Building the Millennium Edition Sparey .8cc (sic)
Click on images to view them in larger size and more detail.
To qualify as a Millennium Edition, the engine must be commenced before Jan 1, 2001, but completed after that date. Since this is the only one I know of that meets these criteria, it is therefore unique and enormously valuable (these are my rules, go make up your own!) I've made a few departures from the plans for what I hope are sound reasons. First the increase in capacity (as an ex-US citizen, I must hold to the simple truth that there is no problem that can't be solved by more cubes!) Second, the venturi/needle assembly is to an ED design and detachable. My first version needed an insert to make it suck adequately. Making the assembly removable permits ready modification and experimentation. Third, the timing has been changed to a set of figures that Motor Boy Stan Pilgrim says are optimum for long stroke side ports--namely inlet 110, transfer 120, exhaust 136 (cf Sparey's design timing of inlet 122, transfer 93, exhaust 111).
The case can be easily made from the solid in the same way that the Weaver cases were produced. The Woking Precision (tempted to say sic again ) is actually a vaguely crankcase shaped lump of aluminum. This is not a bad thing. It's easy to remove metal, but there's not a lot you can do with a case that's undersize in some direction. In this shot, the rear casting stub has been trued so it can be used as a chucking stub to reduce the front bearing housing and true the front face. Notice how much metal has been removed.
Here I'm using a large radius tool bit to blend the two turned sections. I've dispensed with the quick-change tool post to get the tool more rigid and reduce the tendency for the tool to "chatter". I probably should have supported the casting with a center in the tailstock as well, but I got away with it as shown. You can see here how thick the lugs are, as cast, and how they straddle the case center line. After removing all the metal below that line, more still had to come off the top.
At the same setting, a parting tool is turned on its side and the lathe saddle used as a butcher's shaper to get the lower case radius down to size and concentric with the main journal. Note the tendency of the material to "tear" as the tool tip passes the end. This is a common shaper operation problem, especially on soft materials. No problem as the rear is oversize at this time. Next, it goes over to the mill to bring the sides and lugs to size. Then it is gripped by the front in a collet so the internals can be formed concentricly with the front extension. Finally, it's mounted on an angle plate to bore for the cylinder and bring the deck height down to design dimension, less allowance for the gasket material.
I've rambled on (and on, and on...) about bypass construction before in the Prototype Taplin article, so I'll avoid repetition and self-contradiction by saying almost nothing here. Above we see the bits ready for soldering. The jigging needs to be accurate and secure as the final step is to coax the excess solder to a molten blob on an outside edge, then tap the whole kaboodle on the bench so that the blob departs in a safe direction. Naturally, the jig packing pieces need to be non-ferrous to prevent them becoming permanent features. At the same time, they should be small to avoid heat sink effect.
Here's the end result. I always manage to feel smugly lucky when a bypass soldering job comes out clean like this. But as I don't have a bad one to show, maybe it's not luck after all (and I probably should not have written that--pride commeth etc). A job like this does not need a lot of heat. It probably could have been done with a 60W iron, but I used a little butane torch. This torch also provided enough heat to silver braze the inside of the bypass cover, which no soldering iron would manage.
Crankshafts are simple, mindlessly precise work. They also need you to have the rod ready first so the crankpin can be finished to a close, running fit. I forgot this simple fact and had started to reduce the blank for the pin when I remembered. Rather than break off and make a rod, I pulled down my first engine to use its rod. Big surprise time--the rod has bent! No idea when this happened, but the extra friction of the misaligned axes may explain why the engine is so gutless. Fortunately I'd run the engine earlier that morning and know what it will do (3800 rpm on a Top Flight 8x3.5 wood) so when I make new rods, I can see if there has been an improvement. This piston/cylinder is exactly per the drawing. Dig that crazy length! Ken thinks Sparey was trying to achieve a gas seal by inducing terminal boredom in the molecules as they wend their way south. CAD shows the piston and cylinder can be easily reduced in height by 1/4"--but this kills of the anerexic, stove-pipe, Abe Lincon look that gives the engine so much charm.
Getting close now. At this point, the cylinder still needs lapping, pistons and a rod. The tank is the standard yuppie water bottle cap. The needle valve uses a new innovation. Rather than off-hand grind the needle from music wire, I cut down a darning needle and glued this into the brass carrier. Why didn't I think of this before?! My local haberdashery is a treasure trove of inexpensive needles. Also just visible in this shot is an ED style fuel cutout. This is a brass arm/collar on the feed tube that normally covers a hole in said tube. When rotated, the hole opens and fuel draw stops. Seems like it took longer to braze up and file down that sucker than to make the rest of the venturi! Looks nice though and I sure hope it's worth it. The little tool off to the side is a low-torque box spanner for the 10BA nuts that hold the cylinder down. It was made by loktiting a hex head screw into a piece of mild steel, turning both to some appropriate size and applying a bit of knurling. It's neat to have special purpose tools to accompany an engine.
