This is an excerpt from a book I bought in Germany.
Title is "The Wankel Engine Today" 1977 Audi NSU

It illustrates just how difficult it was for Felix Wankel to
devise a practical rotary engine out of the hundreds of possibilities.
It remains an enormous achievement and not easily duplicated.
 
Paul Lamar

************************************************************************

Background to an invention

Rotors instead of Pistons

A smooth-running engine is something that the passengers in a Ro 80 soon
take for granted. They can just hear the engine humming quietly, more like
an electric motor than a combustion engine. They are not disturbed by
vibration or noise; they can enjoy listening to the radio and talk without
raising their voices. At the end of the journey they still feel fresh and
relaxed. But nobody gives much thought to the long story of development
which led to the existence of the rotary engine in its present form.

The history of the engine really goes back as far as the invention of the
wheel. There is no way of telling just how long ago man first put the wheel
to use, but as he watched it going round he was fascinated by the continuous
motion, and since then he has never stopped devising further methods of
exploiting the rotating wheel for his own purposes.  It was only a short
step to the idea of solving his transport problems by fixing wheels to a
cart which was pulled along by animals.  Another fairly obvious possibility
was to construct a mill-wheel driven by moving water to do his work for
him. But later on it first appeared impossible to generate power directly in
the form of harmonious rotating motion.  The question arose when technology
progressed as far as the reciprocating engine with its bulky, oscillating
crank mechanism driven at first by steam and then later by burning gas. 

Ever since the first mechanical prime movers, successive generations of
inventors have visualized a piston rotating without vibration and without
reciprocating masses. This appeared basically quite simple and logical. But
again and again inventors discovered how difficult it was to put something
so apparently simple into practice. The difficulty was that a rotating
piston had to play a part in forming variable-volume chambers to enable the
engine to draw in, compress, burn and expel gas with the correct timing, in
other words to run in a four-stroke cycle.  This ideal solution required a
complex process of thought to visualize the special problems involved. 
Right from the very beginning the almost uncanny fascination of this task
resulted in a flood of inventions and patents, particularly over the
last 150 years. But very little public attention was given to any of these
developments for the simple reason that nobody is interested in failures. 

At that time there were two main reasons for the failure to produce a
practicable rotary engine design. The first, ironically enough, was that
there were simply too many possible shapes which could be used for the
rotors and their corresponding housings. The number of reasonable
alternatives ran into the hundreds, whereas there is only a limited number
of configurations suitable for reciprocating engines.  The scope for varying
the shapes of the rotors proved to be so wide that the problem of selecting
the correct system appeared insoluble.  And anyone who successfully picked
his way through the baffling and deceptive variety of alternatives was then
faced with a fresh obstacle to progress, namely the sealing problems arising
from the annular combustion chambers or other volumetric shapes. They
require multi-cornered sealing lines formed by sealing parts joining from
different directions. This presents far greater difficulties than the
single-plane sealing line for the cylinders of the reciprocating engine.

And so it was not altogether surprising that the number of aspiring
inventors was gradually reduced in the face of such obstacles. One man who
refused to be discouraged was Felix Wankel, who from an early age had
exercised his ingenuity and craftsmanship on research into technical
problems. During the difficult years after the first world war he had been
forced to work as an apprentice in the publishing trade, but his thoughts
were occupied by greater things. He was particularly interested in the
smooth-running properties of electric motors and turbines; the latter were
then being developed as a new type of combustion power unit by the Brown
Boveri Company in Mannheim, not far away from his home town of Heidelberg.

In 1926 Wankel - at the age of 24 - started what was to prove an epic search
lasting for many years and marked by promising discoveries and disappointing
setbacks.  The very first tentative designs on the drawing board were soon
followed by practical work in his own experimental workshops which he had
equipped himself. Although it was not long before his financial
means had increased, the kinetic system which was the object of his search
still remained stubbornly elusive.  

