Construction of
The Mystery Engine

As a Politician said,
"We won't know what's in it 'til we complete it.
21 April 2013:
Yesterday, while fiddling around in the shop, I was attracted to a 4" long piece of 2" diameter bar stock.  The word "CRANKSHAFT" popped into my mind so I headed for the bench when I tripped over the left-over gray cast iron I made the head for the 30-60 out of.  Both pieces ended-up on the bench.  While doodling on a piece of scratch paper, I remembered that I'd bought a set of 2:1 miter gears at Zolfo Springs so I added them to the pile.

The raw materials and sketch.
The shape of the 6" long X 4" tall X 2" thick hunk of cast iron shown has a hopper cooled engine block inside of it.  All I've gotta do is remove all the unnecessary metal.  I now have a sketch that shows a single-flywheel engine with a sideshaft.  I have a hunk of 1" thick steel I may have the flywheel waterjet cut out of.  So far, I have no idea what the valve arrangement will be.  Hopefully, something interesting will come to mind.  Hit and miss governing, probably.  Inertial governor?  Maybe.

Starting on the crankshaft.
After facing the piece of bar stock and putting centers in both ends, I turned the long main to 1.010".  Then it was flipped and the short main was turned to the same diameter.  The finished diameter will be 1.000" but I didn't want to take the chance of dinging up one of the journals in the rest of the process.

Mains done.
With the mains done, I dug out the offsets I used to turn the crankshaft of the 30-60, which has the same main bearing journal diameter.  I deeply center punched them at 0.5625" from the centers of the mains to give a stroke of 1.125".

Offset in place, preparing to turn crank throw.
The setup you see above for turning the rod journal isn't exactly what I ended-up using.  The offsets were fine but I had to grind a special thin tool that could reach all the way to the 0.8125" finish diameter of the crankpin.  That took a while and several experimental tool cutting end shapes to get it done. 

Crankshaft nearly done.
What I ended-up with looked like a wierd cutoff tool made of a 3/8" tool bit.  I ground it to about 0.100" wide and, to minimize chatter, I ground a groove in the center of the cutting edge that made it into a two pointed tool.  This seemed to cut better than anything else I tried.  The last 0.0005" was taken off with a fine file and emery paper.

End of crankshaft bored for gear.
The crank gear was next.  The diameter of the large portion of the small gear shaft is 0.622".  I cut the gear shaft down to 0.375" from the thrust face.  The end of the crankshaft was drilled 0.500" diameter to a depth of 0.400", then bored to 0.623(-) and the gear was pressed into the crankshaft.

Gear pressed into crankshaft and sideshaft laid-up for a look-see.
The only remaining thing to do to the crankshaft is to mill a keyway for a gib key for the flywheel. I think the bore will be in the range of 1" to 1.250" or so, depending on what I've got to make the piston out of and how long the rod can be, etc.  I think I can hog the block out with the bandsaw after I've milled the outsides of the hunk of iron flat and square.  Boring the cylinder may involve making a boring bar to run between centers in the lathe and clamping the block to the carriage. Bearings will probably be bronze and I'll do what's worked fine in the past for a wrist pin - a ground and polished dowel pin. This looks like it will be a minimal CAD project.  I may lay out the block, crank, piston and rod with the 'puter just to make sure I don't have an aw-shoot in the fit department.
22 April 2013:
Not a lot to report today.  Spent most of my time turning that block of cast iron into something square and clean.

And a fine hunk of iron it is!
Now that all the grunge, scale, etc. is flycut off of the piece, I know what I have to work with.
23 April 2013:
Milled the keyway in the crankshaft and did a little CAD.  I think I'm about done for now with CAD and can do some more whittling.  I've decided to try a PTFE (Teflon) "O" ring for a piston ring.  Since it's pretty stiff and a little bigger than my bore, I'll split it and gap it like a metal ring.
Here's the drawing.......

............ so far.
24 April 2013:
Some cast iron sawing got done today.

The puzzle.
While running Engine Number Four to test the Piezosparker or "Piezomag", I laid out and machined the bore.  Then, I laid-out the cut lines for the block and rough sawed the pieces. Dang!  Cast iron's messy.
28 April 2013:
I got the majority of the whittling done on the block. 

It's beginning to look like an engine.
The next thing to get done is to make the main bearing caps, bore for the bearings and hang the crankshaft.  I think I'll make a 1/4" base plate for it so mounting will be easier.  I'll also make a plate that covers the hopper and has a smaller opening in it.  Since I'll probably be painting this engine, I may fill the lines made where the base plate and hopper cover go on and finish it off so, once the paint is on, the joints are invisible. So far, I think I'd like the flywheel to be on the usual side.  That will put the sideshaft on the back of the engine.  This may change as this detail can be modified before mounting the sideshaft.
29 April 2013:
Doing the main bearings is taking a little time. 

Cutting the main bearing caps from some of the sawn-off cast.
The main bearing caps are  made from a piece of the removed cast when the block was roughed-out.

After milling the caps to size, the bolt holes were drilled and tapped.

My method of fitting bearings consists of first accurately measuring the O.D. of the bushings.

Halving the main bearing bushings.
After the O.D. of the bushings is noted, they are accurately marked at the halfway points in the circumference.  In this case, the O.D. of the un-split bushings was 1.128".

I use a Dremel cut-off wheel with the Dremel handpiece mounted in the lathe chuck.  I've clamped a drill press vise on the carriage and, with the Dremel going fast, I slowly feed the bushings into the cutter along the scribed lines.

After deburring the shells, they are held together and the diameter is measured perpendicular to the cut lines.  The measurement is 1.086" and this is subtracted from the original O.D. of the bushings and a shim pack is made-up in this thickness.  In this case, the shim pack is 0.042".

Boring the main bearings in the block.
The caps are bolted to the block with the shims in place and the block is mounted in the mill and the main bearings are bored to 1.125", giving 0.003" of crush on the shells. After the boring is finished, I will put the bearings in place and check the I.D. against the O.D of the crankshaft.  If I'm lucky, all I'll have to do is to hone the bearings a little then lap them to the crankshaft.
1 May 2013: I made a big engineering decision yesterday.  I've decided to go with different sized bushings for the mains.  When I'd gotten the other bushings ready, I came to the conclusion that the thrust face of the bushings was too narrow at 1/8".  Ordered some bushings that will give a thrust face of 1/4".  I'll turn down the O.D. to fit the bores in the block.

Today, I made shims, drilled some holes and made the base plate and the unfinished hopper top.

Waiting for the epoxy to harden.
Tomorrow, I'll clamp the block in the mill and match the surfaces of the hopper top and the hopper.  After that (says here in the fine print) I'll belt sand the top and fillet the edges and, after painting, you'd never know that the hopper was pieced together.  Maybe I can get some "experts" to wonder how I machined out the inside of the hopper. There may be an "Aw Shoot!" brewing.  The piston won't go in from the back and the big end of the rod is too big to fit into the bore.  I'm going to go blythely along 'til I've finished the rod and see if the little end will clear the top of the bore enough to put-in the wrist pin.  If not, I'll have to think on it some.
2 May 2013:
It's slicked-up some and I have the mains made and lapped-in.

The top of the hopper gets levelled.
Overnight, the epoxy hardened very nicely despite it being over fifteen years old.
The sides of the hopper lid are matched to the hopper .
Chamfer applied to top of hopper.
If you look at the photo above, you will see that I subscribe to the axiom, "If you ain't with the tool you need, you need the tool you got" (or words to that effect).  I needed to chamfer the top of the hopper and all I had was a really old valve seat refacer.  it's 60 degrees and made a nice looking chamfer, although I had to feed it really slowly to keep from damaging or dulling it.

Main bearings being lapped.
I won't be showing the steps to get the mains done.  Most of them were the same as I did with the previous bearings.  The bushings were turned to fit the block and the thrusts were adjusted to fit.  Then, after buttering some Timesaver lapping compound on the main journals, the whole works was assembled loosely with the shim packs.  The crankshaft was chucked-up in the lathe and run at the lowest speed while the main bolts were slowly tightened.  It is easy to tell how it's going by simply holding up the head end of the engine and noting the torque required to turn the crankshaft. 

The bolts were tightened until an increase in torque was felt then it was allowed to run for a while, adding thinly mixed lapping compound as the bearings worked-in.  When the main bearing bolts were tight, I substituted oil for the lapping compound and let it run for a while to flush out the compound.  Total time was about a half hour.

The works was disassembled and everything was washed in solvent.  The bearings showed 100% contact and the crank looked good.  The lapping process maybe took off one ten thousandth of an inch from the crank.  After reassembling, it was again put on the lathe and run for about another 15 minutes.  The fit, when checked-out is excellent.

Here we are so far.  Holding sideshaft gear up for ideas.
I'll take a day or two off the project then will start on the rod.
5 May 2013: Moving right along, today the rod was partially completed.

