Global privatisation, and the economic demands of the competitive market, are forcing power plant owners to examine operational efficiency and outage times. Privatisation has also encouraged development of gas turbines. Gas turbines typically require maintenance every two years, and owners must closely evaluate turbine outage times. Swift and efficient bolt removal is vital. Hydratight, which has long argued the case for hydraulic bolt tensioning, also champions induction heating technology.
In a competitive market, turbine outage times are an expensive necessity. Both break out and box up times for the turbine are critical, and operators have to optimise these. In the case of break out, no assessment of required maintenance work on the turbine is possible until the casing is open. If it turns out that the rotor has to go back to the factory, it must meet a time slot, or the outage will be potentially extended.
The box up operation is often on the critical path for outage, so time savings here can make a big difference to revenues, especially for baseload suppliers.
Hydratight has, for a long time, presented the case that hydraulic bolt tensioning is the fastest and most accurate means of tensioning and detensioning bolts. Nonetheless, the company is also supporting the use of induction heating for detensioning of bolts.
The reason for this apparent contradiction, between cold bolt tensioning and thermal bolt tensioning technologies, lies in refurbishment of existing turbines. The cost of replacing thermally stretched bolts with cold tensioned bolts on existing turbines is generally prohibitive. For these stations, the chosen option is to thermally stretch the bolt, either before loosening the nut or winding it down to the casing. In these cases, induction heating offers significant time savings.
The traditional method of thermally heating a bolt is to insert an electrical resistance heating element, known as a cal rod, into the centre of the bolt.
Induction heating, which was first introduced in the USA by Mannings, exposes the conductive bolt material to an alternating magnetic flux, via an electric wand, causing heat to be generated within the bolt. When exposed to an alternating magnetic flux, eddy currents are set up in the bolt. The flow of these eddy currents through a conductive mass dissipates as heat, according to the I2R law.
In addition, if applied to magnetic steel, induction heating also benefits from hysteresis losses – the power losses needed to change the direction of magnetic flux.
Induction heating is faster and more accurate for thermally stretching bolts than conventional heating. In one example from the tests, it took 19 minutes to achieve 0.2 per cent residual strain in a 5 inch diameter bolt using conventional heating. Using induction heating, it took only 3.25 minutes to achieve the same result.
Test results also show that the maximum temperature reached by the bolt is lower with induction heating. This, combined with faster heating, results in less heat being conducted into the surrounding turbine casing. The combination of reduced heating times, reduced cooling periods for the casing and the fact that break open or box up of turbine casings can commence at higher temperatures leads to significant time savings.
There had been concern that induction heating might damage the bolt structure. However, the test results demonstrated that it is possible to achieve similar temperature gradients across the bolt structure compared with conventional heating methods. As a result, it can be assumed that there will be no adverse effects on bolt structure.
There are other reasons to prefer to use induction heating. There is a risk of burn out when using conventional cal rod heaters, as these require contact with the surrounding bolt for heat dissipation. With the induction heater, the wand is water cooled, so the risk of burning up is minimal. There are also health and safety implications regarding the removal of heating elements. The wand of an induction heater is cool, whereas the cal rod can be at a high temperature.
Further questions of reliability relate to general fluctuations in voltage supply, which can damage conventional heating systems. Induction heating does not suffer from such problems. In addition, unlike electric heaters, induction heaters can not be ‘arced’ against the bolts.
Nonetheless, carefully controlled experimental proof does not always reflect the reality of an outage. Proof requires demonstration on turbine outages.
One of the first major applications of induction heating of bolts outside of the USA took place in the box up of the HP stage of a 200 MWe steam turbine at Hendrena in South Africa. This was an emergency outage, and time was critical to enable the station to generate revenue as quickly as possible. It is estimated that the use of induction heating saved two days of outage time.
All 26 bolts of the inner casing were thermally elongated before tightening the nuts to a pre-specified angle to achieve the necessary load. It took 31 minutes to do this for all 26 bolts. It is estimated that the equivalent time using cal rods would have been 130 minutes.
Following bolt tightening, technicians measure bolt elongation to obtain a loading value. To be able to do this, the bolts and casings must cool to the ambient temperature. With induction heating, this took 4 hours. With conventional heating, this cooling process would have taken 6-8 hours. Subsequent measurement then reveals whether further elongation and adjustment is necessary. The cumulative effect of faster and more accurate heating, combined with quicker cooling times, made a significant difference to the total outage time.
The two halves of the outer casing were brought together using a hydraulic cylinder closure system, designed to bring the two halves together evenly. The system simultaneously tensions 25 per cent of the bolts to a base load, which is sufficient to iron out any casing distortion.
Further bolt loading is necessary, but the process means that all bolts start from the same base load. As a result, the bolts only need elongating and tensioning once.
In this outage, the initial loading of the 74 outer casing bolts took 82 minutes. Further adjustment was necessary on subsequent measurement, with the same correction on all 74 bolts. This was because new bolts had been used, and some relaxation of the thread had taken place.
Similar results were recorded for the break out and subsequent box up of a 250 MWe HP turbine half joint at Theiss B in Austria. This outage was a major service for the station lasting 12 weeks, with induction heating of bolts being used.
Using induction heating, 262 bolts were heated for the break out. Work started at 9.00am. By 1.00pm, all the outer cylinder bolts had been heated and detensioned. By 3.30pm, the main contractors Alstom Energie had removed the cylinder cover. By 6.00pm, Hydratight had heated and detensioned the inner cylinder, diaphragm cover and gland cover bolts.
Individual bolt heating comparisons showed that 125 mm diameter bolts took 5 minutes at Theiss B with induction heating; other heating technology took 13 minutes.
Conventional heating technology meant that box up typically took 4 days at Theiss B. This placed the box up operation on the critical path for the outage. The operation took 2 days using induction heating.
Both controlled testing and on-site experience point to the advantages of induction heating over conventional electric heating methods. When financial considerations prevent replacing the existing casing bolting with cold bolt tensioning systems, the choice is between conventional and induction heating, and induction heating has some distinct advantages.
TablesTable 1. Comparison of heating test results