Offenhauser Mighty Midget Racing engine

[Eccentric]

Rear Motor Mount and Bell housing

Today I turn my attention to the rear of the engine. I started to work on the bell housing/clutch assembly, but then realized the bell housing attaches through the rear motor mount and extends beyond the sides of the crankcase. So I start with the motor mount. It is relatively simple as it is a single plate of aluminum that bolts to the rear of the engine and to the frame of the race car.

I import the image of the rear of the engine, scale it and import it into my assembly. Then I lay reference datum lines and take measurements. I build the rear motor mount model and check it in the assembly.

Then I do the same thing with the bell housing /clutch.

This is where the rear of the engine is for now.

As per some earlier discussion with Peter, I changed the angle of the valves from 44 degrees to 45 degrees. This change only impacted two components, the green head and the magenta timing gear tower. I also worked on the purple front cover (below), I found I had made it too thin. I added some detail to the magenta timing gear tower to blend it into the front cover.

I now have the basic structure of the engine and can start thinking about the internal components. I will also start normalizing dimensions and thinking about where the critical dimensions are and what tolerances need to be. Up to this point in time I have only been focusing on putting the lines in the correct spot and have not worried what the dimensions ended up being. For example, if a dimension is now .12734" , I will determine if I can change it to .125" or even .12". this process has to be done systematically as these dimensions will ripple through the entire model.

 

[petertha]

Hmm now I'm even confused myself. Turns out I have the book you referenced, although my cover photo is different & its been a while since I thumbed through it. I assume that's what you are referencing?
https://www.amazon.com/Offenhauser-Legendary-Racing-Engine-Built/dp/1626540411?tag=forumyield-20
On page 54 it says the 255 used a slightly larger block and case than the 270 but the same 72 degree valve angle, early 220 and 255 engines are difficult to tell apart.... If 72-deg valve angle means the same as what we have been wondering about (88 vs 90) that's quite different again. Does the name 'Midget' correspond to a specific model number (like 255) or more of a generation series they produced?

I noticed quite a bit of variation in piston crown shape amongst the various engine models, although the conical combustion chamber seems to be consistent hallmark (closely aligning to valve seat plane?). I'm guessing maybe different aspiration or maybe even fuels used at the time? It would be nice if a detailed matrix of engine specs/features were provided, but unfortunately not present in this particular book. Once you start introducing real world dimensions & components like valves, cages, ignition plugs & target CR.. I'm sure the design iteration will take many twists & turns. All part of the fun. One thing that stands out in my mind when mucking around with radials is how thin castings would scale. Real world cast cooling fins of 0.050" thickness x 2" deep turn into ten thou feeler gauges at 1/5 scale LOL

 

[portlandron]

If you don't have a set of, or want to make the gear cutters, you might consider going with the 0.5 Module gears rather then the 48DP.
I needed to make gears for Rudy's Steam Tractor and tried to find 48DP gear cutters. Ended up going with 0.5 Module instead. Got a complete set for the price of one 48DP cutter. Granted the are Chinese but work fine for the brass gears.

 

[Eccentric]

Revisit Timing Gear Train

I spent today revisiting my choice of 48DP gears for the timing gear train. 48 DP means that a 48 tooth gear will be 1 inch in diameter (really the theoretical pitch circle is 1 inch, the tangent circles where the gears mesh). The alternative is to use Module .5 gears. .5 module means that a 48 tooth gear will have a pitch circle of 24mm(or .9449 inches). That is, 48 teeth * .5 Module = 24 mm pitch circle.

Ron brought up a good point that made me question my original decision. In my earlier post about the gear train, I laid out nine criteria for the design. Ron is suggesting another one: Use gear cutters that are readily available and inexpensive. I did an internet search and I also found that Module .5 gear cutter sets are more available and less expensive (if you are willing to buy from China or Новосибирск in the, Russian Federation) than 48DP gear cutters. I make my own gear cutters, but if another builder wants to buy gear cutters or gears, metric gears may be a better choice, even in the US.

It was actually quite straight forward to pull up my old 48DP gear study and overlay a couple alternative Module .5 analysis.

My previous solution was to use 5 gear types, 16, 30, 32, 40, and 56 with a diametrical pitch of 48. By using Module .5 gears I can still get away with 5 gear types, 18, 28, 36, 40, and 64. teeth. I like this solution for the following reasons:

• .5 Module Involute gear cutters are less expensive than 48 DP gear cutters.
• Still have the same number of gear types.
• The crankshaft Pinion will have more gear teeth, 18 instead of 16 and will be ever so slightly larger.
• The basic geometry is not impacted, for example there are still 4 gears between the crankshaft pinion and the camshaft spur gear.

