Balance Shot

 "For mankind, evil is injustice and cruelty and indifference to a neighbor's trouble, while virtue is brotherly love and justice and beneficence and concern for the welfare of your neighbor" - Gaius Musonius Rufus

The cold start yesterday did not go as I had anticipated.  I was unaware that the steam turbine was going to be given a balance shot.  The purpose - like your car's tires - is to attempt to minimize an unbalance condition.  Unlike a car's tire, this process is complicated by the length of the machine, the potential for misalignment of the bearings, thermal expansion, and the variety rotating components.

Turbomachinery comes with installed probes that allow monitoring vibration levels on each journal bearing.  The X and Y axis in the middle of the machine have been running high for a while now.  

When you have a vibration issue, it's common to temporarily install a very sophisticated high-speed digital vibration computer, and hire a specialist to interpret the data.  Typically the machine will be fed data from the turbine's installed vibration probes for a few weeks to gather as much data as possible prior to a plant outage.

The machine connected to our monitoring system uses the acronym ADRE (spoken as "Audrey").  It can plot the minute movement of the shaft within the bearing, as it rotates at 3600 RPM.  Crazy, the technology we have these days!  Bear in mind that the shaft is not actually touching the bearing - it is suspended on a thin film of oil - so there is room for motion within the bearing metal.  Ideally the rotating part is well enough balanced that the orbit is small.

The data gathered by ADRE allows the technician to determine the what is wrong in the machine and where the issue is located.  As an operator, all I know is that if I can put my coffee on the machine and it vibrates enough to mix in powdered creamer, the balance is not very good.

Even with a few week's worth of accumulated ADRE data, it still takes a skilled and experienced technician to determine how to interpret the information and decide how to proceed.  To my mind, this topic is about 50% guesswork/magic.  After hearing a few conversations, it's clear that nobody knows what is going on vibrationally inside an operating machine.  Mostly it's a bunch of people making educated and uneducated guesses.  My guesses fall into the uneducated category.

One important aspect of gathering this data is called a "Key Phasor".  The key phasor is a magnetic marker on the shaft, which signals to the ADRE when the shaft is at 12:00 each time it rotates.  This allows the ADRE to generate graphs with the correct orientation.  The technician can then identify where in the 360 degree rotation of the shaft on each bearing where the imbalance is located, and understand the phasing of the imbalance with respect to each bearing.

Below:  Image showing phasing of vibration due to imbalance:  Bearing A experiences upward force, while simultaneously bearing B experiences downward force. 

Wikipedia: Dynamic balancing - author Kaboldy

It's necessary to add a "balance shot" weight to the rotor, to see how the change influences vibration.  Once this new vibration data has been gathered, it's possible to determine the vector of the original unbalance and correct it.  So we add a weight, see how the vibration changes in direction and amplitude, then remove the test weight, and add the correct weight at the correct position. 

Like I said, it's kind of an esoteric art-form, especially when the rotational object is longer than a tire :)  I wish it was as easy as hammering on a weight after a quick spin on a tire balancing machine.

 After a couple of failed starts due to issues with our fuel gas supplier, (they were only supplying us from a 2 inch line, and the main valve never opened up)  we got a successful start on the gas turbine.  There was a long boiler warm-up to get the correct steam pressure, followed by long period to reduce the steam temperature to something more appropriate for an ice-cold steam turbine.  Finally we rolled it up to full speed, ran it for 15 minutes or so to gather vibration data, then shut down again.  

The turbine will cool overnight, and the technician will install what he believes will be the correct balance weight on the correct bearing, in the correct position on the shaft.  Even after a brief operating period and a day to cool down, he will still need to use welding gloves to attach the weight.  I hope to hell he gets it right.  It would be nice to get back to a normal routine at work.

Added to the fun of two failed starts followed by a balance shot, I had to shout over the racket of a power plant during start-up at two trainees that were assigned to me.  It's been at least a decade since I had to train anyone, and I'm a bit rusty at that task.  It didn't help that they had significantly different knowledge levels.  Anyway I'm still hoarse.  On the bright side, I got to skip out of the morning all-hands meeting :)