|Will the Engine Really Rev?
The idea of an oval-piston engine was not without a few doubts, of course. Could it achieve the calculated intake efficiency under actual running conditions? What about friction and the sealing of pistons? Could the engine be cooled effectively so that the pistons would not be deformed by high temperatures? Several problems needed to be solved, but no one was more aware of that than the development staff. In their desire to bring something new to the racing world, the voices of concern simply did not resonate with them.
"We didn't think much about whether the engine would actually turn over," Yoshimura recalled, "or even whether it would be practical at all. We weren't worried about those things, since we just wanted to make it work."
Testing with a two-valve, single-cylinder engine indeed confirmed that the oval piston would rev. They gradually increased the number of valves, in time reaching the eight-valve, single-cylinder. However, by this time the development staff had already gone through numerous problems.
The phenomenon of sudden disintegration was a significant obstacle, typically arising when the engine speed exceeded 10,000 rpm. The cause for this was the twisting of connecting rods. Unlike a regular piston, an oval piston has two rods. The rods would distort as the engine speed increased, pulling the piston pin out of its proper orientation and causing the parts to break. To solve the problem, not only would the design specifications have to be modified, but the machining accuracy would have to be improved. Accordingly, the development team worked with the staff at Honda Engineering (EG) to try various approaches.
The piston ring, also, was a real headache, since the oval shape was so difficult to machine accurately. Experiments were repeated over and over, through a process of trial and error. For example, they tested a split-type ring made of two parts and formed the ring into a "walking stick" shape. But after every conceivable alternative had been tried, they found themselves back at the starting point with a conventional, self-stretching piston ring. Even with that configuration they still had to expend extra effort calculating the appropriate dimensions, so that the ring would remain in free form during machining, but produce constant bearing pressure once installed. NC machines were still inaccurate during that period. Therefore, additional effort was required to produce the desired quality of a part by accurately reflecting the specified machining dimensions.
However, through persistent effort the team identified solutions to these problems, one by one. The completion of the piston ring, in particular, was a considerable boost to the feasibility of the overall design. With that, the test target moved from the single-cylinder design to four cylinders.
|<< previous||3 of 6||next >>|