The down side is the magneto drive train is a little close to the edge of the gear tower and the engine is 1/16" shorter.

The other change this will drive is the gear shaft diameter and metric bearings. I had settled on 3/16", but metric gears in this size typically use 5mm shafts, so instead of a 3/16" X 3/8" X 1/8" bearing. Bearings with a 5mm ID commonly have a 2.5, 4 or 5mm thickness. The timing gear tower was designed around the 1/8" bearing thickness, so this also needs to be adjusted. Running some numbers, a gear with a width of 5mm and two bearings with a width of 4mm each will fit nicely in the timing gear tower. So MR115-2RS Bearings with 5mm ID, 11mm OD and 4mm thickness should work.

Original 48 DP option above

.5 Module solution with 6 gear types above, clean magneto gear mesh

.5 Module solution with 5 gear types, not optimal magneto drive gear mesh.



I have three solution:

the 48DP option that has:

• 2 ea 16 tooth gear
• 1 ea 30 tooth
• 4 ea 32 tooth
• 2 ea 40 tooth
• 2 ea 60 tooth

A .5 Module options has:

• 2 ea 18 tooth gear
• 1 ea 32 tooth gear
• 5 ea 36 tooth gear
• 1 ea 48 tooth gear
• 1 ea 56 tooth gear
• 1 ea 64 tooth gear

A second .5 module solution with fewer gear types:

• 2 ea 18 tooth gear
• 1 ea 30 tooth gear
• 5 ea 36 tooth gear
• 1 ea 40 tooth gear
• 2 ea 64 tooth gear

I spent way too much time on this today, I think I will noodle on it a bit more before committing one way or the other.

 

[CFLBob]

The down side is the magneto drive train is a little close to the edge of the gear tower and the engine is 1/16" shorter.

I'm way out of areas I have any experience in, but I'm not sure I see where this comes from in your drawings. If this results from the number of gears with the smaller pitch diameter can't the physical size of the engine remain the same? It would change the positions of features, the shafts attached to those gears, but that 0.9449 vs. 1.0000" diameter doesn't need to impact every other dimension, does it?

 

[Eccentric]

Bob.

I am attempting to maintain the distance between the crankshaft, water pump, magneto and the camshafts. When ever I change the diameter of the gears these dimensions are impacted. I am also fighting the fact that the camshaft gear has to be twice the diameter of the crankshaft pinion, right now I think the crankshaft pinion is too small and the camshaft gear is too big. Also as can be seen in the photo below, the gears driving the magneto should be in a stright line.

The original Offy used gears in two planes to solve the issue with the size of the crankshaft pinion and the camshaft gear. The gear pair in the gentlemans hands helped reduce the ratio of the camshaft/cranksaft.

You are right, the difference between 1" and .9449" is very small and in the end I could ignore it and my diminsions would be off a bit. But there is a stackup where all 7 gears are off by a bit and the error adds up.

I am taking a break from this timing gear train design and will noodle on it in my sleep.

Happy New Year Everyone!

 

[Eccentric]

Crankshaft

I need a break from timing gears, so I am going to turn to the Crankshaft.

I import a scaled drawing and build a crankshaft on top. Notice in the drawing that the two center cylinders are pushed towards the two ends of the bock to make room for the center crankshaft bushing. Also the conrods are offset from the center of the crankshaft journal as well, again to make space for the crankshaft bearings.

As shown below, In my Wallaby engine I used two ball bearings in housings mounted to the end of the crankcase and a bronze bushing mounted in the center of the crankshaft screwed to the upper crankcase half. The ball bearings do not need oil delivered to them, but the center bushing does. This scheme worked well and I am going to use it again and see if I can fit everything in.

My Wallaby engine and its crankshaft and bearing arrangement​

I need to add some features to the crankcase for the crankshaft. These are shown in the picture below:

Then I create a sketch to define the dimensions of the crankshaft. Not sure if this diagram is going to make sense to anyone else but me.

I am offsetting the connecting rod journals on the crankshaft 1/16" to allow for more room for the crankshaft bearings. That is, the center of the connecting rod journals are offset 1/16" with respect to the center of the cylinders. If I can get rid of this later, I will.

