Technology developments in mid-range machines5 February 2002
Legislation, environmental and market pressures, knock-on effects from developments in larger machines, computer aided design - all these factors are bringing about design changes in the 1 - 50 MW range David Cox, Peter Brotherhood Ltd, Peterborough, UK
Technological development of steam turbines has traditionally been led by the manufacturers of larger machines. When a turbine is powering a generator producing many tens of MW even a fractional increase in efficiency or a slight improvement in reliability produces a rapid return.
Mid-range steam turbines - broadly defined as those in the 1 to 50 MW range - have tended to incorporate new developments only after they have been proven on larger machines. Like most generalisations, however, this one is only partly true. Operating at a smaller scale brings its own engineering problems for which unique solutions have been developed. Some developments described here were originally implemented by large turbine manufacturers but adapted for mid-range machines, while others are specific to mid-range machines.
The steam turbine market
There are two major segments of the steam turbine market. The fastest growing is probably that which can be characterised as 'environmental' - in particular, waste-to-energy schemes for the production of electrical power from the incineration of various types of waste material. Examples are schemes that burn hospital clinical waste, domestic refuse, mines gas, sewage sludge and bagasse, the waste material left after extracting sugar from cane. The latter is a particularly 'green' power generation system because all the CO2 liberated during combustion is absorbed again the following year by the new cane crop. Moreover, power generation utilities are increasingly employing mid-range steam turbines in waste heat recovery systems for diesel and gas-fired power stations.
The second major market segment - which often overlaps with the first - is the purchase of mid-range steam turbines by industrial organisations generating electricity for their own use.
Steam turbines are the dominant power source in the industrial generation market. The technology is relatively simple and well-proven and offers good levels of thermal efficiency. Capital and running costs are comparable with - or usually better than - any competitive system.
Many industrial processes make use of steam as a heat transfer medium. This opens up possibilities for integrating power generation with steam production for other purposes - in effect, combined heat and power schemes which offer very high levels of overall efficiency.
Legislation is as important as economic factors in both of these market sectors. CE marking, for example, has brought about a change of attitude and a standardised approach to safety and related issues across Europe, at least in theory, although interpretation of the standard seems to vary markedly with geography.
The Pressure Equipment Directive (PED), which comes into force in May 2002, means that all pressure-containing pipework and vessels used in the turbine package will need to be verified by an independent authority. This will add to cost and administrative effort and will arguably produce little benefit for the end user who purchases from established and reputable suppliers. Disreputable suppliers will simply flout the new regulations, secure in the knowledge that there is little prospect of getting caught until something goes disastrously wrong with one of their machines.
Other pieces of legislation have a less direct but equally significant impact on turbine design. Manual handling regulations, for example, have limited the size and weight of components that a maintenance engineer can be expected to manhandle.
Legislation has affected the market in other ways. Environmental laws on waste disposal, and tax breaks for 'green' energy, for example, render financially attractive some schemes which in other circumstances would never secure funding. At the same time, regulations for the operation of the market for electricity that are making schemes which appeared to be unworkable financially viable have made some otherwise viable schemes financially unworkable.
Because different national governments have introduced their own particular legislation and tax regimes, the market tends to be different in each country. Despite this, both segments of the market for mid-range steam turbines are truly international. Peter Brotherhood is only one of many turbine manufacturers, with competitors in most major industrialised countries and several of the developing ones, yet it has sold turbines in nearly every country and in all continents apart from the Antarctic.
The technological development of mid-range steam turbines is being driven by the demands of this international market. One market demand has been for larger and larger machines. For example, Peter Brotherhood has recently commissioned what is believed to be the largest steam turbine ever supplied to a sugar mill in Africa - a 20 MW back pressure machine for the mill in Hippo Valley near Chirezdi, Zimbabwe, owned by Hippo Valley Estates. The turbine is almost 50 per cent bigger than any other known to be operating in the continent.
Responding to the growing demand for size by extending their product portfolio has meant that traditional mid-range manufacturers have begun to compete at the top end of their product spectrum with traditional 'large' machine manufacturers. This has tended to speed up the adoption of new technology - to compete at this level the manufacturers of mid-range machines have had to match developments in large machines and, having done so with their most powerful machines, have then implemented the same technology further down the range.
New design features
Among the changes which the market has thus encouraged have been the widespread adoption of fabricated casings (ten years ago most mid-range turbines had cast casings) and laser-cut and welded, rather than vacuum brazed, diaphragms. Brush seals, common on larger machines for many years, are replacing labyrinth seals on mid-range machines, making it possible to secure the same pressure drop over a shorter length of shaft and therefore contributing to reductions in the overall size and weight of the turbine. Twisted and tapered blades, designed to maximise efficiency, are now common on even relatively small machines. More design enhancements to improve efficiency further are being introduced, such as flared rather than parallel diaphragms.
