Project Profiles : SSiS

Project Profiles - Pipeline Services

Sahara Audio-Video Successfully Locates Two Leaks on a DN1600 Steel Pipelines

20 May 2012 

Leak detection and simultaneous CCTV inspections were recently performed on a critical transmission main in Gauteng.  Approximately 2.4 km of the pipeline was surveyed and two (2) large leaks were detected.

 Neither leak presented any surface signs i.e. standing water or abnormal vegetation.  The leaks have probably been present for at least five years and would have gone unnoticed for many more years if it was not detected.  Using the simultaneous CCTV inspection capability of the system, it was possible to also confirm that both leaks were located at pipe joints that had previously been repaired. Visual observations also enabled confirmation of the internal pipe condition as well as the extent and severity of biofilm growth inside the pipeline.

The inspection outcome will aid the water utility in accurately planning for the leak repairs.

Leakage may or may not be classified as a structural failure depending on its location and its severity. Corrosion of the pipe wall may lead to perforation and leakage. Leakage can also occur at joints and contribute to mechanical failures due to eroding pipe bedding support. It has been noted on multiple occasions that pipelines often leak before they fail and that many leaks on large diameter pipelines never surface.  Leakage can therefore be seen as an indicator of wall penetration by pitting and the integrity of pipe joints and is hence a useful indicator of the overall pipeline condition.  In addition, other very useful information is generally gathered regarding the overall operation and condition of the pipeline such as air entrapment and movement, location of bends and illegal connections etc.




Sahara Finds 'Undetectable Leak'

Recent hydrostatic pressure tests on a newly completed section of an 1800 mm diameter steel pipe forming part of a major new water scheme, showed a loss of pressure in excess of the acceptable standards. It was evident that the pipeline was leaking and the Contractor was facing penalties if the source of the leak could not be found.

The Contractor first carried out extensive visual and physical checks to confirm that all components were sound. When the pressure test still failed, the services of a leak detection specialist were procured that made use of the latest conventional acoustic correlation techniques. After spending three days on site, trying numerous techniques and technologies and excavating two (2) ‘dry holes’, the specialist contractor failed to locate the leak(s).

SSIS Pipeline Services (SSIS) was contracted to search for the elusive leak(s) using the Sahara inspection technology.

The leak was detected on the very first inspection!

Upon excavation of the leak it was found to be at the invert of the pipeline on a defective joint at the exact location pinpointed by Sahara. During the pressure testing, in excess of 100 000 litres of water were pumped into the pipe section. The leaked water, however, quickly disappeared into the surrounding permeable soil. As with most high-pressure bulk pipeline leaks, the identified leak did not surface nor did it show any surface signs of a leak.


The Case to Assess Large Diameter Pipelines

Within the water conservation and demand management (WCDM) fraternity, there is often the mistaken perception that leak detection on large diameter transmission mains is not important as these leaks will nearly always surface. Nothing could be further from the truth. Based on extensive inspections performed to date on bulk pipelines (i.e. ≥ 300 mm diameter) in South Africa, more than 80% of all high-pressure leaks never showed any surface signs of leakage.

Equally important is the fact that although there are physically many more leaks on smaller distribution pipelines (and the perception that fixing these leaks would be more cost-effective and beneficial compared to bulk pipeline leaks), the actual contribution of these leaks to the total system losses (i.e. distribution + transmission pipelines) could vary from 63% to as little as 13% (see Table 1 below).

The conventional acoustic correlation approach to leak detection is generally accepted and proven within the reticulation and urban environment on smaller diameter pipes with many service connections spaced closely together.

Transmission mains, however, present difficulties for this approach. Sound waves attenuate more quickly as diameters increase, meaning that the larger the pipe, the closer together the access points need to be for these methods to be effective.

Leak noise decays rapidly as it travels down the barrel of the pipe and water column and is exacerbated by decreasing pipe stiffness associated with increasing pipe diameter. On large diameter pipelines the spacing between access points is invariably not sufficient as was also shown in a recent comparison study conducted by Sydney Water. The project again confirmed the fact that even large leaks on transmission mains do not necessarily show surface signs and also proved the inefficiency of conventional acoustic correlation on large diameter pipelines.

