Volume XII Issue 4

We have a Code of Safe Working Practices at Hydro-Québec which covers maintenance, repair and refurbishment activities. When we have to do work on the moving parts of a vertical axis generating unit, the code calls for a visible energy cut-off point that may cause the generating unit to turn. In several cases this cut-off point is the head gate. Often the head gate is not watertight and during work on the alternator rotor, leakage from the gate is drained by the turbine with the wicket gate fully open. Thus the workers assigned to the job do not feel secure. They fear that the leakage will be sufficient to cause the generating units to turn.

In July 1999, workers refused to carry out such work, claiming it was unsafe as the head gate of the unit they were working on seemed to be showing a significant rate of leakage.

We explored different approaches in order to establish a method of evaluating what leakage rate at the head gate was sufficient to cause the rotating part of a shutdown generating unit to rotate. This document presents the method we have chosen. It also presents the results of a series of tests carried out to validate the rates established by this method. The series of tests was conducted on 20 November 1999.

The chosen method calculates the water flow when a change in momentum across the turbine wheel passage produces a torque equal to the frictional torque between the rotating part and the fixed part of the generating unit. The principal frictional torques are at the thrust bearings and the braking surface.

The torque transmitted to the shaft by the turbine will be the change in the tangential component of the momentum of water flow, at the average radius, between the turbine entrance and the turbine exit. The calculation does not require precise measurements but they should be consistent and representative.

When the generating unit is fitted with a vertical lift gate at the intake, the flows calculated may easily be verified by comparing them with the opening required to obtain the same flow. Then the leakage rate observed may be compared with the flows established earlier.

The difference in the water level between the scroll case and downstream, which is equal to the dynamic pressure at the entrance to the guide vanes, plus the head losses at the guide vanes and in the turbine wheel passages, is also an indication of the flow.

Included in this document are the calculations of the torque required to cause rotating parts of a generating unit to rotate; the calculation of the flow required to produce this torque; and the results of a test carried out on generating unit 3 at La Grande 3 power plant.

Want to read the full paper?

To purchase a copy of this Dam Engineering paper, please email the Editor, Carrieann Davies, on: cdavies@wilmington.co.uk or fax: +44 (0) 20 8269 7804.
Please include your name, postal address, credit card number and issue number.
The cost is £20 sterling for each paper, postage & handling free.
Your order will be confirmed by email. We aim to dispatch your order within 48 hours of receipt.