Seeing as the conrod of my original 0.63cc engine had bent in service, some changes seemed in order. I'm tending to favour the ED "dumbell" design these days (Bert Streigler calls these "dogbones"--a term suggesting a certain lack of precision which is certainly the case in the trial version here). These are made from drill rod that is larger in diameter than the larger ball end. Stubs on either end then allow the piece to be clamped in the milling vice for flattening and drilling without distorting the delicate ball ends. This procedure was described by David Owen in his machining instructions for the "Mate" diesel. The circular cross section gives a slight increase over the average area of the original design (0.001 sqin), although circular sections resist compression loads better than rectangular ones, so I'm told. This "trial" rod (meaning I'm ashamed of the finish and the fact the ball on the little end is slightly out) will get polished, hardened and tempered and tested in the 0.63 before I make the real Millennial Rod.
After writing the previous paragraph, I had occasion to visit Mr Russell Watson-Will (he of Weaver fame) for a look at his 5cc Sparey. Russ related how he'd had the big-end sieze on him, serving as a reminder that hard-on-hard is not a grand combination. Russ has forgotten more about this game than I know, so I went home and made a type 3 rod for the Millennial Sparey from 1/8" 2024-T3 plate. In this shot, the holes have been drilled and reamered at the same setup under the mill to assure parallism and the ends center drilled so the sides can be rounded between centers. This was done with a mild, one degree taper mainly to get rod side clearence in the case/cylinder.
Now we see the Exploded Sparey. Note the finished con rod, with a small engraving to denote the back side (the piston has a similar internal mark). Note also the heroic amount of metal above the wrist pin on said piston. Makes for one hell of a reciprocating mass and gives the stove pipe look to the cylinder. I fitted studs (10BA) to the case on the theory that this may reduce the tendency of the fine thread to strip. The screws are tightened up well before the heads are cut/ground off. I've also increased the wrist pin from 1/16" drill rod to 3/32". The original design called for that 1/16" diameter pin to be fitted with brass end pads--proving Sparey was a sadist. I prefer to make my pins a press fit by not fully reaming the hole in the piston. The pin is rounded on the entry end and left flat on the other both to ease assembly and as a reminder as to which way to take it apart!
Every engine project needs some special purpose tooling. Here we see a jig to hold the cylinder for fin cutting that reduces waste as no chucking piece is needed. This was done as I will need to make more cylinder jackets for the Twin. The gnurled ring is a lock for the cylinder (See the Delong Diesel for the full explaination). The other tools are my "standard" expanding cylinder lap and a mandrel to hold piston and contra piston during honing. There is a hole thru this fixture to assist removal of the parts which are a press, or Loktite fit on the mandrel, depending on how the machining turns out . The other shot shows the stuff-ups. On the left are my attempts at a non-adjustable, tapered lap as favoured by Motor Boy Stan Pilgrim. I gave up after having the cylinder sieze on the lap while being rotated by hand! The other reject parts are a shaft that went too small by a thou, or so, and the first attempt at making a cut-off arm.
Now, the moment of truth. Does it run and how well does it run? After finding the compression setting--I just about ran out of screw length before getting there--the Millennial Sparey burst into life, swinging an 8x4 at over 4000 RPM. Several tanks of fuel were run on this prop at a rich setting with the engine starting easier each time. The needle works extremely well. Some of my off-hand ground needles look good, but are uneven in response probably due to concentricity errors in the off-hand grinding process, so for this engine I the cut off tip of a #9 darning needle (the traditional "Gramaphone Needle" being largely unavailable and a collectors' item anyway!) For a final test, the prop was changed to a 7x6 wood and initially gave an easy 6500 RPM. After several tanks of fuel it was turning an 8x4 glass-filled nylon APC at 7500 and the 7x6 TF wood at 8200 with more to give (I have tached it hovering about 9000 on this prop, but there is a noise I don't like at this speed. This may be due to insufficient clamps on the test rig, but...). Note a small change between the "exploded view" and the running shot; I've cut a psuedo-starting cord grove in the spinner to give a period touch to it.
So there we have the Millennial Sparey .8cc(sic) of 1cc capacity. Construction commenced on Friday, December 29, 2000. First test run on January 7, 2001. Finished weight, including spinner nut, 100 grammes. A good looking, fine running engine, except for a *major* stuff-up in the case that resulted in a case top about 20 thou narrower than the cylinder base (damn, damn, damn...). Still, I like it and will definitely fly it in something, someday.
Please submit all questions and comments to firstname.lastname@example.org