Given the high pressures, significant water loss, good probability that no surface signs will be evident and the strategic importance of transmission mains, there is an important case to be made to implement more comprehensive and accurate large diameter leak detection programs.

Apart from the significant water loss that can be avoided (currently estimated at an average of 150 kilolitres/leak/day), leak detection is also an important first indicator of the condition of a pipeline. Early detection and repair can improve the reliability of a system and potentially avoid catastrophic failures later in the lifetime of the pipeline.

Table 1: Comparison between Distribution and Transmission Pipeline Leaks Based on the Number of Size and Leaks (IWA 2010, Transmission Main Water Loss Reduction in Urban Centers)

Proven Technologies

Proven large diameter leak detection technologies include Sahara and SmartBall. In both cases the acoustic sensor is brought to the source of the leak noise by travelling inside the pipeline.


Sahara®is a tethered inline leak location and a non-disruptive condition assessment technology that pinpoints the location and estimates the magnitude of leaks and air pockets in bulk water pipelines of all material types with a diameter of at least 300mm.The latest development of the Sahara® system incorporates the ability to perform simultaneous acoustic leak detection and live Closed Circuit Television (CCTV) inspection on water mains. It is now possible to pinpoint the location of leaks while at the same time inspecting the internal condition of the pipeline using a combined acoustic hydrophone with integrated CCTV camera. The system uses the same Sahara insertion platform and is capable of being inserted into live pipelines.


SmartBall®combines the sensitivity of acoustic leak detection with the 100% coverage capability of in-line inspection. The free swimming device is spherical and smaller than the pipe bore allowing it to roll silently through the line and achieve a high responsiveness to small leaks. SmartBall is ideally suited for leak detection of long transmission pipelines with limited laterals or turnouts. With 12 hours of battery life, 20-30km of pipeline can be inspected per day from one insertion pointThe free swimming device is housed within a foam ball containing an aluminium sphere that contains sensitive acoustic instruments. It is inserted into and extracted from the pipeline under live operating conditions through 100 mm diameter flanged openings fitted with a valve in water and wastewater applications.

Bloem Water Pro-Actively Manages Critical Bulk Pipeline

10 September 2013 

Leaks on small diameter distribution pipelines are the most common leak a utility encounters. However, locating and repairing leaks on large-diameter transmission pipelines is also important in maintaining safe and reliable service delivery. These leaks are often more sparse, and therefore more difficult to locate which can lead to prolonged leakage and extensive water loss. 

Utilities that have a leak detection program in place for their large-diameter transmission mains often achieve greater reductions in Non-Revenue Water (NRW), which is the amount of water lost before it reaches the customer.  Furthermore, a leak detection inspection is a valuable step as part of a condition assessment program.  Utilities can avoid expensive capital replacement programs by gathering real data on the condition of their pipelines, and addressing problems as they arise.

Bloem Water, who provides water services to the central region of South Africa, recently implemented an asset condition assessment project that included a comprehensive leak detection program using inline methods on a strategic 1200mm (48in) Pre-stressed Concrete Pipe (PCP) that supplies the City of Bloemfontein with roughly 60 percent of its drinking water.

Inline leak detection is a very accurate method of leak detection because the acoustic sensor used to identify the sound of the leak is brought directly to the source, unlike traditional methods, such as correlators and listening sticks.  Such techniques lack the accuracy needed to locate leaks in larger pipes because the sound of a leak dissipates rapidly in large diameter pipes.

Both SmartBall® and Sahara® Audio Video inspection technologies were successfully applied.  A total length of 103 km between De Hoek reservoir and Brandkop reservoir was inspected.

To maximize efficiency, the SmartBall® leak detection technology was used first to cover large sections of the distance in single deployments. The SmartBall tool is a free-swimming leak detection platform that operates while the pipeline remains in service. It is equipped with an acoustic sensor that identifies acoustic anomalies associated with leaks; the acoustic signature is then analyzed to determine if it is a leak, air pocket, or an external noise.

The SmartBall® inspections were followed by selective Sahara® AV surveys to provide visual confirmation on the location of the leaks in order to aid in the interpretation of findings and plan interventions.  Sahara® is a tethered inline leak location and condition assessment technology that pinpoints the location of leaks while at the same time inspecting the internal condition of the pipeline and verifying the cause of leakage using a combined acoustic hydrophone with integrated CCTV camera.