Getting ready to lay the rod out.
I got lucky and found a chunk of 1/2" HRS that a portion of which was exactly wide enough for the rod.

Laid out and the radiuses drilled.
After sawing the blank, the major points were laid out.  The radiuses between the big end, little end and the rod are simply 1/2" holes.  After sawing out the bulk of the steel between the radiuses, the rod was put into the mill and the rest of the steel was removed.

Then the rod was clamped in the vise at 45 degrees and the big end radiuses were blended.

Cap separated from rod.
You will notice that the little end of the rod isn't per the drawing.  It is squared off and I've drilled a center in the end.  There's another center in the cap where the grease cup will be.  This is so, if there are clearance issues as the rod oscillates around the crankshaft, it can be put into the lathe and the corners turned off.  I'll wait until the rod's finished and the piston's done so I can assemble it and see if there is any interference between the bottom of the bore and the rod.  If all's well, I'll then finish-off the little end and drill and tap the big end for the grease cup.

At this point, I inserted the rod into the bore from the back and there's plenty of room to assemble the wrist pin from the top.

Ready for shims and boring big end.
The rod bushing was cut in half then measured to determine the shim pack to bring it into roundness.  The shim pack ended-up being 0.040".  The O.D. of the uncut bushing was 1.001" and I want about 0.002" of crush so I'm making up a shim pack of 0.038".  The thick shim is 0.036" so I've got to cut two 0.001" shims to bolt-up.  The bore of the rod will be 0.999" to make sure the bearing is held tightly.  If it's too tight in the rod, I can always dress off a thousandth or two from the halves. After the shims are in and the rod cap is tightly bolted-up, I will carefully lay out the big end spacing from the little end.
6 May 2013:
The rod and the piston are done.

Here's what I got done today.
The work today wasn't technically challenging, just time consuming.  The piston is made of 6061-T1 bar stock.  I've allowed 0.003" clearance in the cylinder.  The bore is 1.125", so it's a little close but should wear-in fine.

You can see the Teflon piston ring.  It's fitted with the ends butted together because I figure the "0" ring will quickly wear a flat where it rubs on the cylinder and this will quickly open up the gap.  We'll see. 

The grease cup is an extra I had from The Homebrew Engine project so it's going to be used on the rod.  I will make some grease cups for the mains and sideshaft.

Running-in the rod bearing.
I just got started lapping the rod bearing.  I'm using the finest TimeSaver compound so it will take a little while.  The way I do it is to set the shim pack just a hair on the snug side, about 0.0005" or so.  Then when I first assemble it for lapping, I only tighten the rod bolts until there is substantial drag.  After a while, I can tighten the bolts a little then run it some more.  When the bolts are fully tight, the bearing is broken-in. After it's run for a while, I'll take it apart and see what the pattern looks like.
17 July 2013:
I ran the engine for a while with the gr
7 May 2013:
Here's what the rod bearing halves and the rod journal looked like this morning.

Rod journal and shells after lapping.
When I took it apart, I could see that the wear pattern was 100%.  No more lapping was required.  Since the bolts weren's fully tight and the bearing was still binding slightly, I made one 0.001" shim.  Back together, it was still just a little tight so I applied more of the finest TimeSaver lapping compound (diluted in oil) and ran it until it was free, about another 15 minutes.  My guess is that the fit on both the rod and the mains leaves only a couple of tenths of clearance.  I'll be using a light Teflon filled grease so it should work out fine.  I guess that, after several hours of running, the fit will be between 0.0005" and 0.001", good clearance for light grease.

Getting ready for paint.
After all that, it got taken apart again and is being prepped for painting.  It won't be super slicked because the paint I selected is an old can of RustOleum Hunter Green that I'll apply with a brush.  In case you ask, if you click on the photo and look carefully, the bearing bores and mating surfaces are masked as well as where I think the sideshaft bosses will go.  On the head end, I've masked where I think the head will be.  To the right of the engine, you can see the piece of cast iron I may be using for the head.  It came off of the block and is the same width as the hopper end.  All that is in question is if there's enough room for the valves and combustion chamber.
9 May 2013:
The paint is applied, the sideshaft and mounts are mostly done and I've made a couple of grease cups for the mains.

Sideshaft and grease cups.
While waiting for the primer and paint to dry, I made the supports and bearings for the sideshaft.  Since I haven't figured out just what the valve arrangement is going to look like, I masked-off a mounting pad on the block for the cam end of the sideshaft.  The other mounting pad is on the back of the nearest main bearing boss.

The grease cups were sketched out and the tops and base are threaded 3/8-24 while the bottom where they screw into the block are 10-32.  The holes are tapped in the mains with a plug tap and the cups thread in like a pipe thread.  I like my new knurling tool although I haven't quite gotten the hang of doing matching knurls.

I just had to take a pitcher of it with it's paint and grease cups on.
Just before I quit for the day, I made the oiler pipe and drove it into the bore in the cylinder block using some LokTite.  I don't think it's gonna leak or go anywhere.
12 May 2013:
While waiting for the paint to dry, I got some more parts made.  Today, I did some assembling.
It's beginning to go together
Do 'ya think I'll have wear problems on the sideshaft gears?
After getting the sideshaft mounts done and having a look, I think the gears are just a little teensy bit out of scale.  I went with what I had!  I don't think I'll have an overloading problem. I greased it up and ran it some more on the lathe.  It now turns nice and smooth.  Also, parts of the oiler (I'm making two - one goes in the oiler drawer).  I'm now waiting for the polycarbonate tube and some brass stock to come in.  The shipment should be here late Monday. I was going to use glass for the sight glasses but, since I was ordering from McMaster-Carr anyway, I got the polycarbonate.  I can cut it myself to any length I need. The flywheel is going to be about 8" in diameter and 1-1/2" thick.  I ordered the ductile iron for it along with the other stuff. I think I can make the head and valve cages out of cast scraps from other engines but haven't gotten that far yet.
15 May 2013:
The last couple of days were devoted to making the oilers and the wood base. 

The oilers are a study in "Once you figure out what you're doing, you're done".  The polycarbonate tube is easy to cut and work on the lathe.  Making the small brass parts was more of a challenge and took some time.

Engine with oiler.
I still haven't totally figured out the head and have several ideas for a cam-stopper but all of them are pretty complicated.  I may just not do that.  While waiting for my brain to grind out a bit more detail, I think I'll start on the flywheel.  It should fit in the lathe.  If so, it will be easier to make than the larger ones that have to be jury-rigged in the mill.
16 May 2013: Working on the flywheel, it took most of the day to lay it out, bore the hub, clean up the O.D. and face it. 

Really more than my l'il lathe is designed to do.
I probably shouldn't try something this big in my little lathe.  It was working really hard, tripping the mechanical overload clutch if I fed it just slightly more than really slow on the larger diameters.
18 May 2013:
Yesterday and today was spent finishing the flywheel and getting the engine mounted on it's wood base.

I like big flywheels!
The hunk of cast iron I got could be ordered either 1" long or 2" long.  Since I wanted a 1-1/2" thick flywheel, I ordered 2".  When I got to machining it and realized how long it was going to take, I decided to just clean it up and use it at a little under 2".  Also, I wasn't planning on the web between the hub and the rim being over about 1".  Again, it was taking altogether too long to o so it ended up being more like 1-1/2".  It could be considered to be an "electric" flywheel.  As you can see, the engine should run real smooth.  If it ends-up being too heavy, I can put it back into the lathe and face it to 1-1/2".  Doing this would take more than a day to do.
20 May 2013: The last couple of days have been partly spent at the CAD  'puter.  I tried to lay out the head arrangement using simple notes and sketches but decided that there were just too many places to go wrong.  Here's a simple view of what I've come up with.

An idea of what the head and valve actuation will look like.
The view above is facing the head end of the engine. Of course, the drawing doesn't show bolts or the valve details.  The valves, springs and keepers are some I rescued from a dead 4-cycle weed eater.  As you can see, there's only one cam lobe driving both valves.  It is shown in the "overlap" position just as the exhaust valve closes and the intake opens.  The duration is necessarily the same for both intake and exhaust, about 200 degrees.  I can change duration by increasing or decreasing lash.   The rocker arms are 1:1 and the cam lift is 0.200, more than enough for the engine to run. The lash adjustment is pretty Mickey Mouse.  Right now, the pivot point for the rockers can be moved around to change lash.  I may  change to tappet bolts if that doesn't work out.
21 May 2013:
In thinking about it last night, I figured I might as well put the adjustments where they are supposed to be.  Here's the new drawing.

