Remote, but in touch

Maintenance and energy managers are under increasing pressure to control costs. Pumps often represent the largest single energy usage in process plants. Keeping costs down, while at the same time ensuring pump equipment is serviced properly to prevent breakdown and extend the life of the plant, can present problems. These are particularly common at sites where there is no dedicated maintenance personnel to check equipment and carry out proper servicing.

Remote diagnosis and monitoring can provide the answer. Although this is not a new concept by any means, new technology and developments from pump manufacturers are expanding the functions and benefits all the time. Devices can now not only provide data on the monitored equipment, but can also provide a recommended course of action.

There are three main operating philosophies for running pumps. The first is to run the equipment until it fails. Action is taken when the damage has occurred; in other words, this is a 'closing the stable door after the horse has bolted' method. This might appear to be the low cost option but, when we consider the total cost of failure, downtime of plant and any consequential damage to the pump, system or plant, this thought is quickly dispelled.

The second option is to adopt a maintenance schedule, either by carrying out regular inspections and maintenance at fixed intervals or adopting a 'let's see if everything is working fine' philosophy. Routine programmes can provide a solution, but this can lead to higher costs due to work being carried out and parts being replaced earlier than necessary.

The third, and best, option is to implement permanent monitoring and prognosis. This results in condition-based maintenance which provides early warning of problems before they occur, allowing for planning of labour and downtime. This accomplishes the goal of reducing the operating costs of the plant and maintenance as well as the life cycle cost of the equipment.

<b>Scheduling maintenance</b>

The principle of monitoring and adjusting service periods based on use and conditions is now being widely used in a range of manufacturing sectors. Depending upon the configuration of the application, remote access devices can maintain and improve process and system reliability as well as optimising control and energy efficiency.

Many sectors are familiar with remote monitoring and its benefits, but with continued pressure to reduce costs and the subsequent decrease in back room activities such as maintenance, the need and range of applications for remote monitoring are ever increasing. These include process engineering, building services, water and wastewater applications.

The role of condition monitoring is too often thought of as simply part of a preventive maintenance system for the equipment, while its role and benefits are much wider. The diagnostic functions can be used to measure the process or system, and can indicate a malfunction before it causes damage to, or failure of, the pump. By the same principle, process reliability can be improved by recognising system irregularities early, therefore protecting the operator, the system and the environment. By the same token, energy efficiency optimisation can be achieved. Early identification of deficiencies in the pump and/or system, such as a discrepancy between design and operating data, enables the operator to take action to maximise efficiency. Such devices can also be used as a tool for monitoring overall expenses (life cycle costs) by providing information on the number of pump starts, periods of operation and the load profile. This data simplifies monitoring of the life cycle costs of a pump.

The provision of an electronic maintenance log is also possible (similar to the technology integrated in modern cars). The ability to acknowledge error and alarm messages, as well as a history function which keeps track of past maintenance, can also be included.It should also not be forgotten how the gathered information can provide an important data base for implementing process improvements.

Condition monitoring systems which not only detect faults, but also provide a recommended course or courses of action, have the added advantages of enabling the maintenance manager to often know what is required prior to the visit. This therefore allows them to supply the right engineering staff, the right level of manpower and potentially the right spare parts.

Another reason for the increased use of condition monitoring is the latest ATEX regulation, 137, which obliges pump owners and operators to permanently monitor pumping equipment installed in hazardous locations and to perform condition-based maintenance. Remote diagnosis equipment can make a substantial contribution towards meeting these requirements.

How the information is used is as critical as the measurement itself. Data can be localised to the condition of a particular pump, system or process which can be collected routinely at site or the data can be transmitted to a control room by means of a field bus or to a PDA (bi-directional) via an IrDA interface. Some companies now use a remote accesss device as a base unit for teleservice-based services (remote data transmission).

The principle of remote access devices is the same as those used on computers. Your service agent will monitor the equipment, will receive early warnings and can interrogate the system remotely. They could also automatically respond and repair the pump.

When selecting condition monitoring equipment it is important to ensure that the system can be customised to provide flexibility for the user. For example, a product which is modular in design allows the client to select a customised version. This enables the client to build from a base unit and add on the functions they require. This modular format also means the client can upgrade at a later date by increasing functionaility.