Visual leak verification was also performed at all accessible components along the pipeline.

Leaks were identified and classified as either pipeline leaks (i.e. on the pipe barrel itself), component leaks (i.e. at valves, air valves, scour valves, etc.) or off-takes.  The majority of the leaks were found to be at components while pipeline leaks were detected mainly on joints (i.e. where the PCP was connected with prefabricated steel joint pieces). Some of the leaks on these joints may be attributed to poor bedding, improper VJ connections, etc. The estimated water loss based on indicative leak sizing categories amounts to approximately 1200 kl/day. Interestingly, of the majority of pipeline leaks detected, only three (3) leaks showed any surface signs of leakage.  This re-iterates the importance of not relying on surface inspections as the only means of detecting large diameter pipeline leaks.

The findings of the latest leak detection surveys were compared to that of a previous Sahara leak detection inspection project performed in 2007-2009 to establish trends.  This information, combined with a first order engineering evaluation of available pipeline design and manufacturing data, the failure history of the pipeline, operating records and the hydraulic behavior of the system were incorporated into a risk assessment model for the pipeline. 

The leak detection findings, engineering evaluation and risk assessment were factored into the development of a pipeline specific management strategy. Bloem Water can now implement this strategy as a guideline to pro-actively manage this valuable asset in order to prolong its remaining useful life, avoiding expensive capital replacement of the asset.

For more information please contact:

Mr Mokutu Kgwale

Bloem Water

This email address is being protected from spambots. You need JavaScript enabled to view it.



SmartBall®  Leak Detection Equipment




Clamping SmartBall® in Foam Shell Prior to Live Insertion



Inserting SmartBall®into Closed Isolating Valve




Insertion Site Setup with Isolating Valve Open



SmartBall® Extraction Site Setup



Sahara® Insertion Tube and Winch Assembly




WCDM Pro-Active Programme

Bulk water supply pipelines rarely fail suddenly or catastrophically.  Small leaks, if left unattended, will however over time turn into larger leaks and ultimately lead to pipe breaks or failures.  It is estimated that repairing a break reactively costs three times as much as to proactively fixing the leak.  Accurate leak detection is therefore an effective means of establishing the condition of bulk supply infrastructure, in addition to effectively managing risk and deferring capital expenditure.

The West Coast District Municipality (WCDM) completed a comprehensive leak detection survey as part of a pro-active program to assess the condition of their bulk water supply infrastructure.  The investigation was performed by SSIS using the precise Sahara® leak detection system.   The following four bulk water pipelines were inspected.   

Approx. Pipe
Length (m)
Pipe Dia.
Pipe Material
1 Besaansklip Reservoir to Vergeleë Reservoirs  11 000 900 Pre-Stressed Concrete
2 Swartland WTW to Kasteelberg Reservoirs  15 500 500 Steel
3 Withoogte WTW to Besaansklip Reservoir 62 500 1200 / 1100 Pre-Stressed Concrete & Steel
4 Misverstand Dam Pump Station to Withoogte WTW  13 500 1000 Steel
  TOTAL 102 500    


Pipeline Nos. 1, 3 and 4 were built as part of the Bergriver-Saldanha Water Project in the mid 1970’s while Pipeline No. 2 forms part of the Swartland Water Supply Scheme and was built in the late 1960’s. All the pipelines are therefore older than 40 years and are approaching the end of their theoretical design life.

Inspection Procedure

Since Sahara® is an in-line inspection technology, access into the pipelines was obtained at existing air valves which are frequently spaced along the pipelines. The under-pressure insertion was achieved by removing the air valve and fitting a specially fabricated flange and an insertion tube on top of the gate valve. The Sahara® sensor head, inspection cable and parachute were then deployed under-pressure through the insertion tube into the pipeline while the pipeline remained in service. The flow of water opens the parachute and pulls the sensor and cable behind it.

Inspection Findings

Seventy (70) insertions were performed at air valve locations after WCDM removed the air valves. A distance of 85 km, or approximately 83% of the combined pipe lengths, was successfully inspected between February 2010 and May 2010. Inspection distances varied from 50 m to a maximum of 1895m with an average inspection distance of 1 258 m per insertion.

