Proportional pneumatic valve with stepper

[ewinek]

Hi all,

I am designing a pneumatic system as follows: There is a container (5 ml volume) which is pressurized at max 1 bar, at the output of the container there is a valve to release the air into a second container, that valve must be operated in a controlled manner with the lowest power consumption. That's the trick.

Do you know any pneumatic valve in a miniature size, controllable for example with a stepper motor? I think that would be the most accurate and low power consumption since no continuous activation is needed to regulate the output. Thanks for the help!

[CaptDon]

You will probably end up building a dedicated PID controller to generate the steps. The problem is also an end of travel limiter like microswitches or opto couplers when your system senses out of range flows and jams the valve in the open or closed position. When a near end of travel limit is reached you use an exclusive or gate with two inputs, first is the direction signal hi or low, the other is the limit signal which when it changes will cause the direction signal at the output of the XOR gate to flip. You may see a two steps closed, two steps open oscillation until the system comes back into proper control but it will absolutely stop the valve from getting jammed!!! I developed and built a water flow control system of 1 to 50 GPM using exactly the system you are looking for. Our stepper was large and consumed power in the hold phase when not stepping and 99% of the time it was stepping as input and output tank levels shifted. My system was 8085 based and all of our sensors were 4-20ma. for safety around water. I still have the first prototype and it runs a modified 5.25 inch floppy drive chassis head position stepper to prove concept with varying sensor signals.

[TurboTom]

You may experiment with a piezo valve since they basically only need a voltage / charge for controlling, albeit at some elevated levels. Here's a link to a model by Festo: https://www.festo.com/tw/en/a/8064292/

How well they work in the proportional regime, probably needs to be evaluated.

[CaptDon]

I added a low frequency 1 bit dither so that the 'average' control was actually better than coming to rest in one position. The 4-20ma. sensors fed an 8 bit ADC so at 1 to 50 GPM each bit represented a somewhat big step so the dither was sort of like adding an extra bit of resolution. It is also why the stepper was generally continuously 'correcting' the commanded flow as compared to just resting at some flow point. 1 gallon per minute as the low flow was determined by a hard travel limit set by an opto coupler and 50 GPM was the max flow the pump could produce with the valve open to the maximum allowed position set by another opto coupler. Microswitches worked but were unreliable in a wet environment. There was an 8 spst DIP switch for system configuration. One switch determined 'level' or 'flow' optimization. One switch commanded a sequence of diagnostics. One switch determined if the 8 segment LED display showed the commanded setpoint or the difference between the setpoint and the actual sensor reading. The remaining switches set some pre-programmed loop ballistics for fast or slow response tailored to stop serious overshoot / undershoot and hunting.

[ewinek]

Thanks TurboTom and CaptDon for the suggestions.

It seems that this kind of valve in the size I need is actually not existing, or not at a budget price.

I think I will try to solve the problem by regulating the pressure in the primary container, I think that would be the easiest and cheapest solution to the problem.