May 31, 2006

Some Thoughts on Piston Valves and Lubrication...
Aside from some occasional calls at the basement shop (the engine is still there...), my main occupation with the engine was as „armchair culprit“.

The Pearl engine, which features a piston valve, has been designed for outer admission. This layout is a complete mystery to me, because

1. the pressure load on the stuffing box is higher than in an engine with inner admission
2. when using the steam inlet connection recommended by the manufacturer (lower opening of the piston valve cylinder), the inlet path is almost blocked by the piston valve when it reaches the bottom position.



Longitudinal section of the cylinder, showing the original piston valve. Piston rod and stuffing box are on top. The piston nearly blocks the steam inlet path.

For these reasons, I am considering to re-design the engine for inner admission. The resulting piston valve is shorter, and does not block the steam exhaust.



Re-arrangement for inner steam admission. Piston valve featuring labyrinth sealing (see further down)

The cylinder lubrication issue keeps me busy. I am aware of the fact that a minimum of lubrication is absolutely necessary with some designs, especially when using cast iron or if the engine features slide valves. Considering the great deal of measures necessary to separate the oil when operating in the condensing mode, the idea of being able to do without it is very tempting. Would it be possible to run my engine without cylinder lubrication?

In this regard, I cannot count on the experience of other Pearl users. No information is available about cylinder lubrication - not even from the manufacturer -, to say nothing of the contact address of further customers. Two single-cylinder engines are presently being built in France, however in the cast iron execution. According to my information, just two twin cylinder engines have been sold to Europe. The second one (hello Hans...!) is being assembled in Sweden from a machined cast-iron set.

My engine features bronze cylinders and piston valves. Due to the design of the piston valve cylinder (annular grooves), no piston rings can be used for the piston valve. Sealing is then reached by means of a tight fit. At first glance, the material matching (bronze / bronze) is not very suitable. But because no friction should actually take place, this material matching is the ideal one with regard to thermal expansion: the fit remains constant at different operating temperatures.

Now to cylinder lubrication. The engine will be operated with saturated steam. Could a kind of labyrinth seal (narrow, 1-mm grooves as shown above) be a sufficient measure to achieve the necessary oil-free sealing?

The main piston is another story . The owner of SL „Felizitas“ (the engine of which is operated without cylinder lubrication) told me that they are testing plastic piston rings (of unknown material...) which do not last very long. Titanium was also mentioned (??), which has a far lower thermal-expansion coefficient than bronze.

What about proceeding according to the motto: „The proof of the pudding is in the eating“? Suggestions are welcome
.

April 30, 2006

Spring 2005 to the Present Day

Description of the Pearl Engine
I had first purchased the bronze castings for the single cylinder engine, as its indicated power seemed to be sufficient for the hull I intended to use. Later on I ordered the missing parts for building the twin which consists mainly of two single engines mounted on a larger bed plate.


All-bronze single cylinder engine



Bronze / cast iron twin cylinder engine


The single cylinder engine is a faithful revival of the engine manufactured during the 1880s by the Edward S. Clark Company in Boston, Massachussetts. It was used to power pleasure launch style boats. The design of this 2 ½ by 3" engine is unique, with its piston and valve rods pointing upwards. Both stuffing boxes are accessible from the top of the engine, thus allowing for easy maintenance. This design calls for exceptionally long, wishbone-shaped connecting rods, which result in increased efficiency through less friction and strain.

The engine is being offered by the Pearl Engine Co. in Sutton, Vermont, in different casting and machined kits (bronze and bronze/gray iron castings), as well as fully assembled. The twin engine is a development of the Pearl Engine Co., as it was not included in the manufacturing program of the Edward S. Clark Co. The engine is a strikingly simple and rugged design. The assembled crankshaft (more on this later on) runs on babbitted main bearings. The twin is offered with a single-lever reversing mechanism which I am not going to install (I don't like it...).


My first task consisted of preparing a metric version of the delivered drawings with the aid of the Autosketch drafting software. Whenever possible, I have rounded up the resulting millimetre dimensions. At the same time, I have increased the cylinder bore to 66 mm (instead of 2 ½" = 63.5 mm) and the stroke to 80 mm (instead of 3" = 76.2 mm).

