The combustion space has long been a target for investigation, design and improvement in internal combustion engines. In particular, the "high turbulence" approach pursued by Sir Harry Ricardo in the early 1900s has morphed through a variety of concepts that, along the way, was sparked by other efforts that include Swiss engineer Michael May and a young man I chanced to meet during the 1970s who literally put his own spin on improving power by enhanced air and air/fuel charge motion. He broke onto the scene by introducing drag and circle track racers and engine builders to his methods under the name of Endyn. His name is Larry Widmer, and he continues his work today.

If I recall, you began your career in motorsports by building and racing go-karts. Was it then that you started looking at "unconventional" ways to build engines?

Actually, I did start out in the go-kart community. In fact, early on I decided to look for ways to treat a package of engine parts as a system that requires integration. I'll come back to this a bit later, but I strongly believe engine builders should consider the impact of how each component must work in conjunction with others. But it was in karting that I got my start.

It was in drag racing where you made a discovery that could easily be applied to circle track racing. Will you expand on that notion?

After spending time on an engine dyno trying to optimize power and finding out there was more to it than this, I realized that these engines only spend a short time in the upper rpm ranges. Most importantly, they must be able to accelerate through a range of rpm. With the possible exception of a superspeedway engine that operates in a comparatively narrow span of rpm, an engine must be built to move quickly through whatever range of engine speed is being used, under load. Stated another way, it must not only make torque, but do so as quickly as possible.

What was your initial approach to creating or helping to establish this ability?

At the time, since I was a Ford proponent and building 351 Cleveland engines, I discovered that Ford appeared to have oversized its cylinder head port areas, resulting in lost volumetric efficiency and the ability to produce what I'll call good ‘transient torque.' You probably remember I approached Edelbrock to do some intake manifold port core filing to reduce manifold section areas while I was putting inserts in the port floors of the cylinder heads. We didn't see any appreciable reductions in peak power but the engine began accelerating through the gears much quicker, so I figured we were on to something.

Space doesn't allow us to chronicle all the steps and discoveries you made for the next few years, but you wound up doing some work for Penske and his Ford NASCAR program back then. However, shortly thereafter you were diagnosed with terminal cancer, correct?

Well, everybody else seemed to think it was terminal but me. I entered chemotherapy while concurrently continuing on my career path, not grinding on heads, but researching where I could go next in improving combustion. I'd already discovered the value of studying engine technologies from different types of engines in different forms of racing. It was also apparent that optimizing the performance of any one component without regard about how it might affect another part was not good, in the sense major components involved in the combustion process (heads, pistons, and intake manifolds) need to work in conjunction with each other before, during, and after the burn.

May we dwell on that for a bit, if you don't mind?

OK, here's an example. I researched the concept employed by Michael May, an engine engineer from Switzerland. He advocated directing as much of the air/fuel charge as possible toward the exhaust side of the combustion space, compressing it with a high ratio, and aiming the spark plug toward the exhaust valve. As I got into this, based on some prior work I had done with an ultrasonic fuel atomization device and system, I realized that once you move the air/fuel ratio well-past stoichiometric, combustion temperatures begin to become cooler. As a result, I discovered it was possible to run much higher compression ratios, the burn rate was accelerated and less spark timing was required to optimize power. Increasing spark timing creates negative power, so this was clearly beneficial. Further evidence of the faster burn was a corresponding reduction in exhaust temperatures, since more heat had been liberated in the form of work on the pistons, before the exhaust valve opened. I chose to call this a ‘biased burn' process since we were favoring combustion more localized toward the exhaust valve.

From some of our prior discussions, it was somewhere in this time frame that you began exploring specific air/fuel charge mixture motion, was it not?

Yes, and since there's not much space here for a detailed explanation, I'll focus on the more important aspects of what we did, because this is when the concept of swirl entered our work. If you'll visualize the air/fuel charge path resembling the threads on a screw with its axis coinciding with the axis of the cylinder, this was the motion we discovered particularly beneficial. A major revelation here was that the air/fuel charge ratio varied from the top of the cylinder to the bottom at BDC on the intake stroke. In fact, it became progressively leaner moving from TDC to BDC during induction. By enabling this condition, biasing the combustion toward the exhaust valve, increasing the compression ratio for a quicker burn and reducing spark timing, the power gains were remarkable…and the engine could run through its transient torque range even more quickly.

Before we button this up, let's return to the concept of making parts work together. What got you into this?

When I built my first flow bench, I did so in a way that allowed me to pull air out of the test cylinder, through the exhaust port. It also let me install a piston in the cylinder that I could position relative to intake and exhaust valve openings. Bottom-line, all this allowed me to study airflow during valve overlap periods, and it was here that we really began to see how intake and exhaust ports needed to flow in only one direction, and the significant role piston crown shapes play in overlap flow efficiency. Integral to all this was the realization that we needed to do valve jobs that did the best job possible of not flowing backward during reversion periods. As a result, while recognizing the inherent disadvantages of a dry, steady-state flow test environment, we were able to see how slight changes to an intake port, exhaust port, valve job, or valve configuration could affect the performance and requirements of the other components. As I previously said, we treat all these parts as an integrated system.

Your cancer is still in remission?

Well, my first chemo was during 1979-1981. It went into remission but returned for two years in 1987-1988 and has remained in remission since. I'm convinced a good attitude and appreciation for life has been a big part of why I'm still around today, and working as hard as ever.

While Larry's approach to the combustion process may not align itself with your particular view of the subject, he has clearly explored the essence of internal engine combustion, while making notable progress in his chosen field of work. Admittedly, he managed to stir some controversy when he came onto the radar some 30-plus years ago, not so much for what he was doing, but for how some chose to view him--a reaction often experienced by forward-thinking persons. The fact of the matter is he's still advancing certain notions involved in the combustion process, and we're hopeful you gained some benefit from what he was willing to share.