30cc Inline Twin 4-stroke Engine based on Westbury's Wallaby

[Eccentric]

Valve Train



I have been working on some odds and ends in the valve train.

This is a rocker arm cut from 1/4" mild steel. I mount it on a mandrel and machine the sides in the lathe.

The rocker arms are handed, that is the spigot is longer on one side than the other

Included in the above photo is the rocker arm pivot bracket.

Below are some shots of the fabrication of the Tappet guides cut from bronze.

Using a tap in the lathe, quicker and easier than cutting threads.

A threaded fixture used to cut the hex heads on the tappet guides.

Finished tappet guides, they will thread into the four yet unthreaded holes in the crankcase



Below is the final operation in the fabrication of a valve guide.

I am using a split bushing to mount the valve guide in the mill vise to prevent marring and deformation while drilling the side hole.

Completed valve guide.

The valve guide is test fit into the head, the small 45 degree valve seat is highlighted. This was cut using the cross slide set to 45 degrees in the same setup used to drill and ream the internal holes.

 

[Eccentric]

Wallaby Valve

This morning I made a valve for my 4-stroke Wallaby.

The valves are machined from 303 stainless steel. I started with a fail. I had too much stick-out when I tried to cut the retaining ring slot.

I had success with the slot by cutting it with no stick out at all:

Then I continued machining in a couple of smaller segments

This works so much better than the last time I made valves, turning the whole length on centers.

I tested the quality of the seating of the valves using a technique I learned from Terry Mayhugh. My method is much simpler and less scientific. I pull the vacuum through the rear of the valve cage, past the stem of the valve. I seal the side hole and then take time measurements to see how long it takes to bleed down in different configurations. With the valve held slightly open it takes 3 seconds to bleed down--this is my baseline. This measures the leakage between the cage and the valve stem. Lightly holding the valve closed with my thumb the time jumps to 30 seconds to bleed down, this represents the valve spring holding the valve closed. If I do not press on the valve with my thumb and just rely on the vacuum to seal the valve it takes 15 seconds to bleed down.

Since I pull the vacuum down past the valve stem, the bleed down times do not mean much in themselves and cannot be compared to times others may get. They do indicate that the valve is sealing against the cage and that there is not a misalignment or major flaw in either the seat or the valve. The quality of the seal really can't be determined by this simple method, just that there is a seal.

I feel confident that I have the techniques down now to build the balance of the valve cages and valves. Back to it.

 

[Eccentric]

Flywheel

The most important aesthetic criteria I have for a flywheel is that it runs truly true. I want my flywheel to appear to be a stationary blur, if that makes sense. So I strive to machine all surfaces in one setup, and a setup representative of how the flywheel mounts to the engine. As far as the functional design, I follow George Britnell's advice: "The flywheel in total doesn't need to be heavy but rather just the weight at the rim. Most flywheels (hit and miss type) are spoked. This is to reduce the overall weight while still providing the force needed to overcome the compression. If the flywheel is too light the engine will still run but it won't run slowly. If the flywheel is too heavy the engine will run but won't respond to throttle openings properly."

I liked the look of my finished gears with the six lightning holes, so I went with that design here. I also like to use a separate split taper collar as opposed to tapering the crankshaft. In the picture below I have highlighted the taper collar, unfortunately it is almost the same color as the flywheel and hard to discern. I use the compound to cut the 10 degree inclusive angle on the flywheel and split taper collar, and use the same setup to do both to insure the angle is identical. I make the taper collar first, then leave the 5 degree offset in the compound, machining the flywheel using just the cross slide until the internal taper is cut.

(Bonus question,"What is wrong with the pistons in the above model?")

Design of the flywheel and split taper collar.

I face the front of the flywheel to create a relatively flat side that I then mount in the centering 4 jaw chuck. I drill the center hole to the small diameter of the taper collar, then cut the taper into the flywheel, using the collar to test fit until it is just even with the faced surface (back side).

I fabricate a mandrel to the same dimensions as the crankshaft. Then I mount the flywheel just as it will be mounted on the engine.

At this point I machine the complete flywheel in this one setup. I even take a tiny skim off the back of the flywheel as deep in the X direction as I can without hitting the collet nut, to true this edge as well.

Then it is over to the mill to drill out the large lightning holes. I drill these in five increments not including the center drill. I use parallels underneath to support the work piece, moving them to insure they don't get hit with the drill bit. I use aluminum packing to protect the outer surface.

Here are a couple of pictures of the flywheel test fit on the engine.

The flywheel ended up weighing 19 ounces. I still need to balance it and I think I would like a larger chamfer on the center mounting spigot.