I draw up the crankshaft according to my napkin sketch and drop it in the crankcase. I have to clear some material inside the crankcase to clear the crankshaft webs. The bearing at the back, flywheel end, looks good. The bearing at the front of the engine is going to be a challenge because I also have to deal with the timing gears at the front of the engine.



I am designing a model of the 97 cu In Midget Offenhauser. It was the larger 255 and 270 cu inch Offys that were used in the Indy cars. The MIdget Offy was used in the Midget Racers of the 1930s and 1940s. I guess they still race them today, but that was the heyday.

 

[Eccentric]

Gear Train Final(?)

It is important to nail down the gear train as it dictates the engine geometry. The gear mesh tolerance has to be perfect and the distance between the gears defines the position of everything else. I think I finally have a solution I am happy with.

These are the Constraints I laid out earlier for the gear train:

• I would like to use a "standard" gear pitch. The smallest that can be considered standard is either a 48DP or a .5 Module (which is about 50DP). Use .5 Module
• The camshaft has to turn 1/2 the speed of the crankshaft. Yes
• Would like to have a single plane gear train for simplicity. The Mighty Midget has a dual plane gear train as can be seen in the photo above. Use Dual plane gearing, that is, use a cluster gear to allow the use of a larger pinion on the crankshaft, 32 tooth instead of 16.
• The camshaft gear, as well as all of the others, has to fit in the gear tower housing Yes
• The crankshaft pinion should have as many teeth as possible. 32 is a good number
• There should be as few bearing sets as possible. There are a lot of bearings, but I think this is the minimum possible.
• A plus would be to have as few gear types as possible, also "standard" tooth counts would be preferred. Yes, 4 gear types and they are standard: 28, 32, 36 and 56 teeth.
• There are other mechanical constraints such as the 88 degree inclusive angle between the two valve banks and the distances between the camshaft and crankshaft. All met, went with 90 degree inclusive angle on the valves
• Also the magneto needs to turn at the same speed as the crankshaft. Yes
• Use an inexpensive involute gear cutter set. Yes, used .5 Module. these are inexpensive and readily available. so are the gears if one want to go that route.

Below is the final(?) solution:

The Camshaft gear can be no larger than 36 teeth, otherwise it would be too big for the gear tower. If I used a single plane gear train the crankshaft pinion would have to be 1/2 of that, 18 teeth. Not the end of the world, but the crankshaft would be pretty small and I had a hard time defining a gear arrangement that met all of the dimensional requirements. By using the cluster gear set as shown above, I am able to keep the camshaft gear to 36 teeth and have a crankshaft pinion of 32 teeth. In the picture above you can see I can have a larger crankshaft diameter at the pinion. The engine geometry worked out spot on with the .5 Module gear pitch.

 

[petertha]

You lucky USA folks have McMaster Carr & similar domestic suppliers. But if you are shifting focus to smaller module gears or maybe even bevel gears or belts for driven accessories, I had good experience with Maedler (Germany). Also, the RC (radio control) car/heli/etc mechanical models make extensive use of metric gears & accessories so you might find some stock from those arenas.

https://www.maedler.de/

 

[Eccentric]

Crankshaft Continued

I continued to work on the crankshaft. I have already established the bore to be .75" and now I want to confirm the stroke. A stroke of .75" is a good place to start. I develop a model representing the volume inside the cylinder with the piston at the top of the stroke and at the bottom. The valves are at 45 degrees with respect to the top of the piston so the roof of the combustion chamber is a cone. The flat spot at the top is where the spark plug will be mounted.

The numbers circled in red in the photos below are the volumes of the two models.

Dividing them gives the compression ratio. .39 cubic inches : .06 cubic inches or a compression ratio of 6.5:1. This is a good compression ratio for a model engine, I personally don't want to go any higher. I think that model engines with a higher compression ratio don't start as well, don't idle as well and have a larger tendency to blow out head gaskets.

I create a simple model of a piston and connecting rod, knowing that these are just to give me something to work out the crankshaft dimensions and clearances. I will revisit both later.

Above is a cut away looking at where the piston ends up at the bottom of the stroke and to see if the cylinder sleeve hits the crankcase.

Here the piston is at the top of its stroke.

I started out with a simple internal space inside the crankcase, but the pictures below show that I have interference and need to open up the space inside the crankcase.

Above the connecting rod hits the top of the crankcase space.

