Calculation
of
Losses

Updates
and
Additions

Home | IFS Content | History | Purchase | Contact

Internal Flow Systems

History

 

1. Loss Coefficient Definition

Research at the University of Munich in the early 1930s was the starting point for the generation of reliable pressure loss coefficients for piping and duct components. A procedure for measuring pressure losses was developed that:

• Generated defined flow conditions at inlet to a component by using a sufficient length of inlet pipe to achieve a developed friction gradient prior to a component
• Measured the friction gradient in the outlet pipe sufficiently far downstream from a component where a developed friction gradient was established
• Projected the inlet and outlet friction gradients to a component and took the difference between the gradients as the pressure loss. In plotting the gradients a component is represented as having zero length.
• Calculated the loss coefficient from the ratio of the pressure loss to the mean dynamic pressure at inlet to a component.

This ideal procedure has to be adapted to deal with factors such as: a short or no inlet pipe, changes in flow area between a component's inlet and outlet, dividing and combining flow, interactions with another components.

The procedure is in effect a means of communication so that researchers and users familiar with the procedure understand what a loss coefficient refers to.

 

2. BHRA

In 1947 the British Hydromechanics Research Association (BHRA later to become the BHR Group) was set up by a number of UK companies with government support to improve fluid related equipment and systems. This led to 30 years of research and development aimed at improving the design of flow systems. I joined BHRA in 1965 at a time when engineers at BHRA had shown that improved design of power station cooling water systems could reduce pumping costs by 20%, without increasing construction costs; a saving in today’s terms of several tens of million pounds over a 1000MW power station’s life.

A contract from the UK’s Central Electricity Generating Board initiated extensive studies at BHRA to provide loss coefficient data for large flow systems. These studies were continued by support from Industry and the Department of Trade and Industry. Research was also ongoing into fluid transients in piping systems, compressible and other complex flows.

By the late 1970s gas compression costs in North America reached a billion US dollars per year, with an estimated 10% of energy input being dissipated within poorly designed compressor yard pipe work. The Pipeline Research Supervisory Committee of the American Gas Association funded studies at BHRA to extend the experimental work on large flow systems to include interactions involving dividing and combining flow.

Whilst research on pressure losses was on going at BHRA, model studies were also being carried out on a wide range of flow systems. These included projects related to conventional and nuclear power stations, hydroelectric and pumped storage schemes, chemical, oil and gas, water and other process industries as well as projects for industrial and defence companies. Involvement across many industrial sectors provided insights into what engineers and designers required from an internal flow system guide.

 

3. Need for a Design Guide

A powerful driver for the writing of Internal Flow Systems was investigating flow system problems and failures in the UK and overseas. My experiences were reflected in the literature with the publication of reports on failures of many different types of flow systems, some of which involved hundreds of millions of US dollars of remedial and consequential losses. Relatively simple changes at the design stage, such as the provision of a well designed diffuser at entry to a heat exchanger or other plant item, would often have avoided failure and consequential losses.

I became convinced that there was not a lack of knowledge for the design of efficient internal flow systems but there was a serious and very difficult problem in getting information to engineers and designers in a form they could readily use. A more disciplined and informed approach to flow system design was needed because of:

 The growth in flow system size and complexity
 Over sizing of large pumps leading to numerous system and pump problems
 Energy savings from good matching of fluid machines to flow systems
 Structural optimisation making systems more susceptible to failure from flow induced vibrations and fluid transients
 The need to intensify flow based processes
 A greater awareness by engineers, designers, and industries of their responsibilities in providing safe and reliable systems.

Even today (2009) articles continue to be published on the poor matching of fluid machinery to flow systems and the resulting consequential costs that are estimated to be in excess of a billion US dollars per year.

University based research on internal flows virtually ceased during the 1960s as funding for experimental work dried up. It was evident that by the end of the 1970s the same would happen for organisations like BHRA. A situation was developing where considerable internal flow data and knowledge would be lost to industry as those with experience to interpret the extensive literature retired or moved on to more active areas of research.

 

4. Writing Internal Flow Systems

Extracting information and data from the literature and putting it in a form to meet a broad spectrum of industrial needs was beyond any contract that BHRA could secure. In view of this and my interests in understanding and adding value to others’ work, writing Internal Flow Systems became a ‘hobby’ for ten years, followed later by a further five years in writing the second addition. In this endeavour I had access to BHRA’s library which was probably the best library for internal flow literature. Organisations around the world had for many years sent restricted readership reports on internal flows to BHRA.

Much of the published experimental data on component pressure losses was not gathered using the Munich experimental procedure. However, where sufficient information was available on the experimental methods used I was able to correct data to conform to the Munich procedure. Reliable experimental data for a few components extended over a wide range of Reynolds numbers and provided a good base for extrapolating Reynolds number trends for other components.

It became clear in analysing pressure loss coefficient data from hundreds of sources that to maximise the value of published data it needed to be classified as to its reliability and suitability for design. The result was the classification system used in Part 2 of Internal Flow Systems. Loss coefficient data included in the book was verified as far as possible against experiments at BHRA and/or the work of a number of groups around the world that had contributed experimental results over an extended period.

In the years between the first and second edition I revised some of the loss coefficient data in the light of new data and comments from users. These changes are outlined in an Appendix in the second addition. Since first publication of the second edition only minor corrections have been made to the text and figures.

Understanding flow behaviour in internal flow passages is essential when flow distribution and stability are important, such as at the inlet to fluid machines and to processes such as heat transfer and chemical reactions. For flow passage geometries not included in Internal Flow Systems, one can make a reasoned estimate of pressure losses based on an understanding of flow behaviour in flow passages of similar geometry. Information on flow behaviour in a wide range of flow geometries is included in Internal Flow Systems.

The Reynolds number is the only flow parameter involved in the flow of Newtonian fluids through a component until:

• Pressure within a liquid flow falls to the liquid's vapour pressure when a cavitation parameter called the Cavitation number is involved.

• Flow velocities exceeds 20% or so of the speed of sound in the fluid when a compressibility parameter called the Mach number is involved

Extensive studies on cavitation in piping system components was initiated in the 1970s by the Metropolitan Water District of South California; the largest bulk water supplier for municipal use in the world. The researchers for these studies adopted a Cavitation number based on a component's inlet velocity. This Cavitation number definition was adopted for Internal Flow Systems as loss coefficients in Internal Flow Systems are generally based on a component's inlet velocity.

The Mach number is well known from its application in aerodynamics but less so in internal flows. A number of approximate procedures for calculating compressible flows are available in the literature for pressure relief and venting pipe work but these procedures are not generally applicable. A general procedure was developed for compressible flows that uses incompressible loss coefficients with correction factors that depend on Mach number.

To accumulate sufficient data to derive the Mach number correction factors in Internal Flow Systems it was necessary to re-interpret a significant percentage of published experimental work on highly compressible flow in components. Static pressure gradients become very steep as Mach numbers approach one making it very easy to misinterpret static pressures measurements. By reinterpreting publish compressible data sufficient data was available to derive compressibility correction factors that cover basic components.

 

5. Software

Early on in the writing of Internal Flow Systems I knew I should be putting information and data into software in order for it to be in a more readily usable and up dateable form. However, this was outside my competence and beyond the capabilities of the existing software techniques. Before the first edition of the book was published I started seeking venture funding for BHRA to develop such software. This was to be a five year quest before initial funding was secured and development started on what became the Flowmaster suite of computer programs, developed and marketed by the Flowmaster Group (www.flowmaster.com).

© D.S. Miller (See Permission to Use)

.