- Joined
- Nov 26, 2009
- Messages
- 3,081
Snotzo - the smooth ramp cams are now in use in several racebikes around the world. I know of one short stroke 500 that revvs 9000RPM no problem. There is a larger displacement motor which does not rev as high but puts out 85HP on dyno.
Mtd-son The JA software is not cheap at $2500 (30 day trial version is available). In my talks with John Andrews he mentioned entering valve train simulation data and how use it but I did not pursue it for reasons I’ll explain below.
There is an option to check the constant jerk box and that will change the ramps dramatically. There is also “uniformly accelerated ramps”. But I chose neither for reasons I’ll explain.
The valve train simulation does not take cam flex into account because this is not usually a problem – but its a problem with Nortons and causes valve bounce which can tangle the valves. When the descending valve comes to a sudden stop, all that inertia is exerted against the cam and it flexes – then when it rebounds it pushes the exhaust valve open and if its too far it will crash against the intake valve. That is why we need a smoother more gradual ramp for closing the valves.
Opening the valves also strains everything but it never seems to break anything. The oscillation of the valve spring is dampened out somewhat when the spring nears coil bind at full lift – but you have a lot of bouncing spring when the valve has to stop and close on its seat. So we can have a more abrupt opening ramp than closing and you see this in modern cams.
The math get a little thick. So I check and verify everything visually by looking at the actual cam curves in Autocad where you can zoom in and compare them. When you manipulate a cam profile to match an existing Norton race cam – you can read and compare all the data. Then you can change it further and see what it does to both the shape and the data. After viewing and working with hundreds of images you get an idea of what’s going on.
In the image below you see several cam ramp profiles. I’m comparing to a sample PW3 cam (suppliers vary) because the PW3 has harsh ramps compared to other Norton race cams. See the alternate ramps. I rejected the constant jerk and uniform accelleration opening ramps as too harsh (uniform acceleration not shown). And I want a smoother opening ramp than the PW3. The closing ramp I’m using will cause less valve bounce than any of the original Norton race cams – but I’m keeping it within reason. Nortons don’t have to rev to 9000 RPM unless they are extremely short stroke and thats a special case anyway. You can see that the black an dark blue ramps are the smoothest.
Mtd-son The JA software is not cheap at $2500 (30 day trial version is available). In my talks with John Andrews he mentioned entering valve train simulation data and how use it but I did not pursue it for reasons I’ll explain below.
There is an option to check the constant jerk box and that will change the ramps dramatically. There is also “uniformly accelerated ramps”. But I chose neither for reasons I’ll explain.
The valve train simulation does not take cam flex into account because this is not usually a problem – but its a problem with Nortons and causes valve bounce which can tangle the valves. When the descending valve comes to a sudden stop, all that inertia is exerted against the cam and it flexes – then when it rebounds it pushes the exhaust valve open and if its too far it will crash against the intake valve. That is why we need a smoother more gradual ramp for closing the valves.
Opening the valves also strains everything but it never seems to break anything. The oscillation of the valve spring is dampened out somewhat when the spring nears coil bind at full lift – but you have a lot of bouncing spring when the valve has to stop and close on its seat. So we can have a more abrupt opening ramp than closing and you see this in modern cams.
The math get a little thick. So I check and verify everything visually by looking at the actual cam curves in Autocad where you can zoom in and compare them. When you manipulate a cam profile to match an existing Norton race cam – you can read and compare all the data. Then you can change it further and see what it does to both the shape and the data. After viewing and working with hundreds of images you get an idea of what’s going on.
In the image below you see several cam ramp profiles. I’m comparing to a sample PW3 cam (suppliers vary) because the PW3 has harsh ramps compared to other Norton race cams. See the alternate ramps. I rejected the constant jerk and uniform accelleration opening ramps as too harsh (uniform acceleration not shown). And I want a smoother opening ramp than the PW3. The closing ramp I’m using will cause less valve bounce than any of the original Norton race cams – but I’m keeping it within reason. Nortons don’t have to rev to 9000 RPM unless they are extremely short stroke and thats a special case anyway. You can see that the black an dark blue ramps are the smoothest.
Last edited:
