[desktoplathes]

Hey all - Here is a little project I posted this weekend. I was trying to figure out how to quickly create a large arc cutout (2.6" diameter) in an aluminum plate in order to fit the curvature of a soda can. Misusing the fly cutter did the trick. Check out the Can Cooler Project details here if interested. Thanks! -Keith

[keebler303]

That's not misuse, that's creativity. If you just bored a hole in a block and cut it in half, you would have gotten 2 pieces with more contact area.

Cool project but if your idea is to cool a can of soda the fastest then a bucket of salted ice water will do it much faster. You have a higher delta T and probably 10 times more contact area with the can.

Matt

[desktoplathes]

Matt - Thanks for the feedback. Agreed... there are much faster and simpler ways to cool a can of soda! With this project, I'm going for a very niche application where I don't have easy access to ice or water and don't want the mess at my desk. Also, there isn't room for a mini fridge but there is plenty of power. Part of my fascination with having fabrication tools at home is that I can make custom devices that would likely never be found in the marketplace. Who would make a drink cooler that only fits 12 oz soda cans right? Lastly.. I had some peltiers and aluminum bar stock laying around so why not have some fun with them? :-) -Keith

[Lynn Livings]

Keith, looks like a fun project! I too have been known to do things that would never be viable for commercial use or sometimes even worth the effort of making two! But, usually I'm more interested in learning about something I've got on my mind or like you, have goods parts laying around and an idea develops.
Hopefully you will have the time and inclination to keep this project going a bit longer and develop it some more. Lynn

[desktoplathes]

Lynn - Exactly! I usually head down a narrow path like this when I'm interested in tinkering with and learning about something new. I'd never messed with Peltier coolers until this effort. I've started collecting some data with an arduino and will post the results in the next few weeks. The next phase will be to finish up an enclosure that can support that massive upside down heat sink / fan combo. Also, I need to keep condensation from dripping and causing future problems. Lots of fun challenges to solve. Thanks for your feedback! -Keith

[keebler303]

Keith, you can prevent condensation by doing the cooling in a vacuum.

Matt

[desktoplathes]

Keith, you can prevent condensation by doing the cooling in a vacuum.

Matt

Haha, yes and that would take care of the losses due to convective heat transfer from the ambient temp air! Time to build a vacuum chamber :-)

[keebler303]

A turbopump would be an ambitious hobby project!

Matt

[Menacer]

Nice project and good job so far!


You could try using less voltage with your peltier... Simple fix but you might not be happy with the time it takes to cool.... Also sand your surface with super light sand paper (On a FLAAAT suffice) till you get a merrier finish on both sides (not the peltier ) and use heat transfer gunk so your air gap is next to nothing. I think brass works better than aluminum for heat transfer but I never looked into this much...

[keebler303]

Copper is the best conventional material for conductivity, aluminum is both cheaper and easier to machine.

If you really want high conductivity, look up highly oriented pyrolytic graphite (HOPG).

Reduced voltage would give you reduced cooling, some arctic silver at thermal junctionswould definitely be helpful.

Matt

[desktoplathes]

Yep... am using arctic silver at thermal junctions. I've collected some data with a thermistor RTV'd in a can of water - just need to plot it out and see what kind of time to chill I'm getting. It's quite a bit more than 10 minutes I'm afraid (guessing ~20). -Keith

[Fixittt]

I did this exact thing many moons ago when I was building watercooling systems. Man I wish I still had those pictures.

I was even asked to help design a cooling vest for motorcycle police using the pelt tech.

One tip I will give you is isolate the pelt between the heatsink and the cooling block. If I remember right I used the foam pack that meat comes on when you buy it from the store. layered it up

[Fixittt]

Copper is the best conventional material for conductivity, aluminum is both cheaper and easier to machine.

If you really want high conductivity, look up highly oriented pyrolytic graphite (HOPG).

Reduced voltage would give you reduced cooling, some arctic silver at thermal junctionswould definitely be helpful.

Matt

If you really want the best metal for conductivity, it will cost you. But silver would be the most efficient, yet not practical

[keebler303]

If you really want the best metal for conductivity, it will cost you. But silver would be the most efficient, yet not practical

HOPG is 3-4 times more conductive than silver in the X-Y plane.

It is lighter than silver, but no cheaper.

HOPG Highly Orientated Pyrolytical Graphite - ZYH, ZYB, ZYA
HOPG or Highly Oriented Pyrolytic Graphite for short. - Overclockers Forums

Matt

[Fixittt]

yes, you are correct, but its not a metal.
I cut a wax block once made the wax into the shape I wanted my water block to be.
Then had it casted using the "lost wax casting" process. Then cleaned it up and lapped it.

most expensive block I ever made.

From what I read about the HOPG is that its simply a "wafer" or plate of material That is to be sandwiched between the heat source and the cooling source. To me this is simply nothing more then 2 more contact points as the HOPG cannot remove any more heat then what the cooling course can deal with.

As with water cooling, your limited to the efficiency of the device (radiator or evaporation cooling) to remove the heat from the liquid.

[keebler303]

Fixxitt

It must be applied properly to be beneficial, as with most engineering solutions. The applications I have been involved in with HOPG use it as a heat spreader. Take the example of a computer CPU, you have a small silicon chip putting out say 50W of heat. WHen that heat flows into the heat sink, the center of the heat sink will be hottest while the edges will not be so useful. HOPG can be used to spread the heat over a larger surface area which can make a heat sink more efficient and reduce device temperatures. Contact resistance is minimal in the devices I have seen as they HOPG wafer is sandwiched (cast?) into the base of the heatsink.

Matt

[Fixittt]

that sir, makes perfect and total sense. Thumbs up on that.

[desktoplathes]

Regarding the motorcycle vest cooling application.. how did it turn out?

I wonder if it would work better to just leverage heat pipe technology and make it a passive solution. You've typically got quite a bit of air flow available on a motorcycle and peltiers seem to limit your heat transfer to their maximum effective wattage rating. Also, they have some outrageous power requirements.

[keebler303]

I've been involved in some basic development of a similar vest for cooling soldiers in combat. The big advantage of a peltier is you can cool below ambient temperature. When you are fighting in the desert at 110F, a heat pipe is not doing you any favors.

Matt