Above the connecting rod cap screw again hits the side of the crankcase.

Here I open up the internal crankcase space and shorten the cylinder sleeve. When I finalize the piston and connecting rod later, I will need to revisit these interference areas.

This is my solution for the front ball bearing holder for the crankshaft is shown above. The green part marked by the arrow will house the bearing and then screw into the front of the crankcase. I have made sure the bearing holder and its retaining screws clear the gears. I should mention that my plan is to split the crankcase at the crankshaft.

The above photo is a cutaway showing how the front crankshaft bearing holder is mounted to the front of the crankcase.

The rear crankshaft bearing holder is much simpler as it does not interfere with anything like the timing gears in the front.



To be continued....

.

 

[Eccentric]

Upper Crankcase Half

Today I worked on the upper crankcase, the crankshaft and its bearings. In the photo below a cut away exposes the detail of the upper crankcase and illustrates how the crankshaft is mounted. The center crankshaft bearing is still missing.

I have highlighted detail of interest in the bottom diagram. The front and rear ball bearings are mounted in aluminum bearing holders (colored green). These are produced on the lathe in a single operation so that all of the features are concentric. The crankcase will be line bored so the mounting points of these crankcase bearing holders and the center bearing will be concentric. I have decided to use metric bearings throughout the engine. The crankshaft big end journals are not centered on the cylinder center axes. By offsetting them as in the real engine, I can get more material at the ends and middle of the crankshaft to make room for larger crank bearing surfaces. In this small model I need as much meat as I can get in the crankcase for the screws that secure the bearings and the two crankcase halves. The real engine has cutouts in the side of the crankcase covered by the crankcase breather covers. Since these will not be seen I will not make these large cutouts, just a single hole for the breather. If I were to put these cutout in, I would lose surface area and screws to secure the two crankcase halves.

Oil will be delivered under pressure to the center bearing through and oil gallery in the crankcase. Internal oil holes in the crankshaft will then deliver oil to the big end journals.

Below is a rear view of the engine showing the rear motor mount and the rear crankcase bearing holder.

Below is the beginning of the crankshaft drawing. I want to drill out the connecting rod crank journals, but the outer most crank web is in the way, I may have to eliminate it as I need these holes to deliver oil to the conrod big ends.

Below is a picture of the 97 cubic inch midget Offy crankcase with the side holes highlighted. I will eliminate these as I will split my crankcase as I have shown in the above pictures.

Next I will turn to the lower crankcase half. I think I will add an oil splash tray in the bottom of the crankcase.

 

[petertha]

Looking good! You probably already know this but depending on how you orient your valves, their size, & where the ignition plug bottoms out in the combustion chamber, collectively the CR can get altered a bit of your head looks like sketch. On my engine I was kind of surprised by how much, but its a function of the engines own geometry. Yours may well be different or not be worth fussing about on gasoline/spark. On my (methanol) engine I erred on high CR side with default geometry. I can more readily add head shim to reduce CR without too many consequences, but more difficult to alter things to increase CR after the fact. Food for thought.

 

[Eccentric]

Peter,

You are right, I need to pay attention to the compession ratio design with a more complete model. I have not thought about the head yet, but when I do I need to insure I am satisfied with the compression ratio. I cannot add a head shim as it would raise the valve box and my timing gear would no longer mesh.

 

[Eccentric]

Crankcase Bottom

Today I turned my attention to the crankcase bottom. Since the crankcase is split I need some method of screwing the two halves together and I would prefer if I could hide the screws, the original engine has a one piece crankcase.

One possible idea is to hid them behind the side crankcase breather covers. I need to make sure: 1. I can insert the screw, 2. There is access for the tool to tighten it. 3. the slots don't interfere with the side cover mounting screws.

I worked on the internal features. I want to have a baffle under the crankshaft, I need to think about adding features for it to mount to. I am thinking two pieces of thin aluminum sheet mounted in a "V" with the slot open between them. Simple. The baffle is intended to help keep the oil from frothing up so it can be pumped out of the dry sump.

I added some features to the front timing gear cover. I am thinking the housing for the magneto bevel gears will be a separate piece.

Spent the day helping the wife un-decorating the house, so not much time was spent on the engine model.

 

[Eccentric]

Top End

I have started looking at the top of the engine, the head, cambox, and camshaft. But I got sidetracked working some of the subtle shapes of the crankcase that bothered me. I did not like the way that SolidWorks generated the fillets circled below. I spent a fair amount of time adjusting the control points and did not like this trial and error approach.

