affect the valve timing, (the engine has a transverse piston valve operated by a link from the vertical shaft). Slight adjustment is possible. The exact dimension from the block centreline is 2 inches = 50.8mm. If E is made 32mm, this dimension becomes 51 mm, a difference of 0.2mm, about 0.008 inch. But we have, as it happens, moved the seating of the bevel gear on the shaft endways by 0.25mm, which compensates almost exactly. So, E can be 32mm with no problems. Finally, F. This figure affects the meshing of the two bevels. The distance is, therefore, rounded upwards, so that shim-washers can be used for adjustment. In all this work it will be helpful to remember that 0.1 mm is almost four thou
(0.004") and 0.01mm four-tenths. For the metric user, one thou (0.001") is 0.025mm when converting drawings in the opposite direction.
References between Figs 86 and 87 will show how the other, minor, dimensions have been settled and you will notice that we have been able to make the final conversion in round figures of millimetres with very few exceptions. We have, in effect, redesigned the four components to S.I. standards.
Agreed, this exercise has taken some time, and to deal with all the lesser dimensions not shown on the sketches -and those on the two or three score other parts - would take even longer. As I said
before, we are really redesigning the engine with the constraint of having to work to the ■existing castings, and this, inevitably, takes longer than designing an engine and leaving the materials supplier to make patterns and castings for you.
On a really large set of drawings, of course, the job would be formidable indeed, and some would say that it was impossible. This is far from being the case. It is, in fact easier than reducing the prototype from full size to model scale; at 1 in/ft 19 inches comes out at 1.583333 in. However, it does take time, and I find that by far the best procedure when a number of really large drawings are involved is, after carrying out the initial steps (identifying rule, mating, and exact dimensions, and then making an exact conversion of all) to get out the drawing-board and make a new set of drawings altogether. I would add a further piece of advice, too. If scaling down a full-sized prototype but to metric dimensions it is far easier (and safer) to use a decimal scale -1/10 instead of 1/12. This cannot be done for locos which must run on an imperial track, of course, but for other model work there are very considerable advantages.
I need not add that in all cases it is vital to check and double-check your conversion drawings. I might add that such checks may often reveal some errors in the original drawings before conversion.
Was this article helpful?