Hydraulic Analysis of Fire Protection Water Supplies
In Part 3 of this series we introduced how to diagnose problems in an underground main when the measured friction loss does not match the actual friction loss. In this final section, we will solve an actual problem that occurred at an industrial facility. The parameters were changed to protect the identity of the facility. You may wish to reread Part 3 before proceeding. You will need the slide rule to work the problems on your own.
This was a first property loss prevention survey being conducted at a newly built industrial facility. Fire pump tests were conducted with satisfactory results. Valves in the loop were checked and appeared to be open. A loop test was then conducted with recently calibrated gauges. The loop is shown in Figure 1.
There are many ways to conduct a loop test but this method has served well in detecting problems. Figure 1 shows that, after following all impairment procedures outlined in NFPA 25, Standard for the Inspection, Testing, and Maintenance of Water-Based Fire Protection Systems, one side of the loop was shut off and then water was flowed all the way around the loop. The idea is to create a long run of pipe so that friction loss issues will be more likely to manifest themselves.
Initially a gauge was not used on hydrant B, The difference between the pitot reading and the pressure on hydrant A was examined. This is essentially the same as the basic fire department pump pressure equation: EP = NP + FL where:
- EP = engine (or pump) pressure (or in this case the pressure at hydrant A)
- NP = nozzle pressure (or in this case the pitot pressure at the flow hydrant)
- FL = friction loss
According to the slide rule, a pitot pressure of 51 psi corresponds to 1200 gpm. You need to account for friction loss in the hydrant. I have used 5-10 psi as a rule of thumb for this kind of rough analysis. Three to five psi is more representative of new hydrants.
Since we are starting at 130 psi and ending with 51 psi pitot pressure, we have an estimate of 74 psi friction loss when allowing for 5 psi hydrant loss. Using the calculator and the techniques discussed in the earlier articles, you can see that this represents nearly 9,000 feet of 8" plastic pipe.
The actual pipe length being tested was approximately 2200 feet so something is obviously wrong.
The next step is to do a more accurate friction loss test that removes uncertainty about the friction loss in the hydrant. It also eliminates an engineering argument that is frequently raised. The argument says that at point A "normal" pressure was being measured and at the hydrant pressure "velocity" pressure was being measured. This engineering argument is entirely correct but in using the rough technique as a screening tool, it has been corroborated by the more accurate test being described next. Please contact the author by e-mail to discuss the details of the engineering argument.
There is also elevation loss between the main and the hydrant that must be accounted for. In this case, the test was in a warmer climate and the depth of bury of the main would at most have a 2 psi difference so it was ignored in the screening. The land was flat so grade elevations did not need to be corrected for.????
In the detailed test a gauge was added to hydrant B and the test repeated. This resulted in 60 psi at hydrant B, which means there was 70 psi friction loss between A and B. This corresponds to approximately 8,500 feet of 8" plastic pipe.
As we discussed in Part 3, it is time to do some detective work. The valves were rechecked to be sure that they were fully open. Once again, they all appeared to be. The next step was to measure the friction loss in shorter segments of the pipe to see if there was a sudden drop somewhere that would indicate a partially shut valve or another obstruction of some type. Remember that the valve to the flow hydrant could also be partially shut. A terberculated main was not considered because it was observed to be plastic pipe when installed.?
?After eliminating local problems in the main, the problem was thought to be the installation of a different size pipe. The rest of the loop test was completed and all impairment checklist items were completed to be sure the system was restored to full service.
The next step was to search for the as-built drawings. This took quite awhile. What we found was that 6" pipe was installed instead of 8" pipe. The 70 psi friction loss corresponds to about 2,050 feet of 8" plastic pipe which is almost exactly how long the pipe was between A and B.
Once that was done, we had to decide what to do about the 6" main. All of the fire protection was designed based on being fed by an 8" main. Recalculations were needed to see what could be achieved by the 6" main. Such calculations can be done by sprinkler contractors, engineering firms, or insurance loss control engineers. Once those results are in the owner, fire marshal, and insurer need to determine a course of corrective action.
Questions can be directed to the author at John_Frank@swissre.com or +1 770-569-7082.
Swiss Re is the world's leading and most diversified global reinsurer. The company operates through offices in over 30 countries.? Founded in Zurich, Switzerland, in 1863, Swiss Re offers financial services products that enable risk-taking essential to enterprise and progress.? The company's traditional reinsurance products and related services for property and casualty, as well as the life and health business are complemented by insurance-based corporate financial solutions and supplementary services for comprehensive risk management.