The history of the turbogenerator already spans nearly 100 years. Charles Brown, one of the founders of BBC/ABB, built the first 2-pole turbogenerator in 1902. Roebel’s invention of conductor bars (BBC patent of 1912) enabled utilisation rates to be improved and allowed designs to be more compact. In 1915, Max Schuler had the idea of using hydrogen as a coolant for electrical machines because of its superior thermal properties. The thermal conductivity of air is seven times lower than that of hydrogen, while its density is 10 times that of hydrogen. It was 1947, when the first European hydrogen-cooled generator (125 MVA) was built in Switzerland by BBC.

In parallel with hydrogen-cooled turbogenerators, the technology of air-cooled turbogenerators was also steadily improving. Air is easier to handle and, for a given output, the cost of an air-cooled turbogenerator is significantly lower than that for a hydrogen-cooled generator. In the early 1970s, the output of air-cooled turbogenerators reached 90 MVA. In 1984 a machine with 188 MVA was commissioned and in 1993 an output of 225 MVA was possible. The 300 MVA limit was surpassed in 1995, with entry into test operation of the first unit at ABB’s GT test centre at Birr in Switzerland. Five units of that size are now operational and more than 30 have been sold so far. Efficiencies up to 98.8 per cent have been witnessed by Lloyds.

The impetus for larger and larger air-cooled turbogenerators has come from the markeplace. Customers facing increased competition increasingly recognised the clear cost, performance and delivery advantages of air-cooled solutions over conventional hydrogen-cooled technologies. This was the catalyst for ongoing developments. Taking ABB’s experience as an example, some 180 air-cooled turbogenerators of 150 MVA and above with over 400 accumulated operating years have come into operation over the last six years worldwide. All these use the TEWAC (Totally Enclosed Water to Air Cooled) air-cooling concept, in which water is used in coolers to cool air circulating in the generator, as opposed to open ventilated air cooling (ie once-through cooling). Large air-cooled turbogenerators tend to be of the TEWAC type.

The rise of air

Simple air-cooled turbogenerators have been used in gas turbine power plants for many years. In recent years the development of air-cooled technology has been driven by rapid evolution in the gas turbine business. More and more hydrogen-cooled technology is being replaced by air-cooled systems because of their inherent simplicity and cost competiveness. As air cooling technology encroaches further and further upon what has been thought of as the traditional domain of hydrogen-cooled systems so equally hydrogen cooling is encroaching further into what used to be regarded as the territory of hydrogen/water cooled turbogenerators.

Developments in hydrogen-cooled technology have also been spurred on by progress in gas turbine plants, in particular use of the single shaft combined cycle configuration, where generator outputs typically 30 per cent higher are needed. In the case of the larger hydrogen/water technologies, the development strides have been more restricted, the one notable exception being use of stainless steel windings.

In some respects, the dynamics of change in the generator area are actually picking up pace. Recently we have seen other innovations that focus on delivering economic benefits and interface advantages to the user. The new Powerformer generator technology is a good example. Powerformer aims to achieve a high enough terminal voltage to eliminate the need for the step up transformer. Current development work is focusing on an output voltage of 225 kV. It remains to be seen how Powerformer will impact on the more traditional generator technologies.

Large air-cooled systems for 60 Hz

Meanwhile, the potential applications of large air-cooled machines are now being extended into the 60 Hz realm.

In 1995 the largest air-cooled generator for 50 Hz, rated 300 MVA, was put through its paces in the test bay. In September 1998 the first 60 Hz unit was also successfully type-tested.

The 60 Hz unit exhibits the same design and cooling features as its 50 Hz brother. Based on standardising principles, the stator, core and rotor diameter have been kept the same for both types. The rated voltage for the 50 Hz machine is 19 kV and that for 60 Hz is 21 kV.

The rotor slot filling is optimised to adapt to the higher speed. Friction and windage losses make a considerable contribution to total losses for air-cooled turbogenerators. This is of even more importance in the case of 60 Hz because friction losses increase with speed to the power of three. Optimisation of the cooling system therefore needed significant further investigations.

As part of the test programme, the following measurements were carried out for both the 50 Hz and 60 Hz machines:

  • Electrical: no-load and short circuit characteristics, sudden short circuits for determining the reactances and time constants.
  • Losses.
  • Temperatures.
  • Pressures.
  • Mechanical/noise.

    For the 50 Hz running tests, approximately 70 pick-ups were installed and 80 pressure and 200 temperature measuring points provided. It is worth mentioning the exceptional efficiency achieved by the 50 Hz machine, namely 98.8 per cent, which is very close to that attained by hydrogen-cooled turbogenerators. In full load operation at 300 MVA, the temperature rise is well below the limits of thermal class B.

    The 60 Hz prototype – like the 50 Hz machine – also fully met expectations. Based on the recorded temperatures, there is scope for increased output while still maintaining adequate margins in relation to the permissible temperature limits.

    The 60 Hz machine attained an efficiency of 98.5 per cent, which is remarkable enough, but studies have revealed potential for further increases in the efficiency. Evaluation of the test data showed that an efficiency of more than 98.6 per cent is achievable.

