7 replies to this topic

[AH1979]

I have recently carried out some very simple pump hydraulic calculations using the standard method of first selecting the fixed pressure points of the upstream and downstream vessels, pressure controllers etc and then back back calculating the inlet/outlet pressures for the control valve on the pump discharge so that the control valve can be sized.

 

My problem is that one of my clients who is the Process Manager for the client company doesn't understand this concept and thinks that control valves are an instrument engineers responsibility and has nothing to do with process, but he does not understand that the instrument engineer needs input from process to size the control valve.  Keeps saying that we are "completely wrong", even though this is a very simple calculation and this is the way any process engineer is going about it.  The problem is that he is saying we must repeat everything done his way at our expense and there are no other process engineers in the client company with the expertise to challenge him.

 

My question is as follows, are there any international standards that outline the standard method of how to conduct a pump hydraulic calculation.  We need a silver bullet that we can use to defend our work and avoid the large expense of having to rework everything his way.  I'm also concerned that he is bad mouthing our company to other people in the industry.

[Zauberberg]

There are different guidelines and standards from one company to another, but no matter which one you follow the common sense should be your best adviser.

 

Control valve sizing is extremely dependent on hydraulic parameters of the circuit where the valve is employed. These consist of: static head difference between the source and the receiver, equipment pressure drop, and piping pressure drop. The available (reserved) differential pressure of a control valve is selected by making frictional pressure loss calculations of all the equipment installed in the hydraulic loop including the piping at the maximum and minimum flow rates anticipated, not including the control valve. The difference between these two numbers is the minimum differential pressure that must be available to the control valve to achieve flow control in the hydraulic circuit. The guideline that one company standard recommends is 20-30% of the overall circuit differential pressure at maximum flow rate. Some experts (see attached article) recommend up to 50% of the piping pressure loss to be available for the control valve pressure drop.

 

Clearly, when process flow rates are at the maximum, hydraulic circuit installed process equipment uses maximum and the control valve is using minimum differential pressure. Conversely when flow rates are at the minimum process equipment uses minimum and the control valve absorbs maximum differential pressure.

 

So, the truth in your argument with the client's manager is somewhere in the middle - all elements of the hydraulic circuit should be sized first (including the control valve), and then the pump should be selected based on total head requirements. Normally you don't start with the pump, unless this is a revamp case where the existing pump remains as given.

Attached Files

•  Avoid Pitfalls When Specifying Control Valves.pdf   520.31KB   153 downloads

[AH1979]

Thanks for your reply. This is a revamp project and the pumps have already been ordered. We have also defined all the pressure drops in the circuit, however the client is saying we don't know what we're doing. Any suggestions?

[AH1979]

Also I forgot to mention. The pump selection and order was done by our client and is not our scope, it is a like for like additional of a third pump in parallel with the existing 2 pumps.

[Zauberberg]

Show them the attached article if they can't understand verbal arguments. If the Process Manager persists with his attitude, you (or your superior) should go to a higher instance and speak to the Engineering Manager or Project Manager of the client. This shouldn't be like children playing, when somebody keeps saying "yes" and the other one "no", just for the sake of arguing.

 

It is very easy to illustrate the point: 0.5 bar pressure drop available for control valve in service where piping pressure drop is 1 bar will be just perfect, and 0.5 bar pressure drop for control valve in service where piping pressure drop is 10 bar will simply not work.

[shan]

Client is god.  There is no silver bullet that is able to kill god.

[Zauberberg]

Perhaps a proper memorandum signed by all parties can be rubbed against the client's nose later on, if everything is done the way they suggested and the system doesn't work in the end. That hurts more than silver bullet.

[Bobby Strain]

My clients asked for stuff. If they paid for it, and accepted a schedule delay, then they got it. For fixed price contracts they got nothing without approved change order. These things must be sorted out at project management level. I'm sure you are in no position to agree to extra work. So, simply kick it higher and move the project forward by giving your instrument engineer the data that she needs. Lots of clients are idiots, and there is nothing you can do about it. You never, however, want to give a client something that doesn't work properly, even if he asks for it. Should you proceed down that path, you may be taking on lots of liability. And, you should be able to get the job done right and make your client happy, too.

 

Bobby