Don Wood's blog

Pressure Surges Caused by Routine Fire Flow Pump Tests Can Result in Sprinkler System Pipe Failures and Catastrophic Damage

Several years ago I was performed a pressure surge analysis of a high rise fire protection sprinkler system which had been subjected to a serious pipe system failure during a routine fire flow test. During the fire flow test in a high rise sprinkler system a cap was blown off a standpipe due to a high pressure surge causing over $5 million in damages. The test involved the opening of a deluge valve in a dry pipe section followed by the startup of a fire pump. A transient analysis of the sprinkler system showed that this action was capable of producing very large pressure spikes at the 9th floor standpipe cap (in excess of 1,000 psi).

By Dr. Don J. Wood and Dr. Srinivasa Lingireddy

Back in the 1960's pipe system transient analysis code using a technique developed by Dr. Wood was written to handle propellant flow transients for NASA. This technique, initially referred as the Wave Plan Method, was further developed and around 1984 the first commercial software (SURGE1) was released for general use. Since then several thousand software packages have been supplied to engineers worldwide and this software has been applied extensively, undergone numerous QA procedures (nuclear facilities) and, in general, been exhaustively tested and verified.

It has been suggested that the Wave Characteristic Method (WCM), a.k.a. Wave Plan Method, is less accurate than the original Method of Characteristics (MOC) for transient pipe flow analysis. This is a totally false representation. The reality is that the MOC is a very inefficient technique for pipeline transient analysis. Handling larger piping systems will often be impossible using the MOC or will take hours compared to minutes to attain a solution.

We often get feedback from our clients. One of the most valued inputs are comparisons of field and model results. I recently was sent a copy of a paper published in the June 2011 issue of the South African Institution of Civil Engineering magazine. The article is entitled “Hydraulic modelling and field verification on the Withoogte to Besaansklip bulk water supply pipeline” and was written by a very knowledgeable Surge2010 user, Kobus Prinsloo, from the company SSIS (Pty) Ltd. The article compares field and model results for this strategic South African pipeline. The pipeline is 62.7 km long with diameters ranging from 1100 to 1500mm.

The following paragraph from the article explains the project.

"SSIS was subsequently commissioned to compile a hydraulic model of the Withoogte to Besaanklip pipeline in order to mimic the steady state and transient behavior of the pipeline following various operational scenarios. The hydraulic model was compiled using the Surge 2010 analysis software and calibrated against actual measured pressure data captured on site using Remote Transient Pressure Monitors (RTPM).”