Turbine failures

09 December 2012
If P&C has over the years been involved in many claims due to turbine failures. To show the complexity and effects of such failures we would like to show you a practical example of one of our turbine claims.

​Description of the loss

One day the turbine in a pulp plant tripped for an unknown reason. The turbine was inspected visually and the quality of the oil was analysed. After consulting the supplier of the turbine it was restarted successfully. The following day it tripped again only 30 minutes after that a vibration measurement indicating normal vibration levels were taken. The restart of the turbine was not possible again due to high levels of vibration.

The turbine G3 is a STAL-LAVAL DM2070- NO B2310 from year 1978. No major modifications have been made. Last overhaul was done 2003 and next planned for 2013. The nominal capacity is 26MW. Current inlet pressure is 37 bar / 420 °C. Back pressure is 4,6 bar / 180 °C.

After inspections it was established that the three outmost blade drums of the turbine were damaged and had to be replaced. Overall this led to a standstill of the machine of around 8 months. Unfortunately there are few suppliers available to replace parts in this kind of turbine. There are extremely few persons in the world with the right competence to handle these old Ljungström-type radial turbines. ​

Case pictures​ ​ ​

Wheel 68 ​ ​

Wheel 69 ​ ​

Wheel 70 ​

​Wheel 71 ​

Possible failure scenario

Of utmost importance in a claim, especially a machinery breakdown claim is to understand the root cause of the failure. This is critical to avoid the failure from happening once more but also to be able to make a proper decision whether or not the claim is indemnifiable. In this claim the possible failure scenarios were:

  1. Deposition of impurities on blades.
  2. Standstill corrosion. Aggressive electrolytes form that lead to pitting corrosion when deposits dissolve as blades become wet.
  3. Fatigue cracking of one blade on inner row (left in drawing) drum 70 due to pitting corrosion on leading edge. The blade is twisted and bent and causes wear on adjacent drums and limited in balance.
  4. Fatigue fracture of second blade on drum 70. This blade is torn off and the loose blade causes severe damages and more blade fractures in the same row. >40 blades are milled to grains.
  5. Outer row (right in drawing) of drum 70 becomes loose due to the loss of blades in the inner ring.
  6. Mechanical damages and vibrations lead to fatigue fracture of free support ring of drum 71. This leads to immediate loss of all blades of drum 71 and they are blown down the steam outlet.
  7. The two loose support rings (the middle and the free rings) of blade drum 70 rupture and are thrown onto the volute casing.

Conclusions

Over the years If P&C has established good working relationships with the best experts in the field to be able to pinpoint the actual root cause of failures like this one. After very thorough examinations If P&C, based on the external expert, came to the conclusion that:

The primary cause of failure is most likely pitting corrosion that has developed over several years.

The conditions necessary for pitting corrosion is believed to be related to impurity deposition on the blades. If blades become wet due to condensation during a standstill the deposits may dissolve and create an aggressive electrolyte that triggers the pitting corrosion process.

Pitting corrosion has caused fatigue fractures on at least two blades. The blade fractures have then damaged other blades and support rings on both drum 70 and the adjacent drums 71 and 69.

The final collapse of the turbine was probably initiated by the fatigue fracture of the free support ring of drum 71 that led to the rupture of two support rings on drum 70.

Summary of damages (picture)

Avoiding similar losses

The turbine had its last major overhaul in 2003. After that the turbine has had 55000-60000 operating hours. The manufacturer’s recommendation for major overhauls is after 45000-55000 operating hours. The overhaul for this age of turbine should be carried out every 6-7 years and not every 10 years as it was planned.

The fatigue fractures could have been discovered some years ago if the turbine had had a major overhaul including modern NDT-inspections. This would have prevented the loss. But it is also possible that the fatigue fractures might have developed over a shorter period, i.e. some months. However the probability that the fatigue fractures starting 1-2 years before the loss is higher than the start and development of fatigue fractures during some months.

If there are many stops, which increases the risk of pitting corrosion due to condensation it is even more important to do the overhauls on a regular basis. The steam quality is also of utmost importance in avoiding pitting corrosion.

Staffan Ljung and Kyösti Korventausta