Case Study – Steam Turbine Rotor Bow

Problem Description:

A Double extraction condensing steam Turbine of 32MW was taken for major overhauling after more than 40000 running hours. Machine was commissioned in the year 2013. After a long run sealing fins of the rotor and guide blade carriers (GBC) were found worn out. During overhauling complete set of sealing fins of Rotor and GBCs were replaced with new set and Dynamic balancing of the rotor was done. During assembly the clearance of the sealing fins were set as per OEM recommendations. Normally during overhauling of the Turbines, clearances were to be set as per OEM  to achieve best specific steam consumption.

After overhauling, during cold start up of the turbine on reaching Rotor speed rated 5000 RPM, Turbine front and rear bearing vibrations shoot up drastically and Turbine got tripped on higher vibrations. Turbine Rotor got jammed (seized) within 4 min of time. After cooling down for 12 hrs, It became free and turbine was put in the barring gear rotation. On subsequent trials, turbine couldn’t reach to its rated RPM due to higher vibrations. Hence, turbine was left for cooling down to investigate problem. 

Fig. 1 Turbine-Gear box- Generator layout schematic

Incident investigation:

On opening of Steam Turbine following were the observations: 

  • Rotor runout measured, +0.22 mm at balance piston area, which indicates Rotor Bow.
  • Sealing fins of balance piston area of rotor and steam chamber were found completely rubbed due to excessive heat generation.
  • Steam chamber sealing fins found completely rubbed & molten in the bottom half.
  • Top half of the steam chamber sealing fins found intact.
  • Ovality in bottom half of the steam chamber by  -0.50 mm.
  • In rest of the part, sealing fins of rotor and GBCs were found intact.

Probable causes:

It was a thermal bow, due to tremendous heat generated at BP gland area by means of heavy friction between sealing fins of BP area of Rotor and bottom Half of Steam chamber. 

Probable reasons for generating friction in BP gland area: 

  • In case of the Turbines having steam entry from top half of steam chamber. Bottom half of the steam chamber will be of solid component. During steam entry, the hollow part will be heated faster than bottom solid half of the steam chamber. It will create difference in temperature gradient between top and bottom portion of steam chamber. Due to uneven expansion of the rotor and steam chamber top & bottom half, radial clearance of sealing fins of rotor and steam chamber reduces. By rubbing of sealing fins, it may create excessive heat in both side contact area, creating ovality due to excessive rubbing. Resulting into the rotor being caught inside the bottom half of the steam chamber.
  • Temperature gradient of BP area created vibration in the Rotor, due to uneven temperature of BP gland area of Rotor and Steam chamber.
  • Smooth expansion of front pedestal is not being taken place may be due to front pedestal spherical washer mounting arrangement. For smooth sliding of the front pedestal molicoated sliding support arrangement will have more efficacy.
  • Accumulation of condensate in the pipeline of gland steam drain and vent, which has given quenching effect on rotor BP gland area. Which may result in to increase in vibrations suddenly.

Analysis :

Steam Turbine is a very sophisticated machinery. Its each and every component is designed by doing a lot of calculations & experiments. The main component of a steam turbine is its rotor, it is the part which is both under thermal & mechanical stresses. In the current times industries using Solid forged shafts for turbine rotors. It has reduced a lot of issues such as machining cost & time, scrap, stiffness & stress in the rotor shaft. In earlier days, a hollow shaft was being used for turbine rotors. In solid rotor the rotor will expand slowly so that the casing & rotor will expand at nearly same rate & jamming will be avoided which will prevent damage to the seals & blades. A large steam turbine rotor must run with low vibration levels during operation and when going through critical speeds. Rotors must be straight and true within a tight tolerance. 

Uneven heating, cooling or quenching of a hot rotor during operation or after shutdown. These operating issues can then result in: 

  • Local rubbing resulting in permanent metallurgical changes 
  • Local expansion and yielding that results in permanent bowing and cracking 
  • Residual stresses and material hardness above OEM limits 
  • Bowing that cannot be balanced out.
Turbine Rotor Bladding arrangement

Rotor Sagging and Bowing

Bow and sag or normally called sagging and hogging occurs on a steam turbine rotor under the following conditions:

A. Sag– relates to the elastic property of the rotor, sagging of the rotor happens when the rotor is at ambient temperature and barring is not in place. so over time the weight of the rotor which is only supported on the rotor bearings allows the rotor to sag in the middle and again the rotor can come into contact with the stator or casing.

Sag is always present, and as it is caused by gravity, is always oriented in the downward direction. This perpetual downward deflection due to gravity is accommodated by the flexibility of the rotor, and is anticipated in the rotor design.

B. Bow– is something entirely different. Hogging or Bowing of turbine rotors occurs when the surface of the rotor shaft is subjected to non-uniform temperature profile, either due to exposure to steam, or localised rubbing contact with stationary parts. In general, three types of rotor bow in rotating machinery can be distinguished, namely:

  • elastic bow, resulting from static load,
  • temporary bow, caused by uneven temperature field or anisotropic rotor material properties, and
  • permanent bow, wherein material yield strength has been exceeded (plastic deformation).
  1. Bowing occurs most commonly when the turbine is shut down faster than normal or under trip conditions and the natural cooling of the rotor is not allowed to happen. Usually happens when the barring system fails to start, the rotor bends in an upward motion in the middle and makes contact with the stator or casing thus allowing rubbing to occur. 
  2. When a rotor is allowed to lie at rest within its casing while the unit is hot. The admission of gland sealing steam to a condensing unit while the turning gear is disengaged for instance. The gland steam is directed at a single point of the stationary rotor shaft, causing localised heating, and thermal bowing due to asymmetrical expansion of the rotor shaft surface.

