Scotch Yoke Engine

A "Scotch Yoke" is a device that takes the place of the common crankshaft in converting reciprocating motion to rotational motion.  In theory, an engine using a Scotch Yoke is supposed to be simpler, having no wrist pin and a shorter piston.  In practice, friction is it's downfall.

I won't let that small fact stop me - I'll build the engine anyway.  

My plan is to build a two cylinder opposed piston four-cycle engine.  It will use eccentrics instead of cams to control the exhaust valves.  I know that this is an "eccentric" idea but you must consider the source.


15 June 2015:
Anyway, here is a pile of some of the stuff I'll use to build the engine.

There's an engine in there somewhere!


25 June 2015:
It's been a few days and I owe you an update.  The CAD is coming along and I've received some of the parts.

After fiddling around in CAD trying to make the eccentric valve actuation work, I've made the executive decision to revert to a regular cam.  A regular cam is MUCH easier to design and has fewer parts between it and the valve.

The revised materials photo.

The crankshaft main shaft is to be a hunk of 3/4" ground and polished dowel pin.  It will run in caged needle bearings.  I've gotten a carbide end mill to make the flywheel keyway.  The pistons are to be made out of a couple of pieces of cast iron that are the cores of holes that were hole-sawed into a piece for another engine. (Never throw away scraps!)

The piece of large round stock will be used to make the crank cheek/counterweight.  The two sealed ball bearings are going to be used as the crankpin of the yoke.  They will be pressed onto a dowel pin that will be pressed into the crank cheek.  The steel blocks that are on either side of the main block are to be made into the heads.  

The timing sprockets and #25 (1/4" pitch) roller chain are for cam timing.  In the background are the cam bearing bushings and the shafting to be used for the cam/ignition drive.  Both exhaust valves will be operated by one cam, mounted centrally above the block.  The cam followers will be mounted at a 45 degree angle to each other so the valves are timed correctly for each cylinder.  If you look carefully on the top of the block, you will see the cam follower rollers, small ball bearings taken from a defunct hard disk drive.  Actuation of the exhaust valves will be by way of pushrods and rocker arms.  The valves will be made using 1/8" drill rod with 3/8" diameter cast iron heads unless I can score some 4-cycle weed eater valves.  Does anyone out there have four weed eater valves?  I'd be grateful to not have to make valves from scratch.

The ignition distributor will be on the end of the camshaft opposite the sprocket and I may use regular automobile ignition points running on flats milled on the hub of the distributor disc.

I've got some more CAD to do before I can start making parts (Torturing metal).


26 June 2015:
I was getting itchy to make some swarf so I dimensioned the main block of the engine and got it sized.

And, the fun begins.
Doing this was a good way to start getting used to the DRO on the mill.  I find that I can get to within 0.001" by measuring the piece then entering the amount that needs to be removed and going to zero.  Easy!

I will have to make a boring bar and figure out a way to hang the block on the lathe cross feed so I can do the cylinder bores but that will be after holes are drilled and some serious metal is removed from the block.

The ignition is figured out (mostly) so, once I get the block done, I can stick dimensions on the parts and get to the fun whittling.


28 June 2015: 

I was going to start the whittling today but, when I went to find zeros on the block, an old problem came up.  

The piece was too big and the back of the vise hit the knee ways well before the end of Y travel.  It missed going far enough by about a half inch.  Since I'd had this problem before, I made an executive decision to fix it.

                   Vise bangs into ways.                                                          Using my new saw to "improve" the vise.

  Problem solved!
The fix isn't elegant but it does give me another 3/4" of Y travel.  Since the vise won't be used for anything else and will always be mounted on the rotating stand, there's nothing to lose.  You may think that I could move the vise to the front slot but that won't work either because in that position, I run out of Y trasvel before I get to the position I need so the problem is still there.  I know - I tried it!

Then, since I'm going to be tapping some holes in the mill, I thought I'd make a spring tap guide.  I know I could have bought one but, since I'm retired, worthless and bone-idle, I spent a half a day and saved twenty bucks (I work really, really cheap).
Using my fancy tool post grinder to reduce diameter of drill rod.                                    Parts ready for assembly.                          
First, I chucked-up a piece of 1/4" drill rod and using the Dremel as a poor man's tool post grinder, "turned" the diameter of one end to allow a spring out of the junk drawer slip over it.  The length of this piece was just long enough to have it bottom out in the housing before the spring was squashed solid.  The middle portion was left 1/4" and the plunger (pointed end) was "turned" to 0.217".  The 1/2" diameter body was drilled with a #1 bit then reamed to 0.250.  A 1/2" flanged cap was made that had an outside diameter of 0.251" where it was to fit into the body and the I.D. was drilled and reamed to 7/32" (0.21875").  The spring was slipped over the back end of the plunger, the two were fitted into the housing and the cap was pressed into the body.

Finished tap guide.
Voila'!  I now have a nice spring loaded tap guide.  Since I was getting ready to drill and tap some 10-32 holes, I took out a tap and found, to my dismay, that the smaller taps don't have a countersink in the wrench end.  Live and learn!  At least I now have one.  I've put it safely in the tap wrench drawer so I'll remember where it is when I need it.  That is, if it doesn't go walkabout.  That seems to happen more frequently as I get more "experienced".

Maybe tomorrow, I can get to making swarf.


4 July 2015:
It's been a while since I've had a chance to get into the shop.  Today, I got the major bit of whittling done.
Main bearing bore roughed-in and Yoke space milled.
I removed a major amount of cast iron out of the block.  The two "Mickey Mouse" ears are to allow the yoke to travel to the corner of it's cutout.  Otherwise, there would be a 7/8" radius from the milling tool.

The metals and fasteners I needed are here and I can soon start on the block extension, which mounts opposite the yoke cutout and locates both the flywheel side needle bearing and the sprocket and cam bearing.  

Since the alignment of these two parts is critical, I will be drilling and reaming for a couple of 1/4" dowel pins.  After the extension is bolted to the block (with the main bearing bore roughed-in), the block and the main bearing bore will be precisely located and the entire main bearing bore will be bored to 1.000"(-) for a light press fit of the caged needle bearings.

After some thought, I may have to forego making the boring bar.  The cylinder bores, located on either side of the yoke cutout are to only be 1.0" in diameter.  Starting with 7/8" rough, I will only be able to use a 3/4" boring bar.  Since the distance it has to cover is going to have to be around six inches, I'm concerned that the bar may be too limber to avoid chatter.

What I may do is to is to drill and bore one bore to size in the mill.  After turning the block over and roughing-in the second bore, I could use my coaxial center finder with a long feeler through the rough bore to center on the first bore.  Since the ends are parallel within a couple of tenths, I should be able to get away with it.  If there is any misalignment, I may have to machine the piston diameters slightly smaller to accomodate it. In that case, a piston ring on each piston will be necessary due to the clearance.


9 July 2015:
Back at it.  Today, I got the cylinder block almost done.  All that's yet to do is the counter bore for the crankshaft and the main bearing bore finishing.  Those operations have to wait until I've got the casting for the outboard main bearing and sprocket end cam bearing done.  Then, I can assemble this to the block with locating dowel pins and bore the mains in perfect register.  The crankshaft counter bore will be done at this time.

Today, I had one of those "AW SHOOT!" moments.

"Snap" went the tap!
As my luck would have it, the 6-32 tap was almost to the bottom of the hole when it broke.  I tried pecking it to unscrew it but it was stuck pretty firmly so I had to resort to carbide to fix it.
Using carbide end mill to hog out the broken tap.                                              3/8-16 tap and screw.                                                               Tapping for the screw plug..          
Screw bottomed in hole with Loktite.                                                                Screw sawed-off.                                                            Milling stump flush with surface.

Tapping 6-32 in plug.
It only took a couple of hours to fix my boo-boo.  I got a lot done in the remaining time.  All the holes are drilled and tapped and the cylinder bores are finished.

Boring second cylinder.
I think I've got the cylinders registered to within 0.0005" using the first finished bore and the coaxial indicator to align them before the finish boring was done.  The bores are finisned to 0.9995" as near as I can tell.