Revised Drawing
22 May 2013:
The head is finished

Cylinder head in place.
This was one of those days when everything seemed to go fine 'til the last step.  The problem today was that two of the head bolt holes in the block somehow got drilled about 0.030" out of position.  There are a couple of ways to fix something like that.  One would be to plug the holes then re-drill the holes in the right position.  I couldn't do that because the block is cast iron and any screws or plugs I would have would be steel.  I was afraid that, even if the holes were plugged and the new holes started in the correct position, the difference between the cast iron in one part of the hole and the steel in the other part would make the drill drift. My solution was number two solution.  That was to drill the two bolt holes in the head oversize so it could be shifted to the correct position. I'm sure you'll never tell a soul what lies hidden behind the top two head bolt washers. After a day or two off, I'll be tackling the valve cages.
27 May 2013:
Today I got the valve cages almost done.

Starting with a scrap from the 30-60 flywheel.
It's good I collected the left-overs from the 30-60 flywheel.  One of the pieces from between the spokes was big enough to make both cages.  Cast iron, even!

Cut into manageable pieces.
First, I cut the piece down the middle and trimmed off what I didn't need.  An end milling operation got them square then, using the 4-jaw chuck, I drilled the valve guide hole, drilled the port and turned the two outside diameters.  Since I turned the valve seats while the cages were still in the chuck, I shouldn't have much trouble getting the valves lapped to seal.

The nearly finished cages.
When I drew the cages, I didn't get the spring height right so I will have to make a spacer for the exhaust to increase the tension on it.  The intake, done later, is all right because I allowed another 0.031" for spring tension.  I think that by tomorrow, I should have the valves lapped-in, the gaskets made and the head on to check for compression.
28 May 2013:
The valve cages are done and I got the valves lapped.

It's now got valves.
The only problem is that they show good seats after the lapping but they leak when the cages are bolted-up.  At first, I thought the copper sealing rings were the culprit so I made new ones out of 0.031" high temperature gasket material.  Same problem.  After spending about four hours fiddling with it, it was quittin' time at Hoyt-Clagwell & Company. Tomorrow, I'm going to do some tests to see if the problem is that the valve cages are distorting slightly when tightened down.  If I loosen the cage bolts, I stand the chance of blowing the gaskets but I can see if the seal is better with less compression on the cage. If distortion is, indeed, the case, I will have to come up with a way to do a final lap with the cages in place.  That will require removing the head but with only four bolts, that's easy.  In any case, today proved that the Teflon piston ring will seal.  With fingers on both ports, it will develop a fair amount of pressure.  Now, how long it will last when the engine is running is another question.
29 May 2013:
Things moved along smoothly today.  The valve seating problem was, in fact, due to distortion in the valve cages when they were bolted to the head.  The solution was to remove the head and, with the valve cages in place, re-lap the valves.  They seem to seal well now.
Next up was to make the rocker arm stand and rocker arms.  These parts were plotted to exact size then the plots were cut out and glued to the steel for the parts.  After that, it was a simple job of following the lines with the bandsaw, belt sander and files.

"O" ring seal on cage.
Because I was worried about leakage around the cage, I cut grooves in the cages and made a chamfer and counterbore in the head as shown above.  This should make a really good seal.  Now, whether the "O" rings can stand the heat is another story.
  Rocker arms in place .
End view of rocker arms.
The only thing left to do with the rocker arms is to drill and tap for the 6-32 adjustment bolts. Now, it's time to tackle the camstopper and cam.  I've got to sort through the ideas in my head and come up with a reasonably elegant design.  Since I've never seen the guts of a camstopper, I'll have to re-invent that particular wheel.
1 June 2013: I'd like to thank the guys at Harry Matthews' SMOKSTAK for their input into the design of the camstopper I'm going to build for this engine.  It took four design tries in CAD before I yelled "Uncle!" and got help.  What we now have is a sort of modified Callahan mechanism. One of the major things I had to work out was how to make the latch go around the sideshaft.  On the Callahan, the camstopper is located at the end of the sideshaft and the shaft ends before the latch crosses it's path.  On my engine, the camstopper is before the end of the sideshaft so complications ensued.  What I've got should work. I'm going to post the CAD drawings (in JPEG form) in case you want to get a headache figuring out what goes where.

Thiis view shows the parts in the unlatched position (sideshaft driving cam - hit cycle)

This view shows the parts in the latched position (Cam idling - miss cycle)

You should be able to click on the above images to get a higher resolution view that you should also should be able to copy to your printer.  That may make them a little easier to read.
This is going to be confusing and maybe I shouldn't even post the drawings.  To give you a little help, some of the colors are: Magenta - Bearings. Green     - Cam and governor carrier. Blue       - Latch, latch arm, screws, rocker arms, sideshaft. Yellow   - Springs, sideshaft bearing mount. Cyan      - Driven part of the camstopper (holds the latch bar). Red        - Shaft collar for adjusting speed spring. Black     - Various pins, washer and short shafts. I haven't gone to the trouble of hiding any lines.  The hiding would make the drawings easier to understand but is time consuming. I'll probably get the materials ordered and start making swarf tomorrow.
3 June 2013: Well, yesterday I found out that I'd gotten some of the drawings reversed and had to spend the day re-dimensioning all of the camstopper and governor parts.  Today, I did get three parts made.

Slowly making progress.
The part that is partially done (awaiting a ball bearing so I can drill and bore the mount) is the sideshaft bearing mount (yellow in the drawings).   The other two parts are the disc the weight arms ride on and the thrust bearing link to the latch rod (a couple of the green parts).  I've scrapped the previous head-end sideshaft mount. Tomorrow, I'm hoping to get some more parts made before I run out of stock.
 4 June 2013:
A little more is done.

Making some progress on small camstopper and governor parts.
There are a number of small parts to make for this engine, especially in the camstopper and governor.  I'm almost at a standstill until I get some more metals and, in any case, I've got some other things to do so I'll probably take a couple or three days off.
8 June 2013:
More progress today.  The governor parts are made and assembled to check for fit.

The governor is about done.
As you can see, I've got it belted-up to a variable speed motor to test the governor and break-in the sideshaft end bearing.  After the camstopper is built, I'll use the motor to check and tweak the governor speed adjustment. It looks like the piezo ignition and the mixer are going to be about the last thing I'll do before seeing if it will run.  That will be a while yet because the camstopper parts are pretty complex and will take some time.
9 June 2013:
I've finished what I call the camstop.  That's not actually what it does.  It's really the driving member of the stop assembly.

Here's the driving part of the camstopper mechanism.
This part is what is solidly connected to the sideshaft via the setscrew.  When the governor allows the engine to run, the latch drops into the slot and, since it's connected to the cam, the cam turns.  At least, that's the way it's -supposed- to work!

This is what I call the latch.
This was kind of tricky to build.  I started out with a piece of flat 1/4" thick steel and used the mill to make part of it 1/8" thick then used another milling cutter to hog-out the inside curve.  The outside curve was rough sawed then filed to the necessary contour.  I left a little extra "meat" on it so it can be whittled on to make it work in the assembly.

The Driving member with the latch engaged.
The latch is shown in the engaged position.  When in this position, it rotates with the driving member and the cam, flipping the bar (in the background) out of the way of the end of the latch.  When the governor shoves the pin into the path of the bar, it forces the latch (against an internal spring)| toward the camera which disengages it from the driving member. Now that I'm done with the governor, I may need to revisit the weights.  It may not have enough muscle to effectively drive the rod into and out of engagement with the bar.  We'll see.
11 June 2013:
After a few "aw-shoots", I now have all the major camstopper parts made. 

Camstopper assembled.
After getting the cam made, I assembled it and found the first Aw-shoot!  The part I call the camstop (the friving member) has a ramp to make it easier for the latch to engage with the slot it.  When I milled the ramp, I put it on the wrong side of the slot so I had to make another. Then, after all was put back together, I found that, for some reason, the intake (lower) rocker arm cam follower doesn't have the right relationship to the cam.  As of now, I don't know what the problem is.  So far, I found that the rocker arms are identical and the rocker arm stand is symmetrical.  Tomorrow, I'll get into it and find out just what's the matter.
12 June 2013:
Things are looking up!
I got the rest of the parts made and assembled.  The short answer is that, after a little fiddling, the camstopper works.  There are still some issues to address, the worst of them is that sometimes when the cam is operating, it declutches itself when the intake valve is ramping closed.  This causes the intake valve to close prematurely.  I believe that by doing some filing on the latch, I can alleviate the problem. Also, although the governor does work, it doesn't have enough power to properly work the latch bar.  I'll probably be making some heavier weights and a stiffer spring to help the situation. The rocker arm problem was fixed by making another intake rocker that has a longer arm from the pivot to the cam follower.  The problem is caused by a small mis-alignment in the sideshaft.  It was much easier to just make the correction with a new rocker arm than to re-do the entire governor and sideshaft mount.  The valves operate correctly and there is plenty of lift so I can adjust the overlap (now about five degrees) by changing the valve lash.  I haven't checked the actual cam timing but, by looking at the crankshaft position as the valves operate, it appears to be pretty close to what I planned.
13 June 2013:
Today, I found out what was causing the jumping of the camstopper.  There wasn't enough engagement of the latch and that made it cam out of engagement when negative torque was applied.  The fix was to machine a groove in the inside of the cam portion of the camstopper that gave the latch another 0.050" to move into engagement.  I think that problem is solved.
Next, since the governor really didn't have enough oomph to positively drive the pin that engages the latch arm (the part on the outside of the camstopper), I made new weights that are quite a bit heavier.  At "quittin' time", another problem surfaced in the form of the thrust plate for the weight arms binding on the sideshaft.  If I can't fix it the easy way, I'll have to make another plate, this one having a longer bearing area on the sideshaft to alleviate binding.