So I decided to use surfaces instead of solid bodies. I have not used surfaces for a couple of years so I spent some time researching and practicing. Below the surfaces are being laid in:

And once the surfaces are created, they are knitted together to create a solid body. I am much happier with the result. Below is the result of my test crankcase using surfaces instead of fillets.

Below is the beginning of my study for the head and the cambox. These are complicated parts. Both coolant and oil travel through the head.

Below is the underside of the head.

I have spent a fair amount of time today reading through Terry Mayhugh's Offy build log; what an absolute mother lode of information. I have a lot of studying to do before I can begin to add the next level of detail the Might Midget 97 cu in. model engine. The model engine will be about 24cc.

 

[Eccentric]

Cylinder Head - Part2

Today I am looking at the head combustion chamber, valve, and valve cage. I start by pulling in a model of a valve I have used before and tweak it to fit. This is used in the Wallaby, a 1" stroke and 1" bore engine.

The first thing I find is that there is not much room for the lip on the valve guide in the Offy cone shaped combustion chamber. In the past I have used a light press fit with Loctite to retain the valve cage in the cylinder head. Terry used a bit looser fit and then relied on a steel pin to retain the valve cage in his Offy.

The second thing I see is that due to the large angle of the where the manifolds mount, and the angle of the combustion chamber, there is not a straight shot for the intake charge or exhaust gasses in and out of the combustion chamber through the valve cage.

The way the side hole is drilled in the valve cage, the edge disrupts the gas flow in and out of the combustion chamber. OK, I thought, I will install the valve cage in the head without the side hole drilled, then drill the hole through the head and the valve cage at the same time. This approach is shown below:

And then the resulting hole through the side of the valve cage would look like the following:

I don't like the shape of the resulting valve cage above. The lip is too thin and could deform over time. Also it is a risky machining operation where the valve seat could be damaged. I rejected this idea.

Another idea I considered for a moment was eliminating the valve cage and fabricating the seats directly in the cylinder head. I rejected this idea as it would be too risky, I would hate to reject a head because I messed up one of the valve seats. I like to test the sealing of my valves and valve cages before I install them in the head.

I also looked at altering the angle of the hole in the head that interfaces to the hole in the side of the valve cage, I tipped it up, tipped it down, but did not find an orientation that significantly improved the air flow.

As can be seen above highlighted by the circle, the valve will hit the top of the piston. I can rectify this by raising up the combustion chamber or by having a more complex shape at the top of my piston. The original had a tent top shaped piston. Whatever I do, my compression ratio will be affected and I need to take this into account.

The head is quite riddled with holes and there is very little area to run coolant. The two red circles above highlight the only area that is really available.

I might be able to get away with a smaller valve and valve cage. The one I started with is from an engine with a 1" bore and stroke, where as the Midget Offy model has a cylinder volume about 56% of that. The diameter of the cylinder is 75% (.75"/.1") so theoretically I could get by with a smaller valve and valve cage. Below is a valve cage from Westbury's seagull that has a .75" bore. This is from the Model Engineer Magazine dated September 14,1950.

I love Westbury's drawings, everything is done in fractions, nothing in thousandths.

I know bigger throat and valves are better, but it is good to have another data point.

If I increase the size of the combustion chamber I need to understand the ramifications on my compression ratio. So I repeat my modeling of the space inside the cylinder at top and bottom of the piston stroke.

When using a flat top piston I get a compression ratio of 4:1, a bit too low. So I model a piston with a tented top. I also add the space at the top where the spark plug is and the area the valves extend into the combustion chamber.

By adding material to the top of the piston I can increase the compression ratio from 4:1 to 9.4:1. What this tells me is that I can increase the size of the combustion chamber to give additional clearance to the valves and adjust the compression ratio within a reasonable range just by changing the shape of the piston top.

So, where did I end up today?

The above cut away sums up where I am with the combustion chamber, valve and valve guide.

 

[Eccentric]

Head Design - Part 3

Continued work on the cylinder head by adding the features shown below. Oil is pumped through the camshaft to the cam lobes. This oil drips down through the cam boxes and collects on the heads where the valve cages are. There needs to be a trough machined to collect this oil, and then drain holes to allow it to drain back to the crankcase.

Also added the threaded mounting holes for the cam box and the manifolds, intake and exhaust. These holes may need to be moved when those other assemblies are worked.

