You could play hockey with a chop stick too. Maybe you'd do so to handicap a much less capable opponent? But it's a false analogy to the pneumatic - electric - (hydraulic) debate.
In applications with high cycle rates and high masses, pneumatic servos can be 25% the cost of a linear motor solution, and 30-50% of the cost of an electromechanical solution - while requiring less space.
For many applications that require a combination of controlled, rapid translation (12" in 100ms) coupled with a high force joining operation (think resistance welding, hot melt, etc.), servo pneumatics again has a similar big cost advantage.
Generally speaking, if you look into many of the positioning, general automation applications in packaging and similar industries, you can replace a linear electric axis with servo pneumatics with the benefit of 50% less installation cost, and a 50% increase in cycle rates.
I would compare moving a load with air to using a rubber bat to play baseball, or a rubber stick to play hockey. You could do it, but why would you? ("Because I can" is not a good reason.)
What are the selling points for such a system?
If the system is not exponentially less expensive to purchase and operate, I don't see the advantage. I doubt if the systems are much less expensive to purchase than a comparable electric solution, and I seriously doubt whether the operation cost is lower, when the cost of clean compressed air is factored in. Then have one incident where the dryer floods the air line with contamination and see what the cost of repair is.
For explosive environments I can concede that air is compelling. Otherwise, I don't get it, and I wouldn't design it in over an electric servo system.
For grippers, clamping cylinders, and a host of other short stroke, bang-bang actuators, I'm all for pneunatics. It is also great for certain counterbalancing systems. I just don't see it for positioning systems.
I/we mostly deal with hydraulics. Pneumatic servo control is a very small part of our business because there are usually better solution. However, we use pneumatics as a means of testing hydraulic algorithms. If one can control pneumatics the hydraulics and servo motors are even easier.
Here is an example of what can be done.
You may need to download the mp4 file before viewing depenind on your browser.
We have controlled pneumatics in some low pressure testing applicaitons.
Besides the energy lost in compressing and decompressing air the second problem is that feedback devices are still needed and the controller can't be a simple cheap PID controller. It takes some complicated math to do the control shown and this isn't cheap. For small jobs little linear motors are tough to beat.
It sounds as if you've had a frustrating experience with a poorly designed pneumatic servo system.
Modern, well designed systems are able to rapidly position heavy vertical loads without overshoot, without high levels of valve control activity, and to high levels of repeatability. The 0.005" accuracy you mentioned is readily achievable in many systems.
The last pneumatic positioner I worked on was in a vertical orientation, which created additional challenges for the actuator. The systems I have worked on used "magnetostrictive" position feedback ("Temposonics" would be one trade name). This feedback system is based upon timing pulses transmitted down a rod that get reflected back by a magnetic ring, or something like that. The time is converted to an analog output of some resolution. The resolution doesn't change as the length of the transducer gets longer, so the system resolution is typically affected by the transducer length.
The controls on this system were essentially analog in nature, comparing setpoints to the transducer position, and throttling air valves to try to keep the actuator in the correct position, and move it to new positions, etc.
The end result was less than what I would have desired (Fortunately I didn't pick the equipment, I was just called on to get it to work.). There was a tendency to overshoot significantly (large fractions of inches at least), especially when moving down. I don't know what the final position tolerance was, but it was not comparable to an electric servo actuator. It is also relatively noisy, due to the air valves constantly fighting with each other to try to position the load. I couldn't help but think that there would soon be a mechanical failure in the valves because of the frequency of the switching, etc.
Having worked with literally hundreds of servo systems of many different flavors, and a small number of pneumatic positioners (less than 10 probably), I would favor an electric solution unless there is some overwhelming reason to not use electric (explosive environment?).
Under ideal circumstances pneumatic positioners talk about 0.005" positioning resolution. That might be achievable under some circumstances, but if I were designing the system, I would be thinking a quarter of an inch and designing in energy absorbers and positive stops at the desired stopping points.
As a general substitution for this type of actuator, I would favor the belt drive style slides offered by a number of vendors. They offer high speed capability in a similar form factor, and the precision of a completely digitally controlled electric servo system.
Pneumatics is becoming an important component in the field of Mechatronics for efficiency and motion control precision. Festo is a leader in pneumatic based robotics with their gripper and arm products. I remember taking a Pneumatics class at community college back in the early 80s. I didn't see the importance of this class because of my fascination with electronics. I which more emphasis on system integration was discussed in class as it relates to Mechatronics, maybe I would have considered working for engineering companies like Festo who makes great and cool pneumatics based robots instead of just electronic/semiconductor jobs.
Chuck, Forgot the second part of your question. I don't think pneumatics is too costly. The precision limits are a reflection of the underlying technology, especially when compared to electromechanical solutions which can achieve very high precision, resolution and repeatable accuracy.
Festo's BionicKangaroo combines pneumatic and electrical drive technology, plus very precise controls and condition monitoring. Like a real kangaroo, the BionicKangaroo robot harvests the kinetic energy of each takeoff and immediately uses it to power the next jump.
Design News and Digi-Key presents: Creating & Testing Your First RTOS Application Using MQX, a crash course that will look at defining a project, selecting a target processor, blocking code, defining tasks, completing code, and debugging.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.