Ohrndorf 5 Cylinder Radial

[petertha]

Then it as just a matter of squeezing the flare die into the tube until the bolt was tight. I dressed the tubing face flat using the tool itself as a sanding guide.

 

[petertha]

Once disassembled from the tool, some cleanup work to remove the grip marks & profile the end. The trumpet flat will reside against a Teflon washer seal in the head port counterbore.

 

[petertha]

With the tube loosely tightened into the head against the seal washer, I was able pivot the tube until it approached the stack tube from the manifold & make a mark indicating where to cut. With a bit of coaxing, they eventually lined up.

 

[petertha]

The installation plan was to first push a section of silicone tubing completely over the end of intake tube, swivel the tube it over the stack, then push the hose down centered over the joint. Not quite sure about hose clamps at this point. Maybe I could modify these automotive wire clamps, but TBD.

 

[petertha]

Basically, the same procedure for exhaust pipes. They are made from 5/16” OD auto brake line. A little bit heavier wall than the intake tubes but the bender had no problems. At some point I may consider a ring exhaust collector that these short pipes would tie into. But at present I lack the tools & know how. Something to aspire to!

 

[petertha]

Before I get on with the pushrod tubes, I’ll back up a step to revisit the bronze lifter/tappet guides because they relate to one another. The plans show a conical profile guide bushing. The idea is to accommodate the pushrod tubes approaching at a 3D angle dictated by the angle/stagger viewed from the front & the different fore/aft position viewed from the side as a function of the intake/exhaust cam plates. The cone also serves to hold them captive. The bushings themselves are inserted into the nose case with Loctite.

 

[petertha]

When I did some test fitting, the fit of the tube to the cone was reasonably OK with some mitering, but I wasn’t keen about metal on metal. Not so much leaking any residue oil, but the metal on metal of a thin wall under normal engine vibration. And then the way the pushrod tubes were to pinned into the base of the rocker box part with a M2 setscrew & a teeny hole into the tube wall to maintain orientation was looking increasingly finicky to my eye. I tried some ideas like heat shrinking the cone. An alternate bull nosed guide to put some kind of disposable tubing boot over the end. Nothing jumped out as a great solution. Note to self, heat shrink is amazingly difficult to stick to metal without some kind of added adhesive. I suppose nothing to fuse like typical sheathing?

 

[petertha]

This ultimately led me to what you see here. I made a new bushing that has 2 captive O-rings. The lower one is a bit wider & thicker; it acts as a snub for the tubing to bottom out on. The upper O-ring self-centers & seals against the tubing ID. After fiddling around with different tubing sizes, I settled on one particular size from the K&S dispenser at my local hobby shop. I’m sure it’s been there from the 80’s so I cleaned him out. Remembering that bronze sets up fast in aluminum, it was Loctite time. Next engine I would make these screw-in or something like that for easier replacement. I’ll show how the upper end of the pushrod tube was similarly O-ringed so the end result is that its captive between rubber on both ends.

 

[petertha]

With the new bronze lifter guides now permanently inserted to the nose case, I made the pushrod tubes. First step is mitering the bottom angle, initially by trial & error. As mentioned, the exhaust tubes are slightly different 3D angle than intake tubes, hence my exotic color scheme. Once my belt sander table & protractor guide was set, then rinse & repeat all the tubes plus spares. With the tubes resting on the bottom bushing O-ring & threaded into a rocker box, rough cut to length with a jewelers saw.

Then I assembled the appropriate coupler fitting on the tube along with the top O-ring, applied some retainer glue to the joint area, then pushed the coupler upward so it had the right amount of compression squeeze. I used some makeshift clothespin clamps until glue cured. The finished tubes seem to sufficiently maintain orientation rotationally because of the coupler miter angle kind of self corrects. Then I carefully filed & finished the tube top to match the rocker box base plus a bit & confirmed the pushrod action functioned without any interference.

 

[petertha]

The pushrod collars were first drilled, OD turned & parted to rough length. I worked out a milling fixture which would hold them at the appropriate rocker miter angle by using dowel pins in the end. One position for the intake, the other for exhaust. I glued the collar stubs into counterbored holes with a fillet of support epoxy. Then milled across them all to a depth offset from the to surface of fixture. Then I heated fixture with torch until the epoxy turns a tan color, gets rubbery & releases. Most of the glue comes off easily but I used the spiral wheel to clean them up.

 

[petertha]

The same fixture can be re-used for multiple part runs after some cleanup. This shows how the upper O-ring fits the underside of rocker box.

 

[petertha]

I forgot to mention & give credit. I learned this epoxy holding trick from Terry Mayhugh pics. He often pots his parts to stabilize them in CNC operations. The only think I'll add is different epoxy glues behave a bit different. The 'good' stuff puts up more of a fight. It goes from brown to black to crud leaving a mess that is hard to remove. I found the bargain brand hardware store stuff turned rubbery & let go easier. Avoid doing this in the kitchen oven, the odor is unpleasant.

CA glue I find is finicky. Maybe because its thinner & will wick into joints often making heat separation more challenging. It also doesn't seem to like continuous planar surfaces curing. I think that's why Clickspring always machines grooves in his tooling plate. I thought it was for overflow but I think it helps with curing. With epoxy you can build up a fillet depending on the part & it has a bit more resiliency with interrupted cutting. But you still cant get too rammy with this kind of fixturing joint.

 

[petertha]

While on the general topic of marrying pushrod tubes to bushings or rocker boxes, I thought I would mention my side effort attempting to mold my own boot. I think that’s what they are called? It wasn’t exactly successful, partly because of the finicky size & what I was trying to achieve. But I’ve always wanted to cast silicone parts from aluminum molds so might be worth showing this for similar applications which may arise on our model engines.