New, heavier governor weights.
I think that, once this is worked out and the rest of the bugs are fixed in the governor/camstopper, I can get on with a mixer and ignition and see if the danged thing will run.
14 June 2013:
Today I made a new thrust collar for the governor weight arms.  The previous one only had a bearing length on the shaft of about 0.400".  Since it is supposed to float on the 3/8" sideshaft, it would bind when moving.  I turned off the old collar in order to avoid damaging the ball thrust bearing and made a new collar with about a one inch bearing area.  It slides a lot easier with no binding.  After a few tweaks (fiddly bits), the governor and camstopper work well enough on the belt to rate a video.  It is below.
As you might note, there is a small crankpin on the cover at the end of the cam.  This will be used to drive the piezoelectric ignition.  No drawings have been made of this mechanism but it will be different from the previous Piezomag.  I'm trying for simple and easy to build.  We'll see. First, probably will be the exhaust manifold and pipe and the mixer which I'll start on directly. By the way, when the engine is motoring on the belt, it has intake suction and some exhaust flow so I'm hopeful it will run when it gets a proper fuel/air mix and something to light it off.
15 June 2013:
The mixer and exhaust went quicker than expected and are done.
Head end of mixer (left) and air intake end of mixer (right).
The exhaust and mixer bodies are made from some 5/8" thick cast aluminum.  The exhaust is simply a chunk of the aluminum milled to match the port then drilled for a tight fit of the tubing as shown in the bottom photo.

The mixer is milled to match the port then drilled through with a 1/4" bit.  From the air intake side, I ran a taper reamer through to almost meet the milled area at the port.  The reamer small end is 1/4".  At the air end, the hole is about 0.312" tapering to the jet. 

The bottom (fuel end) of the jet is made of 1/4" hex brass stock.  The stock is turned and a 10-32 die is run over it to screw into the mixer body.   A  #60 hole is drilled  from the jet end far enough to meet another hole from the fuel end and a length of 1/8" copper line is stuffed into larger hole and soft soldered into place.

The top end (needle valve end) body is prepared like the fuel end and screws into the top of the mixer body.  A hole is drilled and tapped 4-40 through the jet for the needle valve.  The needle valve is made by turning a tapered point on the end of a 4-40 machine screw.  The head is cut off and a brass knob is screwed onto the end of the needle and soldered into place.
Choke closed (left) and choke open (right).
For the lack of a better idea, the choke is made of a piece of 0.030" thick aluminum that is bent to fit over the mixer body and sprung to stay in place.  A slot is cut so it can avoid the mounting screw.  Slide the choke one way and it covers the air intake.  Slide it the other way and it allows the air intake to be almost fully open.  After I get the engine running and tweaked (whichmight necessitate somje changes to the mixer), I may re-visit the choke for something a bit more substantial.

And here it is, almost ready to start on ignition.
I'll make a couple of mounting saddles for the fuel tank and run the pickup tube (with check valve at the bottom) through where the 1/8" NPT plug is now located.  At that point, the ignition will be on the schedule. The next couple of days will be a break so I can do other things and marshal my thoughts on the piezo ignition.
18 June 2013:
Not a lot has gotten done in the last couple of days.  The fuel tank is mounted, the check valve is made and it's plumbed-up.  I spent some time in CAD working out the piezoelectric ignition and have started making the parts to it.

As of 18 June 2013.
The "dangly bit" hanging from the end of the cam is what I've so far gotten made for the ignition.  The top end has a 10-32 female thread in it into which screws a rod (with locknut).  This adjusts the point at which the black cam trips the hammer of the piezoelectric unit.  The longer the rod is, the more advanced the timing is.  The reason the trip cam is black is that I tried to oil harden it.  I don't think it hardened much if at all.  The steel is some that I got from the scrap yard and I have no idea what the alloy is.  If this idea works and the trip cam wears quickly, I suppose I can either armor the wear point with a piece of hacksaw blade or buy some hardenable steel and remake the part.  Right now, since the ignition mechanism is experimental, there's no real reason to go to the extra effort and expense.

There are a few more parts to be made for the ignition but I'm guessing that if all goes well, my first try at running it maybe in the next session. My next session may not be for a couple of days, though.
21 June 2013:
Back at it today.  I thought I'd be able to see if it would run but, because I screwed-up one of the ignition parts and had to do it over, I barely got the ignition assembled before "quittin' time".
Partially assembled ignition.
I don't know exactly why the trip pin was located wrong in CAD but, because of that, I discovered that if you whack the piezo element hard enough, it will make an almost 1/2" spark!  How many sparks it will make before the element shatters when hit that hard is a good question.

After relocating the trip pin by eyeball, it now seems to trip at the proper point.
Ignition installed on engine.
Tomorrow, I'll finish mounting the ignition and maybe I'll try to start it.  Whether it runs or not, I'll makie a video of the attempt.
22 June 2013: As promised, below is the video of today's first starting attempt of The Mystery Engine.  Although I did get it to fire regularly and heat up, it never did make enough power to get off of the belt.  I'm thinking that the problem is due to low compression.  I don't see any evidence of excessive blowby and the valves seem to be seating so the cause might be low compression ratio.Something I noticed was that ignition timing didn't seem to make a lot of difference in the firing.  I adjusted it between a little after TDC to quite a bit BTDC, ending-up with the ignition occuring just before TDC.  Another thing is that, although the piezoelectric ignition works well, it makies an irritatingly loud snapping sound.
Anyway, when the engine is motored at just below governed speed, it can make enough power to boost its speed just enough for the governor to latch out.

Better'n nuthin', I guess.

Over the next days, I'll work to get it to run right.  It's only a matter of fiddling with it 'til it cooperates.  Stay tuned.
23 June 2013: I had an idea last night and, today, made an adapter from a quick connect air hose fitting to a 10mm spark plug.  I used this to do a leak down test and found that the low compression problem was a very poorly sealing (non sealing?) piston rinhg.  As you recall, I tried using a cut-down Teflon "O" ring for the piston ring.  Here's what I found upon teardown.

Teflon piston ring removed from engine.

As you can see, the ring was leaking badly.  Since I had no other ring handy, I made another Teflon ring from another "O" ring.  This time, I turned the O.D. flat then added a spacer to the ring groove to make the edges seal.  After putting it back together, it really tried to run and, a couple of times, when it was running on the belt, I slipped off the belt and it continued to run, going on and off the governor a few times before pooping out.  It didn't run long enough to get a video.  A little improvement may encourage me to do that.

It could be that the ring will need to seat to seal really well but, as of now, it will bounce off of compression although weakly.  Tomorrow, I'll run it some more and see if it improves.  If not, I may try my hand at making a cast iron ring to fit the oddball size ring groove.  That should be easier than making a new piston for a "store bought" ring.


24 June 2013:

Today, The Mystery Engine ran on its own for the first time.

After making a smaller (0.150" diameter) venturi and opening the valve lash a lot to reduce some of the excessive overlap (about 15 degrees), the engine was started on the belt and tweaked until it was helping the motor.  Then the belt was flipped off and the engine continued to run on its own.  It's still not making the power it should and, once I get the overlap sorted out and the cam re-timed, it should do better.

FYI, after yesterday's 30 minute motoring with the modified piston ring, I did another leak-down test and there was no detectable blowby.  I checked the compression when motoring just below latch-out speed (cold) and got 55 PSI, which isn't bad. 


26 June 2013:

For the last couple of days, I've been sorting out various issues.  I've got the valve overlap down to about 5 degrees while keeping the exhaust duration about 190 degrees and the intake about 180.


The governor has been adjusted to as slow as it will go and still govern.  I haven't measured the RPM but it is probably around 500 now.  After optomizing the rest of it, I'll try to work on the governor to make the latch/unlatch speeds more predictable.  Right now, sometimes, it stays unlatched longer, allowing the engine to speed up more.  Then it coasts for a longer period.


Also, I've reworked the mixer a bit with the mixture needle integral with the jet.  Before, the needle approached the jet from the opposite side of the throat.  This, I think, caused the jet to behave erratically due to the needle being in the air flow.


Then, the venturi was way too big at 0.200" diameter.  I made an insert and guessed the venturi needed to be around 0.180" in diameter.  After a lot of fiddling around, I found that there was no way the engine would run with the choke fully open, letting me know that the venturi was too big and the air velocity across the jet was to slow.


