The restored bogies now fitted with air brakes.
I mounted air brake cylinders and brake gear on the bogies and did some experiments in making failsafe continuous air brakes. The system worked, but at the time I used cheap tyre inflator 12V compressors. These units needed to be oiled often and would regularly overheat and become less efficient. This was possibly because some drivers did not quite understand the concept and would often stand in the station with the brakes applied, which continually caused the compressors to cut in. In turn, the batteries would run down before the end of the day. Again, I left the batteries on charge overnight and despite attaching a large notice, warning that the charger leads were attached, the coaches would be drawn out of the shed next morning and the sockets and wires ripped out. Nonetheless, these coaches have since given some years of service, although not continuously braked.
The failsafe air brake module.
The fourteen month ‘Black Five’ boiler examination now made us aware that the tubes had eroded at firebox end and there was a slight weeping from the front tubeplate. Whilst we all felt it was time for a new boiler, John Glaze suggested that we could make a successful repair by inserting new copper tubes, although the boiler would need an ultrasonic test once the old ones had been removed. In addition, the replacement tubes would have to be of C101, or C106 quality; the material source also had to be traceable.
It was left to me to do this work, but there were a number of suggestions from members on how the job would be best done. One was that I should drill the old tubes out. I was not in agreement with this, as it would be all too easy to damage the tubeplate. Any kind of damage and we would be dogged with weeping tubes.
John Graze examines the Black Five boiler.
With the boiler out of the frames I decided to carefully grind off the tube ends that still stood proud of the tubeplate with a mini-drill. Unfortunately, this tool proved inadequate and I therefore had to result to one of the society’s young members, James Gorton, holding my electric hand drill with a flexible shaft having a small grinding wheel attached.
With this job complete, I made a hefty jig to fit over the front of the boiler barrel, also a long ½" pull rod screwed at both ends and collets a good fit into the tube ends with a step just large enough to pass through the tubeplate bore. Copious amounts of WD40 had earlier been applied to the tubes, then with a large nut and ring spanner at the front end of the boiler, pressure was brought to bear on the draw bar. At this point there was no sign of movement and it then became a big hammer job. Several sharp blows with a drift through the firehole, while systematically tightening the Whitworth nut, the first tube began to move. With a series of tube spacers put on the drawbar the tube eventually came free of the boiler.
With the first tube out more WD40 could be applied to the inside of the tubeplate. On the second tube the draw shaft broke and here Ron Scott’s vast knowledge of steel quality came in useful. He quickly produced a bar of EN8 quality steel and this did prove to be of adequate tensile strength. The superheater tubes proved more stubborn, but eventually gave way to an even larger hammer!
The results were finally very satisfactory, as despite the years of service the boiler had already given, the bores in the tube plates were rather like they had been drilled the day before. Close up, the tool marks were still evident.
With the result of the ultrasonic test being positive, only a maximum .025" had eroded in one place on the .250in and .375in platework.
Expanding the new copper tubes.
We were only able to get the use of a tube expander to fit the small tubes, therefore whilst the correct material was being sourced I set about copying the small expander to make one to suit the larger tubes. Another day in my workshop proved well worthwhile to keep our costs down.
The new 14Gauge tubes fitted perfectly and a test made with an offcut proved that annealing would be unnecessary, as had been suggested by some members. Both the expanders worked trouble-free and with the job complete the boiler was given a fresh coat of red oxide; I always feel presenting a job well for the hydraulic test is again worthwhile. We did a preliminary hydraulic test ourselves and there was no sign of any leak; similarly, the results were identical for John Glaze. Not bad for an amateur, I thought. I think even John was impressed.
The boiler well presented for the hydraulic test.
CHAPTER SEVEN
Towards the turn of the century I was thrown a challenge, and this encounter proved to be the most devilish task that I have ever undertaken. This project was to take innumerable hours of my leisure time over the next seven years. Working on this venture was the only time I remember my wife passing any comment regarding my hobby. At one stage she did recognise that frustration was getting the better of me and suggested that I should take a break from working on it. The mere fact that she passed this remark was enough to make me realise that I had indeed become obsessive.
I was asked if I would make a replica 1724 James Harrison wooden regulator clock movement. These old clocks were reputed to keep time to one second a month regardless of the ambient temperature.
The main frame was of oak, as were the clock wheels. Each wheel disc had the gear teeth inserted into them separately with the wood grain running radially along the tooth. This therefore increased the strength of the tooth, as having the grain running across the teeth would make them susceptible to breaking off. The oak gear wheel arbors ran in lignum vitae bearings; this oil-bearing tropical wood made lubrication unnecessary and therefore the clock was deemed to be virtually maintenance-free. The pinion wheels were of the lantern type, made of brass, having inserted small lignum vitae rollers. These rollers rolled in and out of the gear teeth, making the gear train extremely free running.
