Been there, done that. The theory is good, but in application, doesn't work that way. The idea is that cold fuel when atomized in the intake will cool the surrounding air as it vaporizes. This will condense the fuel/air charge for higher density and greater power. Two problems with that - time and vaporization.
With a carb car, the fuel flow is very slow until under high power. That means the cold fuel will be warmer by the time it gets to the carb and into the bowls. Then the bowls are warm from staging and burnout and the fuel gets still warmer. No gains. So, the cool-can would be placed as close to the carb as possible. That helped it stay colder, but then there's the issue that we are trying to vaporize the fuel so it can burn efficiently for power. Cold gasoline is slow to vaporize. So, now we have a cooler intake charge (yay), but raw droplets of fuel that don't burn wel,l if at all (boo). OK, so we richen the mix to get max power (yay), but if it's colder or hotter weather, of there is a delay in staging, or... then the mix is off and power falls (boo). The net gain is about zero overall. Yes, we tried it many years ago on a couple 9 and 10-second bracket cars ourselves, along with a gazillion other teams that all pulled them off.
Note that the effect is different with alky (methanol) fuels, as the fuel itself has different properties. The latent heat of vaporization causes a substantial cooling of the intake charge as the fuel vaporizes, and is one of the benefits to using methanol. The cool-can with gas is an attempt to do the same, but the fuel properties get in the way. There's my 2 cents about it.
David
