2002-055-RES ACCEPT PROPOSAL OF CARTER & BURGESS FOR PROFESSIONAL ENGINEERING SERVICES TO WATER LOST AND UNACCOUNTED FOR STUDY IN CITY'S WATER AND ACCOUNTING AND OPERATING SYSTEM

Home Browse Search

. . Proposal . Comprehensive Water Distribution System Audit . the transfer standard be followed. Additionally, the transfer As an example, for a tracer concentration, (I' of 50,000 . mg/L, and a trocer flow rate of 0.01 gpm, and a standard should be calibrated before going to the field, and . downstream tracer concentration of 5.0 mg/L, the process after being used in the field. flow rate is calculated as, . For pitottubes, the assumption that the average flow in the . 0.05 x 50,000 = (0 + 0.05) x 5.0 pipe is being measured is important. Pipes exhibit a parabolic velocity profile, and multiple measurements are . 500 = 50 + 0.05 required to obtain accurate results. . o = 100.01 gpm After evaluating each meter's installation, the (arter & . The selection of the tracer is critical. The tracer cannot react Burgess Team will use the in situ calibration procedure that is most appropriate for each meter. . with any of the other compounds in the process flow for the procedure to be accurate. Additionally, it is desirable to use Leak Detection . a tracer that is not present in the process flow, because the The location of leaks in the distribution system can be very . need for a background sample upstream of the point of challenging. One of the most common ways to locate leaks tracer addition is eli minated. is to use geophones and listen for leaks in the system. This I procedure is generally quite precise, and it will provide I Mixing in the pipeline is important. The basic assumption is some relative indication ofthe size of the leak. However, thatthe tracer will be uniformly mixed with the process flow the procedure is best employed when the approximate I so thatthe downstream tracer concentration measured is an location of the leak is known. For a distribution system accurate indication ofthe actual downstream concentration. wide study the exclusive use of sonic detection methods is I generally cost prohibitive. I The rate of tracer addition must be known with good accuracy for this method to be used. The rate of tracer addition is I usually determi ned by timing the draw down in the calibrated J tracer container. For this reason a graduated cylinder is usually used as the container. The metering pump must I pump at a constant rate, therefore, multiple volume readings are taken on the calibrated container. I Transfer Standard . J The use of a transfer standard to calibrate flow meters is 1 relatively common place; however, it is important to document that the transfer standard is properly calibrated. I Transfer standards used for flow meter calibration usually Fortunately, there are some steps that can be taken to narrow J include either clamp-on ultrasonic meters or pitot tubes. If down the areas to be investigated. In most cities the night- used correctly, either instrument can function adequately as time water consumption for reSidential/commercial uses is ] a transfer standard. A typical transfer standard installation quite small, especially between 2 am and 4 am. During this 1 is shown in Figure 5. time, portions ofthe distribution system can be isolated and It is important that the recommendations for proving an water elevations in the elevated tanks monitored. Falling J water elevations are on indication of possible leaks. undisturbed flow path both upstream and downstream of ] Carter :~Burgess 18 J ~ . ~, '-',-1':: 'l- 'I' . d._~ -..., ..-. -. .- ....-.. .:... ..;.......