I then bushed the venturi insert and drilled it 0.100".  The fuel mixture needle can now have an adjustment range with the choke fully open after the engine has warmed-up a bit.  Throughout, I've been tweaking the ignition timing and am now at about 15 degrees BTDC.


Because it is at a point that it can -almost- be hand cranked to start, I think I'll take a day off to cogitate my next moves.


Right now, I think I can safely increase the venturi diameter to around 0.120".  This will allow a larger charge and, possibly, more power.  The engine still has to fire several times before the governor latches so there is a bit of improvement to be made.  The mass of the flywheel will probably keep it from single hitting to get up on the latch due to the power input required to raise the speed.


The piezoelectric ignition is operating fine.  Spark plug gap is 0.028" and I can detect no problems with spark.


Compression this morning, with a cold engine and after about an hour of off and on running, was still about 55PSI.  I checked the leakdown and found that there is no detectable leakage past the valves or piston.  The Teflon ring is holding-up so far.


28 June 2013:

Still slogging away.  Over the last couple of days, I've been tweaking the venturi size and think, after wandering around between 0.100" and 0.150" in diameter, I think I've settled (at least for the time being) on 0.110".  With this venturi size, I can run the engine without any choke and still have needle valve adjustment range.


Today, I reduced the cam duration which reduced the overlap. 

Modified cam.

Careful checking found that the duration was about 200 degrees.  Since both valves use the same lobe, it was fine for the exhaust but a bit on the long side for the intake.  You can see how much I took off of the lobe in the photo above.  I did it in two steps and now have about 180+ degrees of duration for both valves.  For an engine that's not going to work for a living, this is fine.


More fiddling got me to an ignition advance of about 20 degrees.


The engine still doesn't just jump up onto the governor but it will take spells where I think it's going to get the idea.  It still has to be started on the motor.


I checked the compression again this morning when it was cold.  It was about 52PSI, a little lower than earlier but that could be due to the changes in intake, cam duration, overlap, etc.  I'll check it again tomorrow morning when it's cold.


30 June 2013:

Today, I got a lot done with some small improvement.  First off, I pulled the head.

Head before milling.
Then, I decided to check the Teflon piston ring.  So far, it doesn't look as if it's allowing any blowby.  That's good.

Piston with Teflon ring.
If you look closely, you will see the copper spacer ring above the Teflon ring.  This is to force the Teflon ring to fit snugly and slightly flatten the sides of the ring where it contacts the lands.  You can also see that I turned the O.D. of the ring flat for a better seal against the cylinder wall.

Next, I stuck the head in the mill and took off
0.075" off of the face to increase the compression ratio.  I connected an ammeter in series with the motor and the controller in order to see if I could see the differences in running when motoring. 

Test meter.
The meter did, in fact, show a slight rise in current as the piston came up on compression and a fall when the cylinder fired.  I could fiddle with the mixture and see when I'd reached the sweet spot.

it ran a little better but was still puny.  I found that the running problem and funky fuel mixture adjustment range (a moving target) was because the spark plug is a resistor type!  I found this out by measuring the resistance between the connector end and the center electrode.  It was about 7.5K Ohms, about enough to kill the spark when the plug collected some carbon.  The piezoelectric ignition just doesn't have the muscle to fire the plug through the resistor and the carbon.

The solution to this was to disable the piezo ignition and substitute a solid-state module triggered by a microswitch.

New ignition setup.
The microswitch is actuated from the crankpin that ran the piezoelectric setup.  The switch is mounted using only one screw so the switch can be moved to change the timing while the engine is running.  The dinner bell rang just as I finished it.  A very short test promises better performance of the engine.  We'll see.

1 July 2013:

It's running better than ever today.  First, I re-timed the valves so the middle of overlap is at TDC.  Overlap is 20 degrees, a bit much I think, for a flywheel engine but I'm planning on running it that way some more while other tweaks occur to me.

I also adjusted the size of the venturi.  Starting at 0.110", I increased it to 0.125" with improvement.  Then went to 0.136" with another improvement.  At 0.141" it was better.  At 0.157", it was hard to keep running without some choke and the mixture screwed way out.  The venturi was reduced to 0.147" and it seems to run fine at this setting.  The way the venturi is changed is by filling the venturi part (the jet end) of the brass part with solder, then drilling the solder to the selected size.  That's a LOT easier than making a new venturi for every test that requires a smaller opening.

Starting is still problematical.  Cranking slowly (about the speed the wheel can be flipped) with the spark slightly retarded and the perforated choke plate completely closed and momentarily putting a finger over the air inlet, it will begin to fire.  I can take it off the belt and most times, by slowly opening the choke, it will pick up speed.  Advancing the spark will make it come up on the governor.

The best it's run, at times it will actually fire once, latch and coast several turns before unlatching and firing again.  Most times, it has to hit between two and five times to get latched-up again.  More tweaking may improve this.

My main goal now is to get it to hand start.  After all of the fiddling is done, I will revamp the ignition to some kind of permanent set-up using the electronic module and then it's ready for the last video and showtime.

Here are some numbers after running the engine today:
- Venturi diameter 0.147" (#26 drill).
- Compression (cold) 65 PSI.
- RPM 475-500.

- After running for about 30 minutes, the following temperatures were recorded:
- Hopper water temperature 170F.
- At exhaust manifold junction with head 220F
- At mixer junction with head 190F

- Valve lash 0.010 both.  (this gives the following valve timing):
- Exhaust opens approximately 170 degrees ATDC on the compression stroke.
- Exhaust closes approximately 370 degrees ATDC (compression).
- Intake opens at 350 degrees ATDC (compression).
- Intake closes at 550 degrees ATDC (compression).

3 July 2013:
Since the engine seems to be running about as well as it's going to until I get some more ideas, I decided to go ahead and make-up a more period-like timer.

Showing timer and auxillary spark gap.

The timer is simply a bronze spring that makes a ground contact with the small crankpin on the end of the camshaft.  The screw with the arm on it is to adjust the timing.  The cam turns clockwise and the sooner the crankpin touches the spring, the more advanced the timing is.  The advance screw bears against the spring to change the position of the contact so the crankpin grounds it earlier or later.  Turning the arm clockwise retards the spark.

Those of you who look for the fine details will see a small electrolytic capacitor (one microfarad) which is there to stop the ignition from triggering due to contact bounce.  It triggers on the first grounding of the contact.  Subsequent groundings during that ignition period are ignored.

While testing the ignition, the engine began faltering, then would quit.  After motoring it for a few seconds, it would start and run for a while before stopping again.  Choking wouldn't revive it so I knew fuel wasn't the problem.  Eventually, when it started faltering, I quickly pulled the high tension lead from the module and made the spark jump a quarter inch or so.  The engine immediately picked-up again.  The problem is in the crummy spark plug.  I'm getting to really hate those things.

As you can see, I've made a "Hot Spark Intensifier" (as they were called years ago).  All it is is a spark gap that allows the voltage from the coil to rise to a higher value before jumping the gap.  Since the voltage is higher and, when the spark appears, the space where the spark is more or less a dead short, the effect is higher current going to the plug.  In normal operation with a non-garbage spark plug, the spark jumps both the intensifier gap and the plug gap at the same time.

If the plug is slightly fouled, the initial path to ground is through the carbon buildup.  With the higher voltage, there is enough current to overcome the shunting of the carbon and allow the plug to fire normally.  At least, that's the conventional explanation.  Makes sense to me.

Anyway, I didn't have enough time to test the "Hot Spark Intensifier" this afternoon and this may have been a good thing.  The clear tube of the intensifier is made of plastic and it's probably not a good idea to have it located at the spark plug where it could get hot enough to melt.  I'll modify it so it can be placed at the coil end of the plug wire.
5 July 2013:
Spent part of yesterday and all of today making piston rings.  I have my doubts about the lifetime of the Teflon ring so I decided to make a new one out of gray cast iron.  I had a hunk left over from making the piston for The Homebrtew Hvid and used that.  Here are the steps.  The result is at the end.

Bore the the relaxed diameter of the finished ring.

Turn the O.D. to a few thoushadths over the relaxed diameter of the finished ring.

Part-off the ring a little longer than needed to fit the groove in the piston.

Here is the parted-off ring.

Using emery paper, lap the thickness of the ring to fit the piston.

Cut the gap in the ring so, when it is squeezed down, the ring will be a little over the finished diameter of the finished ring.

Using wire, compress the ring using a mandrel to keep the ring compressed.

Here's the ring, ready for turning the finished O.D.

Turned to the finished O.D.

Here's the ring on the piston.

Here's the result of almost two days work.
I think the procedure is correct but the execution leaves a lot to be desired.  The ring on the left is the first one I made.  Unfortunately, it didn't have enough pressure on the cylinder wall and wasn't thick enough to seal well in the ring groove.  Result: low compression and a lot of blowby.

I then made another blank using another set of dimensions.  This time, it was too thick and, when I tried to squeeze it down to the finished I.D, it broke. 

