The Ugly Rocker Arm Cloning Project

25 May 2005:

Mr. E.C. Mims has kindly loaned Frank and me the magneto-type ("UGLY") rocker arm and associated parts from his original Webster magneto Aermotor engine.  We are going to make a mold of the rocker arm, the two clevis castings and the trip cam casting to make accurate reproductions.  The mold will be used to cast the parts in wax.  The wax parts will be taken to an investment ("Lost Wax") casting foundry.  When completed, the copies will be virtually undistinguishable from the originals.

I will, as we progress with the cloning, make entries here so stay tuned as we go along.

Here's the whole assembly as it came off the engine.  I think the rocker shaft (the bent shaft assembly) is the same as the battery one except that it doesn't have the taper pin hole in the end where it goes through the rocker arm.  The bolt that holds the clevis casting on the rocker shaft goes through the hole that the battery trip rod was clamped in.  The clamping bolt is still there, although useless.

Here's the trip rod and clevis assembly still bolted to the rocker shaft.

Another view of the trip rod and clevis assembly.  The tension spring goes between the hole in the clevis and the hole in the bent bracket.

The rocker arm casting. Casting Number One

From the left (latch) to the right (tappet bolt) ends, the length is just over 12-1/4" long.

From the bore where the rocker shaft goes through the rocker (bottom of picture) to the clamp end (top of picture) is just over 6-3/4".

The rocker arm casting. Casting Number One

Mark-off all the holes in the new casting from the center of the rocker arm hole to cancel-out any shrinkage that occured in the re-casting.

The distance to the center of  the rocker shaft hole to the cam roller pin hole (the one with the chamfer, [second from right]) is 3.860".

The distance from the center of the rocker shaft hole to the governor lock-out hole (at the far right) is 5.339".

The distance from the center of the rocker shaft hole to the tappet bolt hole (at the far left) is 6.062".

- - - - - - - - - - - - - - -

The approximate height of the casting is a little over 3-1/4".

The governor latch hole (right end) is 0.312"

The cam roller hole (2nd from right) is 0.312"

The rocker shaft hole (center [reference]) is 0.5625" (to fit the rocker shaft)

The tappet hole (far left) is tapped 1/4-20

The two "U" bolt holes are drilled to 23/64"


Note for locating rocker shaft hole center

Since the rocker shaft rests against the rocker arm when it is assembled with the U-bolt (where my thumb is), you must offset the center of the rocker shaft hole one-half of the diameter of the rocker shaft (0.281").  Either use a straight edge to find this point which is at the top right end of the straight edge or use the drill-point I've made in the blank casting.  I think the drill point will be better because, at least, you will be real close.  Look at the picture at 29 May 05.

The "U" bolt and spring. (Fabrication #1 & 2)

The spring is made of 0.039" music wire.  Body of the spring is 0.375" in diameter and 1" long.  Loops extend 0.4" from each end.  (This spring has one end broken off)

The "U" bolt that holds the rocker shaft to the rocker casting is made of .312" iron wire and threaded 5/16-18 for 0.600" on each end before bending.

It is bent in a "U" shape, ending up 1.183" deep (outside to ends) and 1.250" across (outside to outside).


Cleaned up and laid-together, this is what the whole assembly looks like.


Detail of the clevis end and the spring bracket.

The Trip Rod.  (Fabrication #3)

Dimensions are all from the threaded end of the rod.:

Made of 0.312 square stock.

Overall length 6.0".

Thread, 5/16-24, 0.650" long.

To start of ramp, 2.125".

To trip end of ramp, 2.875".

Depth of trip, 0.270"

11 October 2005 note: The trip rod needs to be hardened in the area of the ramp where the mag pawl engages.  We think it will be sufficient to (AFTER machining) heat the rod in this area to a dull red then drop it into a can of motor oil.  If this doesn't make it hard enough, a hole can be drilled from the back to bisect the engagement area and an appropriate sized piece of drill bit shank pressed into place, then ground flat where it engages the pawl.


To remove the cam roller, press out the pin.  Press against the left side of the roller to minimize stress on the casting.  The end of the pin is peined into a shallow chamfer and will push out using a 1/4" pushing pin.   The new rocker arm will have to be chamfered as shown on the right, above.

Clevis Assembly.  (Castings 2 & 3)

Detail, (2 castings and a mild steel pin, peined into place).

The left casting accepts the threaded end of the trip rod and is drilled and tapped for 5/16-24.

Clevis Assembly. Castings (2 & 3)

The casting on the left is 1.650" from left to right (horizontal in the photo) and 0.650" thick (vertical in the photo).

The casting on the right is 1.5" from left to right (horizontal in the photo) and 1.22" thick (vertical in the photo).

Clevis Assembly.  Castings (2 & 3)

The casting on the left is a total of 1.00" high (vertical in the photo).  The hole for the spring is 1/8" diameter.

 The holes for the pin are 0.375".

The casting on the right is a total of 1.0" high (vertical in the photo).

Clevis Assembly.  Castings (2 & 3)

Bottom of clevis parts as would be mounted on the engine.

