Prototype rear induction, Schnurle port Oliver Tiger Cub
| Name | Experimental Prototype Oliver Tiger Cub | Designer | John Oliver |
| Type | Schnurle port, rear induction diesel | Capacity | 1.5cc |
| Production run | not produced | Country of Origin | UK |
| Photo by | Ken Croft | Year of manufacture | unknown |
Notes:
Ken Croft snapped this at a free flight duration comp at RAF Linton-upon-Ouse, near York in October, 2001. He says:The Oliver collectors would go bananas over the Olly. It is the one and only genuine prototype for the schneurle port Cub, and as you see, it is rear drum [or shaft] induction. There is no boss at the front for a front carb, so this was a new 'case pattern. Only one was ever made, and this is it. When the schnuerle Cub eventually appeared, it was front induction like all other production Olivers. Needless to say, this one was not fo sale. [Sorry can't spell schneurle].
Other things to notice are the Cox Tee Dee needle valve and "spray bar" (but not venturi) and the screw-in backplate/drum induction assembly. Now consider that the point where it is fully tight is critically important to engine timing. As thread cutting in the lathe essentially produces a random start (leaving aside multy-start threads), the logical approach seems to be fully machine the backplate, then cut the internal case thread and shave the rear face until the correct orientation of the backplate at the fully tightened point is achieved. Think about it. We can't use a gasket because it would compress over time and screw up the timing, so the mating faces of case and backplate need to be axially and radially perfect for a gasket-less seal. Machining as described (my assumptions, and possibly wrong) maintains the orthoginal properties desired, but when we get the backplate oriented right, the inside face of the backplate is up to one thread turn further "in".
The distance from crankpin end to backplate face is important in setting the case volume and in this case, the backplate face is the bearing surface for the driving flange of the induction drum. Assuming a 32TPI thread, the average error will be 0.016". We could maybe accept this "range of uncertainty" and adjust the drum flange thickness accordingly, but this means every engine would be unique (but what the hell? -- the case/backplate combo already is anyway). If we make the backplate length oversize and try to adjust it later, we will have a devil of a job getting it normal with the drum bore. If we don't finish that bore until after the face has been skimmed, we'll face a similar problem getting the bore concentric with the thread and that has to be cut first. Where I'm going with this is that the arrangement, while perfectly fine for a prototype engine (and I'm certain the Olivers, being superb craftsmen and machinists would have no great trouble achieving exactly what they wanted), this arrangement would be the proverbial SOB to manufacture in any quantity at an affordable price. Hence, it seems reasonable to conclude that we are looking at a magnificant, experimental prototype, not a preproduction prototype.
Late Ademdum Having now thought about it more, Ron's Latest Logical Machining Sequence (replacing his previows stinkin' thinkin') is: finish the case completely and accuratly, then bore the raw backplate casting for the drum (maybe a thou undersize for later lapping). Mount casting on a taper mandrel between centers, inner face to the tailstock and turn to thread OD leaving the nominal backplate flange thickness. Form the normal undercut for the thread run-out (you can't cut a thread all the way into a corner, so threaded type backplates always have an under cut at the sholder so the plate can be tightened). Now cut the thread (this will be over-length because we've not yet faced the inside to the correct length). After the thread is cut, adjust the width of the undercut backwards until the backplate becomes fully tight at the correct orientation. Finally, face back the front (inside face) of the backplate to the correct in-case depth. Still a bunch of work, but easier, repeatable, and gives full control over the location of the engine internals.