Almost done!
I'm going to keep soldiering on but I found out last week that I need to have cataract surgery on both eyes.  The first surgery will be next week and I don't know how long I will be out of the shop.  Recovery is about two weeks for each eye so it looks like most of a month will be shot unless I do exceptionally well.  THEN, I will have to get used to far different eyesight, going from very nearsighted to farsighted.  It doesn't look like I'll be able to get away from wearing glasses after the surgery.  I'll have to at least have "readers" for reading and close work.  I also may have to get used to wearing safety glasses in the shop.  My regular glasses served well to keep stuff out of my eyes but I won't be able to use them any more.  

There are other eye protection options that I will be considering.  One of them is to have prescription bifocal lenses made with no correction on top and "readers" in the bottom.  This way, I can continue wearing glasses all the time as I've been doing since I was in first grade and have the advantage of the safety features and the auto-darkening for outside.

Getting old isn't for wusses but it beats the alternative!


19 July 2015:
Well, here I am, back at it (a little).  The left eye is now 20/20 without lenses, something I can't even remember.  The right eye is going to be cut on as soon as I can get it set-up with the Opthamologist.

Anyway, in the interval, I bought my second Starrett tool, a very slightly used set of telescoping gauges to replace the worthless Chinese set I've been struggling with for a while.

Cheap Chinese on left - NICE Starretts on right.
A little oiling is all it took to have the Starretts looking and working like new.  The Chinese junk gave-up it's springs.  The rest went into the trash.  Good riddance!

Sizing the outer main bearing and cam bearing mount.
I decided to check my eyesight so I sized the cast iron piece that will house the outboard main bearing and the sprocket end cam bearing.  Because I no longer wear glasses, I wore a fase shield to protect my eyes.  What I'll probably do at first is to get a pair of cheap Wal-Mart full sized "readers" for close work and reading.  The "readers", being full size will also act as safety glasses.  I think I can look over the tops of the lenses for the occasional "long" signtedness I will need in the shop.  So far, so good.

It's nice to be back in the shop!


22 July 2015:
A little more got done today.  After finishing the outer main bearing and cam bearing mount, I bolted it to the engine block, expecting to drill and ream for the dowel pins.  When I got it bolted together loosely, the pieces had no wiggle at all so I tightened the heck out of the bolts and eliminated the dowel pins.

The assembly then went back into the mill  and the main bearing bore was drilled and bored to 1.001" to make a medium press fit of the main needle bearings.
Drilling the main bearing bore.                                                                    Boring to size.            
      Mains pressed in.                                                               Main shaft in place for test fit.
The "challenge du jour" was machining the mainshaft.  The dowel pin is really hard so I took my time with this part of the job and used coolant.
                          Milling keyway.                                                       Finished keyway and sprocket setscrew flat.
After doing the keyway and the sprocket setscrew flat, the shaft was removed from the mill and chucked-up in the lathe where the crank cheek end was turned from 0.750" to 0.503".  The 0.003" will be for a tight press fit in the crank cheek which will be reamed to 0.500".  After this diameter was turned (a slow process), the shaft was put back into the mill and a 0.010" flat was milled on this diameter.  The flat is what I hope is insurance against the shaft turning in the crank cheek

Finished mainshaft, flywheel and gib key.
There will be some filing required for the gib key because I figured I was pushing my luck with the carbide end mill cutting the hardened dowel pin and quit at 0.070.  The depth should have been about 0.090".


23 July 2015:
After I'd pressed the main bearings into the block, I discovered that I'd forgotten to make the counterbores for the crank cheek and thrust washer.  Off came the extension and out came the inner bearing.  After I mounted the block in the 4-jaw chuck I discovered that I don't have an indicator that can reach back into the main bore so I can center it.  Not to worry, I just figured that I could use the coaxial indicator mounted to the toolpost.  It worked like a champ.

Using the coaxial indicator as a regular indicator.


24 July 2015:
Worked on the crankshaft today.
Crank cheek blank being squared-up.                                                   Making radiuses for counterweight.
First, a hunk of 1/2" hot rolled plate was blanked for the crank cheek.  After squaring-up the blank so it would hang in the vise securely, the mainshaft hole was drilled and reamed to 0.500" for a press fit of the 0.503" engagement diameter of the main pin.  The crankpin hole was drilled and reamed to 0.375".

The crankpin will be made of mild steel and pressed in.  It will be turned for a press fit of the yoke bearings and the cheek end will be turned for a press fit in the cheek.

Excess material sawed-off.                                                                   Milling to dimension.  
After drilling and running an end mill into the cheek where the radiuses will be for the counterweight area, the excess metal was sawed-off then the lines were milled to dimension.
  Mainshaft pressed onto cheek.                                                          Excess material hogged-off.
The main pin was pressed into the cheek.  Although I'd originally planned to go to a lot of trouble to mill the raduis around the crankpin, I ended-up simply belt sanding the shape.  Again, excess metal was removed by sawing before mounting the assembly in the lathe and turning the counterweight to dimension. 
Turning counterweight to dimension.                                                                     Thrust washers made.      
After turning the counterweight and deburring the crank cheek, I made the thrust washers and laid it up for a test fit in the block. 

Test fit so far.
So far, so good!


25 July 2015:
The crankshaft is done.  Today was one of those days!  I should have taken a nap!  It took three tries to get the crankpin right.
  Two crankpin fails!                                                                    Finished crankshaft.
This was the day for fails.  The first crankpn fail happened after carefully turning the cheek end for an 0.003" press fit.  Then, when I cut it off of the stock so I could turn the other end for a press fit of the two ball bearings, I measured it twice and cut it off short!

The second attempt went fine until I turned the bearing fit.  I took off 0.001" too much so it, too, hit the scrap bin.  The third attempt turned out okay, thankfully because I was getting bored making crankpins!

I guess the next thing I need to do is to make the base plate and some kind of skid tall enough so the flywheel clears the table.  Then I can bore the crankshaft timing sprocket and fit the flywheel gib key.  After that, the yoke must be made.  I looked in the pile and discovered that I don't have any 3/4" plate to make it with.  The closest I have is a hunk of 1" plate but I'm not too excited about hanging onto that moose while the bandsaw cuts off a chunk.  What I may do is to haul it up the street to a welder who can use his smoke wrench to lop-off a chunk.  I'll think about it.


25 July 2015:
It's starting to look a little like an engine now.
Base plate is done and the flywheel gib key is close to dimension.
The base plate is a hunk of 1/2" cold rolled plate.  I made some "feet" out of 2X4 wood so the flywheel doesn't hit the bench.  The flywheel is the pulley off of the air compressor that I turned into a double expansion engine..  Until I figure out where to get the material to make the yoke out of, I will make heads, valves, pistons, rocker arms, a cam and other stuff.


29 July 2015:
After a brainstorm, I took it all apart (what there is of it at this time) and drilled and tapped some more small holes.

With some added stuff.
If you look in the above photo, you can see a small microswitch mounted at the back of the engine, offset from the camshaft.  That will be the timer which will operate off it's dedicated cam.  The cam blanks are sitting on the drawing to the right of the engine.  The top cam is for the ignition.  The bottom cam is for the exhaust valves, both operated off the single cam.

You can also see the three ceramic insulators.  They are for the distributor.  The insulator in the middle is for the coil.  The one on the left is for the #1 cylinder (right-hand side of the engine).   The  insulator on the right is for the #2 cylinder (left-hand side of the engine).  The rotor will be made of 0.0625" thick copper clad printed circuit material with a conductor along about 110 degrees of the outer edge of the disk.

If you look carefully, you will see eight smaller holes along the top and sides of the yoke housing.  I plan to cut a piece of Plexiglas to cover this opening.  The theory is that I can have just enough oil in the well milled in the engine base to be splashed around by the yoke to lube the yoke and cylinders.  Using Plexiglas instead of metal as a cover allows me to see the Scotch Yoke in action and to observe how well it oils.


31 July 2015:
Some days I wonder how I get so little done.  Today, for example.  The "plan" was to make the crank end camshaft tower and the cams.  All I got done was the tower.  Since there wasn't enough time left to do the cams, I just drilled and tapped the cam sprocket for a setscrew and milled a flat on the camshaft for it.

So far today.
I also cut the 1/4" pitch timing chain.  For some reason, my efforts to calculate the sprocket centers for a snug chain were for naught.  I am one-half pitch off.  That means there is a lot of slop in the chain.  As you can see in the photo above, I will be whittling out a Nylon chain tensioner (or "slap eliminator").

