Evolution of Yamaha Two-Stroke Cylinders
Fins!
by Kevin Cameron
Thursday, October 1, 2026
When in 1954 Yamaha had to cast large numbers of finned cylinders and heads for its YA-1 motorcycle, it found conventional sand casting too complex.
During WW II literally millions of such finned cylinder heads for air-cooled radial aircraft engines were cast. For each such cast head, roughly 1200 core pins had to be manually positioned to support the thin sand cores (to become the fin spaces) in place. Yamaha bypassed such difficulties by adopting instead the sand-resin shell molding process, invented by Johannes Croning in 1944. It offered higher dimensional accuracy and smoother surfaces, plus in later versions the possibility of automation.
The sand-resin combination is applied to a heated half-mold (right or left) of the cylinder, on which it cures. The two halves are assembled and, with suitable cores for cylinder bore and the various ports, the assembly can be filled with molten aluminum alloy.
Yamaha’s big discovery in those early days was the hot hardness and strength of hypereutectic alloys of aluminum and silicon. Using such materials for cylinders, heads, and pistons, they were able by 1963 to see their RD-56 disc-valve twin motor away from the Honda four-stroke fours at the fast Spa circuit in Belgium. Honda immediately replied (Aug 1964) with their legendary six but it would take design updating by the famed engineer Irimajiri plus rider Mike Hailwood to put a stop to Phil Read’s 1964-’65 250 title streak on the Yam.
Air cooling has a definite limit – the onset of detonation. Deto is an abnormal form of combustion, occurring long after the spark, as the flame front draws close to the cylinder wall. The heating that the last bits of fresh mixture have received is the trigger – once that hot, compressed “end-gas” reaches something over 900 degrees Fahrenheit it can auto-ignite and burn at supersonic speed. The sound as the shock front hits metal surfaces is heard as ‘knock’ or ‘tinkle.’
Combustion can be seen as a ‘race’ between normal combustion, and the temperature reached by the end-gas. Ideally, combustion reaches the cylinder wall first and no detonation occurs. But the hotter the metal in contact with the unburned mixture, the greater the chance that detonation will win. Detonation damage to pistons occurs at the edges of the crown, for that is where the above-described end-gas is located. The damage looks like some piston metal has been sand-blasted away.
The question is, how much power can we extract from an air-cooled cylinder before its temperature rises far enough to initiate detonation? It is no accident that the first two-stroke motorcycle to win an FIM road race championship was Suzuki’s RM-62 50 cc. (bore and stroke of 40 X 39.5 mm). Why? Because as cylinder bore is increased, the surface area of a flat-topped or slightly domed piston increases as the square of the bore. Therefore the smaller the piston, the smaller the area of piston exposed to combustion, and the easier it is to cool it by simple contact with the cooler cylinder wall.
Searching my shop, I was able to find examples of three of Yamaha’s air-cooled production road race cylinders for the following 250 cc twins;
– TD1-B (1964-’66),
- TD2 (1969-’71)
- TD3 (1972-’73).
“Production” in this case meant based upon a stock crankcase.
All three have nine cooling fins, top-to-bottom. TD1-B, making 35 hp, had fins of constant thickness, root to tip., at 8.9 mm pitch (0,35 inch).TD2, a revolutionary design with four transfer ports instead of two, made 44+ hp, but its fins were now tapered in thickness – thickest at the cylinder, thinnest at the tips, still at 8.9 mm pitch. Weight increased from the B-model’s 1500 gm to TD2’s 1635. As the fins all have the same area, the increased weight is associated with the tapered fins.
In TD3, bore and stroke became 54 X 54 mm instead of the previous 56 X 50, making the cylinder 6 mm taller. The extra height was used to increase the spacing between fins to 10 mm, providing a larger flow path. The TD3 cylinder weighs 1900 gm and total power was 50 hp. Its fins were now rectangular and of much increased area. In addition, fin design returned to constant thickness rather than tapered. Having larger spaces between fins improved cooling enough to deal with the waste heat of 43% more power.
Mr. Kazuhito Yoshida, Kawasaki team road race technician in the 1970s, showed me two coolant radiators of the same gross dimensions (this was in 1976, when Kawasaki was running the water-cooled KR250 and KR750). One radiator had dense secondary surface foils at fine pitch, while the other had wider spaces between its foils. Indicating the rad with wide spacing, he said, “This one, performance better, because more air can flow through.”
Why? Because moving air molecules close to the hot surfaces lose energy from colliding with them, forming a low-energy boundary layer that can be thick enough with close spacing to impede airflow between fins.
To see another design concerned with boundary layer thickness, look at the under-cockpit radiator intake of the P-51 Mustang aircraft. Notice that there is a 1.5” gap between the top of the radiator scoop and the underside of the fuselage. A major purpose of this was to prevent the constantly thickening fuselage boundary layer –low-energy air - from entering the scoop and not doing its full share of pushing air through the rad core.
