There's a couple resources out there re: using gas thrusters for, say, reducing roll. The conclusion is they're very heavy and expensive to get to work. The most successful systems seem to be canards like what is discussed in the "I could just use a little guidance thread" in the staging, airstarts, and clustering forum. Like a song stuck in my head, i keep thinking about using pressurized water for a thruster. Imagine a 12oz, thick, plastic water bottle with a schrader valve on one end. You use a bike pump to pressurize it and then use something like a diy pinch or poppit valve controlled by a microcontroller for release. What you have is basically a small but powerful water gun. Given the density of water and good old F=MA you should get much more force compared to say co2 or N2. You have to lift the water but at least it's not in a heavy pressure tank. So, why won't this work? Surely it's been tried before. I have the programming and electronics experience necessary for a prototype but I don't want to invest the time and money if it's already been done and shown to not work.