New materials and treatments are being introduced. For example, adapting a technology widely employed in gas turbines, Peter Brotherhood is currently using blade coatings of various materials, to improve erosion resistance and to change vibration damping characteristics.
One of the developments that is permitting these new design features to be introduced even on relatively small, low output, low cost machines is the increasing use of computer modelling.
Computational fluid dynamics (CFD) allows us to model and understand the behaviour of the steam at every stage in its passage through the turbine. With increasingly powerful computers and better software CFD is now becoming a design tool rather than solely a research and development aid. The ability to optimise aerodynamics in real time allows us to tailor each individual machine to its intended application.
Techniques such as finite element analysis allow us to calculate precisely stresses in critical areas and therefore to build in safety factors that are invariably more accurate than the 'belt and braces' allowances which our predecessors were obliged to use. This has led to more economic use of materials and reductions in overall machine weight and cost. Piping stress analysis and acoustic modelling ensure that pipework and ancillary equipment that form part of the turbine package meet the same standard as the machine itself and help to minimise noise and vibration.
CAD allows us to produce a virtual solid model of every component and of the complete machine. Problems of pipework clashes and difficult or impossible maintenance access which just a decade ago would only have become apparent on site can now be detected and corrected at the design stage. And we can integrate our turbine design model with the customer's model of the entire plant, allowing such things as hazard analyses to be carried out before a single item is actually built.
The use of CAD has led to dramatic reductions in the time between receiving the order and delivering the product. It also radically reduces the possibility of human error. Once designs have been agreed and approved we are increasingly seeing CADCAM being used to convert the computer models into components with the minimum of human intervention - and therefore the minimum of human error. This is a trend that can only accelerate as we move towards more and more automated manufacture. For example, my own company is currently investigating the precision casting of nozzle segments and nozzle banks using moulds created by 'rapid prototyping' technology direct from 3D computer-designed models.
Advances in control and monitoring
If computer technology has had a marked impact on design and manufacture it has arguably had an even greater impact on turbine control systems.
Even just 20 years ago, the vast majority of turbine controls used mechanical or pneumatic systems or some basic electrical circuitry. Ten years ago, control systems were a mixture of hard wired and PLC systems. Now systems are almost entirely PLC-based, even for turbines being shipped to parts of the world where new technology is adopted only slowly because of problems in finding spare part components and skills needed to maintain it.
A decade ago, also, turbine maintenance was schedule-based. After a defined number of hours running, certain maintenance tasks were carried out. They were done whether they were necessary or not - because there was no way of determining whether or not they were needed without disassembling the turbine to take a look.
Modern monitoring systems - continuously checking such factors as temperature, pressure, vibration and pressure drop - allow operators to determine when maintenance is actually required. Because old-fashioned maintenance intervals were based on precautionary and conservative principles, the advent of requirement-based maintenance has led to less frequent servicing when things are operating normally - and swifter intervention when things are starting to go wrong.
Taken together, these two effects have not only dramatically reduced downtime but also the number of skilled maintenance personnel which an operator requires to optimise plant performance. This in turn has reduced operating costs, justifying the increased capital cost of installing the monitoring systems in the first place.
Alongside this has come a considerable extension of remote operation and monitoring. Starting a steam turbine from cold used to be a highly skilled manual operation. Now it can be completely automated and undertaken remotely from a control room which may be many miles away.
Peter Brotherhood - in common with many other turbine manufacturers - now offers maintenance contracts based on remote monitoring from its own premises. When it detects a problem or a maintenance need the company despatches its own technicians with the tools and components required to take whatever action is required. The cost of training and retaining that highly skilled workforce therefore falls upon the manufacturer rather than the operator.
As the frequency of on-site maintenance visits continues to fall - not just with turbines but with a whole range of plant - there will be increasing economic pressure on operators to contract out their maintenance and repair activities.
Cost and efficiency
There has always been a trade-off between cost and efficiency. It is possible to build highly efficient mid-range steam turbines - but the cost would be prohibitive. New design and production technology, however, is making it possible to get closer to the ideal machine at costs which customers find acceptable.
Partly this has resulted from customers themselves re-appraising the quality versus cost trade-off. Increasingly, turbine users are looking at whole-life costs where the long-term pay-back of improved efficiency and reduced downtime justifies higher initial capital costs. It is not unusual, for example, for purchasers of steam turbines for Scandinavian waste-to-energy district heating schemes to consider a 15 year machine life in assessing the real cost per kilowatt of power generation.
Coupled with this approach is a determination to ensure that turbine manufacturers live up to their promises. Steam turbine purchase contracts now often include clauses specifying maximum machine downtimes and limits on performance degradation over a two or three year warranty period - with significant financial penalties for the manufacturer if the machine fails to meet the specification.
Users are also increasingly putting the responsibility for ensuring machine performance firmly on the manufacturer by taking out long term maintenance contracts which specify required levels of features such as availability. These are trends which reputable manufacturers can only welcome.