This decade saw the greatest change in my career.  After what seemed a lifetime with the firm, we parted company in 1995 and I was cast adrift on a turbulent sea to seek my own salvation.

After six months of searching fruitlessly for a similar position, I came to the conclusion that I would have to leave the Northeast if I wanted to get a worthwhile job.

This led me to (for a Geordie) the deep south, Hertfordshire.

In 1990 I bought my first new car, a Nissan Sunny.  By the time I had passed it on to my son it had traveled the equivalent of seven times around the world—with no trouble!

A chronometer type of pocket watch.  Many such watches were so highly decorated they became works of art.

Life in the Applied Physics Dept. was varied and interesting.  There were the long term projects like developing optical sensors and these were interspersed with urgent programs to investigate any faults or problems which occurred with machines in the field.

 

At this time, optical sensors were in their infancy and as such, were being hyped for all they were worth by their proponents.  As someone with a background in optics I could see that they would indeed prove useful in the future but most of my colleagues were sceptical.  This scepticism I found rather annoying, I know engineers tend to be conservative by nature but few of my colleagues really understood optics and their disbelief stemmed, I felt, from ignorance.

 

A large part of the AP department’s workload was acoustical control.  Turbine generator sets were inherently noisy (earplugs are mandatory in power stations) and this noise had to be kept within limits.  I was never closely involved with this work but, on occasions, I had to use their equipment to check levels on my projects.

 

The main duties of the department were to investigate new technology and integrate it into the turbine where appropriate.  To put this in perspective, turbine generator sets installed in power stations have a very hard life.  The power station operator wants to run the set continuously for as long as possible.  When it’s running it’s making money, when it’s idle it costs money.  This means the machine can run for as long as a year without a pause (and this is at full power).  To do this safely the machine must be monitored constantly for the first signs of trouble.  Now a large turbine generator set is a huge machine comprising high pressure turbine, intermediate pressure turbine and as many as three low pressure turbines followed by the generator.  (Surprisingly, there is not very much information on power stations on the web but this site has some interesting information.)  Back to monitoring; this means hundreds of sensors are needed to measure temperature, pressure, vibration, displacement etc.  The signals from these sensors are fed back to the control room where every aspect of the sets performance is monitored.  Now one place that is difficult for installing sensors is the generator.  The most common sensor is the thermocouple for measuring temperature, thermocouples consist of two wires and the generator is the part where very high voltages at very currents are generated.  Thermocouples have to be installed very carefully to avoid possible short circuits.

 

The generator has a very arduous life.  The core is made up from thousands of thin metal sheets.  Each sheet, in the form of a sector of a circle,  is insulated from its neighbours with a thin layer of varnish.  The centre is hollow for the rotor.  At the centre, slots cut in each sector align to form channels running the whole length of the core.  In these channels lie the conductor bars, these are hollow tubes of copper.  They are hollow because they are water cooled.  Yes, that’s right, water cooled.  This may sound strange, passing water through copper tubes carrying 22,000 volts at 800 amps but its true.  The trick is to purify the water, pure water is not conductive.  Now these conductor bars are subjected to large forces as the rotor turns.  The rotor is really a huge electro magnet and these magnetic fields cause the conductor bars to vibrate as they pass.  To secure the bars special glass-fibre wedges are clamped over them with stiff glass-fibre ripple springs between the wedge and the bar.  The springs allow the bars some movement but restrain this to a safe level.  One of the most common problems with generators is slack wedges.  The vibration slackens the wedge, the vibration amplitude increases until the wedge breaks apart and he bar is freed to wreak havoc.  To avoid this each wedge must be tightened to just the right degree, too little or too much is equally bad.  To achieve this manufacturers have traditionally employed specially trained inspectors who tap each wedge with a small hammer and listen to the sound emitted.  This is similar to the railway wheel tapper who listens for the clear ringing sound which denotes a good wheel.

 

Now this is not a scientific method.  It works (usually) but depends on highly trained personnel.  I was allotted to the project to discover a scientific method of replacing this practice.  What we arrived at was a system where an accelerometer was fixed to the wedge (with beeswax) and a calibrated hammer used to tap the wedge.  The signals from both devices were recorded and processed by a FFT analyser which displayed the response.  By calibration with wedges of varying tightness we could determine what the signal from a correctly tightened wedge looked like.  Testing showed that this method was much more effective that the traditional one and reliability was greatly enhanced.  I left the project just as the next phase began.  This was to design a thin robot which could perform the job without removing the rotor.

 

My next project was fairly challenging.  It was to design a pressure sensor for the tips of the stator teeth.

 

 

 

 

Next decade

Go to technical page