The third ring was dimensioned to be thinner than the first one and thicker than the second one.  The same thing happened to this one.  I'm sure if I tried again, I could get a decent piston ring.  As it was, I was disgusted with the process and put the Teflon ring back in with a corrugated 0.005" brass shimstock expander.  So far, motoring hasn't gotten the compression back up to a point that the engine will run.

Not so much fun today but tomorrow's another day!
15 July 2013:
It's been a while but I've been piddling with piston ring ideas and having little luck.  After fooling with cast iron for the rings with no success, I tried making a ring out of brass.

Brass - not so good.
As you can see in the above photo, the brass distorted and shoved the piston hard against the bore, causing galling.  Scratch that idea!

Then, I tried another one of the Teflon "O" rings, this time with a broader flat on the cylinder wall contact area. 

The second try with a Teflon "O" ring.
This was also unsuccessful.  The engine couldn't develop any compression at all, I think, due to the non-flat contact area on the groove wall.

Then, I ordered some square cross section Viton"O" ring stock. 

Viton piston ring.
I should have known this would be problematical.  Although the Viton had the possibility of making a good seal and had a 250F temperature rating, when the material was formed around the groove in the piston, the width at the I.D. grew enough that it bound in the groove at the bottom and was sloppy at the outside.  The engine did develop about 30PSI compression but after motoring it for a couple of hours, it did not improve.

The next material tried was oil filled Nylon.
Oil filled Nylon piston ring.
The jury's still out on this one.  Since this material grows a lot with heat, I machined a first ring with about 0.010" side clearance.  This was not successful, giving only about 35PSI compression. 

I made another ring, this time with only about 0.001" side clearance.  This one gave about 40PSI compression and the engine tried to run although, when it warmed-up, the ring was binding.

After a couple of warm-up and cool down cycles, it seems to be sealing better and has a better fit although it still binds a little when hot.  Tomorrow, I'll cycle it a few more times to see if it improves.  If it improves substantially, I'll run the engine until I can see that it is going to last a long time or will be yet another failure.

Based on what I've experienced above, I ordered a piece of Teflon bar stock to make a ring out of.  The Teflon is good to over 500F, is very slippery and doesn't grow in heat as much as Nylon.  I'll probably eventually go with the Teflon in any case, as it will make a better material for long life (I think).

Of course, if all else fails, I can always just make another piston and order a 1.125" X 0.09375 thick ring and be done with it.



I've been having some font issues with this page displaying on an iPad.  Nothing I do seems to fix it so, unless one of you knows a sure fix, we'll just have to live with it.  The problem doesn't show up on Windows Explorer or Mozilla Firefox.

I started this website using Mozilla Kompozer.  I'm now back to the old Microsoft Front Page using Arial font, one of the "legacy" fonts that's supposed to work for everything.  We'll see how it works. -------------------------------------------

17 July 2013:
I ran the engine for a while with the green (oil filled Nylon) ring and, althouh it sealed, when the engine would warm-up, it would expand enough to bind in the cylinder. 

Since the Teflon came in yesterday, I made a ring out of that material that fits in the 0.195" wide groove in the piston.  

Teflon (PTFE) ring on piston.

I made it with a snug 0.001" side clearance and it has an 0.001" clearance in the diagonal-cut gap.  Compression before running the engine was an impressive 68PSI.  The piston now runs without binding when the engine is hot.

After motoring the engine, I think I have another problem.  Since the compression is good and the leakdown test shows no leaks, the compression may now be too high.  When motoring the engine, as I advance the ignition timing past about 10 degrees BTDC, i'm getting what sounds like detonation knock.  Since I'm running it on naphtha, which has a very low octane rating, that might be the cause of the knocking.  Just about the time the timing is advanced to a point where the engine acts like it will almost run on it's own, the knocks appear.

Next time I fiddle with it, I'm thinking of draining the naphtha and substituting non-ethanol pump gas.  If that fixes it or makes it better, I'll try a thicker head gasket.

19 July 2013:
Changed over to stinky non-ethanol pump gas and ran the engine for a while.  Although it ran better and actually got off the motor, it still tended to knock when approaching what seemed to be the sweet spot for ignition timing.

I lowered the compression ratio by replacing the 0.031" thick head gasket with one twice as thick.  There was little, if any, improvement.  Another thought occurred and that is that there was a black soot spot on the piston head opposite the intake port.  Could there be a big problem with my oddball combustion chamber idea?  Here's a repeat photo of the combustion side of the head (before milling to raise compression).

Cylinder head showing combustion chamber(s).
To raise compression even more, I pressed inserts in each end of each slot to take-up more space.  The gas flow in the head (I think) causes the mixture to be stratified with the non-purged exhaust gas.  This makes it fire the mixture only when it's correct when passing the plug.  Of course, the mixture is purer (richer) inside the intake passage.  Once the mixture fires, the flame front quickly moves toward the intake chamber where it's richer.  This rich mixture causes the soot spot on the piston.

I tried shaping the intake passage where it enters the combustion chamber proper.  This is in order to get the fuel/air move directly to the plug.  This seems to make the engine run a little better and have more response to the fuel mixture adjustment.  Since removing metal in the combustion area will lower the compression, I may have to eventually go back to the thinner gasket.  Some thinking on it is in order before whittling any more but this might be the right track.

This engine of the few I've made (with the exception of The McVickerish Engine) has been the most reluctant to run well and may show the folly of trying something new without properly designing it. 

If I had thought it out more, I would have given the engine a longer stroke (meaning that I would have had to buy a hunk of steel rather than using a "scrap").  With a longer stroke, the combustion area could have been bigger to give the same compression ratio.  With a bigger combustion area, I think the engine would have run well right after I'd gotten the piston ring worked out.
20 July 2013:
There's a little improvement to report today.  The frustrating thing about this engine is that no one improvement has been really significant.   Just a lot of very small steps ahead, some setbacks then more progress.
Combustion chamber modifications.
Compared with the previous image, you can see the three "improvements" to the combustion area of the head.  First, to raise compression, I've milled the head so there's just enough space left to give clearance between the spark plug and the head of the piston at TDC when using an 0.031" head gasket.  The second (questionable) improvement is the insertion of the four slugs into the port areas to remove some more volume from the head, giving more compression.  The latest improvement (which seems to work) is to relieve the intake and exhaust port openings into the combustion area of the head;

The theory behind this relief is to improve gas flow in the head.  I may do some more whittling on these areas of the head to get burnt gases out better and fresh mixture in better.

It's been a hassle from the beginning of the testing phase of this engine to regulate the choke.  Today, I made a flutter choke.
    Flutter choke on end of mixer.
The flutter choke is nothing more than an extension to the air inlet side of the mixer that has a disc that is spring loaded.   The disc blocks air into the mixer.  The spring allows the choke to open more fully as the engine gains speed and the velocity of the air increases.  I've found that a flutter choke (as seen on a lot of the Fairbanks-Morse Z series engines and others) allows the engine to run at varying speeds and loads without having to adjust the mixture needle.  Of course, most of the engines that use this feature are throttle governed but there's no reason it won't work on a hit and miss engine, too.

Before quitting for the day, I test ran the engine with today's "improvements".  It seems to get up on the governor a little quicker and appears to hit a little less often when tweaked.  Since it was running low on gasoline, I refilled the tank with naphtha and it ran just as well after a minor mixture adjustment.  It's still not nearly ready for "prime time" because I want it to be where it hits once or twice then coasts for several revolutions.  Not nearly there yet.

Another thing I'm seeing is a "sorrying-out" condition when the engine has run for several minutes and is close to a stable temperature.  It begins taking longer and longer to get up on the governor until it finally doesn't latch out at all and slowly loses RPM until it finally quits.  I don't think it's a drag issue (something binding with heat) because the engine spins freely right after it stops.  It could be that the mixer is getting hot enough to boil the fuel but I'm not sure yet.
21 July 2013:
Today there was a little more progress.  First, since the engine ran for a total of about an hour yesterday and sorried-out about quittin' time, this morning I pulled the head and took a look.
   Piston and head after teardown.
As you can see in the left photo above, there is a carbon spot opposite the intake port.  I think this is caused by the rich mixture trapped in the port area blowing into the piston.  On the right is the head, showing the soot pattern around the plug.
Head, after further removal of metal around intake port.
I removed more metal in the port area and reassembled the engine with the original thinner 0.031" gasket to raise the compression to compensate for the increased volume in the head due to metal removal.

The engine ran marginally better and I think it was hitting fewer times before latching.  Since it ran a little better and didn't seem to want to detonate as easily, I decided to remove the 0.031" head gasket to raise the compression.  Since the block and head mating surfaces are really flat, I just applied a very thin coat of high temperature silicone gasket sealer and put it together.  No leaks.

