This is pretty cool, are you developing some software to calculate the lobe shape based on valve motion and follower configuration? I built some software a while back to develop valve motion, but never finished up the part that develops the lobe and simulates the follower system.
Good thoughts about the camshaft grinder! But I have a few points/hints. For the cross slide (and thus the spindle RPM): You have the acceleration of the cam at a given engine RPM. You can use that to calculate the maximum spindle RPM. You can use 1 m/s^2 as maximum acceleration of the cross slide. Thus, say cam acceleration at 3000 RMP (cam) is 6000 m/s^2 makes 0.5 RPM. The max. acceleration of the cross slide with 1 m/s^2 is conservative. But you certainly will have problems at 10 m/s^2. Also note, that there are some geometric distortions and thus the cam accel. is different to the wheel accel. Depends on the rocker arm or whatever your engine uses. F = m * a. Force (along the cross slide) is mass (of the wheel and its drive) times acceleration. Force is in Newton, mass in kg. 10 m/s^2 is about earth's gravity. You need a steady rest. You will have problems to get the cam running true near the center. 0.02 mm of runout is about what is allowed. And this already distorts the cam profile you grind. Not so much for the lift (who cares for 0.02 mm more or less lift), but for the acceleration and even more for the jerk. Keep that in mind. You need coolant. You need coolant as soon as you intend to take off more than a tad. The cam will heat up and expand and take off more, generate more heat, expand more and so on. Also the wheel will clog up and generate even more heat and pressure and push the cam shaft out of center (-> steady rest). This expansion will happen after a few seconds and then get disastrous within a couple of seconds. This is not a tool and cutter grinder where you come off without cooland when you pay attention. I hope you get the math right. The wheel's diameter plays a role as this influences the hight of the contact point that is also influenced by the shape of the lobe. And most of all, that is not what your rocker arm or flat tappet does move along. There are at least 2 transformations. Not wanting to talk you out of your project!
Thank you for taking the time for such a comprehensive reply. In your first point your are looking at the dynamic values in terms of time, that is fine for overall valve train dynamics but when looking at cam manufacture it is more usual to consider those in terms of cam degrees. So acceleration will have units of m/ deg^2 etc. Regarding the steady rest, for long camshafts that is completely true but in an earlier video I pointed out that I am only interested in very short single cylinder camshafts. I only used a long cam shaft in this video for illustrative purposes. Yes I have the maths right, don't worry.
@@MotoChassisByTonyFoale Cam degrees don't help you to calculate the necessary acceleration of the cross slide at a given spindle RPM. Unless you convertit it to m/s^2.. Maybe you misunderstood my explanation.
@@MotoChassisByTonyFoale That the spindle determines the position of the cross slide is nothing new. And it is a fact that the possible acceleration of the cross slide determines the maximum speed of the spindle. The spindle is the leading axis, so it sets the requirements of following axis. And vice versa. So what acceleration of the cross slide do you have?
Very interesting, there are quite a few hidden complexities which you have raised. Like the imprinting of the patten into the grind by the motor using vee or tooth belt. I would not have thought of that. Do you have any good links to people wanting to start looking at adding power feeds (via motors) to small hobby lathes and mills. Starting off very simple mind you before then looking at computer control.
I wrote a reply but it seems to have disappeared. There are a ton of people wanting to do as ask you ask. I do not have any specific links but there will be several on the Home Made Tools forum. www.homemadetools.net/forum/
I have written some synthesis software to calculate valve motion and the it works back from there to calculate the cam profile for any type follower. The inputs to the software are max values of jerk, acceleration, velocity, and displacement and of course duration. So the calcs are tucked away in the software. Several people have asked about the software and I will probably make that available in the future.