The only way you can overcome this is to have effective barring systems to keep the rotor turning as it cools thus not allowing the rotor to hog upwards.

If this occurs, hand baring of the rotor is the only option, ½ a turn every half hour until the rotor becomes uniform in shape again. Once in its uniform shape (could take many hours) the barring system can be put into service.

A bowed rotor is an eccentric rotor, significant potential consequence of a bowed rotor is mass unbalance due to eccentricity. Rolling a thermally bowed rotor with steam can produce dangerously elevated vibration levels, sufficient to inflict severe damage on internal components. Most modern units of any size are provided with rotor eccentricity detectors to warn the operator in the event of a rotor that is not straight. Eccentric Rotor, its associated tendency to reduce radial clearance to stationary objects – attains its maximum at the rotor’s first critical speed. If a shaft rub occurs at or near the critical, and the rotor is not accelerated quickly, the localised heating of the shaft due to rubbing will accentuate; harder the rub, worser the bow.

How to avoid Operationally?

  • During the initial run from cold condition of turbine, allow steam for turbine rolling having somewhat less degree of superheat. It will allow more steam to enter in to the turbine and it will help in uniform heating of rotor and casing. This can be possible in independent units only, in case of common network it will not be possible. 
  • Uniform heating of Rotor and Top/Bottom casing can be ensured by Top and Bottom casing temperature measurements provided by OEM. Delta between top and bottom casing temperature should be minimum.
  • Due to heavy mass of the casing, it will expand and contract at slower rate than a relatively thin rotor during cold start up. Monitoring of Differential Expansion of Rotor and Casing will be helpful to avoid such incidents.
  • For condensing turbines, if gland steam quality having less degree of superheat supplied for the turbine rotor sealing, then quenching of Steam turbine rotor may happen. Which will result into cooling down of rotor sealing area, it may create rotor vibrations. This may lead to reduce radial clearance of sealing fins and resulted in to rubbing of sealing fins. Supply of gland sealing steam quality should be as per standard requirements.

Correction:

When bows do occur, measuring the shaft runout in a lathe is the best way to quantify the magnitude and the location of these bowed regions. The horizontal method involves setting the rotor in a lathe or similar turning device. The rotor is turned when needed to measure shaft straightening progress with dial indicators. This method is typically used when shafts have high values of bowing and can also reduce excess hardness during heating cycles. Machining and Electric heating methods are applied for removal of rotor bow and steam chamber ovality.

Basically, permanent rotor bow causes shaft’s center of gravity to shift from the shaft centerline, which causes unbalance. Permanent rotor bow in steam turbines is not a frequent occurrence. If it occurs, however, it leads to severe deterioration of machine dynamic condition, even if the bow magnitude is still too small to cause problems i.e. with seals and clearances. Removing permanent rotor bow is usually a complex, time-consuming and costly operation. Minor bow can be dealt with by machining and subsequent thermal treatment; if the bow is large, rotor replacement can be the only solution. In any case, long forced outage, large production loss and high repair cost are inevitable. 

References:

  1. M. Chandra Sekhar Reddy. Design and Analysis of Steam Turbine Rotor. International Journal of Mechanical Engineering and Technology.
  2. Robert Goodrick, former Retired Steam Turbine Specialist (1981-2010)- https://www.quora.com/What-are-the-causes-of-sag-and-bow-in-steam-turbine-shaft

Anil D. Vyas, Founder, ADV PowerCon, Energy & Power Plant Consultancy Services

  • Kamlesh Kumar Shukla
    September 27, 2021 at 11:06 am

    Very usefull.
    Thanks,

    • Anil Vyas
      September 27, 2021 at 11:31 am

      Thank you very much

  • Jayesh mehta
    September 27, 2021 at 3:26 pm

    Really very helpful. Currently same problem observed in our plant steam turbine. After MI again rotor was rubbed and damage. This document complete technical information & helps a lot.

    • Anil Vyas
      September 28, 2021 at 3:09 pm

      Thank you very much

  • Alok singh
    September 27, 2021 at 4:15 pm

    Nice artical sir….

    • Anil Vyas
      September 28, 2021 at 3:10 pm

      Thank you.

  • Rajesh
    September 27, 2021 at 4:50 pm

    Deep drive analysis and good knowledge sharing Information.

    • Anil Vyas
      September 28, 2021 at 3:11 pm

      Thank you

  • G J Adroja
    September 28, 2021 at 3:33 am

    Anil,
    The article is very informative and useful for operating as well as maintenance team of thermal power plant
    Congratulations

    • Anil Vyas
      September 28, 2021 at 3:12 pm

      Thank you very much sir. ????????

  • Ketan Vadgama
    September 28, 2021 at 10:25 am

    Useful case study for all power plant professionals.
    Thanks.

    • Anil Vyas
      September 28, 2021 at 3:11 pm

      Thank you.

  • Nirav P
    September 28, 2021 at 4:52 pm

    Very useful case study

  • Viral Shah
    September 30, 2021 at 3:58 am

    Really deeply narrated and informative for those who are in maintenance. Congratulations Sir

Leave Your Comment

Your email address will not be published.*

3 × 3 =