Issue link: https://geospatial.trimble.com/en/resources/i/1415432
64 Spring 2019 intervals, which can be diffi cult to arrange in practice. And even if rotating equipment, and perhaps control valves, control systems and instruments, are made in redundant design, then other elements such as on/off valves and solenoid valves are still vulnerable to failure. In some cases, a degree of redundancy can be obtained at little additional cost by installing 2 × 50% or 3 × 33% capacity on rotating equipment instead of 1 × 100%. This arrangement can enable the entire system to run on reduced capacity instead of being completely stopped in case of equipment failure. Since the total cost of an installation will always be a strong driver, the goal is to maximise reliability while minimising investment and operational costs, including service and maintenance. The following is a list of important drivers for ensuring maximum availability at reasonable cost. A long-term strategy with good maintenance and continuous feedback from existing installations ensures an average VRU availability of above 99%. Technology selection Several different unit processes can be applied for vapour recovery, each having their individual advantages and draw-backs. The most widely used have been condensation by means of temperature and/or pressure, adsorption, absorption and membrane fi ltration, very often in different combinations. Through years of trial and error by different VRU manufacturers and clients, the outcome has been that over 90% of all new VRUs are based on adsorption onto activated carbon, regeneration by vacuum and re-absorption of the recovered hydrocarbons in a steady fl ow of fresh absorbent (typically gasoline). The process is relatively simple and after many years of refi ning the design and controls, the process has proven extremely safe and reliable. Component selection Components should be robust and well-known, with a good track record and long maintenance intervals. It is also important to ensure that chosen equipment can be serviced onsite and does not require regular factory overhaul, since this will severely impact the overall VRU availability. Condition monitoring The VRU control system should not only control the process and alert if any parameter is moving beyond pre-set limits but also monitor and record key process parameters for further analysis (Figure 1). As an example from a pressure swing system based on adsorption of the hydrocarbon on activated carbon, a detailed analysis of the pressure curves can disclose a lot of information about VRU load and status of carbon, vacuum pumps, valves, actuators and instruments. This information should be used as an integral part of the maintenance concept. Periodic preventive maintenance A fi rm scheme for regular, onsite preventive maintenance is crucial for the long-term reliability of a VRU. To ensure reliability and the overall safety of the system, the service job should not only include replacing lubricants and seals but also contain a careful check of all instrument and control loops. On VRUs where suffi cient data from the condition monitoring is available, the onsite services can be supplemented with regular remote or virtual service visits where an expert uses the logged operational data to make an in-depth analysis of relevant key parameters (Figure 2). Such a virtual visit can very often be used to determine which parts are beginning to show signs of wear and should therefore be ordered in advance and brought to the next onsite service. Spare part strategy Even with the best preventive maintenance, a risk of unplanned stop due to failure of a component will always remain. Every VRU owner is different and there are many ways to tackle readiness of spare parts. But it is advisable that a spare part strategy is in place so that at least the most likely situations are planned for in advance. A sensible spare part strategy should – for each group of parts – include an evaluation of the following: Most common type of failure (slow decline or sudden failure). How failure will be detected (via control system, daily inspection by local staff, regular remote service visit or regular onsite service). Figure 1. Logged data from a pressure-swing based VRU. Figure 2. VRU maintenance.