Clevis Assembly.  Casting #3

The hole that goes through the rocker shaft is threaded 5/16-18.

Magneto Trip Cam.  (Casting 4)

Here's the last of the four castings that have to be made.  This is what I call the magneto trip cam.  It has the square hole through it (left-to-right in photo above) cast into place.  Gonna be interesting to figger out how to make that!  In the end, we might have to drill the hole and file it square.

The bolt is a 1/4-20 square head setscrew.  The cam is drilled and tapped for it.

Overall height of the casting (vertical in photo) is 1.125".

Overall length of the casting (horizontal in the photo) is 1.4"


Magneto Trip Cam.  (Casting  4)

Facing the threaded end of the trip rod as assembled.

The hole cast through the cam is 0.330" square and rough.

The overall width of the cam (horizontal in the above photo) is 0.590"

Magneto Trip Cam.  (Casting 4)

Facing the end of the trip rod opposite the thread.


29 May 2005:

Today, amother things, I used some old Bondo to fill-in all the drilled holes in the castings.  After the mold is made, I can easily push the Bondo plugs out of the casting.


Note that I've made a drill center for the rocker arm hole because it would be problematic to get it in the correct position by using dimensions.  This hole should be drilled perpendicular to the rocker arm.  

11 October 2005 note: For the folks who purchaed casting kits, because it's location is critical, we may drill this hole prior to shipping.

15 July 2005:

Since my last entry on this page, back in June, Frank and I did some work on the mold for the parts.

First, we tried to figure out where the mold should separate.  We discovered that, when the rocker was originally sand cast, the pattern was pushed into the wet sand in the bottom of the mold box then more sand was packed around it to where the mold separated.  This made for a multi-level lower sand mold.  This made it hard to do using the RTV rubber because the rubber pours like thick honey before it sets.  


At the left above is our preliminary layout with the steps to make the parting line match up for all of the parts.  At the right, I've cut out the bottom of the box and am screwing flashing aluminum to make a box.  The masking tape on the parts is to keep the rubber from sticking on the pattern parts until the bottom of the box is filled and cuired.


Above at the left, the completed box minus the pattern parts is sitting in a bucket of kitty litter in case the box leaks.  The rubber material is really nasty to clean up and sticks to everything.  On the right, the box has the bottom layer of RTV rubber curing.  After it has cured, the parts will be laid on this layer and more RTV rubber will be poured in to make it reach the level of the parting line.


The parts are laid in place (left photo, above) in the box on the bottom layer of rubber and more rubber is poured to the parting line.  At the right photo, after the previous layer of rubber has cured, another layer of release agent was sprayed on everything and more rubber was poured to completely cover the pattern parts.

Now, Frank had to separate the two halves of the mold which was a real pain.  The Teflon release agent we used didn't work well on the rough iron parts.  By the time he got the top of the mold separated and the parts out of the bottom, he wasn't at all satisfied with the results.  I didn't take photos of this step because it wasn't a pretty sight.  Also, we thought we could make the mold for all of the wax parts in one box, we soon found that this made the mold too complicated for our purpose.  A different release agent will also be used for the second try and each part will have its' own box.

11 October 2005:

Well, after several tries, so far we haven't had a whole lot of luck making our own molds for the wax parts.  One of the problems is that the rocker arm is a very odd shape making it necessary to have the parting line at different heights.  Added to this is that we are having a problem finding both a casting material (the black stuff in the above photos) that has low shrinkage and doesn't stick like glue to everything, in spite of release agent.

Our plan now is for Frank to take the original castings to a foundry and have them use these parts as the patterns, manually packing boxes for sand molds.  This will be done in the next two weeks or so.  Since the original rocker arm casting looks to be about as much too long as the shrinkage will be, we don't have to compensate.  The new castings should be a better fit than the originals.  Don't ask me why!  I think you'll have to contact the Aermotor factory to find out.

We are building-up the casting in the area of the governor latch block (at the end next to the cam roller).  This is because on Frank's engine, the block had to be spaced out almost 1/4" closer to the latch arm due to way too much clearance.  In fact, when I disassembled the engine, I found a thick lead washer in this location, proving that someone else (the factory??) had the same problem.  If there is too much space between the latch arm and block, the engine will run too fast or will not latch out at all.  This location on the new castings will be about 1/4" thicker in the direction of the latch arm and can be filed, milled or ground to fit.

If these castings are successful, we will run enough to satisfy the list of folks who want them.  They will be shipped un-machined except, possibly, for a few really critical things.  The rest of the machining will consist of jobs that can be done with a drill press (or a hand drill if you're careful), files and a tap and die set.

Stay tuned for the next exciting episode of the Ugly Rocker Arm Cloning Project!

9 February 2010:

UPDATE:  The project stalled in 2006 when we got as far as sending the patterns to a foundry and getting a set of really rough castings.  To those who wanted casting sets, I apologize but other projects have taken over the time we have left on earth and, since we'll never finish it all, we have to occasionally shed some stuff we'd otherwise like to do.