For most pumping applications centrifugal pumps are the first choice of engineers. The versatile and simple nature of centrifugal pumps led to the success of this principle that was introduced by Denis Papin in 1689. In today's world many vendors have special designs for specific applications. The selection of a suitable pump is a process in which an engineer first calculates the pumping conditions and then decides which type of pump he wishes to use for his application. After this the search for a suitable pump starts with requisitions, vendor selection and requests for quotations. This is logically followed by evaluation, purchase, delivery and installation of the pump.

Engineering page is independent of any pump manufacturer. The server side software is developed using an array of literature and other independent data about pump hydraulics. The results are neither intended to select nor to exclude any products or manufacturers but to aid engineers with their work.

These results will be useful to establish if a centrifugal pump is a suitable option for your pump application. The calculation results provide preliminary data that will help you to further design the installation long before a centrifugal pump is selected. The data provide an independent benchmark and if the quotations of vendors significantly differ from the data provided by Engineering Page this could be a clear indication of an error.

The operating conditions can be calculated using our Pump Conditions routine. This will provide the operating conditions for the pump. If a centrifugal pump is worthwhile looking at, the Centrifugal Pump routine will provide the useful benchmark as introduced above.

The suction conditions, fluid characteristics and other application data are used to calculate parameters such as the suction specific speed, specific speed, required shaft power, required impeller dimension and volute or diffuser dimensions. A lot of care was put into the prevention of cavitation. The calculation routines are too complicated to describe in detail, and providing all formulae would not help you very much. The important aspects, however, are provided underneath:

To prevent cavitation the NPSH available (Net Positive Suction Head) is the starting point.

with:

NPSH_available= Net Positive Suction Head in Pa

P_{SuctionFlange} = Pressure at the suction flange in Pa

½ rho v² = dynamic head in Pa, units:density rho in kg/m³ and velocity v in m/s

p_vapor = Vapor pressure of the liquid Pa

If the NPSH available is low this can be improved by modifications to the design; for example by relocating the pump to a lower location or reducing the suction line pressure drop using a larger diameter suction line.

The dimensionless parameter that describes the suction condition is the Suction Specific Speed.

This is usually expressed in US units, i.e.:

S = suction specific speed

n = impeller speed in rpm

Q = volumetric flow rate in gpm

NPSH = Net Positive Suction Head in ft

The values have a large bearing on the design of the suction side of the impeller as the impeller needs to cope with the condition without cavitation. A typical value is SSS = 9000. This can be achieved with centrifugal pump impellers of good manufacturers having a flow angle of about 17 ° and approximately 5 to 7 vanes. If your pumping application is a condensate pump you are more likely to end up with a figure of 12000 to 18000. To improve the suction characteristics the flow angle and number of vanes needs to be reduced, e.g. to 10 ° and 4 vanes. A disadvantage of this is that the pump is more likely to run rough at partial load. A low flow angle will also result in a larger impeller diameter and more expensive pump.
There are more tricks that pump suppliers apply, such as inducers, that will help to cope with these difficult suction conditions. The routine will make you aware of how difficult the application is and what measures are generally seen as remedies.

As can be seen from the equation reducing the speed will reduce the tendency for cavitation. This is a solution ... but it will increase the price. Another solution is to put two or more pumps in parallel service which reduces the capacity at the impeller suction eye. (Using a double suction pump provides two impeller eyes and is in fact based on this principle). There is a square root in the equation, so unfortunately this is not a linear relationship. An elegant solution for applications with a low NPSH available and a high pumping head is to put a booster pump with a low speed at the suction side followed by a high speed multi stage pump.

One impeller could theoretically deliver an enormous head. The tip speed would, however, become very high. This would cause excessive wear at the tip of the impeller, so effectively the impeller diameter is limited by the speed and the allowable tip speed.. The simple solution is to put another impeller in series. The software calculates the number of stages that will deliver the required total pressure without exceeding the tip speed.

The program in fact calculates all the important impeller dimensions such as angles, suction and tip diameter, the width and so on to conclude the characteristics. The dimensionless parameter that is used for the shape of the impeller is the Specific Speed:

Ns = Specific speed

n = impeller speed in rpm

Q = volumetric flow rate in gpm

H = Head in ft (for multi stage pumps: this is for one individual impeller)

These are used to calculate the performance of the pump.
Many other parameters are important for pump design such as:

impeller angles and velocity triangles

slip (differences between real and ideally guided fluid velocities)

volute or diffuser design

We refer to the literature for more detailed descriptions, for the interested reader: we use correlations that describe the full velocity triangles including c_m3/u₂, ß₂, ... the complete vane lay-out. A volute or diffuser is selected including throat design, c_thr/u₂ ratios etc.

The efficiency of an impeller is a result of many factors that pump hydraulic engineers balance to design an impeller. Above a number of the aspects were described.

The program will calculate the efficiency taking into account the hydraulic efficiency, the volumetric efficiency, the disk friction and will make an estimate of mechanical losses. These have all been incorporated into the program to provide you with the best estimate of the pumping efficiency and more importantly, the required shaft power to drive the pump.

Pumping theory was historically based on pumping the fluid that is most important to a human being: water. A pump that was designed for water operating with a more viscous fluid will perform differently. Head, capacity, efficiency and required power need to be corrected for this. Viscosity correction graphs have been published by the hydraulic Institute. These have been completely incorporated into the program. It is also possible to get the results of this separately using the Viscosity Corrections routine.

In fact one of the complicated parts of the development of the software was to keep the output simple and not reply with a deluge of parameters. We believe the transparent presentation will help you more and hope that you agree that we succeeded in this.