Parellel Pumping

[madithak]

Hi there sorry I have not been able to make a sketch and post here. I's like to ask the experts a question.
background: I have two centrifugal pump non-identical. One is pumping side-draw off a de-C3 tower (P-1) and other pumping overheads from a de-C4 tower (P-2). The product from both pumps combine and go to MEROX. I feel we need a pressure transmitter at the junction so the flow control from one pump will cascade control with a fixed pressure thus you have a steady state flow going to MEROX.

Seem obvious to me but I have the instrument and controls guy that says there is no need for pressure transmitter as both pumps will automatically (magically) arrive at the junction at the same pressure.

I may not have presented the picture as you need it but thanks for your comments.

Madi

 

[EmmanuelTop]

This setup cannot really be called "parallel pumping" because there are two different sources and only the delivery point is identical. Obviously, two pumps delivering fluids to the same header must deliver them at the same pressure, and that is the header pressure.

Is the pressure controlled anywhere downstream of the junction point? We need a sketch to understand the configuration better.

Dejan IVANOVIC
Process Engineer, MSChE

 

[itdepends]

Well the control guy is right- both pumps will "magically" arrive at the same pressure at the junction. They'll achieve that by a balance being achieved between the discharge pressure provided by each pump and the frictional losses in the discharge pipe work from each pump to the junction. Both the losses and pump discharge pressure will vary as a function of flow (as well as a few other things).

What controls the flow from each pump? Does each pump have a control valve or VSD?

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[don1980]

The pumps are not dictating the pressure at the point where the streams converge - that pressure is completely dictated by the downstream system's resistance and it has nothing at all to do with the pumps. Your confusion stems from a misunderstanding of that point. Whether you have two identical pumps or two dissimilar pumps is irrelevant. All that matters is that the each pump has enough head to overcome the pressure at the point where those stream converge.

You probably have a control valve on the discharge line of each pump (upstream of where the streams meet). Let's illustrate the point I'm making by use of an extreme example. Say the pressure at the convergence point is 50 psig. Now let's say that Pump-A is putting up 100 psig at its discharge. Regardless of whether Pump-B is putting up 200, or 500, or 15000 psig at it's discharge, this has no affect on the pressure at the downstream side of the Pump-B control valve (or at the convergence point where the pressure is 50 psig). Instead, this excess pressure from Pump-B is simply being wasted across the control valve.

Once a stream gets past the control valve, the pressure in the pipeline is only influenced by the resistant in the downstream system. It has nothing to do with the driver, whether that is a pump or a compressor.

 

[gcastan]

if you need to improve your pumping system and your Merox, you can always make a hydraulics study. I could help you on that.

Gabriel Castaneda, P.E.

http://www.gabcheminc.com

 

[dcasto]

The don is right, pumps control nothing, the system controls the pumps! Oh, that's what my coworkers call a Decastoism, no, it's just the fan laws.