Maybe tomorrow, after the "slap eliminator" is done, I can start on the cams.


3 August 2015:
Progress was made today!

Timing chain tensioner in place.
The chain tensioner was easy and didn't even require a rough sketch.  There's a slot for the mounting screw so the tensioner can be slid up and down to adjust the slop out of the chain.

Cams and distributor in place.
Since I've made several cams, making these was easy.  All it required was calculating milling cutter diameter to determine offset, then turning the cranks.  The ignition cam (the one closest to the sprocket) was really easy.  Find center, crank-in X offset then crank the Y to makie the flat.  Turn the rotary table 90 degrees and repeat.

I'm not sure about the distributor.  In order to minimize the chance of crossfiring, I made the rotor conductor a 45 degree arc.  Theoretically, it should be around 90 degrees but, with that much of an angle, both plugs could fire at certain timing angles.  After building the rotor, I probably could have gotten away with nearly 90 degrees because I'll most likely set the ignition timing to a compromise between best starting and best running (assuming it will run at all!) and leave it there.  I can tweak the rotor positon for optimal "distribution".

Tomorrow, I'm going to have to tackle the yoke.  This is supposed to be made out of a single piece of 0.750" hot rolled steel plate.  I think I know where I've hidden a big enough piece of the steel but cutting it out is going to be a slog.  My water jet cutting shop seems to have closed so I've got to rough it the hard way.  When you see what I'm making, you'll first think it's an easy part to make.  Think again.  The outside can be roughed-in using the bandsaw.  The inside will have to be milled out.  It's going to take a LOT of machine time to cut out the middle.  About the deepest I can go with an end mill in steel is about 0.050" and the feed rate will be slow.  That's fifteen times around to get the center out.  Then, there will be some finishing work because the dimension is critical due to the clearance between the ball bearings and the inside of the yoke having to be very small to avoid hammering.


23 August 2015:
It's been a while since I've posted here.  I suppose I should explain why.  About six weeks ago, I had cataract surgery on my left eye.  The "plan" was to get the right eye done about three weeks later.  Right after going off my eye drops for the left eye, it developed some swelling and pain.  The Doc put me back on the drops for a while and we decided to forego the next surgery until we were sure the left eye was good to go.  In the meantime, I am farsighted in the left eye and very nearsighted in the right and because I can't focus up close with both eyes, my depth perception is off.  Now, it looks like it will probably be in November or December before I can get the second surgery done and get my eyeglass prescription filled, giving me back my depth perception.

I can still work in the shop but must be VERY careful when working around machinery.

In any case, I actually got something done today.

Big progress today!
What you see is the yoke blank on the left with the "left-over" piece on the right.

To make the cutout, I will have to drill a bunch of holes then connect them before milling the cutout to dimension.  THAT's gonna be fun.

Thanks for your patience!


29 August 2015:
I finally got back into the shop for something fun.  Today, I cut out the center of the yoke.
Roughing the sides.                                                                           Roughing the ends.                                                                                      Reality!      
I'm really going to have to train myself to see things accurately up close.  I laid out the cut lines, allowing myself some extra material for finishing to dimension.  As you can see on the drawing, I've marked where the cuts actually were.  I sighted on the layout lines to set the mill table and missed one of the lines by a bit, making one side come out to 0.340" so I could finish it to 0.313".  The other side is where I goofed.  It ended-up at 0.270"!  

What I will do, rather than scrap the piece is to just clean-up the 0.270" side making it about 0.265".  Then, I will mill the inside of the other side until the width of the opening is 1.125". the diameter of the crank rollers.  When the inside is done, I will mill off of the outside of the fat side until it is the same as the thin side, making the yoke about 1.655" wide.

To compensate for the narrowing of the yoke, I will make the piston rods about 0.048" longer.  Everything should work fine.

I think this was one of those "King Midas touch in reverse" days!


31 August 2015:
The yoke is done.

Yoke with bearing for test fit.
Everything was going fine until it came time to do the actual fit of the crankshaft bearings to the yoke.  I went to the bench where I had it sitting beside the engine and it wasn't there.  No one else has been in my shop so I figured that I'd just put it somewhere else in the shop.  Although there is no reason for me to have moved it, I looked all over the shop and it is nowhere to be found.

The only good news is that I found the telescoping "snap" bore gauge that's been missing for a while.

Now, everything's at a dead stop until I can locate that furschlugginer crankshaft!


1 September 2015:
I got a piston made.

The first piston and the blank for the second.

I've grooved the piston for a hydraulic "O" ring.  They're supposed to be good to 400F or so and we'll see if they work.  Right now, I've got the fit on the tight side but can modify the fit if it doesn't do well.

If the rings don't hold-up, I can always widen the grooves to 3/32" and use cast iron rings.

I still can't find the crankshaft!  I may have to purchase the material again.  Once I've gotten the second one about half done, the first one will come back from wherever it's visiting.


2 September 2015:
Well.........I ain't a-goin' bonkers after all!  I found the crankshaft.  It was hiding in a box of 1/2-13 bolts.  Another project had me looking for a bolt and I had them spread out on the bench whe
re The Scotch Yoke Engine resides.  When I scooped-up the bolts, I must have scooped-up the crankshaft with them.  The only way I found it was that I was looking for another half inch bolt and......SHAZAM!  There it was!!

Now back to our gripping suspense story.

The pistons are done and, since I found the crankshaft, I now have the rotating/reciprocating part of the engine together and moving.  It's a little tight but I plan to motor it-in.  

Here 'tis, so far.
If there's enough interest, I'll make a short YouTube flick showing it reciprocating.


4 September 2015:
I got it hooked up to the motor today and ran it for about 15 minutes.  It was kinda tight for a few minutes because I had the pistons fitted really close.  After the run and making the flick, it now runs quietly with no tight spots.

motoring away at about 1,000 RPM.

And here's the first flick
Next up appears to be the heads and valves.
5 September 2015:
The heads have been sized from the rough cast iron stock.

Heads cut out and sized.
After getting the heads sized, I looked more closely at the drawings and found that I haven't drawn-in the valves and guides.  It's back to the CAD for a while.  

I plan to make the valver stems out of 1/8" drill rod and the valve heads out of a piece of the cast iron used to make the heads.  I'll probably thread the stems and partially tap the heads and jam them into place.  I'll make the stems a little proud of the heads and peen them to keep them from unscrewing.
6 September 2015:
The hard part of the heads is done.  
The valve seat cutter.                                                                       Cutting the seats.
After boring out the combustion chambers and drilling and reaming the valve ports and valve guide bores, I had to cut the valve seats to 45 degrees.  The way I did it was to find a "retired" half-inch drill bit and sharpen it so one flute had a 45 degree angle.  The other flute was ground back so it wouldn't cut.  The seats are very smooth and are 0.020" deep.

Almost done with the first head.
After cutting the seats, a 1/8" "firing hole" was drilled to intersect with another 1/8" hole drilled in the bottom of the spark plug hole.  Although this arrangement isn't optimal, there was no other practical way to get the 10mm plugs to communicate with the combustion chamber.  I suppose I could have spent the big bucks for a couple of the little 10-32 model engine plugs but I'm a cheapskate.
7 September 2015:
The heads are now done as are the valve guides.
    Heads and valve guides.                                                                                    Engine with heads mounted.
The guides were drilled to one drill number under 0.125" then reamed to 0.125".  The valve stems are to be made from 0.125" drill rod, which is a really nice fit.  Tomorrow, I will do the valves.........I think.

There is a left-over piece of cast iron from making the heads that may be big enough to turn into a bar, from which to make the valve heads.  I think I've got a way figured out to make all the heads from the bar then part them off.  The heads will be drilled and reamed to 0.109" (7/64"), then the ends of the stems will be turned for a press fit into the heads with a little extra length so they can be peened to make sure the heads don't come off.

8 September 2015:
Today, I got the valves almost done.  Here's how I made the valve heads.
   Here's the scrap of cast iron I'm going to use.                                                              Cut it in half.                           
The scrap of cast iron, left over from making the heads, was cut in half lengthwise, giving two pieces about a half-inch square and four inches long.
Mount it in the 4-jaw chuck and turn to valve head diameter.             Set compound to 45 degrees and rough-in valve face.    
I mounted one of the pieces in my junky 4-jaw chuck, banged it semi-straight and drilled a center in it.  Then I turned it to 0.437", the diameter of the valve heads.