I also added raised bosses for the coolant tubing flange. As shown above I do not see this on the original engine but it is nice to have the additional material for the flange mounting screw to bite into. The head coolant passage is just below these holes.

Bottom View of the head with the deeper combustion chamber. There will be many additional holes on the bottom of the head, but these will be defined as part of the block, then their positions transferred to the head. If needed they may be relocated.

Due to my work with the new combustion chamber and recalculating the compression ratio, I have revisited the Crankshaft design. By shifting some of the dimensions around I was able to increase the stroke from .75" to .875". This additional stroke with the larger combustion chamber and a flat top piston give me a reasonable 6.3:1 compression ratio. There will be repercussions to the inside of the crankcase as there are now interferences, but they can be easily addressed.

Latest Crankshaft design - This can be compared to the initial crankshaft drawing to be found in an earlier post. I was able to increase the stroke by going to a smaller connecting rod journal, this also reduced the thickness of the crank stock needed from .625" to .5". I also added the crank pin lightning holes and oil galleries. The ends of the crank pin lighting holes will be plugged to allow oil to flow from the main crank bearing out to the connecting rod journals.

Block Design
OK, now I am turning my attention to the block.

Here is the block with a single cylinder sleeve installed. I need to add some holes to:

• mount the block to the crankcase
• mount the head to the block
• Water passages from the head
• Oil passages from the head to the crankcase
• mount the side plates

I think the best approach is to lay the holes in by eye and then move them around to allow for the most amount of material around them and to insure they do not run into each other. It will be very important to maximize the amount of surface area around all of the holes at the head to block interface. There is a lot of combustion chamber pressure attempting to blow out our head gasket.

Above and below are pictures of my Wallaby block showing the holes that need to be added.

The screws that secure the block to the head will be mounted from the bottom, through clearance holes in the block and threading into holes in the head. I am wrestling with what way the screws should go to secure the block to the crankcase. The outside holes and the center holes are easier to mount from the top, down from the block, because there are webs in the crankcase supporting the crankshaft bearings in the way. The holes between cylinders 1 and 2, then again between 3 and 4, are behind the oil return tube and would be easier to install from the bottom, from the crankcase side. These are identified below as the screws that I may not be able to access behind the oil drain tube.

If I relocate the oil return tube, I can insure that I get a full 60 degree swing on the allen wrench.

I had to redo the holes securing the block cover, these will not match the original, but this combination minimizes interference between holes.

This is the current state of the block, a dizzying array of holes

 

[Eccentric]

Top End

Plan for the day is to define the Cam box end plate, cambox cover and the cambox top features. These include the bearing surface for the cam and a way to collect oil used to lubricate the cam and pass it down to the cylinder head. The cambox end plate may seem like an odd place to start, but this plate will blend with the cambox and the cambox cover and define the shape of the cover and how it blends into the cambox.

Looking at the picture above and below, it can be seen that the cam box cover blends very nicely (tangent entry) with the cambox.

Below is my first attempt at the cover and end plate. I did not understand the relationship of the shapes yet and the cover and cambox do not blend well.

The issue seemed to be whether area between the cam box side and the curvature of the cover was vertical, or parallel with the cambox.

So I loaded a picture of the cam box end plate into my cad software and overlaid a couple of circles and a tangent line between them. Then I dimensioned the drawing and generated ratios I could use to define the shapes. Using this I create a new cam box end plate 3d model.

Below is the resulting 3d model, I am much happier with this look than my first attempt.

Below is a cut away of the cambox and cover showing the way they are joined.

Next I simply transfer the features on the head where it mounts to the cambox over to the cambox bottom as shown below. These features include the large holes that support the cam followers and the small mounting holes. FYI, the cam follower is a part that rides on top of the valve and provides a larger surface area for the cam to work against. It is really just and inverted cup with the valve stem sticking up into the end and the cam riding against the outside.

Then I add the features to the top of the cambox: the oil troughs to collect the oil used to lubricate the cam and the holes to let the oil drain down to the head.

The troughs are cut with a ball end mill, giving them a rounded, funnel shape.

This is where I finished today, I still have some work to do at the interface between the gear tower and the cambox. This area is shown by the red circle and you can see the current interference area.



Next I'll work on the camshaft, then refine the interface between the cambox and the gear tower.

