Some time back MEWS had an article by Graham on his Sunderland type gear cutting machine. The design has now been refined into a practical bench top machine that produces gears at low cost.

Some time ago I decided to attempt to design from scratch a new gear making machine. Some 50 years previously I read about a commercial gear making machine which made use of a simple cutter in the shape of a rack. That machine was a Sunderland Gear Planing Machine. It utilized a cutter in the form of a rack to generate the exact involute curve applied to each gear tooth to enable perfect meshing irrespective of gear size. 

In the case of a very large gear, say of infinite diameter, the gear becomes a rack and the involute curve becomes a straight sided V-shape with the pressure angle determining the angle of the V-form. This was the fundamental approach used by Sam Sunderland who first built a prototype machine using a cutter in the shape of a rack in c1905.  

Operationally, a reciprocating rack cutter makes a cutting pass to a gear blank whilst also moving vertically as if it were in mesh with the gear blank. Thus as long as the cutter rack moves vertically a small amount after each cutting pass and the gear blank rotates a small degree in absolute synchronization then the process will ultimately result in generating gear teeth with the exact involute curve required.

To me this was a beautiful simple and elegant method to manufacture a complex involute curve which is specific to the number of teeth in the gear being made and its PA and DP or Mod specification. The prototype machine I wanted was to be a small bench top machine which once set up would run automatically and the end result would hopefully be a correctly generated spur gear.

The individual mechanisms required would adhere to the basic Sunderland operational principles but be totally different to that of the Sunderland (of which I had no detailed knowledge) and would therefore provide a very challenging design project. Fortunately, I had the use of 3D modelling software to make a digital machine but ultimately it is only when it is made from steel etc. and the trials begin you see if it will  produce a spur gear with a correct involute curve.  Making the prototype in steel was the simple bit and during the many trials it became clear that despite me being the designer and builder I would have to learn how this machine worked to get the best consistent results. 

From the outset I decided the true test of success would be if the machine could produce an acceptable spur gear using hard steel for the gear blank as this is normally difficult to machine and meant that the prototype machine would have to overcome this problem but also meant that in less hard materials the quality was likely to get considerably better. 

This project has certainly proved to be a challenge but the end is now in sight and recent gears made using hard steel are good and the rack cutter seems quite resilient showing no sign of wear. What is pleasing is the design processes used, which in no way are similar to the Sunderland, have been shown to work and also provide sufficient scope to produce gears of a wide variety for very minimal little cost. The making of a rack cutter is both simple and very inexpensive as compared to the commercially available alternatives there is a massive saving (1:160 for single gear specification).

For more details see  gwhengineering.co.uk