I knew that removing the head gasket would cause interference between the spark plug and the piston.  I could have just used a second gasket under the plug but that would have decreased the compression a little. 
   Spark plug before and after relieving.
What I did was to modify a spark plug by cutting off part of the side electrode then hammer it down and file a gap.  My figuring has the plug clearing the piston by a few thousandths.  A miss is as good as a mile!

After putting it all back together, it seemed to run better than ever although I'm still not impressed.  At one point, it was hitting twice between latches but that didn't last long.  After about 15 minutes, when it was completely warmed-up, it began running sorry again.  It didn't immediately quit - it just gradually slowed down until it finally stopped.  I believe the sorrying-out is caused by the fuel overheating.  Tomorrow, I may press a water dampened rag against the mixer when it starts sorrying and see if it begins to run a little better once the mixer cools some.  If that works, it will prove that the liquid fuel is boiling and I'll be tempted to build and test a propane system on the engine.  There's less energy in the gaseous fuel but I don't think the heat will affect it as much.

I'm coming to believe that most of the wimpiness in the running of the engine is due to a poor cylinder head design on my part.  Because of this and in spite of tweaks, I don't think this engine is ever going to be a strong runner.  I'll keep piddling with it to get it as good as it will get. 

At the end, I may also try converting the intake to atmospheric.  Doing that will eliminate any possible intake duration issues.  It would be a shame to abandon the powered intake valve but, if it makes a great improvement in performance, I'll just have to live with it.
22 July 2013:
Well, dang!  This morning, I messed with the mixer with no improvement.  Then, I noticed a little blowby so I pulled the head and piston.
Teflon piston ring after about two hours of off and on running.
The Teflon ring looked good, only showing a little leakage around the gap.  For the most part, the surface finish was shiny and smooth.  I believe that, if sized correctly, Teflon should make good low performance piston rings.  Since I'm bored with doing rings, I just ordered a couple of 1.125" X 3/32" cast iron rings from Starbolt.  I've got the new piston done all except for the ring groove which I will do once the rings come in.

I did say ring(s).  Knowing my luck, I'll probably break one.  If I don't break a ring, I think I know what the bore will be of the next engine!  :-)

We now pause for ring delivery.
28 July 2013:
Well, I've got the new piston made and the cast iron "store boughten" ring in and, although I've motored it for an hour or more, it still doesn't run worth a hoot.  I'm going to motor it a couple of hours more and; once the ring is seated and compression is up, it should run better.

Today, while motoring the engine, I changed intake and exhaust cam timing (as much as you can on such cam and follower arrangement) and saw very little change in the poor operation of the engine.  With the timing set so the exhaust opening is at about 40 degrees BBDC and closing is at TDC and the intake opening about 5 degrees BTDC and closing at about 20 degrees ABDC, I'd think the engine should run robustly.  Although the mixture and ignition timing can be set so the engine fires regularly and the exhaust sounds normal, it just won't "thrive".  It has to be motored almost to governed speed to start and will not continue to run for long after the belt is removed.

I even tried removing the intake rocker and making the intake valve "automatic".  The engine ran worse with this arrangement even after tweaking the spring tension.

I'm beginning to think that the head/combustion chamber design is at fault.  With the present cam and valve arrangement, there's not much that can be done to improve gas flow in the combustion chamber so, if the engine doesn't decide to run well (which it should!), I may have to re-think the head and valve arrangement.

I suppose I could convert it to have a face cam operating the rockers and a conventional valve arrangement.  This would give a conventional combustion chamber.  It would be a lot of work and I'll consider it to be a last resort effort.
30 July 2013:
Today, I pulled the head and piston and cut a second ring groove and installed the second ring I'd ordered.  Originally, I ordered the second ring so, if I klutzed the one, I'd have a spare. 

The second ring is to eliminate any chance that there is no compression loss past the rings.  In setting up to cut the second ring groove, I needed to freshen-up the center and, in so doing, drilled right through the piston head.
"improved" piston with second ring.
Not wanting to make an entire new piston for that small problem, I simply tapped the through hole 6-32 and inserted a screw with a locking nut on the underside.  Works for me!  Also note the spark plug relief in the piston head.  This allows me to use a normal plug when no head gasket is used.

While it was apart, I whittled a little more on the combustion chamber, trying to improve flow and also opened up the throat in the mixer, thinking it needs to breathe better.

As is usual for this project, after I got it all back together and motored it for a while, there was precious little improvement and it's still not to the point where it will run on it's own.  It sounds like an engine but doesn't run like one.
31 July 2013:
This is getting to be riduculi (the plural of ridiculous).  Compression is apparently the problem so I pulled the head and piston.  To eliminate some unused space in the combustion area, I bored the head as shown.
  Modifications to head and piston.
The idea was to take-up more volume using a piston spacer than was added by boring the combustion area.of the head.  After relieving the piston spacer for the valves, I don't think a lot of volume was lost.  Motoring didn't show any noticeable improvement.  I think there is some intake valve leakage which I will look into.  This leakage could be due to a slightly bent intake valve due to the piston spacer not being relieved enough.  We'll see.
2 August 2013:
I've now done everything and nothing works.  I'm getting the sneaking suspicion that because I didn't actually design the engine, something is intrinsically wrong.  Valves are seating fine and I moved the cam and ignition timing all over the place with no noticeable change in performance or the lack thereof.
3 August 2013:
NOWI know what the problem is!

Everything was measured.  Way back, when I made the crankshaft, I said the stroke was going to be 1.250".  Today, when I measured the stroke with the piston in the bore, I got 1.156".  I have no idea how that happened unless I made a command (non-documented) change when I was making the crankshaft.  The displacement is 1.149 cubic inches.

Then, I measured the combustion space including the valve pockets minus the volume of the piston extension and came out with a figure of an ablolute maximum of 3.6:1, probably less.  There is no way I would be able to reduce the combustion volume so that is a maximum compression ratio using this design.  The only way the compression ratio could be increased would be to increase the stroke by more than space would allow.

There will be a donation to the scrap pile and a new head and valve arrangement will be done.  To keep the camstopper, I'll have to work out a face cam and new rocker.  To keep it relatively simple, I'll probably just use an atmospheric intake valve.  This time, I'll actually design it so it will work.

Oh, well - back to the old drawing board.
4 August 2013:
Okay.   I've redesigned the head and valve arrangement.
   Head end view of the new arrangement. Side view of the new arrangement.
This should work fine.  I've got it set-up for around 6.5:1 compression ratio with no head gasket.  If it doesn't seal, I'll make an 0.031" thick gasket.

As you can see, I've found that I can use the original cam by using a bellcrank arrangement.  The cam drives a follower that is guided and contacts the adjustment screw.  This rotates a horizontal rocker arm.  This rotation is transmitted through a shaft to a vertical rocker arm that contacts the exhaust valve (the valve springs are not shown).  The intake valve will be atmospheric.  The exhaust port exits the top of the head and the intake is at the bottom.

The spark plug is threaded into the side of the head.

After dimensioning the part drawings, I can start whittling.
6 August 2013:
All that's left to do on the head is to lap the valves. 

The almost finished new head.
Next, I'll start on the rocker arms and stand.
6 August 2013:
The head is finished and on the engine.
The new head in place.
Presently, I'm making the parts for the rocker shaft.  It'll probably be a couple of days before I can get those parts finished and on the engine.
9 August 2013:
The design I came up with for the rocker arm/shaft assembly is pretty labor intensive.
Rocker arm stand and in place.  Partially finished cam follower guide shown on base board.
The rocker arm stand was whittled out of a hunk of 3/4" thick steel.  If you look closely, you will see that it took some extensive milling go get the necessary shape.  In the same photo, you can see the cam follower guide which is also labor intensive.  The design allows for a 0.150" lift for the exhaust valve and a 1:1 ratio on the rockers which necessitated removing some of the lift from the lobe.

The rocker arms will be made of 1/4" steel and I've found the last piece in my pile which I hope is big enough.  The design has two rocker arms.  One from the lifter to the rocker shaft and the other from the rocker shaft to the valve.  This was the only way I could think of to get the cam motion to the valve in the new head.  I'd considered a face cam but am not sure I can actually machine one that is accurate.

Tomorrow may have the valve train done then it's on to making the camstopper work at the new cam angle.  This will involve some diddling with the latch arm shape.

I'm thinking of making the mixer out of some 1/8" NPT pipe using the needle valve and jet assembly off of the mixer for the failed head design.

All I can say is that, after all this work, the little sucker had better run!.
10 August 2013:
The valve train is done.
   Valve train finished and a look at the re-timed camstopper.
It took a bit of fiddling but the cam follower and rockers are finally finished.  I had time to modify the canstopper trip.

In the photo above right, you can see the cam block I added to the latch arm.  This causes the camstopper "button" (seen at about the bottom of rotation) to be pressed when the cam is straight up instead of at tbout the 2 O'clock position as it was with the previous arrangement.

