Liquid ring vacuum pump control 1

[chicagoeng]

I am trying to find the correct method for determining how much air should be added to the suction of a vacuum pump to control the suction pressure at a specified value within the design range of the vacuum pump. My previous attempts at determining the air flow have not been accurate enough, so I am hoping someone else can shed some light on the correct approach.

 

[katmar]

Unless you have a very unusual application, the quantity of air to be bled into the suction of the vacuum pump will vary because the natural load on the pump will vary too. So you cannot design for a fixed quantity of air.

The way to do it is to put a valve in the bleed line and let it control the suction pressure of the pump. If you do it this way the flow information you read off the pump curve should be plenty accurate enough for you to size the pipe and control valve. Where did you go wrong previously.


Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[chicagoeng]

Thanks for the response. A control valve on the air bleed line with a PIC loop on the pump suction is the approach I have used previously. The issue has been with getting the correct amount of air into the suction to allow the control valve to be put in automatic so it will adjust within a reasonable operating range of the valve. I probably have been using the wrong air flow for the valve design. I have been determining the ACFM requred based on the point on the vacuum pump performance curve that corresponds to the desired pressure. Is this the correct approach? How much additional flow should be added to this value to allow the valve to respond with a reasonable change in valve position?

 

[btrueblood]

...how tightly do you want to control the pressure? A simple spring vacuum relief valve ought to work if you're not too fussy, and there are fancier mechanical regulators that could do better. A PIC loop and control valve is about the last method I'd ever contemplate.

 

[katmar]

In this case I do not regard a PIC loop as overkill because I have seen liquid ring pumps destroy themselves when badly controlled. Especially if the suction pressure is allowed to go too low and the pump runs permanently in a cavitating condition. You can lose a pump in a few hours this way.

I would size the bleed valve to give the full rated volumetric capacity of the pump at the desired suction pressure so that it could cope in the situation where no air is coming from the process. If you need less air than this the valve can close down. We used to say that a valve should never operate at less than (say) 30% open if we wanted decent control, but with modern controllers I have seen butterfly valves operate absolutely stably at only 10% open. Its best not to design for that, but it can work.

Have you asked your pump supplier for help? Here in South Africa I have found the pump supplier's technical backup to be very good. Hopefully you are in a similar position.

What went wrong in your previous application?

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[crjones]

You need to use scfm not acfm. This the mass of gas that you are adding

 

[PISA]

tell us pls. the application, size of the pump, control range (0-100% of the volumic suction capacity is normally not required), we will size a control valve for you. Liquid ring vacuum pumps are easy to control with an air bleed but maybe you could also save energy with a frequency converter (speed control).

Anne Delestre

http://www.cutes-europe.com