 

[Eccentric]

First Assembly

I completed the first assembly of all of the major components on the Wallaby project. This felt like a major milestone. I ran into a minor problem with my CAD model that was realized in the actual build- the pistons interfere with the crankcase at the very bottom of the stroke. If I had done a more thorough study of the CAD model I would have discovered this. So my crankshaft only turns through about 350 degrees.

The two following photos illustrate the area of interference with the second photo being an extreme close-up of the actual interference area. This was very easy to fix as I clamped the crankcase in the mill vise and took off a little extra material. I also corrected the CAD model and drawings, of course.

I have also been working on some of the smaller details such as drilling the oil gallery holes in the crankcase. Oil is delivered under pressure from the oil pump to the side of the crankcase where it is routed to the end of the cam shaft and to the middle main crankshaft bronze bearing. Oil will then travel down the cam shaft and through small holes to oil each of the camshaft lobes. Oil delivered to the crankshaft middle bearing will travel through oil galleries in the crankshaft to the two big end bearings of the connecting rods.

I have also been in a production mode for the rest of the parts in the valve train. I had to fabricate multiple valves, valve cages, rocker arms, push rods, and rocker arm bearings. I slip into a different way of working when I am building multiples of something that I have already developed the process for fabricating. Developing these process is part of making the original prototype where it can take me several parts to get the steps down. When I go into the production mode I will often perform each operation on all the parts before moving on the next. For example, I might complete the lathe operations for a set of parts before I move to the mill for milling operations. One thing I have learned, I should make more than just the right amount of parts; I should make an extra. There are a couple of parts that I made a mistake on, now I have to repeat the whole production process for a single part. You can almost fabricate multiple parts in the same time as a single part.

I also worked on the display stand for the engine. I need to figure out the other items such as radiator, fuel tank, electronics, etc. I have seen many fine examples on this forum.

 

[Eccentric]

Tappets

Today I got back to making parts for the Wallaby, these are the tappets. They were a pretty straight forward lathe job, I used a 1/8" ball end mill to make the pocket for the 1/8" push rods. These are made from 5/16" drill rod, the shank is turned down to 3/16" for a smooth fit in the tappet guide. I have not decided whether I will harden them or not.

I have been distracted from the Wallaby by work in my new design, a 97 cu in Offenhauser Might Midget engine. I have been toying with the idea of creating a YouTube series on its construction. Below is a video of the machining the inside of the Wallaby piston on a CNC router. This video was the second I produced and I am still on the fence whether I want to get into this new way of sharing my projects. Let me know what you think.




Thanks, Greg
GregsMachineShop.com

 

[Charles Lamont]

Have you checked how the tappets ride on the cams? If the faces are only 3/16" diameter, I think it is highly likely that that you will get corner contact.

(Later, for the record: this post was based on a now corrected typo in eccentric's post above. But for that typo, I probably would not have noticed the problem)

 

[ownthesky2010]

When I built my flat twin engine I also machined the internal piston features in my CNC mill. The easiest way was to machine the piston from square and leave about 3mm of excess square stock on the crown to use for indexing and location and then just face it off after.

 

[xpylonracer]

Greg

If the push rods are 3/16" diameter wouldn't it be better to use a 3/16" ball end mill to make the pocket in the tappet or are the push rods tapered to 1/8" at the end ?

xpylonracer

 

[Charles Lamont]

Out of interest, I have now had time to do some calculations on the cams. If the cams are made to Westbury's specification, the inlet cam would have a nose radius (which he does not specify) of 0.040" , and to ensure that the the full 3/16 width of the cam was always in contact with the flat face of the tappet, the latter would need to be 13/32" diameter. Allowing the contact line length to drop to 3/32" at the extreme would mean you could get away with a tappet 3/8" diameter.

Westbury's specified 5/16" is too small, and this is an error he made on several engines. Mind you, he would have been designing with pencil and paper methods, while I arrived at the figures above with a combination of pythagoras theorem and CAD. The whole lot could have been calculated, but the CAD saved me a bit of trigonometry.

 

[Eccentric]

Sorry, I misspoke. The tappets are made from 5/16" drill rod, the shank is turned down to 3/16" for a smooth fit in the tappet guide. I used a 1/8" ball end mill to make the pocket for the 1/8" push rods.



Charles, what you describe about the size of the tappet head is very interesting. Below is a cutaway of the engine:

And below is a close up of the cam lobe contacting the tappet as the lift begins/ends. I can see what you describe, the tappet head is too small and the cam lobe is pushing against the edge of the tappet, not the face.

And with the valve full open, the tappet needs to rise into the hole above it, which now is .406". So your suggested 3/8" should fit at full lift.

Now with your suggested 3/8" top:

Closeup

Even the intake lobe with the smaller cam lobe nose radius pushes on the bottom of the tappet, not the side.