I would not get too carried away with the jerk component, if you become obsessed with this, the lobe will hardly do any lifting ! Finger followers are quite a different case, and require a lobe that is NOT asymmetrical, what type of valve actuation are you designing your lobe for :: SV ,OHV, push rod, OHC ?? @@MotoChassisByTonyFoale
@@bobawatsit With careful design you can lower jerk without affecting the profile by much. Jerk is an important parameter for valve train dynamics and it is foolish to ignore it. There is a difference between considering jerk and being obsessed by it. Initially the lobes will be designed for flat tappets which unfortunately are too small in diameter so later that will probably change to roller followers.
" the cam lobe does not know what JERK is " quote by > Keith Duckworth, Harvey Crane went a couple of iteration steps further with ' SNAP ' and ' CRACKLE ' The amount of change of acceleration / angle is very dependent on the valve actuation system and its ability to either absorb or resist stresses @@MotoChassisByTonyFoale
Just an add on, when you say the f/t are too small because why ? the diameter will only affect the amount of velocity achievable, swapping out to roller followers are heavy and require a totally different profile lobe, and engine design. Exactly what type of engine design are you wanting to achieve ? @@MotoChassisByTonyFoale
There is a wide variety of materials used with differing hardening methods to go with them. Cast-iron is a popular choice for OEMs and has some favourable properties. Otherwise a range of hardenable steels. Hardening methods can be case hardening, nitriding and hard facing etc.
@@MotoChassisByTonyFoale The reason I questioned is, you'll be removing case hardened material when grinding do you intend to re harden? I am of course assuming you are regrinding an original camshaft, and not creating from stock
@@Phantom-mk4kp Your assumption is wrong, I will make the camshafts from scratch. During initial testing I will use some old camshafts to reduce the grinding time. Those will only be used for dimensional checks and will not be hardened nor used in an engine.
Thanks for showing that Tony. That was fun to watch. I bet you're having fun building it too.
Nice job. A real logical & professional approach.
Fun is guaranteed.
Always nice to build your own machines.
I'm really enjoying watching the process.
Good to know.
This is pretty cool, are you developing some software to calculate the lobe shape based on valve motion and follower configuration? I built some software a while back to develop valve motion, but never finished up the part that develops the lobe and simulates the follower system.
Good thoughts about the camshaft grinder! But I have a few points/hints.
For the cross slide (and thus the spindle RPM):
You have the acceleration of the cam at a given engine RPM. You can use that to calculate the maximum spindle RPM. You can use 1 m/s^2 as maximum acceleration of the cross slide. Thus, say cam acceleration at 3000 RMP (cam) is 6000 m/s^2 makes 0.5 RPM. The max. acceleration of the cross slide with 1 m/s^2 is conservative. But you certainly will have problems at 10 m/s^2. Also note, that there are some geometric distortions and thus the cam accel. is different to the wheel accel. Depends on the rocker arm or whatever your engine uses.
F = m * a. Force (along the cross slide) is mass (of the wheel and its drive) times acceleration. Force is in Newton, mass in kg. 10 m/s^2 is about earth's gravity.
You need a steady rest. You will have problems to get the cam running true near the center. 0.02 mm of runout is about what is allowed. And this already distorts the cam profile you grind. Not so much for the lift (who cares for 0.02 mm more or less lift), but for the acceleration and even more for the jerk. Keep that in mind.
You need coolant.
You need coolant as soon as you intend to take off more than a tad. The cam will heat up and expand and take off more, generate more heat, expand more and so on. Also the wheel will clog up and generate even more heat and pressure and push the cam shaft out of center (-> steady rest). This expansion will happen after a few seconds and then get disastrous within a couple of seconds. This is not a tool and cutter grinder where you come off without cooland when you pay attention.
I hope you get the math right. The wheel's diameter plays a role as this influences the hight of the contact point that is also influenced by the shape of the lobe. And most of all, that is not what your rocker arm or flat tappet does move along. There are at least 2 transformations.
Not wanting to talk you out of your project!
Thank you for taking the time for such a comprehensive reply.
