Push Rods
The M5 valve push rods are made from 3/32" drill rid, 1.5" long (though an original Morton part I measured was 1.476" long). As the motion of tappet, push rod and rocker arm are not linear, the ends need to be spherical so the different angles assumed by the 3 parts can be accomodated. The only practical way I could figure to make these parts was using a "form tool"--a piece of HSS ground to compliment the shape required. This shot shows the tool, and to the right, a short trail piece showing two (of several) attempts at the desired shape, along with the lifter the end must be able to fit snugly inside with 5 to 10 degrees of conical movement available. At the bottom of the picture is an original Morton valve rod and above that, my first attempt.
Even though I have a Quorn Tool and Cutter Grinder, setting up a diamond dresser to form a 3/64" Radius on a thin grinding wheel did not work too well. The final shape was made and polished using a cut-off wheel in the trusty Dremal hand tool, taking very gentle sweeping passes around the rough shape until a piece of 3/32" drill rod seemed to fit neatly in when viewed under low magnification. Form tools induce much higher cutting forces than normal lathe tools, so the overhang from the jaws of such a small diameter work piece must be kept minimal to minimise the chances of the part being bent. This shot shows my trail setup in a 3/32" collet. It works well provided the feed is kept very gradual and is never paused with the tool in contact with the part as this quickly work hardens the surface. I've found that applying cutting oil to form tools like this also stops them cutting and immediatly induces work-hardening too (at least that's what I think is happening). At any rate, the finish is good, buffs up well on a Scpthc-Brite belt and appears more spherical than the Morton part (though in fareness, that M5 rod comes from an engine that has run, so it may be worn). The rods will be left in the soft state.
Well, just to show no plan is perfect, here we see the form-tool casulties. The corresponding pile of survivors was a miserable seven (from twenty candidates). Advice from wise machinists (Eric Offen and David Owen) was to reduce the top rake to zero and make sure the tool was exactly on center height. Alas, the reject rate was no better, so with stocks of 3/32 drill rod rapidly diminishing, a better plan was most obviously needed.
Ken Croft suggested a spherical turning attachment and as I'd made one to the Radford/Thomas/Hemmingway design some years ago to make ball handles for the Quorn, I decided to see if it could make real small balls. The photo here shows the general design: the tool bit cuts tangentially and is rotated in an "up and over" motion to form the sphere. The tool carrier is adjusted by a fine lead screw (turned by the Allen key visible in this shot). The problem is that the head requires good clearence around the material and 3/32" drill rod with that much over-hang would never stand the cutting forces.
The solution was simple enough. A 3/8" stub holds the rod supporting it rigidly with only 1/4" of material protruding. After a couple of trials on the many failures to proove the technique, the remaining 14 rods were machined with no failures--even though the process was slower than using the form tool (when it worked!). The seven rods on the right were turned this way. The others are mixed and it's hard to tell which is which. Anyway, I now have two M5's worth of push rods.
Rocker, Big End and Wrist pins
All these parts are made from 0.125" diameter water-hardening drill rod (called "silver steel" in the UK because of its appearence--there's no actual silver in it at all). I thought these would be just simple, boring repitition work. How wrong can you get.
Both the rocker and big end pins are designed to screw in on one end. The thread is 2-56, so the diameter must be reduced and the thread cut to a undercut sholder. To ensure accurate alignment, the female threaded end is counter bored 0.125" for a short distance so that both ends of the pin are supported in a reamed hole. The threaded length is only 0.095" long (under cut 0.016" wide at the sholder). Very fiddly--watchmaker stuff. As this is about the same length as the tapered lead-in on my 2-56 die, first attempts were made running the die on backwards using a tailstock mounted die holder. It was Yet Another Disaster--eight failures in nine attempts. This shot shows, left to right, a failure, an uncut blank and the single success. I'm thinking of changing the design to use full floating pins, retained by E clips at either end. Grrr...
After the rocker pin fiasco, I felt I needed a success or two to maintain momentum and enthusiasm for the project, so the next (and simple) task chosen was the wrist pins. These are just 0.515" lengths of drill rod, drilled through 1/16" (and am I ever glad I took a couple of days out to make that lathe spindle back-stop!)
The wrist pins are full-floating, so their ends should be smoothly rounded to prevent scoring of the cylinder liner, but the pins are drilled through (for lightness, I think), so rounding is not practical. Although the drawings don't show them, I decided to make little brass end pads for the wrist pins. These will be glued in with LockTite and profiled. Ten pins, 20 pads, one hour: 3 minutes per pad that includes a tool change from brass turning tool to thin (0.020") part-off tool. Just as well I don't do this for a living!
The pins that retain the slave (or link) rods in the master rod are called link pins. Morton designed them the same way as the rocker pins with a 2-56 thread to an undercut sholder. As there have been instances of these unscrewing and jamming up on the crankshaft counterbalance cut-outs (I shudder to think of the damage at 5000 rpm), Bruce Satra suggests making them full floating as well, retaining them in the master rod with thin steel washers--the whole thing sandwiched between front and rear crankshaft webs. So that's how they've been made here. The cunning retention method for the fully floating pins is shown in the Link Rods section. This photo shows, as usual, two engines worth of pins.
Valves
Valves are not something I've made before and the Morton valves are tiny! They are 0.797" long with a 0.092" stem and a 1/4" diameter head, 1/16" thick. The groove for the retaining clip is 0.020" wide and 0.010" deep. The plans call for 4130 steel, hardened to Rockwell "C" 33-38. Lacking a heat-treat oven, hardness testing gear and being worried about the stems warping, I decided to make them out of stainless steel as seems to be "modern" practice. The plan was to make one valve to experience all the processes involved (ie, all the things that can go wrong), then devise the mass production process (I need 20 of the little buggers, plus some spares). Next, do a trial of the "production-line process" -- say 2 valves, and finally, make 20 more!
The plan, for a change, worked. To conserve material, the production run will be two batches of 10 made from five 3-3/4" lengths of 1/4" stainless rod. Working on each end produces 10 valves, so after being cut off (in the band saw), the remaining length of rod (about 4") will give another 10 valves and enough in the middle to grip in a collet. Final facing, trimming to length and retaining clip grooving will be done collectively on the 20 valves. The photos show the basic steps. Batched operations were governed by tool, or set-up changes:
- Set over compound 30 degrees; perform Step 1.
- Reset compound to 0; perform step 1a (no explicit photo).
- Change to roughing tool; perform step 2.
- Change to finishing tool; perform step 3 (a, b and c)
- Set over compound slide to 45 degrees; perform step 4
- Reset compound slide, fit 3/32" collet and facing tool; perform step 5
- Fit each part from rear of collet; do step 6
- Change to grooving cutter; perform step 6a