Full size engines obviously have the same issue of pushrods entering their housings at oddball 3D angles. Some shallow, some quite steep. I’ve seen opposed cylinder engines with kind of accordion profile on the ends I assume for this reason. I could not envision any easy way to make this.

Specific to radials, I’ve seen variations of an external rubber boot. I’m not sure if it they are custom molded or segments of straight tubing just deflecting under submission. I tried a silicone tubing cutoff as mentioned. But it was rather thick wall thickness & still didn’t offer me any kind of rubberized gasketing effect under the metal tube, although I thought about positioning an O-ring in there.

 

[petertha]

I thought if I could make a common silicone coupler part with a small internal, integral ledge for the tubing to reside on, it could be used for both the upper & lower ends just by cutting the ends angled to suite. The mold turned from aluminum. The silicone is a low durometer product. Supposedly no degassing or pressurizing required, but I’m a bit suspicious of that. I bought some black color tint & you can see the results. I used the recommended release product but it parted very easily. I can see this stuff being useful for other modelling applications for sure. Silicone is good as an electrical insulator & impervious to methanol & oil (but I didn’t test gasoline).

Mold Max 29NV

 

[petertha]

But for my purposes… Meh. It was kind of a small finicky part to learn on. I guess I could call it a technical success but didn’t like the look of it in the end.

 

[petertha]

The pushrods were pretty simple to make so only one picture to share. I used 2mm diameter O1 tool steel only because it was easy to obtain in metric & I wasn’t quite sure if either end would be hardened. The cam followers are hardened but not the rocker adjuster screws, so we will see. The diameter did matter because the pushrods make a 3D motion inside the tubes & according to my CAD drawings, come close to the edge of the tube wall at certain maximum deflection positions. I could go a little it bigger diameter but not much. The intake & exhaust pushrods are different lengths owing to fore/aft cam plate positions.

To turn the ball end profile, I initially played around with a HSS profiled tool but it had limitations. Even short material stick-out from the collet was enough to see deflection even by feeding the tool axially vs across the stock. The profile wasn’t exactly spherical & harder to control length. So, opted to first part them finish length plus a couple thou, blued the ends with a felt pen & just shaped the ball end profile using a fine file & magnification. I then inserted them into my Dremel collet as far as it would go in & did finishing with paper. Pics showing various stages of completion. You can get a mirror finish but I doubt it will stay that way for long anyways.

In hindsight, because the Dremel can spin up at much higher RPM than the lathe, it might lend itself to grinding the ball end profile on something a stronger/harder alloy like piano wire or similar shafting stock, which I suspect is stronger than annealed state O1.

 

[petertha]

Next up is timing the planetary cam gears & retaining them into final position. A 15T (module 1) crankshaft gear drives a 15T idler gear which is attached face to face with a 10T gear, these two are the idler cluster that run on an intermediary shaft. The 10T then drives the 40T internal (ring) gear which is connected to the cam plates. The CS to ring gear is 4:1 but the cam plates have 2 sets of intake/exhaust lobes 180-deg apart, which yields the 2:1 crank to cam ratio. The cam plates rotate in opposite direction of the CS.

The plans call for using Loctite to retain basically all of the gear surfaces. I was a bit apprehensive about how to make these joints while getting them into timing position & allowing them to cure because access & visual alignment is a bit hidden. I also had this nagging feeling about glue failure one day. For better or worse I decided on a modified path. Loctite the ring gear into the aluminum cam cup because it has a lot of surface contact area around the perimeter. I decided to drill the CS gear with a cross pin running through the CS so the gear could be removed & replaced one day if required (although matching the pin hole would be an adventure for another day). If I did my math right, the pin should be able to take a decent load but I don’t have a good feel of what loads are involved with driving the cam plates. With these CS gears & ring gear now locked into position, that leaves the relative clock positioning of the 15T & 10T idler gears to be rotated between each other to achieve final cam timing relative to TDC & then locked into that position.

CRANKSHAFT GEAR RETENTION
Before drilling the CS gear, I tried to set myself up for future replacement if the unfortunate requirement ever arose in the future. I positioned the CS between Vee blocks in the mill vise & indicated off both sides of the counterweight flats so it was horizontal, the crankpin pointing up. Then I rotated the 15T gear until I could center align between 2 teeth just using a cone shaped tool extended down from the quill pointing into the gear valley. Then I offset a specific distance along the CS axis from the counterweight surface datum, positioning the pin hole on the reduced diameter segment of the gear. With the gear tacked in this position, the hole was spotted & drilled completely through, slowly pecking & clearing, hopefully to keep it straight. The pin itself is made from a HSS drill. Once the pin is inserted through the gear, a brass sleeve was made to cover the hub so the pin can’t fall out. This shows the assembly in progress. There are also some other spacer shims rings in the driveline to make up various clearance distances between components.

 

[petertha]

More CS gear pics

 

[petertha]

How to properly retain the idler gears together face to face caused me some head scratching, mostly due to small dimensional constraints. The smaller 10T gear which has a PD of 10mm & root diameter of ~8mm & runs on a 5mm idler shaft, so there isn’t a lot annular hub material left for mechanical retention. I followed the plans & turned down about half of the 10T gear to 7.5mm OD which basically inserts into the 15T gear reamed out to that same diameter. This becomes the mating surface to which the Loctite would normally be applied.

 

[petertha]

My calculations suggest Loctite should be as strong or stronger than the small diameter axial ‘pin’ key I could accommodate assuming all went well. I did Loctite tests on blanks of 1018 steel mimicking the gears. Face to face bonding was a fail rather as expected. The joint definitely required the step down reduced hub to resist radial shear? It seemed pretty strong just hand wrenching.