sonic nozzles and venturi for pressure control 1

[bevs]

I am presently looking into designing a pressure controller, that will work over the following:
Min Pressure = -1 bar
Max Pressure = 400 bar
Pressure source = Nitrogen, Compressed air.
Pressure supply may be regulated to 110% of the actual working range (i.e. working range = 10 bar, regulated supply = 11 bar).
Internal volume can vary from what would be the internal volume of the manifold (say 10 CC) to an external volume of 100 CC.
Once pressure has entered the system, solenoid valve (supply and vent) will open and close to allow the gas in. The solenoid valves have a cycle time of 2 - 5 mS, and will be PWM to achieve pressure control.
I am aiming for pressure control of approximately 0.1mbar
My main problems I believe will be :
Flow rate of gas entering the system, this will be a big variable due to the pressure it is regulated at, and at what pressure the controller will be working at (i.e. Regulated pressure = 430 bar, set point of controller = 5 bar). Am not sure if a sonic valve (laval nozzle) will help as it will give me a constant flow, but will be dependant on back pressure (controller set pressure) and the gas may well end up at supersonic flow which will not help control.
A venturi will reduce the flow rate and improve pressure control, but not give me a steady flow rate which i believe i will require to obtain pressure control.
Has anybody got any other pointers/ideas that may help, books to read?
Cheers

 

[israelkk]

I assume that you want to control the "average" pressure because the pressure in the volume will be fluctuating at the frquency of the PWM frequency and due to time constant of charging and discharging the volume.

Secondly, the flow rate into and out off the volume is absolutely non constant since it may be choked or non choked. Pneumatic process is a highly non-linear phenomena.

It is impractical to set the regulator to 430bar and the controller to 5bar and expect to control the pressure inside the volume accurately.

For example, if the inlet pressure is 430bar and the controlled pressure is 5bar, the flow rate into the volume will be 86 times higher than the rate out of the volume. Therefore, the orifice area of the discharge valve must be at least 86 times larger than the orifice area of the inlet valve. Add to this that the pressure in the volume decreases when the discharge valve is opened and you will find that the orifice area of the discharge valve should be even greater.

Therefore, the regulated pressure should be closer to the desired pressure in the volume, meaning that you may need a very fast variable pressure controller which I am not sure physically available.

1 mbar accuracy seems to me an almost impossible goal in your case.