In your first point your are looking at the dynamic values in terms of time, that is fine for overall valve train dynamics but when looking at cam manufacture it is more usual to consider those in terms of cam degrees. So acceleration will have units of m/ deg^2 etc.
Regarding the steady rest, for long camshafts that is completely true but in an earlier video I pointed out that I am only interested in very short single cylinder camshafts. I only used a long cam shaft in this video for illustrative purposes.
Yes I have the maths right, don't worry.
@@MotoChassisByTonyFoale Cam degrees don't help you to calculate the necessary acceleration of the cross slide at a given spindle RPM. Unless you convertit it to m/s^2.. Maybe you misunderstood my explanation.
@@MuellerNick In fact it is the angle of the spindle that determines the position of the cross slide.
@@MotoChassisByTonyFoale That the spindle determines the position of the cross slide is nothing new.
And it is a fact that the possible acceleration of the cross slide determines the maximum speed of the spindle. The spindle is the leading axis, so it sets the requirements of following axis. And vice versa.
So what acceleration of the cross slide do you have?
Very interesting, there are quite a few hidden complexities which you have raised. Like the imprinting of the patten into the grind by the motor using vee or tooth belt. I would not have thought of that. Do you have any good links to people wanting to start looking at adding power feeds (via motors) to small hobby lathes and mills. Starting off very simple mind you before then looking at computer control.
I wrote a reply but it seems to have disappeared. There are a ton of people wanting to do as ask you ask. I do not have any specific links but there will be several on the Home Made Tools forum. www.homemadetools.net/forum/
hope you will share your calcs on lobe profile design
I have written some synthesis software to calculate valve motion and the it works back from there to calculate the cam profile for any type follower. The inputs to the software are max values of jerk, acceleration, velocity, and displacement and of course duration. So the calcs are tucked away in the software. Several people have asked about the software and I will probably make that available in the future.
I would not get too carried away with the jerk component, if you become obsessed with this, the lobe will hardly do any lifting ! Finger followers are quite a different case, and require a lobe that is NOT asymmetrical, what type of valve actuation are you designing your lobe for :: SV ,OHV, push rod, OHC ?? @@MotoChassisByTonyFoale
@@bobawatsit With careful design you can lower jerk without affecting the profile by much. Jerk is an important parameter for valve train dynamics and it is foolish to ignore it. There is a difference between considering jerk and being obsessed by it. Initially the lobes will be designed for flat tappets which unfortunately are too small in diameter so later that will probably change to roller followers.
" the cam lobe does not know what JERK is " quote by > Keith Duckworth, Harvey Crane went a couple of iteration steps further with ' SNAP ' and ' CRACKLE ' The amount of change of acceleration / angle is very dependent on the valve actuation system and its ability to either absorb or resist stresses @@MotoChassisByTonyFoale
Just an add on, when you say the f/t are too small because why ? the diameter will only affect the amount of velocity achievable, swapping out to roller followers are heavy and require a totally different profile lobe, and engine design. Exactly what type of engine design are you wanting to achieve ? @@MotoChassisByTonyFoale
where do you find the time???
I work fast.
Are most camshafts case hardened? I don't know. If so how are you going to deal with that
There is a wide variety of materials used with differing hardening methods to go with them. Cast-iron is a popular choice for OEMs and has some favourable properties. Otherwise a range of hardenable steels. Hardening methods can be case hardening, nitriding and hard facing etc.
@@MotoChassisByTonyFoale The reason I questioned is, you'll be removing case hardened material when grinding do you intend to re harden? I am of course assuming you are regrinding an original camshaft, and not creating from stock
@@Phantom-mk4kp Your assumption is wrong, I will make the camshafts from scratch. During initial testing I will use some old camshafts to reduce the grinding time. Those will only be used for dimensional checks and will not be hardened nor used in an engine.
@@MotoChassisByTonyFoale Ah OK I see now. Seeing an old cam in the chuck I thought you were re-profiling. Good luck👍
@@Phantom-mk4kp I just put that in there only for the video.