This tech seems to explain well how exactly electrojet thrusters of Ironman's Suit work! :-)

Cannot measure the thrust? Did they think about putting the whole thing on a scale, and then turning it on?

Whoa, whoa, whoa! Wait a minute there, Skippy!! "but such things are no use in the Earth's atmosphere, as accelerated Xenon ions lose most of their thrust force to friction against the air"? The thrust of ANY rocket engine or jet engine is due solely to the Delta P of the pressure inside of the engine and the outside of the engine. The forward "wall" of the combustion chamber has force being applied against it, and the atmosphere outside of the engine has virtually NONE being applied against it! Thus, the forward wall of the engine moves in the same direction as the force being applied to it. The exhaust gasses doe NOT "push" against the various gas molecules that are outside of the combustion chamber of the engine as they are nearly totally mobile, and there is not much to push against, the gasses get "left behind" as the engine moves in space, it's not like when you lean upon a wall. If you increase the surface area of the wall of the engine, or increase the pressure/temperature in the engine, or decrease the resistance to the flow of the gasses leaving the engine, you will increase the thrust of the engine. It's a fine balancing act that keeps engines running at optimum thrust, and not blowing up!

I am Drocketman!

Whoops! I forgot to mention that it would be easy to measure the thrust of the contraption using a load cell under the whole thing, and take a base reading with it stone cold, then with just the air on and then with air and full power applied.

it sounds as if a big part of what makes it work is transfer of energy from the plasma to the unheated air around it. If they can improve that mixing process (a lot) then you might be able to go with something that moves a bunch more air (a little like high-bypass jet engines) and has a lower exhaust temperature.

I do worry a little about the experimental setup for measuring thrust. If there's significant RF leakage, you could get heating of that steel ball that might show more of an effect than you actually have.

Did I miss a part where hooking this up to a fission reactor on a spacecraft wouldn't make sense?

If the exhaust is at 1000*C this thing is not going to be efficient. That heat needs to be converted to kinetic energy to generate motive power. Need to maximise exhaust momentum to maximise thrust.

Perhaps application possibility as ICP-MS/OES heat source.

A little thing called VASIMIR has done this and more.

Isn't there a blast of compressed air screaming up that tube? Wouldn't that be enough to bounce that ball around?