My question is about the more appropiate method to control the instrument air backup.

The instrument air is normally generated in the IA package and flows towards the IA header of the unit. In case there is a problem in the generation package, there is a backup from the IA refinery header. The two possible options are:

1) Installing a pressure control valve in the backup line. The pressure will be measured in the unit header. If the header pressure falls under 5 Kg/cm2 g the pressure valve starts opening and tries to control the header pressure with the refinery header air. When the generation package is recovered the header pressure rises and the backup valve closes.

2) Installing an on-off valve in the backup line with a solenoid. A pressure switch will be installed in the generation package outlet, so when the pressure falls under 5 Kg/cm2g the valve opens allowing the refinery header air in. When the pressure in the generation package is recovered the on-off valve is closed.

A control valve in the backup line with the pressure controller set at 5 Kg/cm2 g (normal instrument air header pressure is 7 Kg/cm2 g). When the pressure falls into 5 Kg/cm2 g the pressure controller opens 50% of the control valve travel and then it tries to control the pressure at 5 Kg/cm2 g. We do so because maybe the pressure control loop is not fast enough to mantain the header pressure.

We think that the problem with the on-off valve is that if the refinery instrument air header pressure is very high, our system may be damaged with the valve fully opened. Another possibility is that if the valve is fully opened and there is a problem in our instrumen air system we may demand so much air from the refinery header that we could trip the refinery instrument air package.

A self contained regulator on the back-up line is the most cost effective, and VERY reliable method.  Used through-out the industry for this service.  But an upgrade if you like would be to user a transmitter controlling a control valve.  Question you would then have is which way to fail the valve on failure (and no I do not mean air failure, but I mean on control signal failure).  Regulators you do not have this issue.  Given you HP back-up air supply I would fail last, but thats me.

Another thought on your solenoid idea (if you really like that) is to use a downstream orifice to cut any flowing pressure.  (I would assuem once the header got back to pressure and the header starts to "not flow" (i.e. pressure on each side of orifice starts to equalize) that the solenoid would close the valve

MORE NEWS