Heisler Castings
Nelson Riedel, Nelson@NelsonsLocomotive.com
4/17/2004, last updated 07/09/2006

This page provides background information on  the processes used to make the castings for the Heisler locomotive. 

The initial plan was to fabricate the Heisler parts from stock using techniques described by Kozo Hiraoka and Bob Reedy.  However, early in the design process I became acquainted with Charly Wilkins who was interested in building a Heisler and also had an investment casting business which he was in the process of turning over to his children.  We quickly made a deal, I'd make the initial patterns and he'd make the molds for the wax patterns and secure the castings.   If the design works out as planned, hoped and if we're lucky, the locomotive will be attractive to others.   If so, the castings will be available via his family business.  I have no financial interest in this business venture.  It's pure fun for me and I don't want to spoil the fun with worries about profit & loss, liability, etc, etc.

The status in the spring of 2006 is that all the patterns have been finished..  The Molds for all engine castings except the crankcase are at the foundry.  The first 5 casting types arrived in late June 2006 (see photo at end);  hopefully the remainder will arrive shortly.    About 12 more molds are required.  They should be finished by the fall with castings available near the end of the year or earl;y next year.. 

Anyone interested in obtaining information related to purchasing the castings should contact Charly  at:

The investment or lost wax casting process is thousands of years old.  Rather than trying to explain the process here, I direct everyone to the following links for nice graphic explanations.

 http://home.wxs.nl/~wakke007/negatief/investme.htm

http://www.hitchiner.com/HIMCO/HIMCO_Library/Intro_to_IC.pdf

http://www.eskimo.com/~delvest/inv_casting.htm

These websites talk about a pattern tool used to make the wax  pattern.  The pattern tool in those descriptions is an aluminum die and the wax patterns are produced by injecting the wax in the die.  This is an appropriate process when thousands or tens of thousands of parts are to be produced.   For the low volume parts such as a locomotive, a  pattern tool (mold) can be made by using liquid epoxy around a metal pattern rather than machining a mold.

 The way the Charlie and I have split the job is that I make the initial metal patterns and he makes the molds.  He then makes the wax patterns using the mold, assembles trees of the wax patterns and takes them to the foundry to be cast.

Metal Patterns:  The rules for the metal patterns are that they should be 3% oversize to allow for shrinkage.  The pattern must be metal.  Aluminum is preferred because it is easy to machine.  Auto body filler such as Bondo can be used to fill small holes, make fillets, etc.  The shape must be such that the mold can be pulled apart ----  no hooks.   Cavities, hooks, etc can be made using cores.   The cores are also metal and placed in the mold before the wax is injected.  After the wax is injected and cooled, the wax pattern is removed with the cores and then the cores are withdrawn from the wax pattern leaving a cavity  in the wax. The photo on the right shows the pattern for the main bearing cap.

Molds: Charley and son Wade sent some photos of the process they use to make a mold --- in this case the main bearing cap mold.    Charley had described the process to me but it never really sank in until I saw the photos.

 I've had experience with sand castings where standard top (cope) and bottom (drag) boxes are used to hold the damp sand that is formed around the pattern which is attached to a board that is sandwiched between the two boxes.  After the two boxes have been filled with the damp sand and the sand tamped, the boxes are separated, the pattern removed and then the boxes put back together.  There is a cavity left by the pattern at the seam between the two boxed which is then filled with the molten metal.

A strong metal box is used for the investment casting mold..  The box must be made to close tolerances so that the resulting castings are also to close tolerance --- on the order of 0.010" for our locomotive castings.

The first part of the process involves:

1. Selecting a pair of aluminum blocks that are large enough to hold the metal pattern.    These blocks are usually sawed from a larger block of aluminum. 
2. Next, the blocks are squared
3. Next, the surfaces that will be the joint between the two blocks are surface ground so they have a good seal.
4. Next, holes are drilled between the two blocks for index pins and screws to hold the blocks together.
5. The index pin holes are reamed
6. The screw holes are tapped