Ten (10) leaks were discovered on the pipelines with an additional eleven (11) leaks found at components like air valves, scour valves or farm connections. Seven (7) of the pipeline leaks were found on one of the pipelines and all the pipeline leaks were found to be on the pre-stressed concrete sections of the pipelines. The leaks were clearly marked on the surface and GPS coordinates taken for record and future reference purposes. Once again, the leak location re-iterated the important fact that even large leaks do not always show tell-tale surface signs such as moist ground conditions or standing water, with four of the large leaks showing no surface signs and another ‘showing’ approximately 15 m upstream from the actual position of the leak.

Worldwide, the average number of leaks discovered on transmission mains with the Sahara® inspection technology is 1.4 leaks/km while in South Africa, the figure is 0.42 leaks/km based on surveys performed to date. The overall leakage rates found on the WCDM pipelines are therefore well below worldwide and local averages. The leak sizes were estimated based on the acoustic signature detected to allow prioritisation of repairs. Based on the estimated sizing of the leaks, the combined losses from the WCDM system is approximately 800 - 1000 m3/day.

Two of the leaks were excavated shortly after the inspections confirming the accuracy of the technology. The remaining leaks were repaired during the upcoming maintenance schedule.

In addition to the leaks detected, the inspections also documented the location and condition of the pipeline components like air valves, scour valves and isolating valves. Some components were found to be in a poor condition and in urgent need of repair and/or refurbishment. This provided valuable guidance on the mechanical maintenance requirements of the pipelines for future planning, budgeting and asset management purposes. The presence, movement and approximate extent of entrapped air in the pipelines were also determined using the Sahara® technology. Although air movement was detected along a number of pipe sections, no evidence of stationary air columns were found.


Fact Box

  Client: West Coast District Municipality
  Project: Physical condition assessment of four bulk water pipelines
  Region: Western Cape
  Inspection Distance: 85 km
  Inspection Duration: February to May 2010
  Pipeline Material: Steel and pre-stressed concrete
  Pipe Diameter: 500 mm to 1500 mm
  Number of inspections 70
  Average insertion distance: 1258 m

Key Project Outcomes:

1.  Discovery of twenty one (21) leaks of varying sizes 
2.  Assessment and documentation of the condition of pipeline components such as air valves, scour valves and isolating valves 
3.  Establishing a baseline for the pipeline condition against which future inspections can be measured to assess trends. 
4.  Improved understanding of the condition and operation of the system enabling pro-active management and more focused maintenance       




Leak Survey on DWS Strategic Hendrina to Duvha Pipeline

SSIS initiated a leak detection and location inspection on the strategic bulk water steel pipeline linking the Hendrina and Duvha Power Stations. The precise and real-time inspection was carried out on the in-service pipeline using the revolutionary Sahara® inspection technology without any flow or water quality interference. The 32 km long steel pipeline of 1397 mm O.D. was laid in 1979. Access into the pipeline was obtained at thirty (30) insertion points provided at existing air valves. No special drilling or under-pressure connections were required to gain access into the pipeline. A total length of 30 118 m of pipeline was successfully inspected between 9 June and 6 August 2008. The completion of the project was delayed by a couple of weeks due to two insertion points (valve chambers) being located directly underneath Eskom high voltage power lines. For safety reasons Eskom requested that no work should proceed underneath the high voltage power lines until the lines have been decommissioned.

Pipeline Pressures, Flow Rates and Leak Simulation

At each insertion point (i.e. under-pressure access point into pipeline), both the operating pressure and the flow rate were measured by removing the air valve and fitting a pressure gauge and by inserting an insertion-style flow meter into the pipeline. In addition, GPS coordinates were also recorded. Measured pressures typically varied between 110 kPa and 600 kPa while the flow rates varied between 158 Ml/d to 220 Ml/d. Measured flow rates were converted to flow velocities in order to correctly size the Sahara® drag chute. The calculated flow velocities typically varied between 1.23 m/s to 1.7 m/s. In most cases, the flows were more than sufficient to launch the under-pressure inspection equipment and to carry the inspection cable and sensor head to sufficient distances along the pipeline. Of interest, the resulting average inspection distance per insertion was 1000 m with a maximum inspection distance of 1556 m.