It took aviation a long time to understand that finned engine cylinders, just sticking out into the airflow cooled badly because it was easier for the air to flow around them than to flow through their fin spaces. It must have seemed “intuitively” that on an airplane flying at 200 mph, the air would somehow sail through the fin spaces at that same speed. Not so.
Suzuki discovered this in a different way. When they needed improved cooling for one of their early cylinders, they added extra fins at closer spacing. Instead of more cooling effect, they got less. Build-up of stagnant boundary layers in narrow fin spaces reduces the flow-through of cooling air.
Worst of all were the integral machined fins of the rotary engines that powered most WW I fighter aircraft. As the crankcase with its cylinders whirled around while its crankshaft was bolted to the firewall, the leading faces of the cylinders were cooled relatively well, but the negligible airflow reaching the aft sides of the cylinders left them very hot.
The result was cylinders too out-of-round to be sealed by conventional piston rings.
It was gradually realized that high-power aircraft cylinders could not be cooled unless cylinder and head fin space was made the only flow path through the engine nacelle. In this way, considerable ram pressure (around 12 inches of water, would pressurize the air in front of the cylinders. Even so, a safe cylinder head temperature (CHT) could not be maintained in climb unless airspeed was high enough to push the necessary airflow through fin space.
Some air-cooled aircraft engines had to be fan-cooled to give satisfactory performance at altitude. One was Germany’s WW II FW190 fighter and another was the “Big Stick” B-36 bomber.
Because motorcycles do not cruise at aircraft speeds they cannot develop the ram pressure necessary to force cooling air through the more closely-spaced fins of aircraft engine cylinders (0.132”). Their fins must therefore be spaced more widely – such as the 10 mm fin pitch (0.394 inch) of the TD3 cylinder.
For racing motorcycles, when power per square inch of piston area reached a certain level it soon became more effective to adopt liquid cooling (water being 800 times denser than air) than to continue to struggle with cylinder distortion, high piston and cylinder temperature, detonating combustion, and lubricant breakdown.
On the other hand, air-cooling has worked satisfactorily for great numbers of classic production motorcycle designs.
MORE FROM KEVIN CAMERON ON SOUPTASTIC
Marc Marquez Keeps Winning With The Other Two Things
Kevin Cameron on the less obvious ingredients behind Marquez's continued success.
Belly Button Girl
A Kevin Cameron story that could probably only be called Belly Button Girl.
Fillets: The Shape Of Strength
KC explains why the humble fillet matters when engineers need strength without unnecessary weight.
Ask KC: Taking Their Own Sweet Time
Kevin answers another reader question in the particular way only Kevin can.
The Stone Tablets Of Muzzy
Kevin Cameron on Rob Muzzy, horsepower and some hard-earned rules about making motorcycles go fast.
Why Suzuki's TL1000R Never Became Ducati
KC looks at Suzuki's big V-twin Superbike and why it never became the Ducati-beater it was intended to be.
DFA
The sand-resin combination is applied to a heated half-mold (right or left) of the cylinder, on which it cures. The two halves are assembled and, with suitable cores for cylinder bore and the various ports, the assembly can be filled with molten aluminum alloy.
Yamaha’s big discovery in those early days was the hot hardness and strength of hypereutectic alloys of aluminum and silicon. Using such materials for cylinders, heads, and pistons, they were able by 1963 to see their RD-56 disc-valve twin motor away from the Honda four-stroke fours at the fast Spa circuit in Belgium. Honda immediately replied (Aug 1964) with their legendary six but it would take design updating by the famed engineer Irimajiri plus rider Mike Hailwood to put a stop to Phil Read’s 1964-’65 250 title streak on the Yam.
Air cooling has a definite limit – the onset of detonation. Deto is an abnormal form of combustion, occurring long after the spark, as the flame front draws close to the cylinder wall. The heating that the last bits of fresh mixture have received is the trigger – once that hot, compressed “end-gas” reaches something over 900 degrees Fahrenheit it can auto-ignite and burn at supersonic speed. The sound as the shock front hits metal surfaces is heard as ‘knock’ or ‘tinkle.’
DFA
Combustion can be seen as a ‘race’ between normal combustion, and the temperature reached by the end-gas. Ideally, combustion reaches the cylinder wall first and no detonation occurs. But the hotter the metal in contact with the unburned mixture, the greater the chance that detonation will win. Detonation damage to pistons occurs at the edges of the crown, for that is where the above-described end-gas is located. The damage looks like some piston metal has been sand-blasted away.
The question is, how much power can we extract from an air-cooled cylinder before its temperature rises far enough to initiate detonation? It is no accident that the first two-stroke motorcycle to win an FIM road race championship was Suzuki’s RM-62 50 cc. (bore and stroke of 40 X 39.5 mm). Why? Because as cylinder bore is increased, the surface area of a flat-topped or slightly domed piston increases as the square of the bore. Therefore the smaller the piston, the smaller the area of piston exposed to combustion, and the easier it is to cool it by simple contact with the cooler cylinder wall.