I took on this project being poorly versed on the history of the Harrison brothers and having already made two long case clocks certainly made me rather more than overconfident. I had not accounted for the genius of these old clock makers and looking back, it may well have been simpler for me to have attempted to counterfeit any Tate masterpiece!
The Harrison clock movement. The small lead weight is the ‘maintaining power’.
The proposer, Peter Jordan, had initially obtained a very basic set of drawings from the Time Museum, Greenwich, but they were hardly comprehensive. Shown were the number of teeth on the wheels and the pitch circles, there was no gear teeth profile detail and there was also no reference to the front motion work whatsoever. The motion work is the important section that gives the hands their twelve-to-one ratio. I never ever shone at maths, so calculating this gear train certainly became an enigma for me. There was no way that I could arrive at these figures by associating this clock with any normal long case clock, as all the ratios were entirely different. It is common in clocks for the drive for this motion work to be taken from the centre arbor which comes through the dial. To complicate matters for me, Harrison took the drive from the second wheel of the ‘going train’ which is off-centre.
The bit that makes the clock tick, the escapement, was also a completely different concept for me, this important section being of the ‘Grasshopper’ variety. The scape wheel is the only brass wheel in the clock and the arbor that it revolves on does not run in bearings, but sits loose in the valley of two pairs of small lignum vitae rollers; the arbor is held in situ entirely by the weight of the brass wheel and the arbor itself. How Harrison even assumed that this concept would work is far beyond my understanding; I still find it beyond belief that the scape wheel arbor will remain in place on the rollers whilst giving the heavy grid iron pendulum enough momentum to swing in an extraordinary wide arc.
To add to the complications of the clock, the pallets, which engage the brass scape wheel, are of lignum vitae also. The escapement is so-called as these wooden pallets hop in and out of the brass scape wheel with no rubbing action whatsoever. Therefore, technically there should never be any wear. Again, the brass frame holding the pallets is also loose and just sits in two ‘V’ groove
s ground into glass for bearings. The grooves in themselves have to be sloppy to allow the frame to rock and give the pendulum its momentum. During initial tests, it was therefore not unknown for this unit to jump out of the grooves, adding further to my difficulties.
Having the components running loose reduces friction to the minimum and therefore the escapement needs less weight to drive it. I also initially had great difficulty with the scape wheel moving off the rollers, this did tempt me into trying normal lignum vitae bearings. I then found that I had to increase the driving weight out of all proportion to get the clock to run. The winding drum ratchet wheels were specified to be of oak, but I found that being cut from discs of this wood there were places where the grain made the ratchet teeth vulnerable to breaking off. I eventually made these components from lignum vitae, which did solve the problem; even so, every effort had to be made to reduce the weight required to drive the clock as much as possible to protect these vulnerable parts.
Fly cutting the clock wheel teeth. I doubt Harrison would have done it this way!
Lignum vitae, being a protected tree species, is not the kind of timber that can be purchased from B&Q, or any local timber yard. I was able to get hold of small amounts for the bearings, etc., but when I wanted to make the discs for the ratchet wheels there was a problem. I mentioned my difficulty to a member of our society, Peter Evans; lo and behold, a few days after a section of the wood came through the post out of the blue from someone that I had never met. What wonderful camaraderie model engineers have! Thankfully, I was able to thank Brian Young personally for this extraordinary kind gesture at the Fosse Way Model Engineering Exhibition, whilst at the same time admiring his incredible carpentry skills on one of the stands.
This clock also needs what is called ‘maintaining power’; this device keeps the escapement running whilst the clock is being wound. Taking the torque off the scape wheel whilst winding the clock would cause the pallets to miss and become damaged. I find, it is not surprising that despite the concept of this type of escapement being virtually frictionless, few clockmakers were to adopt the ‘Grasshopper’ escapement in later years.
From the onset I had to be very careful. Whilst many parts of the clock had to be very accurate, on the other hand, I did not want the finished job to look too ‘clinical’.
Although a very drawn out process, making the gear wheels was to prove the easy bit, that’s if you can stand the monotony gluing all the gear teeth in separately!
With the ‘going train’ eventually mounted in the frame I found that it took far more weight than I was led to believe was actually necessary to drive the escapement. As you would expect, there are very few of these wooden regulator clocks now in existence, but I imagine that somewhere it is recorded how heavy the driving weights should actually be. However, any enquiries that I made seemed to be met with a shroud of secrecy. Had I known this weight for sure, it would have given me a yardstick and I would have known what to aim for; this may well have saved me many hours of time.
The grid iron pendulum and adjustable cycloid cheeks.
I only had old photographs for reference, therefore my thoughts were at first that the excessive weight required to drive the escapement must be because I had the tooth profile wrong. I then attempted to convert the gear profile to a more modern shape. This aggravated the situation considerably and I had little option but to replace the gear teeth on every wheel and start afresh.