When planning any system it is important to embrace the latest technologies and use them to your advantage. Pump users should be encouraged to call upon the expertise of manufacturers who are committed to developing new products and equipment which will help in meeting today's demands and challenges.PE

Alan Bell is marketing manager at KSB in Loughborough.

Watching the seals

Although it might seem a fairly simple piece of kit, a pump is a collection of components: the pump itself, the drive, the bearings, and the seal, Stuart Nathan writes. Each of these are complex and have many properties which can be monitored. But over-monitoring can cause as many problems as under-monitoring - the huge amounts of data have to be collected, logged and interpreted. So which component is it best to monitor?

According to William McNally, a retired engineer from the US Navy's submarine fleet and now a pumping and seal consultant, the answer is a simple matter of pragmatism. 'Most premature pump shutdowns are related to seal and bearing failure,' he says. Moreover, 'no-one wears out seals and bearings - they always fail prematurely.'

Jim Malcolm of seals specialist John Crane agrees. 'In reality, if you look at failure analysis of critical rotating equipment, the heirarchy is seals first, bearings second and pumps last.' Seals last from three weeks to two years, bearings two to five years, and pumps two to five years. And when a seal fails, the results can become very serious, very fast, particularly in offshore applications. 'If a seal goes down on a production pump, the losses can very easily run up to £18million per year, which is a recent case.' But, as Malcolm says, most pump users are more concerned with monitoring the pump and drive. For the most part, he says, this is a matter of tradition.It would seem to make sense to monitor the condition of the seal.

There's certainly a great deal of information that can be gleaned from the component: the temperature, pressure and flow rate of the fluid between dual seals, for example, as well as the speed of rotation and the movement of the shaft. The condition of the seal fluid can have a particularly powerful effect on the pump's operation: under certain conditions, it might vaporise, crystallise, become more viscous, solidify, form a film on the sliding parts of the seal - any number of effects than can reduce the effectiveness or even destroy the seal.

Commercial seal monitoring equipment does exist. For example, German firm Burgmann Dichtungswerke produces a three-level modular diagnostic system (MDS) which provides continuous seal performance data. The most basic system, MDS1, works with pumps handling a constant-temperature medium, and measures the seal seat temperature to provide a warning that the pump is running dry.

The next level, MDS2, also measures the temperature of the medium; while the most advanced, MDS3, also measures the pressure. MDS3 also analyses the performance data to provide a prediction of the seal's useful life, and triggers a warning light when it seems the seal might fail within 150 hours.

But in John Crane's case, Malcolm says, most seal monitoring is done in the testing laboratory in Aberdeen, rather than in the field. 'We have developed a condition monitoring system that not only includes traditional sensors, but working with some very clever acoustic engineers we have developed acoustics that monitor the seal fluid film - something never done before,' he says.

The system looks at the degradation of the film of fluid, 2-3micron thick, between the seal faces. The acoustic monitoring allows the engineers to look at the behaviour of this film in a way they've never seen it before. Normally, condition monitoring of rotating equipment analyses vibration of the whole body of the machinery, looking for an abnormal vibration pattern or frequency which indicates the machinery isn't working properly and might be about to fail.

'Acoustic technology monitors the plastic deformation of a structure,' says Malcolm. 'The signals are VHF - very high frequency - and are filtered and demodulated.' John Crane's acoustic engineers have developed a signal processing technique which 'allows us to "see" the acoustic signals and film degradation just before it starts to have a detrimental effect on the seal itself,' Malcolm says.

This could have major benefits for pump users. 'It not only prevents seal failure, although that is important,' Malcolm says. 'Being able to combine all the signals, both traditional and acoustic, gives us an unprecedented picture of the reasons for failure. On production pumps offshore, pumping £1million per day, preventing a second failure is extremely important.'

The system isn't just confined to the seal. 'We can monitor the whole pump skid, including the bearings and fluid characteristics, so for the first time we can monitor the whole pumping package,' Malcolm says. So could this become a commonplace arrangement on the plant or production platform? Malcolm points out that, despite the complexity and sophistication of existing monitoring techniques, it's quite rare for them to be used to their full potential. 'People are nervous of buying another monitoring tool,' he says. 'However, if you align monitoring to performance, and performance to rewards, then it is a different ballgame altogether.'