Using the parting tool, I partially parted the valve heads at their thickness.  Then, after setting the compound to 45 degrees, I rough cut the valve faces.  The hole for the stem was drilled and reamed to 1/64" below 0.125 (0.109").  After that, the valve heads were parted-off.
             Heads after parting-off.                                                     Press stem into head and peen for safety.
After cutting the 1/8" drill rod valve stems to rough length, The head ends were turned to make a press fit into the valve heads.  The turned part of the stems is proud of the heads by a few thousandths of an inch and the tops of the stems were peened over to keep the valve heads from coming off.
Use poor man's toolpost grinder to finish faces.                                              Cut stems to length using Dremel.  
Next, the valves were put into the lathe and, with the compound still set at 45 degrees, the Dremel tool was used to grind the valve faces true.  This works fine if you go slow.  The excess length of the stems was removed, again using the Dremel.

 Valves lapped-in.
The valves were lapped using fine compound.  The keepers still have to be worked-out.

The keepers may not be exactly like the drawing because the stems are too short  to get that dimension when mounted in the lathe chuck.  I'm going to think over how to do the keepers.  I may simply thread the ends of the stems and thread the spring washers on, using locknuts.
9 September 2015:
Because I'm running out of drawings, I spent the morning isolating the rocker arms and brackets, the followers and brackets, the pushrods, the yokes and pins, etc.  I'll get them dimensioned and printed to send out to the shop for the crew (me!) to work on.

In the shop this afternoon I figured out the valve keepers.  Pretty easy.  I rooted in my collection of snap rings and found four that had about a 0.250" O.D, a 0.185 I.D and were about 0.015" thick.  

Then, since I already had the Dremel mounted in the lathe so I put the valves in the chuck reversed (there was -just- enough sticking out to do the deed) and ground grooves 0.020" deep in the 1/8" stems using the same wheel.  The snap rings fit nicely.

The spring washers were made from steel bar stock that was turned down to 0.420".  Drilled with a 1/8" bit, an 0.063" deep ledge was turned to seat the springs.  The total thickness of the washers is 0.125"

Using a 1/4" end mill in the lathe, I counterbored the top ends of the washers 0.030" deep to retain the snap rings.

Kinda blurry but you get the idea.
(It's too blurry to suit me so I'll re-take the photo when I have a head off again.)
The exhaust valve springs are some springs removed from hand lotion pump bottles.  I wound the intake valve springs using a 1/4" mandrel and some small diameter music wire.

I made a couple of head gaskets so I could mount the heads to check the valves.  When I motored the engine, it seemed line one valve on each head was leaking.  I'll look into it further and see id I have to give them another lick of lapping.
12 September 2015:
A little more is done.  I got the CAD of the valve gear done in the last couple of days and have started on the parts.
    Full scale drawings glued to steel.                                                                           The partially completed cam followers.
As you can see in the photos, I made these parts the "easy" way.  I glued 1:1 drawings of the parts to the steel they were to be made out of.  Then, the hole centers were center punched for reference.  After a lot of bandsawing, the parts went onto the mill to get the lines cleaned up.  The corners were belt sanded.  The holes were laid out and drilled on the mill for accuracy.  

As you can see in the above right photo, I decided that the big radius was going to be a pain to do so I left it off.  The little ball bearings are out of dead hard disk drives and will serve as the cam rollers.  I still have to mill for the pushrod clevises and make the roller pins for these parts.  Oh, yes, these parts are the cam followers.

A rocker stand on a head.
The rocker arm stands were easier to make and turned out well.

Next on the docket is to finish the follower arms, make the rocker arms then the pushrods and clevises.  I hope to get those tasks done tomorrow before taking a few days off the project.
13 September 2015:
The valve train is almost done.  As usual, the parts that look simple are sometimes the most complicated to make.
The follower arms, partially assembled.                                                    And mounted on the engine.  
Those of you who are eagle-eyed will see the red marks on the follower arms.  These mark where I have to mill some relief there so the cam lobe won't hit the arms before contacting the rollers. 

Rockers and pushrods and the valve train is done.
22 September 2015:
Well, I'm back.  The rockers are done and I'm nearly finished with the pushrods.
           A pushrod ;and clevises.                                                                          Pushrod and rocker in position for test fit.
I've had to make a few running changes.  One is to the shape of the rockers.  After finishing them, I found that the arms themselves interfered with the spring keepers.  Milling a notch solved the problem.  

I had an idea that would allow more accurate adjustment of the valve lash.  With the original pushrod and clevis arrangement, the same 6-32 threads were on both ends where the clevises were attached.  With this arrangement, the only increments of adjustment were 0.016",  This was because the only way to change the length of the pushrods was to unhook one clevis and turn it one half turn.  On a little engine like this, 0.016" makes a BIG difference in the valve timing and lift.  What I decided to do was to thread one end of each of the pushrods and the clevises to 8-24 and the other end to 8-32.  This would allow adjustment due to the difference in the pitch of the of the threads on the ends.  It would also allow adjustment without having to disconnect one clevis to make the adjsutment.  Adjustments would be far more accurately made.

When I was getting the taps and dies together to make the pushrods and clevises, I found that I don't have an 8-24 die.  The solution was to go to a larger diameter for the pushrod and clevises and use 10-24 and 10-32 threads to allow for adjustment of the valve lash.  I have those taps and dies.

This part is all done except for preparing the yokes of the clevises for pinning.
23 September 2015:
It's really starting to look line an engine now.  The pushrod clevises were a bit frustrating to finish.  I broke off a 6-32 tap in one of the clevises and it took about an hour to finesse the broken part out without damaging the threads.  One of these days, I'm going to have to bite the bullet and order a bunch of new taps.  The one that broke was worn and I think metal fatigue finally spelled it's end because I wasn't gronking on it when it broke.
Two views of the engine as it is today.
The valves are timed pretty close now and, I still have to time the breaker cam and the rotor.  While motoring it, I found that two of the valves aren't sealing so it's off with the heads for a rework.

I still have to make the plexiglass cover for the yoke so I can put enough oil in it for splash lube.  Right now, there's just enough oil to get picked-up by the yoke and, even though it doesn't throw it high enough to do any good, it still splashes out.

After I have compression, I'll attack the intake manifold, mixer and exhaust pipes.  I'm still considering running it on propane, at least at first.  Since it will be throttle controlled without a governor, there may be "interesting times" ahead.
5 October 2015:
It seems that every time I think I'll get something done on the engine, something else comes-up.

Today, I got a little done on the project.  I pulled the heads and lapped the valves again, this time, being more aggressive.  It appeared that the "valve job" was an improvement because,
with the plugs in, and motoring, it sounded like it would be willing to try to run if I got the ignition done.

While I was in the mood, I made the plexiglas yoke cover and put it on.  

Motoring with oil in the yoke case.
I started motoring the engine and then added oil until it was sloshing around pretty well.  As you can see, at about 500 RPM, the oil gets around pretty well.  Then I worked on the ignition.

Ignition done and ready for carburetion.
After getting the wiring done, I set the ignition in time and did a little modification to the center (coil) contact of the distributor.  It had to be widened in order to allow both plugs to fire and give a little margin for timing changes.

I got it motoring and decided to see if it would act like it wanted to run.  I turned on the ignition then took a propane torch and, without lighting it, held it to the intake port for each cylinder.  The #1 (left) cylinder actually made putt-putt sounds but when I went to the #2 cylinder it was only half-heartedly firing.

The next time I get a chance for some shop time, I'll see if one of my compression gauges will fit and will check compression.  I may have to visit the valves in #2 once more.
19 October 2015:
Finally, I've had a chance to do some more on the engine.  

The first thing I did today was to check the compression.  #1 is making about 65 PSI while motoring but #2 is only making about 40 PSI.  I stuck the rubber tip of the air gun into the spark plug holes and the exhaust valves are still leaking.  More lapping is in order.