At the time of the inspection, the flow in the pipeline was reversed to gravitate from Hendrina Power Station to Duvha Power Station as opposed to pumping from Duvha to Hendrina. As a result, the operating pressures under gravity flow were very low varying from as little as 1.1 Bar to a maximum of 6.0 Bar. As expected, the reported inspection sensitivity is affected by the pipeline operating pressure making it a little harder to detect small leaks (say less than 100 l/hr) at very low pipeline pressures. Therefore, several “trial leaks” were simulated by barely opening gate valves on the pipeline to determine the minimum threshold values. Under these conditions with pressures of about 3 Bar, leaks on the order of 75 l/hr could still be detected. Since the gate valves were offset by at least 1.0 m from the top of the pipe, the expected leak sensitivity on the pipe barrel would be substantially better presumably well below 50 l/hr even at the very low pressures. Leaks at pipe risers and pipe connections such as at off-take pipes, cross-connections, gate valves and scour valves are generally less audible due to the faster dispersion of acoustic energy and to a smaller extent, the reduction in pressure head. Based on the trial leaks, It was concluded that the inspection technology was more than capable of locating leaks of a size that are of a concern.

Leaks Detected and Verification

Only two leaks were detected and located on the entire length of pipeline inspected. Leak No 1, detected during Insertion No. 26 at pipe chainage CH27+707 m was about 250 litres/hour (Medium leak) while Leak No. 2, detected during Insertion No. 31 at pipe chainage CH3+553 m, was very large and estimated to be well over 1000 litres/hour. GPS coordinates were recorded at both leak locations.

From a water loss point of view, it was suggested that both leaks should be excavated and repaired. From a pipeline durability and structural integrity point of view, it was suggested that Leak No. 2, situated in a farmer’s maize field, should be repaired as a matter of utmost urgency. Also, because Leak No. 2 was so large, it was difficult to determine with absolute certainty that there was no other leak immediately next to or close to this leak (within say 2 m to 4 m) although two leak signals were faintly distinguishable. Therefore, it was recommended to open the pipeline say 4 m upstream and downstream of this leak as well. In fact, DWS (Department of Water & Sanitation) recently excavated the pipeline at Leak No. 2 and located the leak at the indicated position. Also, DWS did in fact found a second leak within about 2 m from this leak as suggested. Both these leaks were successfully and expediently repaired without affecting Duvha Power Station’s water usage. However, Leak No. 1 is scheduled for repair during the next available shutdown due to the high current water demand by Duvha Power Station.

Anomalies Detected

Anomalies are defined as sounds or events detected other than leaks or unexpected events encountered during inspections such as “broken glass” sounds associated with entrapped air, sudden flow changes (i.e., reducer in pipeline), pipeline route changes etc. Four types of anomalies were encountered namely:

Vertical and horizontal pipeline bends were easily detected.
An open bulk off-take on the pipeline was easily detected as the sensor approached it. 
Entrapped air pockets were only detected during two of the insertions indicating that the air valves were still functioning properly. 
A reduction in the pipeline diameter was detected where an old flow meter used to be connected. This resulted in heavy localized flow turbulence along a short section of pipeline.


Summary Remarks

Given the generally high water table and saturated soil conditions along several parts of the pipeline, and the fact that the cathodic protection (CP) system is not currently operational (as it was vandalized and stolen on a couple of occasions), the current condition of the pipeline from a water loss point of view seems to be remarkably well. This can only be attributed to a well-designed pipeline, properly installed and maintained over the years (i.e., functioning cathodic protection system). However, as evidenced by recent leaks, the integrity of the pipe barrel is questionable and could be deteriorating fast due to the inactive CP system.

Unfortunately the pipeline traverses through several vlei areas with high water tables, and possibly corrosive soil regions, coupled with overhead power lines running parallel to the pipeline in places (in addition to rail crossings), making it absolutely critical to reinstate the CP system urgently. Although critical sections along the pipeline may have to be identified and re-surveyed to establish the possible rate of deterioration, it may be more prudent to design a robust CP system that cannot be tampered with in future!






Created by 1021 Media & Design