Condition Monitoring

[Mahesh]

Can anybody tell me weather we always need two sensors to measure the differntial expansion on a ramp surface?

We'd installed only one 25.5 mm tip dia sensor on a ramp with angle of 14Deg with XM-320 module. But now we are getting errorinious readings (too high values than actual condition). When we measure the gap voltage, then the values matches with the callibration chart. We'd configured the XM-320 for normal mode and 14 Deg angle.

There are two differential measurement in the 210MW steam turbine. One for HP rotor and other for LP rotor. Both on Ramp. HP Diff. Exp. value is correct one and is around 5mm. But LP Diff. Exp. is showing the error with value of 28mm (more than the limit defined for this measurement).

The sensor is mounted at the parting line (Horizontal direction). Is this error due to radial displacement of rotor because of the jacking oil? what may be the solution?

[ClearEternity]

1) On a ramped shaft like you have shown, it is normal to use two transducers so you can compensate for oil jacking. Some will tell you this is not necessary but most will tell you it is best practice to use two transducers.

2) Your problem is on the LP side and you state the error is the pickup is reading 28mm. I have some ideas/questions:

a) What is the voltage you are seeing at the XM320 from this pickup? Is it 0, or -2 or -22V?

b) Can you see the pickup while the machine is running? If you can does the space between the pickup tip and the shaft seem to be appropriate for the voltage from the transducer?

c) What is the voltage coming out of the probe driver?

d) Can you see the bracket you mounted the pickup on? Has it slipped?

[Mahesh]

Attached is the callibration table for that sensor. The sensor was working fine for one day after rolling the turbine to 3000 RPM. But after that started its upward trend and that also very rapidly. There was one sharp step of 2 mm in the trend near its 17 mm value. The value kept on increasing untill the voltage reduced to sensor fault condition. The voltage we are reading at probe driver is near to zero voltage. If this issue is due to jacking oil or radial shift of the shaft, then it should not move so much. Also the reading is well beyond the possible expansion of the rotor. There might be some other issue. As the turbine is in running condition, we cant see the physical gap and condition of the bracket. Is there any possibility of this behavior due to oil leakage into the Sensor-Ext Cable connector? We took care to prevent oil leakage, but cant ignore the possibility.

[ClearEternity]

1) Have you measured the resistance of the extension cable and probe. You can do this by un-threading the extension cable from the probe driver. Then take a DMM and measure the resistance between the center pin and the outer hull of the extension cable connector. You should read close to 5 ohms.

2) With the extension cable disconnected see what the voltage is at the signal output of the driver. If it is -22 or greater then this tells you if the probe or extension cable was cut (creating an electrical open) that you should be reading a very high negative voltage. This would also say that if the probe is very close to the target or the probe system is shorted you would tend to get an output close to 0V.

3) As to whether this output is caused by oil jacking you would need to find out if the shaft can physically move (clearances in the bearings and supports) that many mm. Most turbines I have worked with did not have 28mm horizontal bearing clearances.

4) If oil penetrated the connector between the extension cable and the probe, then you should see some difference in the resistance measurement.

5) Question, did you isolate the extension and probe connector so that the outer hull of the connector would not physically touch (creating an electrical short) to the machine?

6) Have you verified the probe driver is still isolated from the junction box? (RA sends isolating shoulder washers with the probe drivers to do this).