Before torturing the valves again, I wanted to get the intake manifold done.
            The materials are selected.                                                                                   Layout of the flanges that go on the heads.
I'm using 3/8" copper tubing and fittings for the manifold.  The flanges that go on the heads are made from some 1/4" thick copper buss bar that I salvaged out of an old elevator control years ago.
Flanges and adapters.                                                                                                            Semi-finished manifold.    
The outside diameter of the tubing is too large to fit into the flanges and have room enough for the mounting screws so I had to make adapters.

I wanted to have room to remove the yoke case cover without removing the manifold and this made it necessary to have the junction of the two runners lower than the ports in the heads.  Then came the fun part, bending the tubing to form the runners of the manifold.  The radiuses of these pats are very short so with the proper amount of straining, hammering and swearing, they were made and soldered.  Although I'm sure it will work, the next time I get to work on it, I'll see if I can make it look a bit better.  The mounting flange for the mixer has yet to be made and soldered-on.

Another running change I'm contemplating is the use of liquid fuel, probably naphtha.  Since there is no governor on this engine, I'll control the speed by simply opening and closing the throttle.  I may experiment with a barrel type throttle because I think it will be easier to make than a butterfly valve.
25 October 2015:
In the last few days, I've gotten a little more done.  First, I pulled the right-hand head and re-lapped the valves.  Now, I've got about 75 PSI on both cylinders.  That's good enough for now.

I also finished the intake manifold.  It isn't a work of art but it will get the fuel/air to both cylinders.

It's gettin' close!
While fiddling, something just didn't seem right.  The engine turns counter-clockwise when viewed from this end of the yokeshaft and I've got the left-hand cylinder as number one and had the firing order, naturally, 1-2. Today, I noticed that number 2 cylinder didn't follow cylinder number one. The only way I could ger it to fire one cylinder 180 degrees off of the other was to make the right hand cylinder (#2) fire first.  Then it worked out.

What I had to do was to cross the spark plug leads (I had to make a new, different set) and move the points cam around to correspond to this order.  Now, the cylinders fire right but the firing order is 2-1......  Oh, well......

The exhaust and the mixer are next then it will be the moment of truth to see if it will actualy run.

Stay tuned!  We'll eventually get there.  Sorry this one is dragging out so much but other things have taken precedence.
26 October 2015:
Today's project time was spent making a drawing of the mixer.

Drawing of mixer.
This time, I'm going to experiment with a "barrel" type throttle.  Doing butterfly valves is technically interesting but not much fun.  The way I'm going to arrange the jet is so I may not have to have a venturi.  We'll see how that works in practice.  Although the barrel will not be self balancing, with a stiff enough return spring, it won't try to go wide open if the throttle shaft is let go.

Maybe tomorrow I can start on it.  I don't think I'm quite ready to take the tripod out to the shop for the first startup video but it's getting close.

28 October 2015:
The exhaust pipes are done and installed.

The ezhaust pipes.

The mixer is a little more complicated than the exhaust.  I've got the body done and still have the throttle, throttle cover and mixer needle and check valve to do.

Engine with partially completed mixer.
I'm not sure when I'll get back into the shop.  It could be in a couple of days and could be in a week.  I'm getting itchy to see if it will run but other stuff getsd in the way.

31 October 2015:
I'm getting close.  Maybe one more day and I can see if it will run.  

Today, I worked some more on the mixer.
Ready for throttle shaft assembly.                                                                     Fuel jet orfice.          
This is what I consider what my British friends call the "fiddly bits".  It took a couple of hours to make the check valve and fuel tube and another couple to make the throttle and the throttle cover.  I wanted the throttle to fit the mixer body very close and used my shear tool in the lathe to turn the last couple of thousandths off of the diameter.  At that point it was aaaaaalllllmost a fit so a twirl with some worn out 400 grit emery got a nice fit.

The next time (tomorrow?) I get to work on it, all that needs to be done is to make the needle valve assembly.  That shouldn't take very long if Mister Murphy doesn't come to visit.

2 November 2015:
I finished the mixer today, lashed-up a temporary fuel tank and gave it a spin.
        The completed mixer.                                                            The engine, ready for first trst attempt.
As you will see in the video below, the test run was a qualified success.  It made the proper sounds with a little coaxing but I came to the conclusion that a bit of tweaking with the ignition and valve timing and adding either a centuri or a choke is in order.  This is because it wouldn't keep running after I quit finger choking it.  By the sound of it, I didn't think it would run off the belt yet so I didn't try.
Here's the flick.
3 November 2015:
Today, after a few modifications it ran on it's own for the first time.

One thing I did was to go ahead and install the hydraulic cylinder "O" rings in the previously cut grooves in the pistons.  Compression was 75 PSI on both cylinders before the rings were installed and between 85 anf 90 PSI afterward.  That's not much of an improvement but it is significant.  If the rings don't self-destruct, the compression should improve as the cylinder walls become smoother from running.

"O" ring ready to be installed on a piston.
Then, after some thought, I made a sort of venturi for the mixer instead of a choke.
Venturi ready to be set in place.                                                       Completed venturi assembly.
Since the throtle is not balanced and the jet is just drilled into the throat of the mixer, I decided that the venturi could be simple as well.  All it is is a piece of steel turned to the throat diameter (3/8").  I milled a flat at an angle on the intake end tapered from 0.100" at the air intake to partway down the slug.  Then I set the slug in the mill straight and milled it so there's a flat with an opening 0.050" close to the jet.

The difference in the running of the engine is remarkable.  After a little fiddling with the fuel mixture and throttle opening, the engine took off.  After tweaking, I turned the motor off and it pulled the motor just fine.

In the video below, you will see that it does run but roughly.  I think the problem here is that it is over-oiling and that the excess oil is making it a little hard to start and causes it to run sloppy.  At least, we know that it is getting sufficient oil!  It's sufficient enough that liquid oil droplets are coming out of the exhaust pipes.

I plan to run it several times over the next few days and let the oil level decrease to where it doesn't smoke so much.  Then, I will have a place to mark the oil level.  

Now, I wish I'd included some kind of a cooling system.  The engine seems to like to run but I can only let it enjoy itself for a few minutes at a time before it gets too hot to touch.

Here's the video.
6 November 2015:
So far, I've got about 30 minutes of run time on the engine.  It starts easily now that I've retarded the ignition to TDC and I can run it for over 15 minutes before the heads get too hot to touch.  I think the block acts like a heat sink.

A couple of days ago, the engine started knocking and quit running on one cylinder.  The compression on #2 cylinder was way low so I thought the rubber "O" ring was at fault and, this morning I ordered a couple of cast iron rings to replace them with.

Today, after I took it apart, I found out that the #2 piston bolt had loosened from the yoke and it was losing compression by way of the bolt.  While I had the head off, I pulled the piston to check on the rubber "O" ring and it looked good.  When I put it back together, it ran fine.

There are still some issues to work out.  One of them concerns carburetion.  It doesn't want to run at a constant speed with no load.  At all needle valve settings, it will run for a few seconds, then peter-out.  Cracking the throttle open will cause it to start running again.  I think the reason is because of the location of the jet at the edge of the air flow.  The engine will accelerate strongly and, if a load is placed on the it, it will continue to run smoothly when the throttle is opened to hold the RPM up.

Another bothersome item is that #1 is really pumping oil and smoking.  The exhaust pope gets wet and it pukes droplets of oil.  I'll pull that head and piston tomorrow to see if it's an "O" ring issue.

Maybe tomorrow between runnings, I'll get around to making a gas tank for it.

7 November 2015:
The first thing I did today was to pull the #1 head and piston and here's what I found with the "O" ring.

Kinked "O" ring.
I -think- the kink in the "O" ring is what made it puke so much oil.  That was the first ring I installed and must have forgotten to carefully check to see that it was properly seated in the groove.  It got replaced with a new one.

The next thing I did was to make a fuel tank.  It's pretty small but, since I can't run the engine for long periods, it is adequate.
The new fuel tank.
I gave it another run for about fifteen minutes and, at the end of the run, the engine wasn't so hot I couldn't hold a finger on it.  For some reason, it is running a little better but I still think I will have to address the fuel jet location.  It's still puking some oil and I think I will need to drain off a little more oil from the "yokecase".

Oh, yes, I changed from the 5W30 synthetic oil to straight 30 weight detergent oil.  Somewhere, I read that the synthetic is designed for engines with roller tappets.  This engine has roller tappets but the Scotch Yoke does have some sliding component at the ends of the stroke.  I thought it prudent to keep this in mind.