Below is a closeup cutaway on the other axis and it can be seen with the larger tappet face, we are still OK.

Thanks Charles, I never would have noticed this. Well I guess I'm back in the workshop making new tappets.

 

[Mechanicboy]

And the cam is in off side to tappet to prevent wear out the cam and tappet? With cam off side to tappet help to rotate the tappet and lasting long time.

 

[Charles Lamont]

As you see, I have updated my post with a drawing. The exhaust cam will have a larger nose radius, and that makes the situation easier.

 

[CFLBob]

I have been distracted from the Wallaby by work in my new design, a 97 cu in Offenhauser Might Midget engine. I have been toying with the idea of creating a YouTube series on its construction. Below is a video of the machining the inside of the Wallaby piston on a CNC router. This video was the second I produced and I am still on the fence whether I want to get into this new way of sharing my projects. Let me know what you think.

If it's not too late to open this can of worms, I tend to like videos, but you're spot on about needing a whole bunch of skillsets to pull off good ones. Good video editing software is a must. As is writing a script and reading an audio track in once you get the video sequences you like. You need a place with both good acoustics and good lighting. There's a guy whose stuff I've watched (Russtuff?) who generally doubles or triples the video playback speed for things like hollowing out that piston (as you did when you showed the cutter digging in at the bottom of its cut).

I've put up many videos of some operation or another but have yet to take the plunge on a microphone and recorder. Because of that, they'll be like one pass around a part or something simple like that. Nothing like a complete look at how to make a particular part.

Good videos can condense the information: the old "a picture is worth a thousand words" line. Getting good videos isn't trivial. Too many kids just turn on the camera. All the hard stuff needs to be planned in advance and I think it's likely to be a pile of work.

 

[awake]

If it's not too late to open this can of worms, I tend to like videos, but you're spot on about needing a whole bunch of skillsets to pull off good ones. Good video editing software is a must. As is writing a script and reading an audio track in once you get the video sequences you like. You need a place with both good acoustics and good lighting. There's a guy whose stuff I've watched (Russtuff?) who generally doubles or triples the video playback speed for things like hollowing out that piston (as you did when you showed the cutter digging in at the bottom of its cut).

I've put up many videos of some operation or another but have yet to take the plunge on a microphone and recorder. Because of that, they'll be like one pass around a part or something simple like that. Nothing like a complete look at how to make a particular part.

Good videos can condense the information: the old "a picture is worth a thousand words" line. Getting good videos isn't trivial. Too many kids just turn on the camera. All the hard stuff needs to be planned in advance and I think it's likely to be a pile of work.

As someone who has made a grand total of two (two! count 'em! two!) tutorial videos, I agree with everything CFLBob says above. I will add one other thing: consider carefully whether the videos actually need to show your face. In general, a "talking head" is less appealing to me than other options (e.g., a voice-over of clips or even stills of the machining process). But IF it is important to show your face, consider investing a few bucks in a simple teleprompter. It is astonishingly easy to put one together, using anything from cardboard to 3d printing, that works very well indeed. The key ingredients are a piece of clear glass or plastic at a 45° angle, through which the camera looks, but which reflects the image of a tablet or even a cellphone. Here are a couple of pictures of one I made, that I use with a surplus Android tablet and the nanoPrompter app in my fancy (?), high-tech (?!?) home studio:

 

[Eccentric]

Oil Pump

Made the oil pump today. Both the oil pickup from the sump and the oil gallery fitting in the crankcase are on the same side of the engine. So I had to do some creative oil routing in the pump so the fittings could be on the same side of the engine to simplify the oil tube routing.

I fly cut a piece of stock on both sides to the correct thickness, then cut the inside and outside features on the CNC router.

When I made the oil pump gears I beveled the edges of the gears, not a good idea for the oil pump gear as you do not want any leakage around the gear corners. Fortunately the gear were a little to tall, so I put them in the oil pump housing and fly cut one side....

then flipped the gears over and fly cut the other side flush with the oil pump housing.

The gasket between the oil pump housing and the oil pump cover should provide enough space for smooth operation.

Then I reamed the shaft holes...

Then I milled off the oil pump housing from the base stock. Drilled and tapped the holes for the oil line fittings.

Still need to deburr the holes with the counter sink tool, but I think the pump will pump.

The gear with the shaft turns counter clockwise, delivering oil from the bottom fitting to the top.

 

[Eccentric]

Wallaby – Distributor​

The first run of the Wallaby was done without a distributor, I used two separate ignition systems triggered by the same hall sensor on the fly wheel firing both spark plugs simultaneously. This was just an expedient to get to first run, but now I want to complete the distributor and use a single ignition system.
I planned to machine the distributor body, cover and rotor from Delran, and 3D printed the parts first as prototypes. But the quality of the parts are good enough to use. The distributor body is separated from the engine by the timing case so should not get too hot.