    The difference in efficiency betweeen hydrogen and air cooling at full load is around 0.2-0.3 percentage points in the case of the 60 Hz machine and for the 50 Hz generator the difference is only 0.1 percentage points. As load falls, the gap widens, with the efficiency of air-cooled machines falling off more steeply than is the case with hydrogen-cooled units. This is due to the fixed friction and windage losses, which contribute to about 50 per cent of the total rated loss.

    Reliability issues

    The 180 ABB air-cooled turbogenerators above 150 MVA already in operation have consistently shown good reliability. The average forced outage rate over more than 400 unit years in the period 1993 to 1997 has been as low as 0.05 per cent. Together with the additional advantages of air-cooling during planned outages, this amounts to an excellent availability record.

    After having been type-tested, the first unit of the 300 MVA series went into operation in the Birr gas-turbine test plant in 1996, where it has been cycled in real load conditions. With nearly 100 additional temperature and vibration probes remaining from the running test it was possible to verify these measurements. This first unit was brought to nominal speed without vibrational problems. In operation the stator end winding and rotor vibrational behaviour was completely satisfactory. So far over 250 start-stop cycles have been performed.

    Looking at commercial units, the three air-cooled turbogenerators in the Rocksavage plant in the UK have accumulated over 16 000 operating hours (as of February 1999) without problems.

    Using WIDIPRO diagnostics supported by the on-line diagnostic program GOLD, a large body of expert knowledge has been accumulated based on over 2000 diagnoses on about 550 units. In the ABB case the TEWAC configuration and the Micadur insulation system have been primary contributors to the excellent operating experience.

    Air v hydrogen: comparing the costs

    As already noted air-cooled turbogenerators are today increasingly used in output ranges where previously only hydrogen-cooled turbogenerators were installed. The slightly lower efficiency is more than compensated by reduced first and running costs.

    The market demands not only high efficiency but also:

  • Short lead times from ordering to commissioning. This releases tied-up capital and permits a quicker return on investment.
  • High flexibility, ie standard solutions for all applications (conventional steam power plants, gas turbine plants, combined cycle power plants, thermal power plants etc), thereby reducing total manufacturing costs.
  • Good reliability, availability, and maintainability, thereby reducing the secondary costs.

    A comparison of a 300 MVA TEWAC generator with existing hydrogen-cooled machines shows first cost savings of around US $1 million. The reasons are:

  • Less additional equipment needed, such as hydrogen, carbon dioxide, and sealing oil systems, with their associated monitoring and control systems.
  • Less space required in the power plant because additional equipment, tanks etc are not necessary.
  • Lighter foundations and smaller building required.
  • Less pipework required.
  • Fewer interfaces to other power plant components such as motors, water supply etc.
  • Lower engineering costs associated with plant layout (no equipment underneath the shaft train), lower documentation costs.
  • Reduced number of trained personnel.
  • Quicker commissioning and easier monitoring.

    An estimation of the secondary (running) costs shows savings of US $ 0.5 to 2.5 million per turbogenerator, depending on the application. These derive from:

  • No servicing or operational costs associated with hydrogen and carbon dioxide tanks and associated reserve systems.
  • Fewer spare parts and less sophisticated spare parts.
  • Simpler and safer operation (no special hydrogen safety precautions required).
  • Easier maintenance (fewer parts, fewer preventive measures, simpler monitoring).
  • High availability.

    Aiming for 500 MVA

    The rapid evolution in gas turbine and combined cycle plants point to a need for air-cooled units beyond today’s 300MVA.

    To assess the benefits of higher ratings conceptual studies have been performed. These suggest further improvements are needed entailing developments in a number of areas. Among the measures that are likely to be necessary to get to larger sizes of air-cooled turbogenerator are the following:

  • Reverse cooling (ie sucking air from the air gap as opposed to blowing air into it).
  • Stator: refine the modular multi-chamber cooling system, using more parallel air paths rather than serial ones.
  • Rotor: increase air-flow, improve temperature distribution along the rotor.
  • Insulation: higher temperature (class F instead of class B thermal conditions); reduce thickness by equalising electric stress.

    Based on these measures conceptual solutions for air-cooled turbogenerators up to 500 MVA have been developed, retaining all the advantages of the current technology and further improving utilization. This will be the subject of a future article.

    The field of application envisaged is primarily combined cycle single-shaft plants, where the customer already tends to specify air-cooled turbogenerators.

    But we also expect new possibilities for use of air-cooled technology to arise in steam power plants too.

    Designing higher-rated air-cooled generators: back to basics


    More Relevant


    Sign up to the newsletter: In Brief

    Your corporate email address *
    First name *
    Last name *
    Company name *
    Job title *
    Vist our Privacy Policy for more information about our services, how we may use, process and share your personal data, including information of your rights in respect of your personal data and how you can unsubscribe from future marketing communications. Our services are intended for corporate subscribers and you warrant that the email address submitted is your corporate email address.

    Thank you for subscribing

    View all newsletters from across the Progressive Media network.