The next step is to mill out pockets in the blocks.  This photo shows milling a pocket in one of the blocks of the main bearing cap mold.
This photo shows the two blocks that make up the Main Bearing Cap mold.  Note the 4 holes in the corners of the blocks.  Two holes in each block have been drilled and reamed for index pins.  The other two holes have screw clearance holes in one block and are tapped for the screws in the other block.   Note that are some holes visible in the bottom of the pocket in the right block.
The right block has been flipped over in this photo showing that there are 5 holes into the pocket. The holes have been counter sunk on the outside. These holes are used to retain the epoxy that will line the inside of the pocket.
This shows the left half of the mold  has been filled with green clay and then the pattern pressed about half way  into the clay.
The two halves were then screwed together with the half with the clay on the bottom.  Clay was then pressed against the seam between the two halves to keep the epoxy from leaking out.   The clay on the top forms a wall (or dike) to hold the epoxy.
This shows the block epoxy had been poured into the top half of the mold through the five holes.
The mold has been taken apart and the top half (the half with the epoxy) is being heated to ease removal of the pattern.  The greens is residue from the clay that was in the other half.
This shows the top half of the mold with the finished cavity.  The green around the edge is clay residue that will be cleaned off.  The next step will be to finish the bottom half of the mold.   Holes will be drilled in the bottom half to retain the epoxy.  This time, rather than using clay, the pattern will just be pressed back into the epoxy in the top half.  The pattern and the set epoxy will be coated with wax to to prevent the fresh epoxy from sticking to either.
This shows the cleaned up top half of the mold.  The next step is to machine the sprue and gate --- the passages used to inject the liquid wax into the mold.  These passages will be along the seam between the two blocks.

Photo above shows the mold fabrication status as of the end of January, 2005.  The main bearing cap mold in the center of the photo is complete.  Others such as the connecting rod strap on the right side and the exhaust manifolds near the center of the photo have the epoxy in one half the molds.   Note that this process requires considerable effort and lots of aluminum to make the molds.   Hence, with the very low volume, the cost of a finished casting is nearly all due to the mold fabrication cost.  Another way to say this is that a casting with a few dollars of iron or steel might cost $100 or more.  As the volume increases, the mold cost can be spread over more parts and the cost per part reduced. 

Foundry: Charly sent some photos of the foundry operation with a description which I've included below.

This is a wax injection machine.  The hot molten wax is injected from behind.  The hydraulic press holds the mold together.  The flat piece with the dowel pin sticking out the side of the mold is a core.
This is a mold being opened with another mold in the machine being injected.  The two rods in front of the mold are cores which have been pulled out of the mold.   The operator uses two screwdrivers in slots to pry open the mold. The aerosol spray can to the rear is mold release that is sprayed into the mold before injecting.  The blue pieces in the lower left corner are consumable cores ----- probably similar to the sand cores used in sand casting.
This a finished wax pattern.  The operator has cut off the sprue from the injection process ---- it is laying on the table.    The rectangles on the top of the wax are gates which will provide a path for the molten metal.   The gates will be sawed off after the casting process.  The groove at the bottom of the wax part was formed by the aluminum core with the dowel pin described earlier.  The two round holes were formed by the core pins on the table under the piece.
These are wax part trees.   The brown wax is the part.  This is a different part than the one  described above.  The light green column is the tree (or tree trunk).  The top is a ceramic funnel and the red wax is a safely riser. The brown wax parts are stuck to the tree trunk.
This shows an investment mixer and a dipping tank.  The wax trees are dipped and then hung on a rack to dry.  The dipping process is repeated until the investment coating is the proper thickness.  A couple of the coated trees can be seen in the background
After the parts are coated they are fired to harden  the investment and also to melt the wax leaving a cavity.   These are parts ready to be poured.
After pouring and cooling the investment is removed from the parts by hammering, tapping, etc.
One of the last steps is to cut the parts off the tree and then grind off the gates.

The castings: (Updated 7/9/2007) Upon return from five weeks volunteer teaching in Latin America my wife greeted me with a few words about things that needed repaired (YUK!) and a box of castings that arrived during my absence (YA!). 

The photo shows the 5 castings types received.  (The foundry has 18 molds so hopefully the remainder will be coming shortly.)  The valve stem heads are on the left,   the rocker arms in the center top and the rod bearing strap on the upper right.   The upper and lower outer eccentric straps are on the bottom.  All these are cast in 1020 steel.  That is a bit of surface rust on the one lower outer eccentric strap   The parts meet my expectations and will require little machining.

Hopefully I'll have photo of more casting types in the near future.