9 November 2015:
Well, it looks like I'm at about the end of this project.  All that's left is a little tweaking.  Once the cast iron rings come in, I'll widen the ring grooves in the pistons and fit the new rings.  I may machine a second groove for "O" rings to try to cut down the oil consumption.

Yesterday, I took the mixer off with the intention of changing the fuel jet.  When I took it apart, I found to my dismay, that I'd put it together wrong.  I had the throttle valve 180 degrees out, which messed up the air flow.  After reversing the throttle, it runs a lot better now.

One of the runs today was 30 minutes and the engine was just a little too hot to keep a finger on.  I figure that the block is massive enough to be a good heat sink and 30 minutes is longer than demonstration runs would be.  Here's another movie showing how it runs today.
It's running better now.
11 November 2015:
Yesterday, I enlarged the piston ring grooves to fit the 3/32" X 1" cast iron rings.  Before tearing the engine down, I did a cold compression check with the "O" rings.  At a cranking speed of about 625 RPM both cylinders measured  70 PSI.

When I took the "O" rings out, I compared the "oldest" ring (about three hours running) to a new one.

New ring on left, 3 hour ring on right.
I guess if I used a magnifying glass or microscope, I could see some wear on the used ring but, as far as I'm concerned, this type of ring, properly fitted and well lubed, would work fine in a total loss oiling system engine as long as careful attention was paid to make sure the piston was always wet with oil.

After re-assembling the engine, and a couple of minutes of motoring it to "introduce" the cast iron rings to the cast iron cylinder bores, I did another cold compression check at the same 625 RPM.  Both cylinders measured the same 70 PSI as the "O" rings.  

When I ran the engine with the new rings, the first thing I noticed was that it was no longer passing oil out the exhaust.  Since then, I've run the engine under both no load and a light load (pulling the turned-off motor) for about three hours.  Since I no longer have to worry about melting the rings, I have run it as long as 45 minutes continuously.  Head temperature is just a bit too hot to touch (my el-cheapo I.R. thermometer says it's about 120 degrees F after the run).

This afternoon, after the last run of the day, I measured the compression hot.  Same 625 RPM.  #1=75 PSI, #2=72 PSI.  That's probably about as good as it's going to get.

The mixer, for some reason, gets slobbery after about ten minutes and the mixture has to be leaned-out to make it quit dripping.  Then, after a while, the engine will begin to run lean and I have to richen the mixture to keep it running.  After a while, it gets again.  I'm not sure why this is happening but it could be because of imperfect atomization causing fuel to condense in the manifold, making it load-up.

I'm not sure what I'm going to do about it but I may go ahead and re-design the jet to stick proud of the mixer bore by a few thousandths of an inch to get it more into the air flow.  Also, I could heat the tee fitting in the manifold and rotate it 90 degrees to make the mixer into a side-draft.  I don't know if this would help but somewhere in my little gray cells, something is telling me it would help.

29 November 2015:
Over the last couple of weeks I've been fiddling with the carburetion, trying to get the engine to run well at all speeds and power settings.  I haven't been having a lot of luck.

First, I thought I needed a smaller jet so tried using a hypodermic needle.  There was no improvement and it plugged easily.
Hypodermic needle was too small a bore.
At this point, I made a new needle and jet, designed to stick out into the airflow a bit more.  Then, the trouble began.

Another jet and needle idea.
The new needle and jet were made.  Even after a lot of re-work, I couldn't get the fuel mixture to lean out enough for the engine to run smoothly.  The problem was in the seating arrangement of the seat assembly in it's bore.  Fuel was leaking past the seat into the throat so, obviously, even with the needle valve completely closed, enough fuel leaked to make the engine run a little rich.  After sorting that out, the engine still didn't want to run well at all speeds.  I finally decided to plug the first jet bore and move the entire thing over to the other side of the mixer and locate the jet about 0.100" farther upstream, in the throat of the venturi.
A flutter choke and the jet moved to the other side and upstream a bit.
I got the jet re-located, then another problem showed-up.  At first, I thought it was going to be fine but, after running the engine as little, I found that, after the needle was set for acceleration, it would first run a little rich at idle then, within a minute or so, it would lean out and quit.  No matter how the needle valve was set, the same thing happened at idle.  My theory was that, at small throttle openings, there wasn't enough suction to pull fuel from the tank, past the check valve and to the jet.

Okay, how about a compensating air valve?  I made an adjustable spring loaded flutter choke as seen above.  At small throttle openings, the flutter choke is mostly closed, only being pulled off it's seat a little on the intake strokes and making some vacuum.  At open throttle, there is enough air flow to pull the disc off of the seat enough to allow ample air.  Adjustment of the mixer at this point is a juggling act between needle valve opening for full throttle and flutter choke spring compression for idle.  Usually, these two settings can be adjusted for a fairly good range of engine speed/power with good fuel mixture.

In this case, Mister Murphy reared his head.  No amount of adjusting of the needle or choke, including trying various spring strengths, worked.

At this point, the engine runs steadily at full load or high speed but falters and quits at idle.  Since I'm pretty disgusted with it at this time, I think I'll let it sit in the corner with it's Dunce cap on.  I may have to try a completely different mixer or, at least, go with a float chamber to keep a constant fuel level at the jet.  I'll think on it for a while.

Oh, yes - I'm a little surprised that the engine can be run for about an hour without getting more than just a little too hot to touch.  It must be the large amount of heat sinking the block provides coupled wiht the small breeze provided by the flywheel fan.

7 December 2015:
For the last few days, I've been designing a propane carburetor.  The plan was to have a variable gas valve working on the same shaft as the throttle.
The propane mixer.
A whole new mixer body was made.  It has a quarter turn 0.100" deep spiral 1/8" groove in it.  The copper disk has a matching 0.100" groove but it is concentric with the throttle.

One end of the groove in the body is drilled through to the carburetor throat, upstream of the throttle.  The groove in the disk is simply a channel that connects the inlet gas line witth the spiral groove in the body through the groove in the disk.

The copper disk and the carb body have been lapped to a gas-tight fit.  Springs on the cover plate screws serve to keep the disk and the body pressed together.

Since I didn't want to get into the physics of propane mixing, I took a wild guess as to the port sizes, etc.

Just at the time I had to quit for the day, I put the new carb on the engine and motored it while fiddling with ignition timing, etc.  Using a grill regulator from the tank, the system is WAY rich but I did get the engine to run reasonably well by using a C clamp to squeeze the rubber gas line to the carburetor.  I think I need a needle valve in the gas line as well as some kind of choke in the air inlet.  

Experiments will continue.

31 December 2015:
It's been a while.  I finally got my second eye cataract surgery done and am now able to see the same in both eyes.  That means I can once again see close in stereo with reading glasses.  

I made the needle valve to see if I could get the engine to run better on propane.
      Parts of the needle valve.                                                                           Needle valve hooked up to mixer.
It still doesn't run well.  At a fixed throttle setting, the engine lopes no matter what the fuel mixture is.  What I probably need to do is to just bite the bullet and design a demand regulator to have a chance of a proper mixture no matter what the throttle setting.

Another thing that may be making the engine lope could be different ignition timing betweeen the two cylinders.  The timer I made-up after the microswitch bit the dust may be the cause.  I'll have to mark the flywheel and make a pointer so I can check the ignition timing.  While I'm at it, I'll accurately tram the valve timing.  It could be that I don't have the valve opening and closing points averaged out between the two cylinders.  Since I'm using one cam for both cylinders, if the cam is shifted one way or the other, the valve timing will not be the same between the two cylinders.  It's things like that that can result in funky operation.

At one point, I had the engine running constantly for over an hour and it didn't overheat.  At some times, I was using the cranking motor as a generator and was powering a 50 Watt light bulb.  At this rate, Hoover Dam it ain't!  At the end of the long run, I could still hold a finger on the block for a couple of seconds.  There must be enough mass in the block and heads to soak-up the heat and radiate it away.

16 January 2016:
I'm back at it again.  Today I started on a demand regulator to try to get the engine to run more reliably at varying speeds.  

