Part two by A Evan Lewis

My Hero’s Engines

Version 1 was my Grandfather’s engine. For safety reasons my Hero engines, referred to as versions 2 and 3 used a small elliptical boiler. Version 2 was built
with ball bearing races, but because they had to be sealed to prevent steam from escaping they created too much friction and it would not run. So I scrapped that idea and rebuilt it with plain phosphor-bronze bearings. But they too seemed to create too much friction.

So I looked back at my grandfather's engine and the Greek drawings and found that they used small pivots with much less surface area and thus reduced friction. So all subsequent versions were built with pivots and they worked. This convinced me that the Greeks actually built these machines and knew exactly what they were doing, and so did my grandfather.

Experiments with various pivot designs

The two pivots I used initially were conical pins that were stationary, inserted into cone shaped recesses on the ends of the rotating shaft. One of these pivots has steam entering through a hole in the center of the stationary pin and moving into the cone shaped recess with a hole in the centre.

As an added safety feature these joints were surrounded by a metal ring or sleeve so that if the joint came apart the shaft would be retained in the ring. This actually occurred several times.

From the pivot the steam moved into the rotating shaft, into the ball and out through the jets. The pivot on the opposite end of the shaft does not have to provide steam but is mounted on a screw that can be tightened.

This increases pressure on both pivots and reduces the steam that tends to escape around the pivot, and this allows the steam pressure to increase which tends to produce increasing speed. But as the screw is tightened the friction increases and the rotation slows down. It is difficult to get the perfect balance between these opposing factors. Also if the rotating shaft was not in perfect alignment with the axis of the pivot it would leak steam.

I looked back at my grandfather’s engine and realized that his pivot was not a perfect cone but slightly rounded like part of a small sphere about 6mm in diameter. I had thought this had been roughly machined, but then realized that the shape may have been by design and I changed mine to match.

A sphere fitted into a cone shaped recess will seal with minimal surface area touching and this minimizes friction. It also meant that alignment was not critical. On the other a hand, a cone seated in a cone can create an enormous amount of friction especially if the angle of the cone is shallow. On lathes and other tools we make use of this fact with Morse tapers which have a cone angle of about 1.5 degrees and can lock in so tightly that they can hold a drill chuck while drilling holes in steel.

Originally I used a cone with a surface at 30 degrees to the axis (using a centre drill). This design worked quite well and it ran at speeds up to 3,300 RPM. A speed of 5,400 RPM was recorded with no load on version 4 which was balanced more precisely. Despite improved balance it still rattled and by watching the steam escape from the pivot I could see it was pulsating in time with the rattle. This was caused by the shaft oscillating backwards and forwards along it's axis. If the adjustable pivot was tightened the rattle stopped but the sphere slowed down due to increased friction.

It was very difficult to get the tension between the pivots perfect. They needed to be a bit loose to allow it to pick up speed, but at high speed it was possible to tighten the pivots so that less steam escaped and the pressure increased causing the speed to increase further. With bigger jets it was necessary to minimize steam loss but if the pressure applied to the pivots was increased the friction actually made it stop spinning.

It took another careful examination of my grandfather's Hero engine to realize that the angle he used was shallower at about 45 degrees. I thought the shallower the better, so I tried the tip of a drill which has an angle of 88 degrees from the axis. The pivots jumped out with this angle. Finally I followed my grandfather's design exactly: ball shaped stationary pivot on both ends, seated in a cone shaped recess in the rotating shaft cut at an angle of 45 degrees. This gave the best results. The steam escaping through the pivot actually provides a cushion of pressurized steam for it to float on. Consequently steam loss cannot be completely stopped. But to reduce steam loss I placed the pivots inside sleeves so that the steam had to escape through a small gap between the pivot and the sleeve (see the diagram).

All the modern designs I have seen appear to allow the steam to enter through one pivot while using the other to adjust the amount of pressure applied between the pivots, as described above. If steam was allowed to enter through both pivots it would allow both to ride on a cushion of steam and may help to minimize friction. Will this be version 5?

See part one here.

For more information see Evan’s website here.

The YouTube below shows development of the engine - Note: Version 3 produced 0.055 Watts not 0.55 Watts as stated on the video. The best power output measurement with the model was 0.1 Watts or 100 milliWatts. 



the only free and the only weekly magazine for model engineers. 

Editor: David Carpenter