs I screwed the two sides to a scrap of ply wood and drilled a hole using the center seam as a guide

Step 3: Some Time Savers

Picture of Some Time Savers
13, 11:23 PM.jpg
13, 11:23 PM.jpg
13, 11:23 PM.jpg
13, 11:23 PM.jpg
When I was building the box I set some spare prices of wood under the floor plate this helped me determine the correct vent height as well as the best place to locate the switch, and on light.
Since I discovered an old switch that required more than a round hole I hung the box over a clamped piece of ply wood this prevents tearout and allows you to put pressure on the bit with out breaking the box.
I recommend building your box a little larger than you think is required because stuff almost never fit to perfection

Step 4: Make A Logo

Picture of Make A Logo
13, 11:23 PM.jpg
13, 11:23 PM.jpg
I just bought an engraver so I thought a cool logo would be nice, I got the idea from Edge lit acrylic sign from this site.
Sorry I can't post a link I'm working on an I Pad.
Just do a search here and the process is explained quite well, to cut costs I ripped a salvaged overhead fan blade made of clear plastic, if you decide to have a trophy shop cut your design don't polish the edges until it's been lettered, smooth edges slip out of the engraver.

Step 5: The Power Supply

Picture of The Power Supply
13, 11:23 PM.jpg
13, 11:23 PM.jpg
Peltier chips are power hogs, I am using a 12 volt 8 amp video monitor transformer and it's warm after an hour, not hot but warm, I was going to list the power requirements but when I went to measure the current through the Peltier chip it pegged my swing meter.
Judging by the temperature of the transformer it's safe to assume it's pulling near capacity of it.

Step 6: Final Thoughts

Picture of Final Thoughts
This is a fun project, when you first connect voltage to a Peltier it seem like the potential is great, but getting the heat away and figuring out how to use it effectively is more of a challenge than you would think.
I strongly suggest buying several chips at one time, this reduces the fear of "breaking" your only one, plus after you experiment with your first one you'll want to use your new found knowledge to make a better project.
I used JB Weld to secure the chip to the heat sinks, some brass powder was mixed in to help conductivity.
buddarinc1 month ago
Can you post specific size for the box layout
killbox8 months ago
you can even stack Peltier coolers for a slight coldness improvement, (assuming you have the ability to deal with the increased hotside waste heat!)
I reckon you'd have to increase the number of coolers exponentially with each layer if you stack them. All that waste heat from the first cooler has to be "pumped" away, so it might take something like 8 coolers (in parallel) for the second stage to keep up with the first stage. (Unless I'm overlooking something or just deeply confused, again.) If so, efficiency is not only out the window, but halfway to Mars. However, doubling the temperature drop should help chill drinks faster, and keep them cooler when the ambient temperature is high.
longwinters (author)  bassbindevil8 months ago
You hit the nail on the head. Peltier chips were used for a short time to cool over clocked micro processors, fell out of favor rather quickly
One reason may have been that when they get to hot, ( as in a stacked configuration or over clocked MP.) they simply quit working, I have not read the technical data on them, but from casual observation they do seem to lock up when saturated with high temps.
Consider the power consumption of 9 chips all at once, 8 cooling 1 at about 3.5 amps each 12vdc
ac-dc killbox8 months ago
Without the inherent space limitations in a professionally engineered device needing to be as tiny as possible, it is better to arrange peltiers in parallel (thermally regardless of how they are wired) across a hot plate rather than stacked in series.

This reduces the maximum temperature they reach besides also doubling the thermal junction area to the heatsink which lowers temperature further.
longwinters (author) 8 months ago
Wroger-wroger,
Most of these issues were addressed in my steam punk version, you may be pleased to see the processor cooling fins from a commercial Dell computer.
Between the two projects I ordered an infrared thermometer with that it was a no brainer to find the points of in efficiency,
If they were not such power pigs is build a drink cooler with 4 of them, but I have no desire to burn the bearings out of my electric use meter when a few ice cubes would do it better at nearly no cost.
Jedrokivich8 months ago
That acrylic sign is a nice touch :)
starphire8 months ago
Looks cool, though your heat sink suffers from a few significant limitations:
-brass is a poor choice for thermal conductivity. Aluminum has twice the conductivity and copper is far better if soldering is required. Aluminum heatsinks are pretty easy to find as scrap from old PCs, etc.
-the base with fins should be placed as close to the Peltier device as possible. It'll work much better if you flip that heatsink upside down.
-steel is even poorer than brass for thermal conductivity. Consider replacing that 'tin' can with aluminum or copper treated to keep from oxidizing.
-putting a layer of insulating material between the hot side and the cold side of the assembly will help prevent thermal leakage between the two sides through the air. A thin sheet of foam with a cutout for the Peltier device would be effective. You have to watch out for condensation collecting there, though - it could short out the device eventually.
longwinters (author)  starphire8 months ago
Back in 1975 I actually got an associates degree in commercial refrigeration, steel was the most common choice for high side condenser coils.
Evaporators were almost always made of aluminum, I'm sure cost was a major factor in manufacturing choices.
In my next project the steam punk chiller both of these problems are addressed.
After doing 3 of these projects 2 here, I would offer to any trying to use a Peltier that insulation and cold side materials are the primary areas that need special care. As starphire has pointed out.
My apologies for not being more diligent in my explanations it is a major failing of mine.
It isn't as much of a problem in thin tubing because it doesn't have that far to travel. Most refrigerator condenser coils are still steel. In a heatsink, the heat has to travel lengthwise through the fins, so thermal conductivity becomes more of an issue. That big block of copper is doing most of the work. Copper is the best because it has high heat conductivity and specific heat.
I have been thinking about the use of steel in condenser coils, (unrelated to this project) I think the reasoning is that the black painted steel gives up infrared energy quickly, this should not be confused with thermal conductivity although somewhat related.