Demand regulator so far.
AND, there was an AW-SHOOT! today.  I was tapping the next to the last hole in the regulator body when the 4-40 tap broke.  DAGNABBIT!!  There was no way I could get the pieces out so I started again, this time with a re-design.  

I have some 0.015" neoprene to use for the diaphragm.  The "plan" is to make a thin brass button with a hole drilled into it that acts as a moving seat for a needle valve.  The button will be glued to the center of the diaphragm and a hole punched through the diaphragm to move the gas from the needle (supply) side to the engine intake side in response to the engine suction pulling the seat away fromt the needle.  I will have a compression spring with an adjustment screw to (hopefully!) set the mixture.

I'll probably finish it in the next couple of days and see if it works.

17 January 2016:
I finished the demand regulator today and just had time for a preliminary twiddle.

New regulator to the right of the mixer.
I made some wild guesses as to spring tension and orifice size and will probably have to increase the orifice at the needle valve.  I'm not sure about the spring rate.  It's pretty weak now but I'm really not sure which way to go with it.  The engine will run with about 3/4 of the air intake taped-off but the idle to full throttle mixture balance leaves a lot to be desired.  There will be a bit of "vacation" time until I get another session in the shop.

25 January 2016:
Today, after experimenting with different springs in the demand regulator, I got the engine to run pretty well although it still has to be started off of the motor.

Then, I decided to change the ignition timer to a variable version so the timing could be changed on the fly.
Variable timing arrangement.
After I got it done, the engine ran a lot better and I could tweak the timing but it still had to be cranked on the motor.  Note that I have changed the spark plug type.  The little plugs fouled out all the time and, as another experiment, I used some long reach plugs with spacers.  They seem to be doing just fine.

Then, I went and messed it all up.  I was tired of the roaring sound of the roller bearings on the crankshaft and, looking around, I found a couple of bronze bushings and a lip seal.   Now, the engine is mostly a pile of parts.  It has to be totally disassembled to get the crankshaft out but I suppose a good cleaning needed to be done anyway.  

The roller bearings were in good shape and the hardened crankshaft was fine so the roller bearings went into the bearing bin and the bronze bearings were installed.  The fit is a bit on the close side so I've been motoring it with just the crankshaft and a lot of oil and it is running-in fine.

I may have it back together and running by tomorrow.

27 January 2016:
Well, I got it back together yesterday and, with limited time, did a test run.  After the "improvements", it doesn't run well enough to pull itself off the motor.  Well now, ain't that special??!!

My theory is that unseating the rings has left the compression lacking.  Later, I'll motor it for a while to see if seating the rings will help.  

While tramming the engine, I used a dial indicator to determine the -actual- opening and closing points of the valves.  Taking out the valve lash of 0.008", I find that the duration is only about 180 degrees.  I've got to go back to the CAD and see where I messed-up.  Making a new cam with an actual (taking lash into consideration!) duration of 210 degrees may help make the engine run more robustly (says here in the fine print.)

31 January 2016:
I'm getting to think that I shoulda left well enough alone.  After replacing the main bearings and after motoring for about an hour, the engine will barely pull itself.  This is after I re-made the cam, this time with about 20 degrees more duration.  I've also lengthened the cam for better bedding on the camshaft and provided a larger setscrew.

New cam on left.
The exhaust valves now open about 40 degrees before Bottom Dead Center and close right at TDC.  I've also put the flutter choke back on the mixer air inlet and fiddled with the demand regulator, making a new needle for the demand valve.  There are getting to be too many tweakers for fuel mixture.  In addition to the throttle, there's the flutter choke adjustment, the needle adjustment and the diaphragm spring tension adjustment.  All of these seem to allow better control of the mixture but it's a PITA to get them all working together.  

When motoring, adjusting the spark timing makes the exhaust note change like it should but there doesn't seem to be a sweet spot where the engine can accelerate.  The engine sounds strong but can hardly pull itself.  As time goes by, I'll motor it until it gets hot and see if re-seating the rings will improve things.  I've bypassed the fuel delivery disk on the mixer, allowing the demand regulator do all the work.

An idea just occurred.  I may have too much restriction in the gas passages within the mixer.  Maybe, because it breathes better, it's not getting enough gas to run strong.  If the engine doesn't run better with more motoring, I'll drill out the internal passage between the inlet fitting and the throat of the mixer.  Maybe that will help.

5 February 2016:
Today, I opened-up the passages in the mixer and plugged-off the taper valve on the throttle shaft with no big improvement.  Next, I decided to accurately map the valve timing.  First, I made a pointer then used a dial indicator to find TDC on each cylinder and mark the flywheel at that location.

Finding TDC & BDC.
I indicated TDC on one cylinder by pressing the intake valve down with the indicator, then turning the engine until the piston pushed it back up.  BDC was determined by indicating TDC on the opposite cylinder.

Finding valve timing.
Next, I used CAD to make a couple of +- 50 degree strips for the 10-inch diameter flywheel.  The valve timing was detrmined by again using the dial indicator, noting the start of opening and closing of the exhaust valves.  Here's what I got.

#2 cylinder, open at 33 degrees before BDC and close at 8 degrees after TDC (duration 221 degrees).  #1 was virtually the same.

I then changed the timing so the valves closed at about 2 degrees before BDC and tried running it.  There was a very slight improvement so I decided to go for it and move the timing so the exhaust valves opened at about 48 degrees before BDC and closed 10 degrees before TDC.  The engine came to life!  Still not perfect but it now runs a lot better.  I may try to move the timing to close the exhaust valves even farther before TDC but I may be on the wrong side of the curve.

Making a manual choke plate to replace the flutter choke also got the demand regulator to work a little more like it should

While fiddling with the mixture to get it to run well at all speeds (it's pretty close now), I began experiencing something odd.
Ignition arcs.
Occasionally, the spark would jump almost a half inch to ground from the distributor and/or the plugs.  As you can see, I even changed back to the little plugs and then gapped them at 0.015", which is closer than I like but the arcing persisted.  There must be something about propane that insulates the plug gap and causes the ignition voltage to build-up until it arcs externally.

I think this is changed slightly by differences in ignition timing and fuel mixture but, right now, I'm at a loss to know how to fix the problem.

Anyway, for a while there, it chugged along happily at about 500 RPM never missing a lick for several minutes until the arcing started.

15 February 2016:
I finished the new carburetor today and found that it's no great improvement!
     Throttle assembly.                                                                                Throttle assembled..
I'm using a bellville spring washer sandwiched between the knob and the retaining plate so there is some drag on the throttle.  I've also milled a flat on the back of the knob that is stopped at idle and full throttle by the Allen head cap screw.

Here are the drawings for the mixture needle and jet.  Showing the venturi.
Drilling the #70 hole in the jet.                                                             Needle/jet assembly.
Drilling the jet was fun.  A #70 drill bit is easy to break so I modified the drawing so the #70 hole only went to 0.100 from the end.  I drilled it with a #65 from the other end until the holes met.
Needle/jet assembled.                                                                     Downdraft trial.
After all was said and done, the carburetor still slobbered and dripped as an updraft.  As suggested by a viewer, I reversed the manifold to make it a downdraft.  It doesn't drip but I think the fuel is pooling in the manifold because I can shut off the mixture and the engine will continue to run for some time until the excess fuel is used-up.  

At this point, I'm just about out of ideas.  I'm going to make a new ignition cam, one that I can adjust the quadrature on so both cylinders are timed the same.  I don't know if that will help, though.

9 March 2016:
After fiddling with the carburetor, the engine didn't run any better as a downdraft so I put it back that way.

In the meantime, I thought that low compression could be the problem.  As it was, the compression was about 80 PSI so I decided to mill some off of the heads.

Milling one of the heads.
I made the decision to remove 0.035" to make the combustion chambers smaller.  The result after re-installing them is about 98 PSI compression.  The end result is no improvement I can see.  Experimentation stopped before cam timing changes were tried due to the timer contact breaking off.  I think I will make an entirely new timer using either a Hall-effect transistor and magnets or an optical interruptor.  That depends on which device strikes my fancy at the time.

13 March 2016:
After some chip making, I now have a new optical interruptor ignition system.

Camshaft, cam and ignition parts.
First of all, since I got tired of setscrews on the camshaft boogering it up and causing it to be difficult to adjust the cam and rotor, I turned areas on the shaft where the setscrews bear.  This eliminates the burrs and makes life easier.

