I chose to turn the eccentrics from 2 3/4" HRS round stock, rather than using the Allen castings. Figure 1 shows the drawing for an eccentric. The main reason is that I had the steel available. Even if I had had to purchase it, the place I bought it from sells by the foot, and for this size charged $10/ft. + $2 for the cut.  After turning the parts I still have a decent stub and another 7" of material for some future project.

Figure 1

    Figure 2 shows the drawing for the initial turning of all four eccentrics from the bar (the dimensions for all but the diameters are not close tolerance). Note that the drawing is for just four eccentrics.  I am not going to be using a traditional axle pump, so I did not make a fifth eccentric to drive the pump.  I you will be using an axle pump, add a fifth eccentric to the drawing.
    The initial turning is done with all four parts as part of the bar. After this machining the bar is removed from the chuck and the individual pieces cut apart. Except for the two faces the rest of the surfaces can be used as reference surfaces. The red outlines in the drawing represent the outline of a finished eccentric. You will notice that I left a generous 1/4" of cutting allowance. Those who have a higher level of skill, or better bandsaws can decrease this measurement as they see fit. This cutting allowance also comes into play in determining the overall length of the blank. To prevent wasting the stub I reversed the work in the jaws at a later stage and faced and turned the stub surface, so it will be ready to be chucked for some future project.

    If you only have a 3 jaw chuck, I will describe a simple jig (in another section) that can be made from the stub to allow you to machine the axle hole and collar.

Figure 2

Figure 3

Figure 4

    I do not have a steady rest for my lathe, so I used the following method to chuck the bar centered for turning. To start the blank is cut off and the center of one end found using a set of hermaphrodite calipers. Figure 3 shows a set of this type of caliper. To use them you set the separation between the ends to a little greater then the radius of the bar. Then by placing the curved tip against the side of the bar just below the edge scribe an arc at the middle of the top surface (Figure 4). Repeat this for three other points at the rim about 90 degrees apart. What you will have is a "square" with curved sides. Scribe two lines diagonally between opposing corners. Where the two lines cross is the center of the circle. Figure 5 shows the marked and center punched bar, and Figure 6 is a close-up of the marks.

                         Figure 5                                                                    Figure 6

    I center punched the intersection of the marks and placed the bar loosely in the chuck with the center punched end towards the tailstock. The tailstock (with a center installed) was then run up and the point of the center placed in the punched mark (Figure 7). With the bar thus aligned the chuck was tightened. The bar was now properly aligned, but the bar is long enough that end had to be supported by the tailstock during turning. For this, using a drill chuck in the tailstock, the end was center drilled with a small center drill, at a slow speed and gentle feed. Once the center drill starts it will help to hold the end of the bar as it continues to cut. The center was then installed and used to support the end. An out off square end can tend to throw the end of the bar slightly out of alignment. To correct this the end of the bar was faced as closely to the center as possible (Figure 8), and then the end was redrilled with a larger center drill, one large enough that the little bit left at the center was drilled away. The tailstock center was then reinserted and the machining started.

Figure 7

Figure 8

    First the outside was turned until all the scale was removed.  To aid in the rough turning the bar was coated with layout dye, the edges of the spines were measured and small marks scratched at the locations.  A tool bit was then set at each mark and a light line scratched in the circumference with the work piece turned by hand.  Figure 9 shows the results.  The lines in the non-blued areas were virtually invisible with the work piece stationary, but once it started spinning they "poped" out of the background of machining marks.  The rough cuts were brought up close to theses marks, and then when the finishing cuts were made those cuts were made until the line was hit.  These marks were close enough for rough cutting the spines as the 1/8" thickness leaves plenty of "meat" for the finishing cuts, once the parts are separated.

Figure 9

   The "valleys" diameters were turned to the finished size, and lightly run over with a file and sand paper to produce a smooth finish. The finished should be smooth, but don't get to the polished finish stage, until all the other machining is done.  These surfaces will have to be gripped in the chuck jaws during later machining. The valley floor should be the same diameter across the whole length. The collar end will be used to chuck the part in while the strap end is faced. Whenever any type of abrasive is used on a piece in the lathe cover the bed to prevent any grit getting onto any of the ways, and clean up thoroughly before and after removing the covering!! The faces on either side of the spine also were accurately (perpendicularly, not to thickness yet) turned.

    I started with a triangular (threading type bit) (Figure 10 center).  A round nosed left/right bit would also be good, probably better, but I used an indexible carbide bit, so triangular was the only option. I used this bit to turn as much of the valley as possible, leaving the floor about 0.010" to 0.015" oversize.  I then switched to a left and right tool (depending on which side of the spine I was facing) (Figure 10 left and right).  I used these bits to both face the spine and cut the valley floors down to being 0.001" oversize.  This extra was left as a filing and sanding allowance.  Once again there is room on both sides of the specified diameter for error.  As long as all the parts are the same diameter, any over or under error can be corrected when the straps are bored.

Figure 10

    The face of the spine on the strap side is the finished surface, and the other side will be placed against the top of the chuck jaws to align the inner face of the strap end for final facing. The spine is left wider than the final thickness to allow for any correction that may be needed to bring the strap bearing surface to the correct thickness.   The spine will be machined to final thickness when the collar is turned.  The diameter of the spine is not critical. It serves to keep the strap from wandering off the eccentric, so as long as the diameter is close, it will operate satisfactorily.

Figure 11

Figure 12

    Once the initial turning is completed (Figure 11), the bar is removed and the first 3 (three) eccentrics cut off. Return the bar to the lathe and chuck it by the end with the remaining eccentric (Figure 12). You will notice that I forgot about the need for brass shims to protect the eccentric surface during this operation.  The reusable ones I made are described later.  Face and turn the circumference of the former stub, truing those surfaces up.  The stub is now ready to be used in some future project, rather than ending up as "a too short to use chunk" in the scrap box.  Alternately it can be used for the the simple 3 jaw eccentric turning jig (described in the section so named).  Remove the bar and separate the last eccentric.

Figure 13

    To keep from marring the finished surfaces I made three simple reusable brass jaw covers as shown in Figure 13.   These should be made so that they are shorter than the jaw height, so that the eccentric spine can set down on the jaw top.

  Return the eccentrics to the 3 jaw chuck and face each side (Figure 14)(Note: Again before the brass jaw protectors). Starting with the strap end, bring the strap bearing surface to the correct thickness. Machine some of the spine thickness away if the area is too narrow after the facing operation. Reverse the part and face the collar end. You can either turn the spine to the proper thickness now, or wait until the axle hole and collar are turned later.

Figure 14

More to Come!

Return to Valve Gear page.

Return to Minnie page.

Return to the Live Steam Construction Articles page.

Return to Home