Any successful running of the escapement does rely entirely on the smooth running of the going train; any variation in the torque whatsoever tends to cause the scape wheel to lift off the rollers. In which case, one of the delicate wooden pallets would narrowly miss the scape wheel tooth; the brass wheel would now be free to spin just as the pendulum brought the second pallet to bear on the scape wheel. The scape wheel teeth then effectively become a circular saw, severing both pallets. This leaves the gear train free to run wild at the mercy of the heavy lead driving weight.
The motion work – all the gear teeth inserted with the wood grain running radially along each tooth. The lower wheel will operate the date ring.
Losing the pallet ends during the early stages of experimenting with the ‘Grasshopper’ escapement was not too much of a problem. Having modern adhesives around I could quickly glue the pallet ends back on again in many cases for further tests. Also, I was usually in attendance and quick to stop the gear train running wild by removing the weight. However, this did later become the nightmare scenario as the clock got towards completion and was therefore left unattended. I was all too often to come downstairs in the morning to find, not only the pallet ends on the floor, but the gear train running wild had also caused the wooden lantern pinion rollers to disintegrate when running at high speed. The brass pins holding the rollers in situ had then come into contact with the gear teeth, severely damaging them also. After countless hours of work, this clock would literally destroy itself in seconds.
After many hours of observation and listening to the escapement working, I did find the reason for the variation in torque. Most of the pinions had eight rollers or more, but the third pinion in the going train has only seven. This pinion works in a rather large diameter wheel and I found that the escapement lost some momentum as each of the seven rollers began to take up the drive. Again, I spent countless hours re-meshing the pinion, altering the size of the rollers and profile of the gear teeth ends, but never really felt that I had achieved best performance. Although the clock did eventually run for long periods, keeping reasonable time, I still felt that more work was necessary.
Unfortunately, as mentioned earlier, there was no reference to the motion work. I then had to calculate the size of the wheels and number of teeth required for getting the twelve-to-one ratio for the hands.
After hours of pondering I felt sure I had cracked the problem, and therefore made what I felt was the correct setup. Alas, revolving the minute hand round once didn’t quite bring the hour hand to where it should have been. I then had no option but to recalculate and make another complete set of gears. I went to bed with the problem on my mind and got up the following morning still chewing it over!
The strike train, showing the count wheel.
The whole principle of this clock was an absolute minefield for someone like me; however, I floundered on with information which seemed to me to be rather more speculation than fact.
The grid iron pendulum is built up of brass and steel rod; this design keeps the pendulum length constant in all temperatures. As I mentioned earlier, the pendulum swings in a very wide arc, so much so, that the trunk of the clock is cut away and boxed in to allow for this. The rather long suspension spring swings between two cycloid cheeks designed by Harrison. The suspension spring wraps around these cheeks which are adjustable and can alter the arc that the pendulum swings in. Having read in detail the reason for these cycloid cheeks, I have to admit that my 1940s Secondary Modern education does let me down here. I will therefore have to leave it to those who are better versed to ponder over what Harrison’s design does achieve.
Unfortunately, the only convenient place I had to site the clock whilst under construction was in our conservatory. Here I could keep a watchful eye on it and often hear if it did begin to run wild. Whilst historians insist that these clocks were happy in all conditions, my assembly most certainly rebelled against the extremes of the conservatory. In summer the wood dried out and cracking became evident; on the other hand, the humidity of the tumble drier and washing machine in winter caused swelling of the timber. I liberally waxed the oak in an attempt to keep out any moisture, but this did little to improve matters. Yet finding a place where the clock would not be, at the very least nudged, really was a problem.
I dismantled this complicated mechanism innumerable times for adjustment, but whilst the clock ran for long periods and kept reasonable time, I began to run out of ideas of what to do to improve its reliability. I had never started another project before completing the previous one, but
on this single occasion I had indeed reached stalemate. There again, I longed to see if I could eventually achieve the ultimate timekeeping of one second a month. Fully intending to go back to it, I did break the habit of a lifetime and start another venture.
The finished clock. Note the box sections which allow extra swing of the pendulum.
However, after some seven years of waiting I felt that Peter Jordan had been patient for long enough when he asked me if he could borrow the clock so that his carpenter could make the hood for the clock, the trunk being already made. Peter is well acquainted with the history of Harrison and also did run his own clock restoring business for many years. I was therefore happy for him to take it.
Peter mounted the clock on the trunk and this was screwed to the wall in his living room – this was an ideal place for it. Things came to Peter’s notice that I was not aware of, so with his years of expertise he was able to do perhaps a little more than fine tuning.
The carpenter and Peter did a lovely job on the clock case and with the clock mounted securely there was less chance of it ever getting bumped and therefore destroying itself.
There was also an error on the drawing for the pendulum which came to Peter’s attention; this fault rendered the fine adjustment useless. I later modified this to his instructions.
The Mettle for Metal Page 6