The new parts in the photo above from the right are the electronic part of the optical interruptor and it's mount and screws.  This part consists of an LED, a gap and a phototransistor.  When something passes through the slot, the light beam from the LED is interrupted, causing the transistor to turn off.

To the left of the electronic interruptor and it's mount is the 0.062" thick phenolic swing arm that the optical interruptor mounts to.  The aluminum part next to it is made to fit loosely into the bore for the sprocket end bearing.  This has a step that fits snugly into the big diameter on the swing arm.  The small bracket fits over the large diameter of the aluminum part and clamps it to the bearing housing, giving the swing arm some drag so it will stay put where it is adjusted.

The next aluminum part is the interruptor itself.  I used the rotary table to accurately position the two slots exactly 90 degrees apart.  The plastic part is the distributor rotor, unchanged from before.

Assembled ignition system.
With everything assembled, you can see how it works.  The handle of the swing arm is moved to the left to retard the spark and to the right to advance it.

After setting the cam timing as it was before with the exhausts opening at 42 degrees Before Bottom Dead Center (BBDC) and closing at Top Dead Center (duration of 222 degrees), the engine was motored over and it was tweaked for best running.  Although it didn't just jump off of the bench, it did run better than before.

I then re-set the cam so the exhausts opened at 34 degrees BBDC and closed at 8 degrees ATDC.  This resulted in better performance.

The last cam timing setting had the exhausts opening at 40 degrees BBDC and closing at 2 degrees ATDC.  The engine seemed to run best at these settings.

Interestingly, after tweaking the spark timing for best running, I checked it with the degree strip and was surprised to find that it was firing at about 10 degrees ATDC.  I would thought that it should have run best with the spark occuring few degrees BEFORE TDC.  The engine bogs down if the spark is adjusted to TDC or before.  I can't explain that one.

I also fiddled with spark plugs, ending up with the CM-6 plugs gapped at 0.022".

I noted that as the engine warmed-up, it got sluggish.  I believe this is because of the close fit of the bronze main bearings.  As the engine heated up, I think they were tightening-up.  Maybe this will go away after a few heat/cool cycles.  Ohterwise, I may have to go back to the roller bearing mains and just put-up with the noise.

More development will follow.  I still need to do some tweaking to the carburetor.  It doesn't want to suck up fuel at low speeds and partially closed throttle.  I've changed the flutter choke spring several times and haven't really found one and settings that will allow a full range of RPM and throttle position.  I may end-up simply eliminating the check valve and making a float chamber so the fuel level is constant at the jet.

I'm going to give it a bit of a rest now.

18 March 2016:
I've been thinking about why the ignition timing is wierd, having to be timed substantially AFTER TDC for the engine to run.  Today, I took the time to draw out both a crankshaft, rod and wristpin arrangement and a Scotch yoke arrangement with the same geometry.  I was in for a surprise.

First, the crankshaft/connecting rod/wristpin configuration:

Crankshaft/connecting rod/wristpin configuration.
Then the Scotch yoke configuration:

Scotch Yoke.
Here's the comparison:

The first thing I found is that the crankshaft, rod and wristpin configuration doesn't yield a true sinusoid in piston motion.  I should have figured it out years ago but just never thought about it.  The non-true sinusioid is due to the changing geometry of the piston and rod with crankshaft position.  Note that the halfway point in the stroke is not 90 degrees on the crankshaft.

The Scotch yoke obviously puts the half-stroke piston position at a 90 degree crankshaft angle and the piston motion appears to be a true sinusoid.

On the power stroke, expansion occurs later in the stroke compared with the Scotch yoke.  I think this is the cause for ignition timing having to occur past TDC.  

Who'da thunk it??!!

23 March 2016:
FINALLY!  The engine runs almost like it's supposed to.

Tweaks I've recently made include new valve springs and a better demand regulator.  I made the new springs because one cylinder ran a little richer than the other.  My theory was that mismatch of the spring rates caused the problem.

New valve springs.
After experimenting with winding a spring to find out what the "relaxed" diameter of the spring was, I machined a piece of bar stock to the dimension I thought would work.  It came out pretty close and I then counted off the same number of turns for each spring and stretched them to the same length.  This was to insure that both valves worked the same way.  It seems to work because now, both cylinders richen up and lean out at about the same time.

New demand regulator pallet valve.
Next, I reworked the demand regulator I used earlier.  Instead of a needle valve to turn the gas on in response to manifold vacuum, I turned the needle down to a pin then used some of the rubber sheet I made the diaphragm from to make a button that fits over the pin and is glued into place.  In operation, the valve acts like a pallet valve and is sort of either on or off.  This seems to be an improvement.

Propane fuel system.
The fuel system consists of three parts.  Connected to the manifold is the throttle body.  It is fed propane from a needle valve which sets the maximum amount of gas allowed under full throttle.   The needle valve is fed from the demand regulator which, in turn, is fed from the grill regulator on the tank.

Adjustment is tricky.  I'm using the spring loaded flutter choke in the air inlet to the throttle body to create some suction so the demand regulator will operate.  The demand regulator turn-on point is set by turning the screw that compresses a light spring to push the diaphragm into the pallet valve until manifold vacuum can lift it off..

While motoring the engine with the throttle and needle valve wide open and the flutter choke adjusted about half way, the gas is turned on at the tank and the demand regulator screw is backed-off to where the engtine starts firing.  It is set a little on the rich side and the needle valve is closed until the engine runs steadily.  The motor is then slowed-down and the throttle is set to approximate idle.  The demand regulator and flutter choke are then diddled for steady running.

After steady idle is achieved, the motor is sped-up again and the throttle is opened.  The needle is adjusted for smooth running.  Of course, the job is not finished and several speed-up and slow-down adjustment cycles are performed until idle and full throttle can be achieved with no misfires.  

One of the interesting things about this engine is that, even after nearly an hour of running, the temperature at the heads is only about 150 degres F.  Granted, the engine isn't running at load but I guess the reason for the relatively low temperature without a cooling system is the mass of the block and base plate and the slight breeze from the flywheel spoke fan blades.

It's still not ready for prime time but, since I've got it running better than ever, I've made another YouTube video.

And here's the latest flick.
31 March 2016:
Well - I'm a-chasin' the problem with carburetion.  What I found out is that the first cylinder to fire in the firing order runs richer than the second cylinder.  That means I have to adjust the fuel mixture half-way between and the adjustment is touchy.  After doing some head scratching, I came to the conclusion that since the gas was being fed constantly, in the 360 degrees of crankshaft rotation between the second cylinder and the first, fuel builds-up in the manifold.  The first cylinder sucks up this mixture but, since there's only 180 degrees of rotation between the two, the mixture hasn't had time to richen-up again before the second cylinder begins sucking fuel.  That sorta sucks!.

My attempt at a solution is to make a bigger intake manifold with higher volume that allows enough mixture to be present in the manifold to make the difference in mixture between the two cylinders less.

The new manifold.
The new manifold does, in fact, make the problem go away from a practical standpoint.  I still have the issue of getting the demand regulator to work right. and to control the idle mixture better.

21 April 2016:
I thought an update would be in order.  After putzing with the fuel system, I went back to the liquid fuel mixer for simplicity's sake.  The engine still didn't thrive.  Maybe I'm expecting too much, running it slow.  If I speed it up, it seems to run better but I'm not comfortable running it at over 1,000 RPM.  Anyway, I've got the ignition rev limiter set to cut-out at a little over 1,000 for safety's sake.

Before making the "show" skid, I tried reducing the valve duration from about 230 degrees to 200 degrees by increasing the valve lash.  After adjusting the cam so the exhaust valves close right at TDC, the engine runs better but I can't reduce ia more by increasing lash because the valve opening is getting smaller.  When I get back to the project, I may make a cam that I can change the duration on.  Because of the way crank angle/piston position works with the Scotch Yoke, I may need to reduce the duration to just a shade over 180 degrees for the engine to run at it's best.

Here it is as it sits now.  It's covered with a couple of old towels and is resting comfortably in a corner until the spirit moves me to tinker with it some more.

The engine, semi-ready for Prime Time.

Boy!  This Is Fun!  

Comments?  Email me

Unfortunate individuals have stumbled upon this page

Since 18 June 2015.