Under pressure

Materials World magazine
,
8 Apr 2015

Turlough Guerin, Environmental Manager at First Solar, analyses the causes of fluid spills from plants and equipment on resource construction sites, and offers some possible solutions.

More than 370 million litres of oil leak from hydraulic equipment every year, excluding all other sources of plant fluid losses. That’s the equivalent of the volume of Cologne’s beer production at its peak, or enough to fill 150 Olympic-sized swimming pools. Even a minor leak of one drop per second will empty almost all the contents of on 205L oil drum over 24 months. Apart from the downtime and productivity losses, the safety and environmental ramifications of such leaks should be of concern to project managers.

Sealing technologies in hydraulic systems have advanced considerably over the past 30 years, but so have hydraulic system operating pressures and response times. The construction of resource project infrastructure such as liquefied natural gas (LNG) plants and pipelines requires extensive earthworks, deploying large, fixed and mobile plants under demanding project timelines, and often operating in environmentally sensitive locations. The problem of leaking hoses, lines and their connectors has been around for as long as hydraulic systems themselves, but with more system pressures and the need for greater HSE diligence, it is imperative that attention be paid to this aspect of operations. Plant and equipment product losses pose a safety hazard and can compromise machine reliability, as well as contribute to soil and groundwater problems that could otherwise have been avoided – the rectification of which consumes finite resources. 

With more than 20 LNG projects under construction globally and even more in the planning stages, the challenges of loss of containment from plant and equipment spills in the upstream oil and gas and resources sector alone are obvious. Environmental and safety requirements as well as the expectations of operators are on the increase. 

Studying the problem

In a study to examine the distribution of spills across plant and equipment types, and sources and root causes of equipment fluid spills at an LNG construction site, it was discovered that the main product released across all 86 spill incidents was hydraulic fluid. This was present in 74% of the 86 incidents, diesel 16%, engine oil 7% and the remaining 2% degreaser and brake fluid. An estimated 3,000 litres, or 2% of all oil used (dispensed into plant and equipment), entered the environment inadvertently during the period studied.

Over 14 months, the study examined fluid spills in plants with more than 300 pieces of equipment and 400 personnel, during 24 hour-a-day operations covering the main (earthworks) phase of construction on the site. Each incident was investigated to determine the root cause or causes of the spill. 

The spills were predominantly in the main earthworks area of the site. There was a total of 20 different types of plant and equipment involved in the spill incidents. Of these, eight types accounted for 80% of these incidents. These were, in decreasing frequency of involvement – loaders, excavators, haul trucks, a surface miner, water carts, graders, dozers and a drill rig.

Loaders and excavators were the most likely items of plant to be involved in spill incidents, accounting for around 40%. The most frequent number of spills occurred on loaders (17), excavators (16) and haul trucks (8).

Of the 14 root cause descriptions ascribed to all 86 events, four could explain 60% of the spill incident causes. These causes were typically associated with failures in the hydraulic system, such as connectors and o-rings. These findings suggest that at least a portion of the spill incidents could be readily controlled or reduced. For example, the cause ‘incorrect procedures followed’ can be addressed by increasing the effectiveness of training, particularly in the use and handling of heavy equipment. The cause ‘impact with an object’ could be reduced by targeting safer handling and use of the heavy equipment, reducing collisions and accidents.

Learning the lessons

In summary, hydraulic fluid spills from failed hoses and fittings are the most common sources of spills. The three generic types of plant and equipment most likely to be involved in a spill event are loaders, excavators and haul trucks, accounting for half of all incidents.

Of the root causes, around 50% of spills were from defective equipment parts or manufacturer’s designs not anticipating the conditions experienced by operators using the equipment. These causes are considered to be difficult to control and reduce by contractors on site. 

A smaller proportion (around 25%) of root causes were identified as incorrect procedures followed, equipment impacting with other objects, no inspection, operators allowing procedural deviation, mistakes (or mental slips), inadequate process hazard analysis, and operator’s understanding of equipment needing improvement. These root causes were considered to be within the control of the contractor company on the site and, specifically, its operators, the workshop/field service maintenance team and field supervisors. A greater awareness of the hazards from potential fluid spills can help eliminate and minimise this costly operating expenses item.

Based on the findings from the study, there are a variety of steps operators and contractors could take to reduce spills from plant and equipment on similar construction projects, in addition to an increased sharing of data and best practice for all spills, using pre-start discussions and toolbox talks, across the industry.

The adoption of preventative measures that eliminate or reduce the size of spills, including return to work debriefs rather than a focus on clean-up, and the development of lead indicators that assist in the early detection of oil spills could make a significant impact. This is particularly the case when early signs of failure can be detected, such as observable weeps and seeps (and aggregation of dust and dirt).

Hydraulic hoses, o-rings (within the hydraulic systems), and hydraulic hose couplings (including failed crimped ends) represented 50% of the spill sources. With this in mind, inspections should, as a minimum, cover as many of these as practical. Attempts to improve the quality and rigour of visual inspections of hydraulic hoses and hydraulic hose couplings, particularly during pre-start checks, should yield benefits. Owners should ensure effective training in this activity is provided.

This economically and environmentally costly problem could be substantially reduced, and the need to do so is only becoming more pressing.