Xtremetechuk<---Hopefully I finally have this right
Here's the deal. The reason my chiller grossly underperformed (well maybe not that bad but there is way more in a 4 x 50mm set up then I was able to show) Was in fact the sandwiched cold side of the chiller. I didn't run channels in the way you have shown we (Mindchill and I) used a more open flow approach. Even then the problem was the flow was slowed down to such a degree that it was flash freezing the medium to the corners at the turns in the block no matter what I ran. (Although I never tried straight methanol) It would slush up restrict even more and it would just lose efficiency from there. Now if you look at Mindchill's 4 x 40mm chiller you will see one large difference well other than the plate designs but in the end that wasn't what really made a difference IMHO. Cold side wise you'll want as least restrictive design as you can muster. I would have continued on and tried to correct mine but with all the chaos at that time things just didn't work out that way. One quick solution I had always wanted to try was to make a nylon or delrin layer in between the sandwiched cold plates which would serve as nothing else but a way of opening up the flow path a glorified spacer if you will. Ideally this would be metal but the costs involved get to high well unless you use aluminum and an anti corrosive which will probably occur anyway (glycol). You could also nickle plate and or anodize either way it will work. Acrylic is out as whatever you use to keep the fluid from freezing will destroy the acrylic. As for bolts the least amount you can use is best. In the end I used 10 in total for the 4 x 50mm chiller.
p.s. You may also want to consider a square design as it will easier allow you to open up the flow path as with that many 90 degree bends you will need one hell of a pump.
Must admit, I hadn't considered the cooling icing up. Personally I was considering a mostly ethylene glycol, probably to the tune of 50:50.
Just dug some numbers up for the viscosity:
So at -17oC, the viscosity of a 50:50 mix would be about 22 times that of water...
Yup you need a damned strong pump. I see why you've been saying that in other threads now.
Info from here:http://www.engineeringtoolbox.com/et...col-d_146.html