Cam Grinding Hypothetical Design Concepts

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Well-Known Member
Staff member
HMEM Supporting Member
Global Moderator
Jul 2, 2021
Reaction score
MidSouth, USA
I am spinning this thread off of this topic:


There has been a great deal stated about cam grinding machines, but I must admit I can't follow the logic.

So this thread will be my attempt to create a cam grinding machine in 2D CAD and/or 3D modeling (Solidworks), with the intent of either verifying the design parameters a the SW motion study, or verifying it in 2D CAD by making progressive rotations on 2D drawings.

I have no significant knowledge of cam grinders, so this will start from scratch.

I will start with a hypothetical cam lobe that is 1" diameter, and perhaps 1.5" at the top of the lobe (I will draw something up).

I will assume that the camshaft being ground will move horizontally towards the grinding wheel, as if it were mounted on drawer slides, to simplify things.

I will pick a nominal pattern that is scaled up 6 times larger than the cam on a copy machine, and then will compare that with a trace that would be made on the pattern wheel if a finished lobe was tracking against the grinding wheel.

I will use a horizontal point source that tracks against the pattern wheel.

The pattern wheel could actually ride against the end of one of the drawer slides.

Case #1:

Assuming your cam was already ground to final size, and your pattern wheel was the exact same size as the cam, then rotating the camshaft with the cam riding against the grinding wheel would produce a motion that would be exactly me mimicked in an identical sized pattern wheel.

1:1 ratio

Case #2:

Scale up the finished cam lobe on the copy machine by 6 times, and print.

Install a pattern wheel on the end of the shaft, round and nominally sized slightly larger than 6 times the cam size.
Rotate the finished cam lobe against the grinding wheel, and with a fixted marker, trace the path taken on the blank pattern disk.

Compare the trace with the lobe that was scaled up on the copy machine 6X.
According to what others have said, the copy machine scalled lobe will be too small.

Last edited:
The stop that rides on the pattern wheel should be capable of being made either with a sharp point, or a rolling wheel.

Either way there will be a point contact with the pattern wheel, or I guess in the case of a wheel it would be a horizontal line contact, which would act the same as a point contact.

The drawer would be spring loaded towards the grinding wheel.

If you added a linear actuator to the drawer, you could program its movement to grind the cam, but we won't go there.

The height of the camshaft above the drawer slides should be kepts as small as feasible, since the taller you make it, the more moment it will create (twisting force), which is going to cause some deflection of the framework.

Perhaps the camshaft could be on the same centerline as the grinding disk.

The "drawer" could slide on round shaft supports, similar to what is often used for 3D printers.

Synchronizing the camshaft rotation with the pattern wheel rotation could be tricky if the pattern wheel contacted one drawer slide.
I guess the belt would have cantilievered off one end of the drawer.
Camshaft would be mounted just about the slides.

Or let the belt go over the drawer frame, and put the cam on center with the slides.
So this would put the centerline of the drawer frame, the centerline of the guild rails, the centerline of the camshaft, and the centerline of the grinding wheel shaft in the same plane.

Pattern wheel would protruct towards the front of the machine.
Last edited:
Your concept of moving the cam horizontally is a new one on me the ones that I have seen rotate the cam and master cam.

One of the main reasons to rotate is that you get a constant flowing surface, any horizontal movement will result in a series of faccets as the cam is indexed to the next horizontal position which would then really want to be blended. Most people grinding cams do it by rotation specifically to avoid faccets which result from milling with software like camcalc
Laid out flat like this, very similar to how a 3D printer moves in the horizontal axis.

All shafts in the same horizontal plane.

The cam would rotate via a cog belt, as would the pattern.

The idea is to make the equations linear.

And you could drive the carriage with a linear motor, and use a computer to control position, but would otherwise manually control the finger contacting the pattern wheel.

The motor and grinding wheel could slide from side to side.

If a linear motor was added to the grinder motor carriage, you could move it side-to-side, from cam lobe to cam lobe, all automated, CNC-style.

Last edited:
At the moment I can't see how the grinding wheel's movement is reduced to the proportion of master to model cam.

There will be more up/down movement of the master due to it's larger size, there will also be more horizontal movement due to it's larger size
Your draw runners will move both teh follower on teh master and the wheel by the same horizontal amount.
I see what you are saying.

I need to think about what is happening with the cam grinder Peter is using in the other thread, in particular the pattern wheel.

The 6X pattern wheel would need to be moved over onto the right side of the rotating camshaft.

Slow speed camshaft drive motor could be mounted on the right side above the crankshaft.

Last edited:
I think this camgrinder design will be more for simplifying and understanding the equations involved.

From a practical standpoint, Peter's style design may be more practical to construct.

You could put a linear actuator on the right side of Peter's style design.

I would not worry about anymore motors than the one doing the grinding at the moment, just crank the linked master and blank by hand and alter position along the camshaft by hand. Infact just work out the concept for a single cam for now to save undue complications.
Lets say you enlarged the profile of the cam 6X in CAD, and then enlarged it again some fraction, would that produce the correct shape/size pattern wheel ?

The white paper seems to indicate that this would not work.
If your linkages give the same reduction of movement of the wheel as your master is enlarged then they should do.

If you want to have things moving horizontally then you would need the master follower on one slide and the grinder on another linked with a 1:6 ratio

Not sure if draw slides don't have too much play, ground rods with oilite bushes may be better but again that can wait until the concept is sorted out.


  • 20240605_121046[1].jpg
    1.9 MB · Views: 0
But you need to understand that his pattern wheel is not just the cam profile enlarged X times like you are suggesting, If I understand it right it is X times base radius but lift is the same as the finished cam
I am following along because it is interesting to see a project start with a concept and then move on to design: I am not interested in building a cam grinder at the present time.

Quote from Jasonb: "Not sure if draw slides don't have too much play, ground rods with oilite bushes may be better but again that can wait until the concept is sorted out."

In my experience drawer slides have a lot of play. Some type of machine slide is called for, like the ground rods and bushings. Perhaps there might be off-the-shelf linear bearings that could be used.

AFAICT there is nothing to be gained using linear bearings, its an added complication plus they are expensive, AFAIK everyone uses a pivot, and when I say everyone I mean everyone from full size professional automotive cam grinders to our teeny tiny model engine cam grinders, everyone. The two competing designs, master on same axis as cam blank like I use, and master and cam blank on two different parallel axies like Charles uses, both use pivoting mechanisms.

and if you use linear bearings you are forced into using the Don Bell style "additive" master cam that I use, you will not be able to use a "scaled" (6X or whatever) master cam that Charles uses. <--- think this through before going any farther, make sure you understand why <---

in any event, you're learning by doing, excellent way to go !!!
Last edited: