Making dry ice

Does anyone have any experience making dry ice "snow" from compressed co2 canisters? Is this a cost effective way to make dry ice? I've heard you can just put a wet sock over the outlet of a co2 tank and crank it open and the sock will fill with dry ice snow. I'm sure there's a lot lost. But if my options are buying pelletized dry ice for $1/lb vs. filling co2 tanks at $0.25/lb, is it possibly worth it?

I do not think the "snow" approach would have nearly the efficiency, as it would simply melt too fast to be effective for heat transfer. Buying a pelletizer or compressed brick machine that is connected to c02 tanks might be a better option, but unless you can get a legal delivery service for large gas canisters, I am not sure that it would be worth it. That is why I made a joke about setting up a co2 gas plant in the other thread. I think an ln2 generator might be more practical than co2/dry ice, but that is a serious engineering feat in and of itself.

So I've looked into several methods of producing dry ice onsite. There are several cheap ($100-1000) products that just clamp onto a co2 tank and make snow (the cheapest) or softish blocks between 250 and 1000 grams each. These seem to be in the 15-20% range for efficiency. Unless you can recollect the bit that comes off as a gas, 46% seems to be the best efficiency I'm seeing. Then there are block and pellet machines that range from thousands to hundreds of thousands, but there seem to be some decent used ones at $10-15k at http://www.dryiceclassified.com/used

I'm not sure how efficient something like that is. I think a 50lb tank here is $20 and that's really cheap from what I hear. So it's looking less and less likely that I'll be able to make dry ice any cheaper than I can buy it locally (95 cents/lb). Although if it were close in price, the convenience might be worth it.

Another option is use two part marine foam and build a bad ass dry ice storage bin at your site, and another smaller unit for transfer, and then load up 100 pounds or more at a time. I fucking hate lugging heavy tanks and $1 per pound is actually pretty good. If you really need hundreds of pounds of per week, than a machine with a co2 delivery service could make sense. If you have to fill up the tanks yourself, there is not much a difference than buying the ice. Also factor in bulk tanks (food grade or above for direct material contact), and you will be laying out some coin. It could be worth it. LN2 is different because it is much easier to extract from the air, but you have to have some high level engineering skills to put one together on a budget. There is a video were a guy uses a scuba compressor to build one.