The Knucklehead's case will require machining three relatively large chunks of aluminum most of which, depending upon the work-holding margins, will be returned to the recycler. I spent an inordinate amount of time trying to come up with a machining sequence that wouldn't require a lot of special fixtures but would guarantee the required alignments among the finished parts. I'm sure there are better ways to go about machining these parts without generating as much waste as I did, but my tired brain eventually had enough and began pleading with me to start making chips.
All three parts will end up with final shapes that will make their work-holding a bit sketchy especially during the heavy cuts involved with pocketing their voluminous interiors. The cam box, in addition, will require a number of precise boring operations to locate its various bearings, gears, and shafts. My plan included performing as much of their machining as possible while the workpieces still had straight sides and could be easily held in a vise.
For cosmetic reasons the complex peripheries of the final parts, especially the crankcase halves, should wind up closely matched to each other. Because of a very tall workpiece stack-up, it would make sense to machine the two halves in separate setups and then file and sand away their differences after assembly. My tentative plan, though, is to temporarily assemble the two halves and machine most of their exteriors in a single operation - mainly to see if I can. With a large enough diameter cutter this wouldn't be a big deal, but a spindly 3/8" diameter end mill with 2-1/2" long flutes will be needed to reach all the crankcase's outer features. My current work-holding strategy will handle either approach but, before finally deciding, some experiments will be done to check the effects of tool chatter and deflection.
Back in the shop, my first steps involved truing up the long sides and both faces of three 1-1/2" thick chunks of aluminum. These roughly 6" x 7" blocks came out of a couple weathered fixture plates acquired from a salvage yard many years ago. Taken two at a time, they were bolted together using a temporary 1/4" bolt through their approximate centers so the holes for a pair of dowel pins as well as the SHCS's that will eventually hold the finished parts together could be drilled through them. Deep drilling parabolic drills were used as pilots for the dowel holes before they were reamed. Prior to installation, an end of each dowel was center-drilled with a tiny divot so its center could be precisely indicated under a spindle microscope.
Remarkably, the y-axis differences in the dowel hole exits on the opposite sides of the workpieces measured less than a half thousandth, and the distances between them varied only a few tenths more. If the holes actually aren't straight, then they're off by the same amount. With the dowel hole pairs being so closely aligned with the fixed jaw of the mill vise, corrections for workpiece rotation won't be necessary after all. The workpieces were all marked so they can be consistently returned to the fixed jaw of the vise for additional machining.
The dowel pins in each workpiece lie on the x-axis centerline of the crankshaft. Its y-axis centerline is centered between the crankcase dowels but had to be offset from the center of the cam box dowels. Later on, the three blocks will be temporarily assembled, and a crankshaft center hole will be bored through the temporary center holes in the entire stack.
As mentioned earlier, it was important that the long sides of each pair of workpieces be parallel to the fixed jaw of the mill vise during the dowel hole drilling even though their widths were unequal. To achieve this, the bottom workpiece in the two-piece stack was clamped in the vise while the top workpiece, after being traversed with a dial indicator, was clamped to the bottom workpiece using the temporary center bolt. Unfortunately, I found that the 1/4" bolt, even with a large beveled washer under its head, couldn't supply enough clamping force to keep the two in perfect alignment even after a light tap from a plastic mallet. This issue arose because the workpiece's 'flat' surfaces actually ended up convex by a couple tenths, and there just wasn't enough contact area around the bolt. I was concerned that the top workpiece might rotate out of alignment during drilling, and so I augmented the bolt with a bead of JB Weld along the rear edges of the parts' excess stock. In retrospect, a sheet of paper between the two might have cured the problem so I wouldn't have had to put up with the overnight epoxy cures. This pairing and drilling procedure was performed on the workpieces for the left/right crankcase pair and then repeated for the right crankcase/cam box pair.
The next step was to pocket the interiors of the three workpieces after indicating their dowels or dowel holes. All three parts were pocketed, and the o-ring grooves were machined on the two workpieces with already finished surfaces. The o-ring groove on the outside of the right half crankcase will be machined later after its surface is finished. At this point its surface contains some excess stock since it might be used later as a machining fixture for the cam box periphery.
The pocketing of the crankcase halves was uneventful, but things got pretty exciting inside the cam box. For the pocketing operations I used a 3/8" three flute HSS steel corncob rougher followed by a 3/8" carbide four flute finishing tool. I ended up destroying two roughing end mills while machining the bottom of the pocket inside the cam box's workpiece. I could tell from the sound that abruptly started that the first cutter had suddenly and dramatically become dull, and so I quickly paused the Tormach before damaging the workpiece. A brand new replacement cutter lasted only a minute or so before it too began making the same noise. This time, however, before I could halt the machine the cutter jammed against an inside wall of the pocket and broke off. Inspection of the broken end showed it had dulled exactly like the first cutter. The depth of cut was only .150", plenty of coolant was being used, and there was no sign of aluminum welding - just a badly worn tip. I eventually had to conclude that the cutter's long stick-out combined with the unsupported 1/8" thick floor of the cam box probably triggered a resonance that, once started, quickly took out the cutter. Since the roughing pass had left plenty of excess stock for the finishing pass, the finishing tool nicely cleaned up the pocket except for a gouge on a top inside corner that occurred when the cutter broke. The gouge didn't extend into the o-ring groove and will be hidden by the cam box cover, and so I'm trying to ignore it.
All three workpieces were then temporarily assembled, and the reference hole for the crankshaft was bored through the entire stack. This hole, by itself, will be sufficient to indicate the locations of the counterbores for the bearings and seals inside the crankcase. The various shafts, bearings, timing pulleys, and gears inside the cam box, though, will also require the the dowel pins as references.
The next steps should include the machining of the cam box cover as well as the crankcase and cam box peripheries. - Terry
