Centrifugal pumps control

[MTQ80]

Hi all
I need your expert feedback on below problem. This is related to newly installed pump station on the cross country pipeline.

One new pump station with three pumps are installed mid way of existing cross country 1200 km pipeline. Pumps are going to commission. We are worried that during start-up pumps will run at end of curve.

Pumps are with VFD with max speed 5300rpm with no discharge control valve. Normal pressure in the pipeline at this point is 1200psig. Pumps will start first on recycle mode with suction pressure 950 and discharge pressure 1500psig at 3400rpm. Pumps will shift from recycle mode to operating mode by opening discharge ZV. During this shifting time pump resistance is lower than pump discharge pressure. Under this condition can pump move to end of curve.
This case was simulated and came to know that pumps go to end of curve by achieving maximum speed 5300 rpm during shifting from recycle to operating mode. This result very low pressure at suction less than vapor pressure of NGL fluid.

Please give your opinion how this situation can be handled

 

[Artisi]

It amazes me how a pump system as described can now be subject to questions as to the mode of operation- I can not believe that such a system can be designed and built and now someone comes online to ask how to operate it.

Or am I missing something?

 

[MTQ80]

I will add that system includes all protections like low suction pressure, high flow, high high discharge pressure and recycle control valves. the only issue is that how we can avoid pump to reach end of curve during shifting time from recycle to operating mode without shutting down the pump.
If you need specific information, please let me knwow.

 

[BigInch]

You need either to limit pump speed with a low suction pressure override, or via using the discharge control valves. If you have those protections like you say you do, how is it that you're having this problem with the simulation? Are you running the model without having tuned PIDs and/or using bad time constants on VFD and CV actuators?

Artisi, happens all the time. Many pipelines are designed without ever considering actual operations. They figure if the horsepower is there, you can run it.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[rmw]

I too agree that way too often design engineers will look at the rated flow conditions of a system and design for it not ever considering how you get from point zero to design flow or what will happen if operation happens to occur off of the design point.

A lot of my engineering career has been fixing just such problems. It is worth $$$ to someone.

Right answer, BigInch.

rmw

 

[BigInch]

Indeed $$$$ it is. I've spent a lot of time working out start, failed start, product change, receipt, delivery, recirculate and stop sequences for pipelines for multiple pump stations with long pipelines from refineries to storage plants and distribution or marine terminals reaching over a couple of mountain ranges and with widely varying SG products in the lines, things can get a bit complex at times with all the settings one has to develope. A good pipeline simulator helps.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dabluffrat]

start it against a closed valve, or 3/4 closed valve, or use the VFD as a soft start.

 

[BigInch]

MTQ80,

Why not start and run the pump up to 3000 rpm to give you about the same 1200 psig discharge pressure you have now? You could bring the control valves down to that corresponding %open and once the line has about stabilized at its original flowrate, start increasing rpm to build speed to 5300 rpm? I don't believe you can't do that. If not, I'm sure you must have your control valves acting too slowly, or maybe your ramp rate is too fast. Increase the actuator's stroke speed, or use a slower pump ramp rate.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/