Diffusers, Free Jets, and Engine Airflow

[Admin]

Kevin Cameron has been writing about motorcycles for nearly 50 years, first for <em>Cycle magazine</em> and, since 1992, for <em>Cycle World</em>. (Robert Martin/)

A diffuser is a device or a flow situation that causes fast-moving air to decelerate, converting what had just been its kinetic energy into pressure. No diffuser can be 100 percent efficient—some of the air’s velocity-derived kinetic energy is lost as heat. A subsonic diffuser is typically a divergent duct.

A diffuser is an integral part of any centrifugal air compressor. In such compressors, a rotating disc has radial vanes whose height decreases from the central entry to discharge at the OD. Such compressors may have a tip speed of up to 1,500 feet per second, so they accelerate air to a high velocity. The next step is to decelerate that high-speed flow so that its energy, instead of being molecular velocities mostly in a single direction (the fast outflow from the impeller), acts in all directions as pressure. The impeller is therefore surrounded by divergent passages (becoming wider) which smoothly decelerate the flow and make an efficient job of transforming kinetic energy into pressure energy.

Related: Four-Valve Beginnings

By comparison, a free jet is a very inefficient way to decelerate a high-speed gas flow. Anyone who has worked with a flow bench has tried to measure exhaust flow by blowing through a cylinder head with its intake valve(s) closed and exhaust(s) at some desired lift. The flow in this condition is very disappointing.

How a Diffuser Works

Why? The pressure in the fast-moving flow is reduced because the molecules in the flow are moving mostly in one direction, rather than in all directions as they do at rest. That leaves less energy in the flow to act as pressure. (This is what old Bernoulli was talking about in his famous principle: That moving air has less pressure than still air.)

The surrounding pressure is that of the atmosphere, nominally about 14.7 pounds per square inch at sea level. Therefore the atmosphere, acting inward on the low pressure in the flow, squeezes it down to a size smaller than the port from which it is flowing. This throttling or pinch effect reduces the flow quite a bit, typically about 30 percent. As the free jet interacts with the surrounding still air, that still air is entrained by the jet in the form of turbulence, what you hear as a modest roar.

Related: Four Valves Per Cylinder, Part 2

Now turn away from the bench while it’s still blowing and pick up a piece of paper. Roll it into a cone whose small end will just fit into the exhaust port. Stick it in. The roar is quieted and the flow increases nicely. You have changed the inefficient free jet outflow from the bare head into a smoothly decelerating and more efficient diffuser flow. As the air enters the paper diffuser it slows down in a controlled manner as the cone’s cross-section becomes larger.

It’s for this very reason that the instructions for a flow bench suggest that the user bolt an exhaust pipe to the head during exhaust flow testing: It prevents formation of a free jet at the port, and the loss of flow associated with it.

This same “pinch effect” occurs when a rocket engine designed for operation only in vacuum is operated within the earth’s atmosphere. Atmospheric pressure pinches the outflowing free jet from the engine’s thrust skirt, causing it to separate somewhat from the inside surface of the skirt, thereby reducing thrust by some percentage.

Reed Valves

Now let’s turn to something with more relevance to the pleasures of motorcycling. While continuing to play with the flow bench, I set it to blowing through a six-petal reed valve from a two-stroke engine. This was disappointing; the valve petals hardly opened. OK, this was to be expected because the atmosphere in the room was pushing in on the petals, while the flow emerging from under them formed a low-pressure free jet.

Now I decided to form a crude diffuser by bringing in two short pieces of 3-inch round-section aluminum bar so that the free jet from the reed assembly passed between them. This caused the reeds to open visibly more and the flow to increase by 20–30 percent. The curious thing was that if I took away one of the rounds, creating what we might call “half a diffuser,” I got half as much effect. The free jet from the reed was attaching itself to the curved surface, which was giving it some protection from atmospheric pinch effect.

Related: Four Valves Per Cylinder, Part 3

After that I thought about the high-speed flow coming out from under a four-stroke engine’s poppet intake valve. Is that, I wondered, a free jet? When I sketched a classic hemi chamber with a single intake valve in it, at some lift I found the explanation of something I had known but had never understood: Why is the flow coefficient of a single intake in a hemi combustion chamber substantially higher than that of the two intake valves in a pent-roof chamber? (The flow coefficient is not a measure of flow but rather of how well a given geometry makes use of the flow area it has.)

Hemis and Airflow

Because of the bowl-like hemi chamber’s curving surface, it too forms half a diffuser, so that the flow from under the valve is partly protected from pinch effect. This is what gives the single intake valve in a hemi chamber its excellent flow coefficient when measuring its flow in cubic feet of air per minute per square inch of valve-head area.

Sketch out one of the two intakes in a four-valve pent-roof head and it’s clear that, with this geometry, it’s harder to cause the outflow from the valve to attach to the flat rather than curved surface of the pent-roof combustion chamber.

Related: What Does It Take To Achieve Fast, Efficient Combustion?

And just maybe that’s why Sadao Shirasagi’s development group at Suzuki surrounded each intake valve of their 1978 TSCC four-valve head with its own miniature bowl. Could part of its purpose be to give the intake flow a curved surface to which it can attach?

History provides a caution. I can hear the words of Hall of Fame tuner Rob Muzzy from our 1982 interview: “You can lose your mind on the flow bench. I know people who have.”

View the full article