Home | IFS Content | History | Purchase Internal Flow Systems
Experimental Measurement of Loss Coefficients
References
1.Bettis Programme Bettis 1 - Coffield, R. D., et al. Piping Elbow Irrecoverable Pressure Loss Coefficients for Moderately High Reynolds Numbers. WAPD-T-3064, Bettis Atomic Power Laboratory, West Mifflin, Pennsylvania. Accessed on www.osti.gov/bridge/servlets/purl/78568-yAJEII/webviewable/78568.pdf Bettis 2 - Coffield, R. D., et al. Irrecoverable Pressure Loss Coefficients for Two Elbows in Series with Various Orientation Angles and Separation Distances. WAPD-T-3177, Bettis Atomic Power Laboratory, West Mifflin, Pennsylvania. Accessed on www.osti.gov/bridge/product.biblio.jsp?osti_id=471350 Bettis 3 - Coffield, R. D., et al. Irrecoverable Pressure Loss Coefficients for a Short Radius of Curvature Piping Elbow at High Reynolds Numbers. WAPD-T-3190, Bettis Atomic Power Laboratory, West Mifflin, Pennsylvania. Accessed on http://www.osti.gov/bridge/servlets/purl/650318-LFAoSE/webviewable/650318.pdf. Bettis 4 - Coffield, R. D., et al. Irrecoverable Pressure Loss Coefficients for Two Out-of-Plane Piping Elbows at High Reynolds Numbers. WAPD-T-3252, Bettis Atomic Power Laboratory, West Mifflin, Pennsylvania. Accessed on www.osti.gov/bridge/servlets/purl/7647-8LcJVH/webviewable/7647.pdf Bettis 5 - Sigg, K. C., Coffield, R. D., Qualification of a Method to Calculate the Irrecoverable Pressure Loss in High Reynolds Number Piping Systems. WAPD-T-3467, Bettis Atomic Power Laboratory, West Mifflin, Pennsylvania. Accessed on www.osti.gov/bridge/servlets/purl/807254-JWHZrb/native/807254.pdf
As discussed in Part 2 the Bettis 2 bend loss coefficients are test facility dependent and this applies to Bettis 3, 4 and 5. For reasons discussed in Part 2, the loss coefficients cannot be corrected to be compatible with loss coefficients measured under the defined conditions used for the loss coefficients in Internal Flow Systems. The Bettis 5 project involved a new test facility but this still had restrictions on bend inlet pipe length. A multi-hole flow straightener was located at the start of the bend inlet pipe to remove disturbances and because it was thought it would provide developed flow at the bend. In reality high intensity turbulence after the flow straightener would prevent the start of the growth of a pipe flow boundary layer to eight or so pipe diameters downstream of the straightener. As a result friction gradient measured for the inlet pipe would have been considerably steeper than for developed pipe flow. Also the flow profile entering the bend would only have been developing towards a profile under which the loss coefficients in Internal Flow System were measured. The Bettis 5 flow meter was located down stream of the test bends. On a drawing of the test facility, a flow straightener is shown installed prior to a bend connected to the inlet pipe to the flow meter. If a flow straightener was installed at this location it would have caused unknown, very disturbed and possibly unstable flow conditions into the pipe proceeding the flow meter. The flow meter and the flow straightener were sent for calibration but no mention is made of the amount of pipework included in the calibration. The miss understanding over the role flow straighteners reinforces the need for guidance for internal flow experimental projects.
2. Utah Water Research Laboratory (UWRL) A series of reports on piping system components carried out at the Utah Water Research Laboratory of the Utah State University by Rahmeyer, W. J, for the American Society of Heating, Refrigeration and Air Conditioning Engineers (ASHRAE). Reports available through ASHRAE bookstore as:
The UWRL projects were carried out under contract from the ASHRAE. The heating, refrigeration and air conditioning industry has historically quoted and plotted loss coefficients against pipe velocity for air and water flows. In other industries this method was superseded over 50 years ago by adoption of the non-dimensional fluid flow parameter the Reynolds number. The UWRL projects involved one of the most extensive programmes of work carried out to determine loss coefficients. It is unfortunate that ASHRAE did not consider other industries needs and require the results to be made available as loss coefficients versus Reynolds numbers Loss coefficients were based on static pressure measurements made too close to component outlets to include all the loss due to a component. This is particularly so for the tests on large bends where loss coefficients need to be increased by 30% or so to make them compatible with the loss coefficients in Internal Flow Systems. Importantly, bend loss coefficients were shown to become independent of Reynolds number at Reynolds numbers below 10E6.
3. St. Anthony Falls Laboratory (SAFL) Extensive comparative tests of steel piping fittings for the ASHRAE.
Like the UWRL projects the project was carried out for ASHRAE. Again the report is a factual account of the loss coefficients measured with little attempt to relate the loss coefficients to others work. Loss coefficients are again presented in terms of pipe velocities but at least in an Appendix they are plotted against Reynolds number The UWRL and SAFL measured loss coefficients for a number of geometrically similar components. This was an ideal situation for establishing loss coefficient correction factors that users could apply to components whose geometry differed from a reference geometry. The cognitive process involved in determining correction factors causes one to gain much greater insight into experimental measurements and to identify inconsistencies in measurements. Using a correction factor approach to presenting loss coefficients requires that a researcher establishes reference loss coefficients. Neither the UWRL or SAFL projects established reference loss coefficients that could provide a definitive link to others work. In effect the projects generated loss coefficients for their worlds. The UWRL and SAFL project reports contain loss coefficient data for component geometries not covered in Internal Flow Systems. Data in the reports needs to be reanalyzed and put into a form that user's of Internal Flow Systems are familiar with. When time permits the author hopes to do this, unless others have already done the analysis and published the results so they can be referenced on this web site.
4. Zanker, K. J., Brock, T. E., A Review of the Literature on Fluid Flow Through Closed Conduit Bends, (1967), TN 901, BHRA, Cranfield, Bedford. 5. McKeon B. J. et al, Friction Factors for Smooth Pipe Flow. (2004) J. Fluid Mech, vol. 511, pp 41-44. Cambridge University Press © D.S. Miller (See Permission to Use) |