After setting the cam timing and as the day ended, I couldn't resist hooking-up the belt and motoring it a little.  With only one valve to operate, the camstopper is now surprisingly quiet and seems to latch-up more smoothly now.  It has very good vacuum and the exhaust flow is much better.  Now that the compression ratio is up, I can hear blowby past the rings which should finish breaking in once the engine is running.

I'll be taking a couple of days off from the project and when I return to it, all I'll have to do is to rig-up a mixer and hook the ignition back up.
14 August 2013:
Well, the mixer's done and everything is in place.

All dressed-up and no place to go.
The improvement - - - - NIL!

Leakdown shows that there are no compression leaks at valves, head mating surface or rings.

It's got good compression - 85PSI.
it's drawing fuel nicely to the mixer.
The intaike valve flutters properly (tried several springs).
Cam timing (at "optimum" performance) is opening the exhaust at about 20 degrees BBDC and closing the valve at just a shade (5 degrees or so) ATDC.
There is no binding of the bearings or piston.
Intake and exhaust are free of obstructions and are sized for good flow.

It sounds like an engine running (and it is, indeed, running!) and the spark timing and mixture adjustments do what they're supposed to do.  The only problem is that it just doesn't want to be weaned off of the motor and it doesn't seem that there is any certain speed it's happier at.

At this point, I'm open to suggestions.  Otherwise, I'm gonna shelve it for the time being and work on getting the 30-60 engine to run better.
15 August 2013:
Last night, I had the thought that the flywheel may be too heavy.  My theory is that the mass of theflywheel can't be accelerated faster than the combustion gases cool.  Although there is good combustion and the heat generated causes the pressure to rise nicely, the piston can't move down the cylinder (allowing expansion of the hot gases) fast enough and the heat is lost to the cylinder wall.

This morning, I had two emails stating about the same thing.  David Jones thought the flywheel had too much mass and Denis Basson offered some calculations showing that a flywheel weighing about 2 lbs at the rim with a diameter of approximately 19 inches and about 5/8" wide would work.

Barbell weight next to the original flywheel.
Not having that size of material and not having the space between the crankshaft and the base plate of the engine, Ihad a look at ye olde junkpile.  I came up with a couple of 3lb barbell weights, about 5-3/8" in diameter and 3/4" thick.  Although it wouldn't provide the ideal momentum, I figured that I could at least prove the theory.  I made a bushing twice the width of one weight, broached for the gib key and bored one of the weights for a press fit of the bushing.The original flywheel is about 8" in diameter, 2" thick and weighs 21.2 lbs.  After whittling, the barbell weight and hub weigh about 3 lbs.  When I tried this on the engine, it appeared to not have enough flywheel because it wouldn't carry-over and as soon as the belt was removed, the engine would quickly come to a stop.

Plan B was tried and the second barbell weight was bored and faced to be pressed onto the hub.  I got in a hurry and got the fit too tight and here's what happened.

Now, I've got to go back and do it the hard way, starting with a large hunk of 1/2" steel plate, cutting a circle out of it and making a hub.  I'll rely on Denis to calculate if a diameter of about 8" will give enough momentum.

One thing I did do with the original flywheel back on the engine was to try to test the theory by blocking out the governor and spinning the engine really fast to see if it seemed to run better at higher speed.  This would test the heat absorption theory.  When I spun the engine as fast as the motor would go (about 1,200 RPM), the engine seemed to run a little stronger and would almost hold it's own off of the belt.
22 August 2013:
Nothing much is happening with the engine.  I've been trying to get a flywheel blank waterjet cut out of 1" thick steel and, so far, the shop hasn't had time to do the job.  They're waiting until they have some other work for the waterjet machine so, since I'm getting a rate based on slack time, I will have to wait.

One thing I have done is to re-make the cam follower using a smaller diameter ball bearing.  This reduces the exhaust duration to about around 200 degrees.  It seems to try to run a bit better although still not weaned from the motor.
29 August 2013:
The shop called yesterday and, today, I picked up the blank for the new 1" thick flywheel.

Semi-finished flywheel.
I still have to hone the bore to fit the crankshaft (it alllllllllmost fits) and broach the keyway then the engine will be ready to start testing again.

In the meantime, I reworked part of the governor to try to make it a little more responsive.  Also, fiddling with the cam timing and duration along with adding a flutter choke to the mixer seems to have the engine running slightly better.  I have my doubts that the new, lighter flywheel will make the engine spring to life but I would welcome a pleasant surprise.  I should find out tomorrow.
30 August 2013:
The flywheel is on and I've test run the engine with it.

The new and lighter flywheel is mounted.
After warming up a bit to get it into it's best running condition, the engine actually does run a little better.  Since all of the changes have been incremental, I have to be happy with little bitty improvements.  I think that, if I was lucky, I could flip start the engine after it's warmed-up.  Since the motor's there, I just start it on the belt.

I think the improvements came from most of the changes made.  The new head allows a much higher compression ratio along with better gas flow.  Reducing the valve duration to about 220 degrees also helped.  Other changes include a better mixer with flutter choke, refinements in the governor and, of course, more break-in time.

I've shot a last (for the time being) video before I put the engine on the shelf.  Developing this one to the point where it's almost ready for prime time has been a trial.  If, in the future, the problem with it hits me in the face, I'll take it down and fiddle with it some more.

29 August 2013:
I said that I had some "left-overs" or parts that didn't work out.  Yesterday, I collected them so I could box them up and put them away with the engine.

Left over parts.
On the lower right of the photo is the piezoelectric ignition which I think I'll box separately.  That part worked fine but, since I was having spark plug problems along with everything else, I used a known hot ignition system.  I may use it on another engine in the future.

There's a lot of time and effort in all of those parts which goes to show that starting off with a design is good.

13 June 2016:

Yesterday, I took the engine off the shelf after better than two years and decided to try to make it run better. First off, I gave the flywheel some time to make it look better.  A day and a half and a waste basket full of swarf and you can see the result in the last photo for today.

I took the head off and re-calculated the compression ratio.  By dividing the stroke by the head-space (a short cylinder shape), I got about 8:1 which seems on the high side.  When taking the volume of the spark plug and it's port into consideration, the ratio is nearer to 6:1.  Thinking that a bit more compression couldn't hurt, I made an 0.062" spacer to fit on the piston head.
Piston at TDC as before.                                                                      Spacer ready to install.
Piston with spacer installed.                                                                   Piston at TDC after spacer.

Running today.
The rings and cylinder bore looked okay so I lightly honed the cylinder.  After putting it all back together, I tried to hand start it but ended-up having to use the motor to get it tweaked until it wanted to run.  I think it runs better now but it still takes the motor to get started and, once latched-out, it takes several hits to get back to latch speed.  There doesn't seem to be much blowby but I will run it some more to see if it does any better with the rings re-seated.  It sure doesn't feel like it's got nearly 7:1 compression ratio.

More work is needed.
1 July 2016:
It's been a while but my excuse is that I had to get my stub finger whittled on again and had to stay out of the shop.

In the meanwhile, I've been fiddling with the engine.  The piston spacer worked - sort of - but caused the exhaust valve to interfere with the spacer when the engine was latched-out.

To try to fix that, I removed the spacer then milled 0.040" off of the head.  THAT sort-of worked but now, I had a lot of blowby.

Since the piston fit wasn't optimal, today, I made another piston.  This one is made of cast iron instead of the 6061-T1 aluminum.
Turning the piston from a piece of scrap.                                                                                      The finished piston.            
Checking the bore several times, I came up with a dimension of 1.125"+.  The piston was turned to 1.126" then sandpapered down to where the piston is a slip fit in the bore with slight tightness.  I plan to motor the engine with the head off for a while to let the new piston get to know it's bore and make sure nothing binds up.  The piston clearance should remain the same regardless of temperature because the block is also cast iron.

I'm using the old rings because now, I'm relying on the close fit of the piston to do most of the sealing.

Tomorrow, I will put it all back together and see if it wants to run any better.
2 July 2016:
The short answer is no (it doesn't want to run better).

After about an hour of motoring and fiddling, it was running but not on it's own.  In desperation, I decided to revamp the exhaust cam.  Because of valve/piston interference, I had to run enough clearance that the duration was only about 120 degrees.  

The exhaust cam before modification.                                                                                  After modification.      
I modified the cam by turning down the lobe enough to have the valve just clear the piston with zero lash.  After turning the lobe, I put it in the mill and milled a new detent dip so the cam would stop in the open valve position.

Did another test run.  No joy!  The engine still doesn't want to get off of the belt.
I checked the compression and it was a little over 100psi which is good.  There is no valve leakage.  There is very little blowby.  The engine turns smoothly with no binding.

I guess the next thing is to hog out the mixer to see if I can get more mixture into the engine, although it -looks- about the right size.  I may also fiddle with the flutter choke, maybe putting a fixed choke in it's place.  I need enough suction for the mixer to draw fuel but any more than that and I'm just throttling the engine.
Head back HOME
As Red Green says,
BOY, This is fun!
[email protected]
Hits since 25 April 2010.
web analytics