|
Parallel Pumping Operation
Started by drdave, Mar 29 2005 11:01 AM
11 replies to this topic
Share this topic:
#1
Posted 29 March 2005 - 11:01 AM
Sir,
One centrifugal pump delivers chilled water (at 4 deg_C ) 43 cu.m./hr at TDH of 37MLC(Meter liquid column).
Running of standby pump in parallel just increases flow to ~50 cu.m./hr at TDH of 39 MLC.
WHY DOES PARALLEL PUMP RUN NOT INCREASE FLOW RATE TO ALMOST DOUBLE (ie ~82 cu.m./hr)?
Does it only because of system resistance?
May you please guide me on this?
#2
Posted 29 March 2005 - 11:21 AM
should be check the standby pump maybe a bad performance impeller or restriction on pipe/checkvalve .
Check this first to take futher steps. Or try 't with trottled discharge to increase pressure/flow.
les
Check this first to take futher steps. Or try 't with trottled discharge to increase pressure/flow.
les
#3 Guest_mvancleave_*
Posted 29 March 2005 - 12:52 PM
Dave,
You have it right in that system resistance (headloss) is the reason the flow does not double.
As you probably know, a centrifugal pump operates on a curve based on the flow rate and delivered head. The curve goes from shut-off head (the head at zero flow) and the delivered head decreases as flow increases. The shape of this curve looks somewhat parabolic, but it is a function of pump design and construction (impeller, volute, etc.).
The system headloss is a function of the square of the system velocity plus any static head. Since you are talking about a chilled water system, this is a loop and there is no static head.
If you plot the pump and system curves on the same axes, the will cross at a specific head and flow. This is the system operating point.
Two identical pumps running in parallel will produce twice the flow of a single pump for a given head. Thus, the pump curve for parallel pumps is constructed by doubling the flow-rate of the single pump for each head condition. The new operating point is where the system curve crosses the parallel pump curve, which will be at a higher head.
The attached diagram may help. If you can get the model number of the pump, you can probably contact the pump manufacturer to get the pump curves, which will also shed some light on the subject.
Hope that was the answer you were looking for,
Mike
You have it right in that system resistance (headloss) is the reason the flow does not double.
As you probably know, a centrifugal pump operates on a curve based on the flow rate and delivered head. The curve goes from shut-off head (the head at zero flow) and the delivered head decreases as flow increases. The shape of this curve looks somewhat parabolic, but it is a function of pump design and construction (impeller, volute, etc.).
The system headloss is a function of the square of the system velocity plus any static head. Since you are talking about a chilled water system, this is a loop and there is no static head.
If you plot the pump and system curves on the same axes, the will cross at a specific head and flow. This is the system operating point.
Two identical pumps running in parallel will produce twice the flow of a single pump for a given head. Thus, the pump curve for parallel pumps is constructed by doubling the flow-rate of the single pump for each head condition. The new operating point is where the system curve crosses the parallel pump curve, which will be at a higher head.
The attached diagram may help. If you can get the model number of the pump, you can probably contact the pump manufacturer to get the pump curves, which will also shed some light on the subject.
Hope that was the answer you were looking for,
Mike
Attached Files
#4 Guest_mvancleave_*
Posted 29 March 2005 - 12:58 PM
The image in my previous post was missing some annotation. Hopefully this one works. I have indicated the two operating points and labeled the curves.
Mike
Mike
Attached Files
#5
Posted 03 April 2005 - 05:32 AM
mvancleave
Thank you very much for your educative feedback in nutshell.
Will only Increase of discharge header sizing help to get benefit of almost double capacity by parallel run of pumps?
What preventive measures to be considered later on for occasional single run?
Thanks once again.
#6 Guest_Guest_Ravi_*
Posted 03 April 2005 - 06:23 PM
Yes, you can increase the flowrate by increasing the pipeline size but it is not practical always.
You should take care not to overload a single pump motor when it is run independantly.
Please note that when your system has constant resistance(like a partially closed valve), the system curve will not start from the origin and thus slope of the system curve changes. You should be careful when plotting the system curve.
Regards,
You should take care not to overload a single pump motor when it is run independantly.
Please note that when your system has constant resistance(like a partially closed valve), the system curve will not start from the origin and thus slope of the system curve changes. You should be careful when plotting the system curve.
Regards,
#7
Posted 23 April 2005 - 03:42 AM
Cooling water pumps in parallel performs well during test in workshop but in actual operation ,There is 20% less flow available obviously with higher discharge pressure indicating system resistance.My queries are :
[1] What are the ways & means to increase capacity in the available pumping system as shown in the sketch(keeping two pumps in parallel)?
Say, each pump capacity is 10000 m3/hr @ 3.5 bar_a.