According to the manufacturer, the twin engine develops an indicated power of 4 HP at 300 RPM, whereby no mention is made of the corresponding steam pressure. It is one of the suitable engines to power the Elliott Bay launch, however, in the lower power density range.


Available Tool Machines

For many years I have been using two table top, small but precise, EMCO tool machines: the Compact-8 lathe and the FB-2 milling machine. They allow for most of the necessary machining work, with the exception of the cylinder bores for the piston and the piston valve. As they feature manual feed drive only, some of the machining steps call for a lot of patience...

I have subsequently equipped the lathe with a quick-change tool holder and a digital readout. Used milling cutters, side milling cutters, reamers (all of them eBay purchases) have been a great help.


Work Carried Out

There is no imperative sequence for the machining process. In order to "slowly reach the operating temperature", I started with the easiest tasks. Machining bronze casting parts is more pleasant than it is for gray iron castings. However, they require a lot of attentiveness.

Machining the bed plate with the small milling machine was certainly tricky. The range of the X-Y table was not enough to reach all surfaces in a single setup.


Bed plate

Clamping the large flywheel was possible on one side only. I still have to turn the other side after inserting a short piece of rod as a temporary shaft.


Flywheel


Machining the four eccenters and their straps was a joy

The cylinder heads with stuffing box and mounting holes for the slide rods were a demanding job. Not so, the slide rod tie.


Cylinder head with slide rods


Both slides were also very demanding. Sintered bronze bushings are provided as a sliding surface. As the slides of my engine are bronze, no bushings will be installed.


Slide

Both pedestals were almost too high to be machined properly, but they were no problem.


Pedestal

Both brackets, which hold the linkage for the piston valve drive, were also demanding, just like the individual parts of the linkage.


Bracket



Upper link

I had to prepare a simple but effective jig to machine the Stephenson link levers. A final grinding step is still necessary.


Stephenson link lever


Machining the Cylinder Bores

Machining the cylinder bores was carried out in a shop where a friend of mine, a model aircraft flier, works. The bore hole for the piston was machined with a headstock-mounted tool, while the bore for the piston valve was drilled and reamed. All of this, of course, was done with a huge NC machine, whereby the programming job took almost half a day... The result is fantastic, and the price for the job was bearable.


Machining the cylinder bores with the NC milling machine


Back again in the basement workshop: Drilling and tapping the fastening holes for the bracket


Manufacturing of the Crankshaft

Now to the crankshaft. According to the original drawings, the crankshaft is made of different parts (throws, pins and shaft) which – after assembly – are secured by means of taper pins. I am, however, a burnt child and think little of assembled crankshafts. The slightest torsion of the crank throws at the pin causes the crankshaft to knock (a frequent damage of model aircradt engines after a crash landing...). Therefore, I decided to machine the throw units (webs and pin) out of a single chunk of stainless steel. Not an easy job with my small lathe. It took me a whole day to machine each of these parts. Pre-machining the shaft sections (stainless steel) was an easy job.

Previously, I had discussed the issue with a shop in Kiel which has specialized in assembling and grinding chrankshafts for different internal combustion engines. No heat shrinking, but brute force: the parts are pressed together with a 60-ton hydraulic press. For proper assembly I had to provide for the precise fit, exactly 0.05 mm.

A borrowed inside micrometer gauge showed that the bore holes for the shaft studs, though turned in the same setup – had developed their own personality, with differences of +/- 0.01 mm. The shaft studs had to be machined to the required diameter, which demanded a lot of concentration and patience.

After assembly, however, the crankshaft ran absolutely true. Thereafter, pins and shaft journals were ground to the final dimension (26 mm instead of 1"). A nice piece isnt't it?.


Detail view of a crank throw



... and of the whole piece


Here is the result of the work carried out to the present day, all parts in loose assembly. A lot of work remains to be done, including important steps like babbitting the main bearings, etc. But the engine takes shape.


All parts in loose assembly

Finishing the hull and building the outer shell for the boiler are the tasks for the next couple of months. Updates of this blog will follow as soon as I proceed with the machining work.