No announcement yet.

Homemade cam grinder

  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by Dan Dubeau View Post
    Will it really matter all that much though? Yes, in a theoretical sense the tangent point will change as the cam rotates against different shape followers, but in a practical sense how much does it actually matter to the end result. Especially for something like say a model engine camshaft.
    That depends on the amount of error you are willing to accept "in the home workshop", making "home-made" parts for a "small model engine", just to hit a couple of the typical buzzwords.

    Depending on the size of the cam and wheel, it could be a error of several tens of thou. Even ten thou of error may totally change the valve timing, especially on a smaller cam.

    Using a 6" wheel, if the follower is flat, and the cam is such that the contact point is 1/4" higher than the line between centers, the error is just about exactly 10 thou large. The wheel at that point is 10 thou farther away than a flat surface.

    The smaller the cam, and the larger the grinding wheel, the more the wheel approximates a flat surface, and the less the error. If you used a flat follower, and ground the cam against the flat side of the wheel, then all would be well. A large wheel would get closer to this.

    Imagine a 1/4" wheel, and 2" long cam. With the follower on a flat surface, the cam might not even touch the wheel some of the time, because the (theoretical) contact point would be far above the small wheel. (Can't call it a "real" contact point since it would not make a contact.)

    However, if you used a 1/4" radius follower, you could grind a perfect cam with the 1/4" wheel, since they would both contact the master and cam blank at the same place.
    Last edited by J Tiers; 11-20-2021, 12:54 PM.

    Keep eye on ball.
    Hashim Khan

    Everything not impossible is compulsory


    • #32
      Originally posted by strokersix View Post

      Had to chuckle a bit with your comment. How do you plan to check your cam lobe meets the print tolerance? Without an understanding of the geometry involved with the manufacture and checking you can't do much more than check the maximum lobe lift.

      Agree with you that it's application dependent and likely makes little difference for model engines in this context.
      With a ball nose indicator tip perpendicular to the shaft and a degree wheel you could properly check a cam. I truthfully never done it, but that's how I'd set it up if tasked with the problem. Seems like a logical way to dimension and measure a cam no?

      Ok, I quickly drew up some cams in cad, and played around with a couple rotations at various points, comparing the center line points, vs the tangent points of contact vs a wheel, and a smaller point of contact and there is a big difference at times. Much bigger than I was envisioning in my head. Sometimes I need to hold/see something to fully understand/visualize it.

      I WAS WRONG. Foot in mouth disease strikes again.....

      Thank you gentleman for the education. I will quietly go and sit down with open eyes/ears and a closed mouth.


      • #33
        Autodesk Inventor has what it calls Dynamic Simulation. It allows you to generate an accurate pattern for a cam by tracing the actual cutting tip. I use it to generate the rosettes (cams) for my rose engine based on the desired shape of the finished piece, the shape of the follower, and the shape of the cutter. It's interesting to see how the pattern has to change to account for the cutter cutting above and below center (tangency point).


        • #34
          Originally posted by J Tiers View Post
          ...........................The smaller the cam, and the larger the grinding wheel, the more the wheel approximates a flat surface, and the less the error. If you used a flat follower, and ground the cam against the flat side of the wheel, then all would be well. A large wheel would get closer to this..............................................
          However, if you used a 1/4" radius follower, you could grind a perfect cam with the 1/4" wheel, since they would both contact the master and cam blank at the same place.
          Jerry's words reflect the exact issues and correct approaches to the issue of building a Cam grinder.
          You must look at the tappets( cam followers) in the engine that the cam will be used in.
          Having flat tappets is the most common for model engines and the easiest to design to .
          Remember , the cam has two major functions--- determine the max "Lift"-- and determine "Duration " of lift in % of rotation, ie "degrees'
          You could also consider the Duration period as two separate function ..Max lift rotation, and total Lift rotation, and this affects the "Ramp"
          It is the "Ramp" section that is mostly affected by wheel size on a straight flat follower , whether in the engine (tappet) or the cam grinder Master Cam Follower.
          For model engines , the ramp design is almost of little value as the primary purpose is to get the engine to run for our hobby. We are not in High horse power competition.
          If you were building a commercial weed wacker or saw, then careful ramp designs must be honored to achieve max HP or RPM.
          Access to CAD allows one to consider the follower design and compensate for grinding wheel conditions.

          In real life , if the Master Cam follower determines that the lift will be .100" , then the Grinding Wheel regardless of OD --will deliver .100"
          And the duration will be the same as the Master Cam calls for .

          The Ramp will be slightly different , but I seriously doubt it will have an impact on the engine.


          Green Bay, WI


          • #35
            Yes, the ultimate lift is the same, since that occurs when the high (contact) point of the lobe is exactly on the line from the cam shaft to the grinder wheel.

            That may NOT be the single most important point on the cam

            The RAMPS are what determine the opening and closing points, as well as the "shape" of the valve movement over time. They control duration of the open time, the point in the stroke wher the cam opens, etc. Those are normally important points.

