# PWM flow control and thrust

If I have a water pump and a fast solenoid valve (can be open or closed) and turn on my pump at a constant voltage and apply pulse width modulation to my valve. Would the resulting thrust created at the water output be constant and can it be controlled nicely by the pulse width modulation process?

What would be the equations that could model the reaction of such a setup?

• I seriously hope you're suggesting reducing the speed of opening/closing the valve, and not to cycle the valve open/closed so rapidly as to make the water stream pulse-width-modulated. As for equations, you'd better get some feedback of the outgoing water flow and just apply PID or such, because attempting to guess the curve of water flow vs duty cycle in such a setup (nonlinear flow, nonlinear actuator, input pressure and valve torque mutually dependent, plus all the parasitic induction and capacity) is a doomed endeavor.
– SF.
Sep 25, 2015 at 22:28

Yes, above some frequency, the mechanical system actuated by the solenoid won't "see" individual pulses, just the average. Usually at a higher frequency, the solenoid itself will smooth out the pulses and maintain are more average steady current.

Electrically, this is basically creating a switching power supply that controls the current thru the solenoid, with the solenoid being the inductor of the switching power supply.

This is assuming your "solenoid nozzle" (do you really mean "valve"?) is intended for other than binary on/off operation, often called a proportional valve. Trying to drive a binary mechanism to in-between states may not end well. Constantly banging it between on and off may be even worse.

• Yes sorry I meant valve, I edited my question. And yes that is exactly what I try to do , having a direct control on the water flow with a very low response time. Last time I checked there is no such a thing as cheap proportional valves. And I heard that fuel injection in car engines is also controlled done using binary valves and pwm. Sep 25, 2015 at 9:33