 

[Eccentric]

Camshaft

My design criteria for the camshaft includes the following:

• Turns on two ball bearings (all shafts turn on ball bearings, crank, gears and camshaft)
• A center bush in addition to the two out board ball bearings
• Oil delivered to the four cam lobes and the center bearing
• Cam Lobe definition - covered later (lift 3/32")
• Firing Order - (1-3-4-2)
• Direction the cam turns - (counterclockwise when viewed from the front)
• Center bearing diameter - (1/4")
• Ball bearing selection - (same as the timing gears 5mm X 10mm X 4mm)
• Method to deliver oil to the camshaft-
• Method to retain cam timing gear - key and nut
• Method to allow cam timing adjustment - the timing gear can have multiple key slots, each with a slightly different timing.

My tentative solutions are above in parenthesis.

Also my current thinking is not consistent with the cambox design and will require changes to other major components to deliver oil to the camshaft.

I need to get oil down the middle of the camshaft and there are three ways to do this: Through the back end, through the middle bushing and through the front end. Getting oil delivered to the middle bushing does not look possible as there is not enough material in the cam box to drill an oil gallery. Sending oil down the cambox cover likewise does not seem feasible. Bringing oil in through the front of the camshaft is complicated with the presence of the cam timing gear, so.. First I look at brining in oil through the back of the camshaft

Below is an example of a technique I have used in the past to deliver oil to the center of the camshaft. Oil is delivered to the bearing holder, then to the center of the camshaft through the back.

Pros of this method are it will be easier to get oil through the backend, but the cons include the extra distance the oil has to travel from the front of the crankcase to the rear. Also I am a little concerned using the 1/16" thick rear motor mount as a cover for a very long oil channel up the back of the engine.

The other alternative is to bring oil up directly from the oil pump and into the camshaft through a bushing. Does it replace the ball bearing or work in conjunction with it? Do I just use the aluminum cambox as the bushing? Then do I abandon using ball bearings on the camshaft?

That front camshaft bearing wants to be mounted in the rear gear tower to maintain predictable gear meshing. But on the other hand the camshaft needs to be securely mounted in the cambox. Now I am thinking that using ball bearings on the camshaft is causing more problems than it is solving.

I am going to think about the bearing and oil delivery system a bit more so I am going to move on to the cam lobe design. The design parameters are:

• Lift - 3/32"
• Major diameter - 7/16"
• Base Circle - 1/4"
• Bearing diameter - 7/32"
• Exhaust Lobe duration - 110 degrees
• Intake Lobe Duration - 125 degrees
• Exhaust Lobe nose radius - .025"
• Intake Lobe nose radius - .050"

These data give the exhaust cam lobe shape shown below

I would like a way to adjust the camshaft timing after the engine is built. Becasue there are 36 teeth on the camshaft gear, I can only get 10 degree resolution by moving fromone tooth to another.

If the camshaft timing gear is held on with a nut and its registration to the camshaft is achieved with a key, there could be 5 keyways cut in the gear. This would give us 2 degree resolution in our adjustment of the camshaft timing. 36 teeth times 5 keyways divided by 360 degrees. This can be seen by looking at where the lines emanating from the center intersect the gear teeth.

I have deleted the camshaft ball bearings and am having the camshaft ride directly on the cambox, at least for now.

This is where I ended up for the day, an exhaust camshaft sitting in the cambox. Alignment between the cam box and gear tower needs to be perfect to get correct gear mesh. Does not look much different than where I started.

 

[Eccentric]

Top End Assemblies Finalized

Today I have gone back and attempted to "finalize" the assemblies that have no known design issues. I put "finalize" in quotes because nothing is ever final. I have decided to machine some coolant cavities into the head to allow for more coolant volume and better coolant flow through the head. The cavities will be machined and then sealed with an aluminum plate. I will used high temperature structural adhesive to bond in the cover plate.

Some of the final details incorporated into the head assembly are the plugs sealing the ends of the lengthwise coolant passages and the dowel pins securing the valve cages. The oil passage was added from the timing gear front cover through the head to the cam boxes. I borrowed liberally from Terry Mayhugh's Offy build.

I added all of the holes to the bottom of the head to match the block, the coolant passages, the oil drain and the head bolt holes.

The block assembly is complete at this point.

As is the cambox assembly.

I think I can make the cambox cover from 5/8" aluminum tube with a .049" wall thickness. I will have to slit it, remove some material, then mount it in a mandrel for the bottom machining.