            If you are uninterested in the power produced, OK.. If you are happy if it will keep "popping away" when on display, the only important point may be that the valve opens more-or-less at the general time it should.

            But, as has been seen with Brian R and the engines lacking power to run other models, power produced can be important even with a model. More so if it powers a boat or airplane. Cams are a part of that.

            Keep eye on ball.
            Hashim Khan

            Everything not impossible is compulsory


            • #36
              For those interested in all things of Cams for a engine, I highly recommend you read
              Carl Wilson"s series in Model Engine Builder magazine titled A Series on Cam and Master Design and Grinding.
              It is a the epitome of cam discussions in volumes 21,22,23, and 25
              Also He continued in depth in issues 33 to 36 which cover Three Arc Cam Design


              Those issues are available for sale from the publisher
              the Issue number or volume number are in Red or Blue for ease of access
              Green Bay, WI


              • #37
                Having worked with camshafts for various race engines, the only numbers we ever concerned ourselves with
                was valve lift at the valve relative to cam position in degrees.

                In a DOHC with direct acting bucket tappets, you'd measure at the tappet, where you could, and off the cam lobe centerline
                if the indicator couldn't get a consistent reading off the tappet.

                With rockers, it's off the valve itself, because the geometry of the rocker, lifters (if it has them) and especially tappets
                all affects how the valve moves.

                One of the worst cases is the engine I race regularly- it uses a curved slipper tappet that directly bears on the
                cam and is part of the rocker, so the contact point is constantly moving (and changing the rocker ratio)
                and is not symmetrical around TDC. So our cam grinds end up looking pretty odd to get a good valve motion profile.

                there will be distortion unless the guide and the grinding wheel are the same size, so that the contact on both is similar.
                I would think, looking at the geometry of that system, that the guide and the wheel would want to be in scale
                with the work- so if the master is 3x the cam, then the guide would also want to be 3x the size of the wheel.
                But I haven't quite gotten to the 'grind my own' phase yet, as there's a local grinder whose work is excellent,
                and is quite inexpensive.

                rusting in Seattle


                • #38
                  I arrived at this party late but better late than never. Thanks to Rich Carlstedt for pointing to my articles in Model Engine Builder Magazine. I did a quick survey of the previous posts and I'll begin with this:

                  Professional cam grinders use large wheels: as large as 21 - 24". And they are allowed to wear to a minimum diameter. IIRC, a 21" may be used down to 19". This will make some change in the shape of the surface but leave the lift unchanged. I would have to review my articles to comment upon the duration. I believe it too does not change. Some grinders have reduced diameter followers that allow the wheel diameter to be reduced more. What does change as the diameter of the grinding wheel is reduced - the follower diameter remaining unchanged - is the shape of the cam surface. That changes the velocity and acceleration of the valve gear which may affect the cam follower.

                  Home shop cam grinders use one of the three basic systems: rocking bar, geometric reduction and reciprocating which uses a linear slide. Reciprocaters are not common; I have seen only one. It used a hardened original which operates against a surface at the end of a spring loaded machine slide. It is necessary that the contacting surface is the same shape as the cam follower and that the original has the same profile and size as the cam.

                  Rich Carlstedt's cam grinder is a geometric reduction. The 'master' is the cam profile enlarged by a multiplying factor. The geometry of the machine reduces the master by that factor to the cam. The master looks exactly like a magnified cam lobe. A major advantage is that defects in the surface of the master are also reduced. It is also a simple geometry to understand and implement. The design published in Strictly IC does not have correct geometry. That being said, it grind cams that operate model engines very well. Model engines should run quite well, at least for awhile, with a dowel pin inserted into a shaft and rounded to the nose radius on the end, With the correct diameter flat follower, the dowel would function as a flat tangent cam profile. A flat follower is not the best follower for a flat flank cam - acceleration is very high, but Elmer Wall's engines run with this combination. The takeaway is that the cam profile is not critical for engine show service.

                  Both geometric and rocking bar cam grinders utilize an oscillating element to move the cam blank toward and away from the grinding wheel. The rocking bar carries the cam blank and the master on a common centerline. All commercial analog cam grinders, to my knowledge, use an enlarged master which does not look anything like a cam lobe. It is much rounder. These masters are generated by adding a constant amount normal to the curvature of the surface at each point. This addition is subtracted during grinding. The purpose of an additive master is to increase the surface length and radius, particularly the nose radius, to reduce wear. Also, the follower is mounted on precision bearings so the contact is not sliding. In general, the master follower is the same diameter as the grinding wheel or within a tolerance. It is also possible to grind cams from a model lobe, which is exactly the same size and profile as the finished cam. This eliminates the step of making the master and is suitable for short production runs.

                  Generating the originals is done on conventional machine tools and may involve some math up to simple trigonometry. Masters may also be ground from model lobes on the cam grinder. It would take too much space to do justice to this subject and I must refer you to my articles. I have published the originals of my articles on cam design on my website: Three Arc – Carl Wilson ( I had hoped to publish the originals of the article on cam grinding also, but that has not been done yet and may not be done. So back issues of Model Engine Builder are the best sources for these.