Without fins the copper block will get hot in about 5 minutes, the chip is really removing very little heat , and generating a great deal of waste heat.
I think I said on th steampunk chiller that heat removal is the lesser concern and getting the cold area well insulated is a bigger problem.
Yeah, the black body radiation and whatnot. Black objects absorb heat more readily, but theoretically they also give off heat more easily. Every calorimeter I've seen was painted black or black anodized aluminum.
Yeah your KIND of right about the thermal conductivity or assorted materials.

But the RATE of thermal conductivity through different metals is a ratio between the quantity of energy, and it's concentration, the AREA and THICKNESS of the conductive material, and the sort of partial absence of the energy in the cold zone.

The laws of entropy and conductivity, surface area etc.

Heat is like water, it only flows one way, from more to less, or from up high to down low.

For MOST situations, the relative differences between the rates of conductivity do not matter, like say a thick cast iron based pan, and a aluminium pan and a copper pan - with a modest heat input, the egg fries more or less evenly.

Inside the nozzle of a rocket engine - burning oxygen and assorted fuels under intense pressure - cryogenically cooled copper is the only way to go.

Or with really power computer chips... I mean look at at 386 or a 486 chip and their heat sinks... and compare that to the super multi core CPU's of today...

With the Peltier modules - they are inefficient - so while a 20W chip may in fact pull 20W, the actual amount of energy being drawn across the junction might be like 5 or 8W (sort of making this up = slightly ignorant and they come out with different effiencies and configurations).

To simplify the issue, big refrigeration plants - and most domestic refigerators use steel piping, because it's cheap, works well and they need to dump the heat, from a GAS that is flowing through the pipe - so that takes time and distance.

Though like the modern CPU, the Peltier module is more of a concentrated heat source and energy producer as well as a energy extractor - which is different to dumping heat from a gas.

And Peltiers do differ significantly in their power consumption AND their capacity to draw heat out of a particular system.

So the mixing of different materials and process's and cross referencing them, as the solution to the other, is not appropriate.

The solution has to be done on a situation by situation basis.

It comes down to this.

"Am I dumping the heat from a match or an oxygen fuel gas welding torch?"

That is this issue of the necessarily efficient process needs to be considered - in terms of conductivity, cost, efficiency, design, space, volume, power supplies, running costs, thermal mass to transfer per time unit etc.

But OLD computers and FREE heat sinks - with thick aluminium / copper slabs...

Bingo!

usMudack19988 months ago
"Sorry I can't post a link I'm working on an I Pad."
Which is way that particular device does not sit on my desk. xD
tioshrek8 months ago
great!!! very nice!!!
Junophor8 months ago
Hi Longwinters

I am deeply impressed!!!
Great idea!!!

greetings from Germany
Yours Aeon Junophor
XboxModz8 months ago
Another great way to attach the Peltier to the heat sink is with my homemade thermal adhesive.

Get some epoxy resin (the kind that comes in 2 parts you mix together), and some Arctic Silver (AS-5). Then find yourself a mixing platform (I used 2 Pringles can lids). Drop equal amounts (about 50/50) of your resin and hardener, each on a separate lid (as to not yet mix them). Then mix in equal amounts of Arctic Silver (also about 50/50) into each epoxy component. When both sides are mixed very well (no streaks), clean/prepare your surfaces to be bonded. The final step, when ready, is to mix the 2 halves together to activate the epoxy. Apply the mixture quickly and evenly and then press the parts together. Allow to sit for about 24 hours, weighted if possible.

Ratios are about 1:1 for each epoxy component. Estimation is fine. It doesn't have to be exact, but try the best you can. I made this stuff so that I could attach some heat sinks to the CPU/RAM chips of a wireless router that I had overclocked, but it could be used on any project where both adhesion and thermal transfer are needed.
longwinters (author) 8 months ago
Not sure if it's in this instructable but I think the draw is around 6 amps at 12vdc so that would be a great power supply
This is great! I don't know why it didn't show up on my following updates?!
longwinters (author)  bricabracwizard8 months ago
Me neither, I'm posting mostly from an I pad and it looks like some of my comments are getting lost, you are one of my favorite fellow inventors here so if I'm silent always assume I did not get the message
astral_mage8 months ago
just a thought, have u tried a 12 volt 15 amp ac - dc converter?
longwinters (author) 8 months ago
Mr Dream I am just finishing a new steampunk version that takes into account your suggestion.
13, 6:37 AM.jpg
Dreamchronic8 months ago
Cool! :) If may i suggest a cover (maybe aluminium),might help to chill your drinks faster ;)
Very nice idea! I like the design. Great!
longwinters (author) 8 months ago
I agree, but I guess how to shuck garlic beat me out.
nerd74738 months ago
This should be featured!
longwinters (author) 8 months ago
Yes sir it is, I'll try and find a small thermometer to get so specs.
gman2000watts8 months ago
Looks awesome... did you use a tin can for the cupholder?
Pro

Get More Out of Instructables

Already have an Account?

close

PDF Downloads
As a Pro member, you will gain access to download any Instructable in the PDF format. You also have the ability to customize your PDF download.

Upgrade to Pro today!