[2] How system resistance curve will be developed in this case? Here, 15 meters back pressure of CW return header to top of cooling tower will be there.
[3] What are effects of suction velocity,submergence and capacity control during parallel pumping of two with the third one idle in common suction pit?
Attached Files
#8
Posted 26 July 2007 - 10:59 PM
[1] What are the ways & means to increase capacity in the available pumping system as shown in the sketch(keeping two pumps in parallel)?
Say, each pump capacity is 10000 m3/hr @ 3.5 bar_a.
In practice, for two parallel working pumps ,we can not obtain two times of capacity of a single pump , and for increasing total capacity of two existing parallel pumps (going toward double capacity) we must decrease the dynamic resistance of the system , for example , increasing the piping size.
regards
Say, each pump capacity is 10000 m3/hr @ 3.5 bar_a.
In practice, for two parallel working pumps ,we can not obtain two times of capacity of a single pump , and for increasing total capacity of two existing parallel pumps (going toward double capacity) we must decrease the dynamic resistance of the system , for example , increasing the piping size.
regards
#9
Posted 27 July 2007 - 03:46 PM
You can increase motor frequency
Regards
Regards
#10
Posted 28 July 2007 - 01:55 PM
Dear A A Mishra,
I would like to bring into your notice that double flow we don't get because of the system resistance and not because of the Pumps or its drives are limiting in this case. Only option to get the total flow for what these two pumps are designed is to design your discharge line/system that can sustain or take this much of the quantum and same is achieved through increasing the discharge line size for the required flow and pressure.But in case the frequency of running both pumps at a time is rare then need not to go for increase the line size as we can achieve some increased flow with an existing line when both pumps will run i.e. with an existing flow design margin of the discharge line.
Now to check or change in the pumps' drives or its parameters will come into picture only when there is problem to get the rated flow through pump for what it is designed then you have to have an inspection of pumps or the cross-check of its drives provided rest of the things are ok ( like density,viscosity,temperature of the liquid.suction and discharge conditions like line sizes and all)
I hope you will agree with my opinion,Please if you have any opinion different from above I will like to see in this forum.
Cheers......Njoy........
Regards,
Padmakar
I would like to bring into your notice that double flow we don't get because of the system resistance and not because of the Pumps or its drives are limiting in this case. Only option to get the total flow for what these two pumps are designed is to design your discharge line/system that can sustain or take this much of the quantum and same is achieved through increasing the discharge line size for the required flow and pressure.But in case the frequency of running both pumps at a time is rare then need not to go for increase the line size as we can achieve some increased flow with an existing line when both pumps will run i.e. with an existing flow design margin of the discharge line.
Now to check or change in the pumps' drives or its parameters will come into picture only when there is problem to get the rated flow through pump for what it is designed then you have to have an inspection of pumps or the cross-check of its drives provided rest of the things are ok ( like density,viscosity,temperature of the liquid.suction and discharge conditions like line sizes and all)
I hope you will agree with my opinion,Please if you have any opinion different from above I will like to see in this forum.
Cheers......Njoy........
Regards,
Padmakar
#11
Posted 28 July 2007 - 05:01 PM
Say, what does parallel pumping look like? Is it something like the following ASCII art?
| -------------------- |
| chamber |
= <-- pump1 |
= <-- pump2 |
| -------------------- |
- Bryan
| -------------------- |
| chamber |
= <-- pump1 |
= <-- pump2 |
| -------------------- |
- Bryan
#12
Posted 01 August 2007 - 03:19 PM
The performance of two pumps operating in parallel depends strongly on the shape of the pump curves. As an extreme, the pump curve of a positive displacement pump is essentially a straight vertical line - the flowrate is the same regardless of the discharge pressure. In this case, employing two pumps in parallel WILL DOUBLE the flowrate. For a centrifugal pump, and especially one with a "flat" curve, increases in the system resistance will decrease the pump's capacity, perhaps quite quickly. If a control valve contributes significantly to the system resistance, it could open to allow for a larger flowrate. Alternately, specify pumps with steeper curves and you can get a more significant capacity increase by operating your pumps in parallel.
Doug
Doug
Similar Topics
Operation Of Heat Exchanger Above The Overall Heat Transfer CoefficienStarted by Guest_Papps_* , 13 Jan 2024 |
|
|
||
Two Parallel Jt Valves For GasStarted by Guest_Mohamed_Mamdouh_* , 19 Jun 2023 |
|
|
||
Aspen Edr - S&t: Process Side Series / Cw Side ParallelStarted by Guest_mitch22_* , 06 Feb 2023 |
|
|
||
Switching From Parallel Mode Into Series ModeStarted by Guest_ahmed abd elmonem_* , 17 Jan 2023 |
|
|
||
PumpingStarted by Guest_engineer24_* , 05 Jan 2023 |
|
|