DFA
Searching my shop, I was able to find examples of three of Yamaha’s air-cooled production road race cylinders for the following 250 cc twins;
– TD1-B (1964-’66),
- TD2 (1969-’71)
- TD3 (1972-’73).
“Production” in this case meant based upon a stock crankcase.
All three have nine cooling fins, top-to-bottom. TD1-B, making 35 hp, had fins of constant thickness, root to tip., at 8.9 mm pitch (0,35 inch).TD2, a revolutionary design with four transfer ports instead of two, made 44+ hp, but its fins were now tapered in thickness – thickest at the cylinder, thinnest at the tips, still at 8.9 mm pitch. Weight increased from the B-model’s 1500 gm to TD2’s 1635. As the fins all have the same area, the increased weight is associated with the tapered fins.
DFA
In TD3, bore and stroke became 54 X 54 mm instead of the previous 56 X 50, making the cylinder 6 mm taller. The extra height was used to increase the spacing between fins to 10 mm, providing a larger flow path. The TD3 cylinder weighs 1900 gm and total power was 50 hp. Its fins were now rectangular and of much increased area. In addition, fin design returned to constant thickness rather than tapered. Having larger spaces between fins improved cooling enough to deal with the waste heat of 43% more power.
Mr. Kazuhito Yoshida, Kawasaki team road race technician in the 1970s, showed me two coolant radiators of the same gross dimensions (this was in 1976, when Kawasaki was running the water-cooled KR250 and KR750). One radiator had dense secondary surface foils at fine pitch, while the other had wider spaces between its foils. Indicating the rad with wide spacing, he said, “This one, performance better, because more air can flow through.”
Why? Because moving air molecules close to the hot surfaces lose energy from colliding with them, forming a low-energy boundary layer that can be thick enough with close spacing to impede airflow between fins.
To see another design concerned with boundary layer thickness, look at the under-cockpit radiator intake of the P-51 Mustang aircraft. Notice that there is a 1.5” gap between the top of the radiator scoop and the underside of the fuselage. A major purpose of this was to prevent the constantly thickening fuselage boundary layer –low-energy air - from entering the scoop and not doing its full share of pushing air through the rad core.
It took aviation a long time to understand that finned engine cylinders, just sticking out into the airflow cooled badly because it was easier for the air to flow around them than to flow through their fin spaces. It must have seemed “intuitively” that on an airplane flying at 200 mph, the air would somehow sail through the fin spaces at that same speed. Not so.
Suzuki discovered this in a different way. When they needed improved cooling for one of their early cylinders, they added extra fins at closer spacing. Instead of more cooling effect, they got less. Build-up of stagnant boundary layers in narrow fin spaces reduces the flow-through of cooling air.
Worst of all were the integral machined fins of the rotary engines that powered most WW I fighter aircraft. As the crankcase with its cylinders whirled around while its crankshaft was bolted to the firewall, the leading faces of the cylinders were cooled relatively well, but the negligible airflow reaching the aft sides of the cylinders left them very hot.
The result was cylinders too out-of-round to be sealed by conventional piston rings.
It was gradually realized that high-power aircraft cylinders could not be cooled unless cylinder and head fin space was made the only flow path through the engine nacelle. In this way, considerable ram pressure (around 12 inches of water, would pressurize the air in front of the cylinders. Even so, a safe cylinder head temperature (CHT) could not be maintained in climb unless airspeed was high enough to push the necessary airflow through fin space.
Some air-cooled aircraft engines had to be fan-cooled to give satisfactory performance at altitude. One was Germany’s WW II FW190 fighter and another was the “Big Stick” B-36 bomber.
Because motorcycles do not cruise at aircraft speeds they cannot develop the ram pressure necessary to force cooling air through the more closely-spaced fins of aircraft engine cylinders (0.132”). Their fins must therefore be spaced more widely – such as the 10 mm fin pitch (0.394 inch) of the TD3 cylinder.
For racing motorcycles, when power per square inch of piston area reached a certain level it soon became more effective to adopt liquid cooling (water being 800 times denser than air) than to continue to struggle with cylinder distortion, high piston and cylinder temperature, detonating combustion, and lubricant breakdown.
On the other hand, air-cooling has worked satisfactorily for great numbers of classic production motorcycle designs.
DFA
MORE FROM KEVIN CAMERON ON SOUPTASTIC
Marc Marquez Keeps Winning With The Other Two Things
Kevin Cameron on the less obvious ingredients behind Marquez's continued success.
Belly Button Girl
A Kevin Cameron story that could probably only be called Belly Button Girl.
Fillets: The Shape Of Strength
KC explains why the humble fillet matters when engineers need strength without unnecessary weight.
Ask KC: Taking Their Own Sweet Time
Kevin answers another reader question in the particular way only Kevin can.
The Stone Tablets Of Muzzy
Kevin Cameron on Rob Muzzy, horsepower and some hard-earned rules about making motorcycles go fast.
Why Suzuki's TL1000R Never Became Ducati
KC looks at Suzuki's big V-twin Superbike and why it never became the Ducati-beater it was intended to be.
— ends —
