OK, lets begin with the subject of dry air.
It is important in any situation, from air tools not kept in a heated environment to paint and media blasting.
Dry air isn't impossible to acheive, and can be accomplished fairly cheap even if you can't weld. By can't weld I mean you aren't competent to weld on a pressure vessel, and I have no intention of getting into welding discussions on this site or welding certification either.
Long ago I developed a contraption that you can find on over 100 sites called the "Franzinator". Feel free to google it. 90% of the people who say it can't work have never tried building it. Rest assured, if built to specification it does work. 90% of the people who say it doesn't work did NOT build correctly or did NOT place it in the correct location in the air system.
http://z6.invisionfree.com/ToolBoxTalk/index.php?showtopic=1461
I have developed the device over 40 years and maximized its possibility in my use situation. Slight improvements can be acheived for people needing more seperation because of where they are located. Feel free to ask questions on Toolbox, because I see no point in posting what I have to say on multiple sites.
There have been a boatload of immitators, they didn't adhere to the dimensions and got crap results. For nonweldors there is a plan on Toolbox for the No Welding Required version.
The "Franzinator" is not a 100% air/water seperator.
Subscequently I had a second idea for what is called the Jonsiematic Air polisher. It's fairly well discussed at
http://z6.invisionfree.com/ToolBoxTalk/index.php?showtopic=1835&st=0.
Again, feel free to ask questions, same rule applies.
To save somebody typing, copper does not give superior or even equal results to iron in the "Franzinator" because copper is nonradiational by comparison to steel in heat situations.
With a polisher behind a "Franzinator" you will have sufficiently dry air for most aplications, unless you dwell in a swamp.
Since my presumption is nobody here is running a 100CFM compressor and pushing a ¼" tip I see no point in addressing that level of machinery.
In any media blasting situation, water in the air is a problem, regardless of the media. Soda is a 10 on the PITA scale when it comes to moiusture being a problem. Ground glass is about a 3. There is nothing short of dumping rock into a truck that rates a 1 or 2.
Coated sodas are lowering their PITA rating, but not by much.
Soda is easily fracturable, and that means in the hose and gun. Blowing dust out of a hose acheives nothing, unless you're a freak of nature who can't get enough of the taste.
Some of the small soda blasters I've seen in the market appear to attempt to employ the same basic design as a powder fire extinguisher. It's a good idea, BUT, the Bozo designing the machine forgot a few things. Extinguishers are pressurized with either DRY AIR or Nitrogen, and are a limited duration fast dump device. Extinguisher powders also contain a considerable amount of lubricant, generally ground talc. Soda doesn't contain lubricant.
Soda blasting requires metering the media into the airstream. Were the media in a nonpressurized hopper this would be very easy to do, BUT the "designers" have become fascinated with pressurized hoppers because they lack common sense. When you factor in that they pressurize the hopper with the same airsource they use for the blast stream, they screw themselves right out of the box. I've proven this on commercial units.
Metering into an eductor isn't hard to accomplish, most commercial machines do it with a pinwheel, as is shown on thepage I previously posted on dry media eductors. It works well until clumping occurrs in the pot. Most commercial machines try to solve clumping by employing a vibrator or thumper. It works some days. The clowns marketing homeowner soda blasters for a few hundred bucks attempt to solve clumping and metering with sized ports and dump valves. I don't know your level of frustration before you kick something out the door, but mine is failry short. I have no desire to stop blasting to clear a clump every couple minutes, and I long ago stopped listening to excuses from manufacturers and vendors.
Bottom line at this point can a small soda blaster for home use be made in somebody's garage? Sure it can. How many hours are you willing to spend making it? I've got a few ideas on what will work, but no inclination to spend 100+ hours making one and I have a fairly complete shop available.
My first experience shooting soda was more than 30 years ago. I told the "engineers" and the people who wrote the spec it was the wrong media for the job. They knew more than I did, and they signed off on my objections. The smart people were the first to run when the cloud lofted into the sky, and they were noplace to be found when the firetrucks arrived. They also paid the bill for all the load of soda, as well as the cost of doing the job correctly.
Blast media has come a long way over the last 30 years, and there are many purpose specific media, including granulated dry ice. For some unknown reason in the last 2 or 3 years a fascination with soda has developed, particularly in respect to cars. I can't see a single benefit to blowing soda toward a car, it's the wrong damn media. If you want to clean caked on grease soda is a good choice, but not for removing paint. Paint manufacturers warn against using soda on a vehicle, and won't warranty product if soda is used.
BTW, don't ever let an insurance carrier talk you into mold remediation by sodablasting. Soda doesn't kill the mold, but it gets the carrier off the hook when the mold consumes your hovel.