The companies sponsoring his work (BMW from 1932-1936 and the German
aeronautical research institute from 1936-1945) were in fact not directly
interested in Wankel's own objectives. They needed his experience for
sealing rotary disc valves, i. e. for engines in which the conventional
poppet valves, which are subject to banging and to high thermal loads, were
to be replaced by rotating discs used to control gas flow. Wankel had really
only worked in this area because he was interested in the difficulties
involved in devising seals for hot gases under high pressure. The knowledge
he gained was to prove indispensable when designing rotor seals for his own
engine.  After several failures Wankel had reached the conclusion that his
project was only going to be successful if he approached the problem from
two different directions: firstly he had to ensure adequate sealing and
secondly he had to devise a rotary system which would allow a fourstroke
cycle.

In 1951 the piston ring manufacturer Goetze expressed an interest in his
experience of sealing problems and in the same year NSU approached him about
his work on rotary disc valves. Appropriate research contracts were
concluded with both companies.  By then Wankel's investigation of the
numerous possible rotor combinations had become more systematic.  He had
completed a comprehensive analysis covering all the designs he had
investigated himself and all the proposals put forward by other inventors
which he had encountered by that time. This was a laborious process which,
like his work on gas sealing, no other engineer had until then attempted. 

At the end of 1953 Wankel started to follow a promising line of research: an
arrangement with a pointed oval rotating inside an almost circular housing
which itself was allowed to rotate. This system induced NSU's chief research
engineer Dr.  Walter Froede to recommend an extension of his company's
contract with Wankel. Dr. von Heydekampf, who was then Chairman of the
Board, gave his blessing to the new agreement. But while initial development
was still under way Wankel, who was persistently continuing his researches
because he was still not quite satisfied, came up with a new design which
was obviously even better. He combined a rotating, slightly constricted oval
chamber with a triangular-shaped rotor which fitted inside the contours of
the chamber. This was the basis configuration of today's Wankel engine,
discovered in March, 1954.  The way was open for a new line of development,
but at the same time a new decision had to be taken: should both components
rotate or should the rotor housing remain stationary? Both alternatives were
possible, each involving different consequences which will be outlined
later.

It was decided to adopt the "pure" rotary engine proposed by Wankel which
had rotor and rotor housing capable of turning at high speeds, although this
"single rotation machine" is now largely forgotten.  After laborious and
detailed development work - in which Wankel's chief designer Ernst Hoeppner
played a leading part - this novel engine came to life on February 1st,
1957.  It was first started up on a test bed in what was then the NSU
motorcycle factory and ran for a short time under its own power.  

In 1958 NSU decided to simplify the engine for use as an automobile power
unit, an application which had been envisaged right from the start.  The
rotor housing was kept stationary and now became the engine housing. The
triangular rotor turned inside this housing, although it now followed a
planetary orbit in order to form the variable-volume chambers as before.
This orbiting motion, instead of simple rotation, meant that the attainable
engine speeds were to some extent restricted. This was a limitation which
was readily accepted due to the simplification involved and because these
lower engine speeds were more compatible with conventional attachments
(power transmission, gearbox, pumps, alternator etc.). The orbiting rotor
now had to be balanced by means of additional balance weights, although
perfect balancing could be achieved, unlike the situation with a
reciprocating engine where a certain amount of dynamic imbalance cannot be
prevented.

The development of rotary combustion engines now continued on a visibly
larger scale. The first engine with a planetary rotor and a chamber volume
of 125 cc was quickly followed by engines displacing 250 ec and then 400 cc
and 500 cc. Smaller engines with a chamber volume of 60 cc were also tested
for independent drive applications.

In 1960 an automobile was driven by a rotary combustion engine for the first
time in the history of automotive technology. The car was an NSU Prinz
driven by a rear-mounted 30 HP rotary engine with a chamber volume of 250cc.
The first rotary powered production car, the NSU Spider, came off the
assembly lines in 1964; the power unit was a 50 HP rotary engine with a
chamber volume of 500 cc. 

The AirCraft Rotary Engine NewsLetter.  Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/            
Copyright 1998-2003 All world wide rights reserved.