The distributor design is straight forward, the rotor mounts directly on to the end of the camshaft and there are a number of precision brass parts that make up the electrical distribution system.

An exploded view of the distributor is shown below.

I first printed the plastic parts on the 3D printer. Below is the Distributor body still adhered to the 3D printer bed. I used ABS filament which does not get soft until 105 degrees C.

Now its time for the brass parts. I made the center rotor contact by first drilling the three mounting holes and rough cutting the part from a 1/16″ sheet of brass.

Then I mounted it on a prepared fixture and turned the outside diameter to spec on the lathe. Finally the flat was machined back on the mill.

The three mating contacts were first turned on the lathe, the center hole drilled and countersunk.

A fixture was made that would allow the inside and outside diameters to be turned on the lathe. First I turned a bit of scrap aluminum with a center hole for a zero reference point on the lathe, then transferred it to the mill. On the mill I centered the part and placed mounting points for the contacts equidistant from the center reference point.

Then back to the lathe where I first turned the outside diameter.

Then bored the inside diameter.

Below is a test fit of the rotor contact.

Looks good.

Below are the finished contacts:

The contacts fit well in the distributor housing.

The wire terminals were then machined. I drilled and threaded the holes in the brass rod first.

Then turned the outside diameter and drilled the small hole for the spark plug wire.

I am using 22 Gauge wire with 40KV rated insulation for the spark plug wires. Below is a test fit.

The terminals were then soldered to the wires.

Above is the final installation on the engine. I am pleased with the smooth motion and tight tolerance of the distributor contacts. We will see how the 3D printed parts hold up.

 

[xander janssen]

Very nice. How big is the gap between the contacts?

 

[Eccentric]

Xander,

I went for .005" gap, but in the real world, the rotor is not perfectly restrained in the distributor housing and the center rotating contact is not perfectly concentric with the outer fixed contacts. I found that the center contact was touching one of the contacts and had about .010"-.005" clearance on the other two contacts. But there was no resistance when turning rotor and the engine ran fine. It is OK if there is a small spark gap between the contacts.

 

[Eccentric]

Below is a short video of my 30cc Wallaby running. The Wallaby is my second IC engine build and my first to actually run successfully. I developed a set of plans for the engine from a construction article by Edgar Westbury published in "The Model Engineer" in 1962. Mine is built completely from bar stock and thus does not use any castings. The design changes include the use of ball bearings on the crankshaft and camshaft, but most changes were to ease the manufacturability of the components from bar stock.




I still have a few issues to iron out. The aquarium pump I use to circulate the coolant stops once the engine starts running, and when the engine quits, the pump resumes running. It is a DC brushless drive pump and I assume that my ignition system is interfering with the internal drive electronics of the little motor.

The DC brushless fan used on the radiator runs fine.



Most of the finishing work left is on the display stand and the packaging of the support equipment such as coolant system and ignition system. I designed my own electronic ignition using an Arduino UNO and I would like to package it nicer.

I created two versions of the plans, one assumes the use of a CNC router on some parts, and the other simplifies a few components to allow construction with only a mill and a lathe. If anyone is interested in building my version of the Wallaby, please feel free to contact me.


Greg

 

[jamesmattioli]

Below is a short video of my 30cc Wallaby running. The Wallaby is my second IC engine build and my first to actually run successfully. I developed a set of plans for the engine from a construction article by Edgar Westbury published in "The Model Engineer" in 1962. Mine is built completely from bar stock and thus does not use any castings. The design changes include the use of ball bearings on the crankshaft and camshaft, but most changes were to ease the manufacturability of the components from bar stock.




I still have a few issues to iron out. The aquarium pump I use to circulate the coolant stops once the engine starts running, and when the engine quits, the pump resumes running. It is a DC brushless drive pump and I assume that my ignition system is interfering with the internal drive electronics of the little motor.
View attachment 139896

The DC brushless fan used on the radiator runs fine.



Most of the finishing work left is on the display stand and the packaging of the support equipment such as coolant system and ignition system. I designed my own electronic ignition using an Arduino UNO and I would like to package it nicer.

View attachment 139897


I created two versions of the plans, one assumes the use of a CNC router on some parts, and the other simplifies a few components to allow construction with only a mill and a lathe. If anyone is interested in building my version of the Wallaby, please feel free to contact me.


Greg

Ciao, mi piacerebbe fare un motore come il tuo, chiedo se hai tutti i disegni per usare tornio e fresa , grazie mille