DESIGN CONSIDERATIONS - POWER PRODUCTION AND TRANSMISSION
Firstly, it would be desirable to design any man-powered device or thresher for easy conversion to electric motor or petrol or diesel engine power at a later date.
Of the conventional systems for utilising human power, treadling or pedalling would appear better than hand-cranking, for moderate or high power requirements.
Pedalling would appear to be more efficient than conventional treadling, but the system required is probably more expensive, especially for several operators. It could be worthwhile considering a seated operator or operators with single-acting or double acting treadle.
** Addendum 1999 - Weir - it could be interesting to extend the trolley mechanism used in my 2 Man Trolley Pump report to 4 persons, or even to 6 or 8 persons, and to link that reciprocating action to a rotary drive for threshing, milling, maize shelling, winnowing etc.. That path was never followed during those years. We didnt really do enough brain storming at the time or thinking seriously in terms of reuseable modules.
It has already been suggested above that a pedal thresher should utilise 4 pedallers and 2 feeders, operating in rotation. The conventional treadle rice thresher is usually operated by only one person at a time, but some Chinese treadle rice threshers are fairly long machines, designed for operation by several people at one time, and able to absorb the available power if motorised. This use of several operators at any time is bound to result in lower unit costs.
Several power transmission techniques were found in the literature - chains, V-belts, flat belts and spur gears. Prototypes at Morogoro originally used a type of flexible V-belting (Eaton) of hollow circular polyurethane extrusion, which could be cut to the length required and joined with a bayonet type small aluminium (fixed) or steel (unscrewable) connector, but shock loadings resulted in excessive belt slippage which could have been fixed by increasing drum mass, flywheel effect and angular momentum. 2 stage chain drive proved fine for the model described in this report, provided all sprockets were properly aligned, ran true, chain tension was kept correct, and some occasional lubrication was used. Power losses in chain drive are usually quite low compared with belts.
Wooden bearings were used for the first prototype Morogoro man-powered thresher, resulting in high relative power losses. Therefore this design uses mainly ball bearings. We tend to go for double-metal-sealed (e.g. type 6204.2Z), since they have some protection against dust and lack of lubrication, but they have slightly increased drag and therefore energy loss when compared with unsealed bearings. We dont recommend double-rubber-sealed (e.g. 6204.2RS) since their drag and energy loss is higher.
It is always nice when building or designing a machine like this to make it modular, so it can drive other equipment which one just bolts on. The idea of a multi-purpose pedal system was proposed by Wilson (1971) who has since developed several pedal devices. At Morogoro our program (but it was never a formal Program) developed the following: a 2 man pedal piston pump, 4 man pedal centrifugal pump, one-man pedal grain winnower, one-man pedal grain mill, and one-man pedal maize sheller. We really didnt do enough to develop multifunctionality, modularity and interchangeability. Rodale (1978) as well as Wilson did some work on the Dynapod multiuse concept.
When considering the plate mill, it is important to be able to remove the complete mill or at least the plates, since the dry rubbing of plates together for example when threshing not only loses power unnecessarily but also wears out the plates. The plate mill can be an integral part of the layshaft (as per the drawings in this report) or it could be driven from the layshaft by a flexible coupling or by a solid or semi-rigid bush. That flexible coupling / semirigid bush design option would just require that the mill was located by a torque arm of some kind, and would make removing and fitting the mill a quite simple operation.
Note that in fact the design as proposed here with the integral plate mill was never built - but we did plenty of work with pedal plate mills, on which this design is based. The integration of the plate mill is an attempt towards the ideal of modular design....