Can Cooler Project - Part 1

This is part 1 of the Can Cooler Project (as of this post, still a work in progress).  The plan is to make a small standalone soda can cooler that can rapidly chill on demand, a single 12 oz can of soda.  Here is what I have so far!
 

can cooler prototype assembly

I'm shooting for a 30 deg F temperature drop in under 15 minutes.  The basic design approach is to sandwich a Peltier element between a CPU heat sink and an aluminum cold sink that I will machine to the contour of a can using the Taig Micro Mill.  I've seen the mini fridge style coolers out there and from what I've read, they don't perform well at all.  The USB powered versions are just silly as they're limited to ~0.5 A at 5 V (~2.5 Watts).  Also, conductive heat transfer occurs where a flat plate contacts the base of a can.  As for the power supply, I'm using an old computer PSU and using the 12V rails to drive the peltier element as well as the cooling fan.

Here is the peltier element I'm using (model # TEC1-12710):

peltier

 

This is a CPU heatsink-fan combo from my junk box.  It didn't work so well but I took more pictures of it than the other so...

cpu fan and heat sink

peltier installed on heat sink

 

Now we're getting to the machining.  Here is a chunk of aluminum bar stock that will become the cold sink:

aluminum bar stock

 

Milling the surface flat for contact with the Peltier element:

facing bar stock

 

So far so good!

facing with endmill complete

 

Squaring up the ends:

sqaring bar stock

 

I used the fly cutter to create the can diameter curvature.  Setup is easier than you might think.  First I used an edge finder to locate the surface.  Next, I moved the y axis half the can diameter away from the fly cutter.  In order to set the cutting tool for performing a boring type operation, I rotated the cutting tool as shown and slid it out until it just barely touched the material; tighten up the the cutting tool and it's ready to go! 

setup fly cutter with bar stock

 

So far so good; I had to increase the spindle speed to 5500 RPM and take very light cuts (.002 to .004):

curvature cut into the bar stock

 

The 2.6 inch diameter "boring" operation (with fly cutter) is complete:

2.6 inch diameter arc in bar stock

 

Chips everywhere!

chips everywhere

 

Cold sink is almost complete - just need to cut off the end:

aluminum cold sink

 

Cutting off the excess on the bandsaw:

bandsaw cutting excess

excess material cuttoff

 

Back to the mill and trimming of the excess with an end mill:

trimming up the excess

 

Ready for first power-on test!  Thermal compound is used between both contact surfaces.  The Peltier element and CPU fan are hooked up to the 12 V rail on a PC PSU.

ready for power

 

It works!  Well, it gets cold enough to create condensation anyhow: 

cold sink getting cold

can on cold sink assembly

can fits nicely with cold sink

 

So this is where I realized that the heat sink was not going to cut it.  The heat sink was almost too hot to touch after just a few minutes of run time which means the cold side wasn't getting as cold as it could.  Peltier elements run more efficiently when the hot side is not so hot.  I removed the fan and placed the heatsink in an icewater bath to see where things could go given a better heat sink.  This resulted in much cooler cold side temps and so I switched heat sinks for the one shown in the first picture (with heat pipes, a 120 mm fan, and lots of fins).  Now the hot side doesn't feel hot at all and the cold side is much cooler.  In part 2 I plan to build an enclosure and post some actual performance data.  That's all for now!