The Edwards Two-Banger Project


For your information, the Edwards engine is a 2-cylinder horizontal hopper cooled engine with the flywheel between the crank throws.  It is a 180-degree engine in that it fires one lung then a half revolution the other one fires.  There is a one and a half revolution pause before the first cylinder fires again.  It's like a John Deere thumper.  Another interesting thing about this engine is that it has a separate mixer for each cylinder and you are able to 'turn-off' one cylinder when you don't need the power.

Albert York Edwards built his engines from about 1920 until about 1926.  I don't know if the serial numbers were indicative of true production but, if they were, Frank's engine would indicate that several thousand were built.  I, for one, doubt that near this number were made because  complication and expensive production most likely kept them from being widely sold. 

Bore is 3" and stroke is 5", making the displacement 70.7 cubic inches.

Edwards advertisement

Rick, one of the guys from Harry's, sent me this scan of an original Edwards advertisement from 1924.  I wonder why there are so few of these around?  Was it because it was a real oddball, not much like the other farm engines of the day?

Frank's engine has a Bosch magneto and spark plug ignition.  The above ad shows an engine with ignitors and a low tension magneto.  I've been told that the ignitor version is a lot rarer of the two.  The ignitor setup uses studs on the pistons to trip the ignitors and, unless their method of timing is different that any other piston strike setup, the ignition timing can't be advanced any farther than slightly after TDC.  I can't but think that this version of the engines put out less power than the spark plug versions.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Here's a group of pictures of the Edwards engine as it was as of May 1, 2004, before we started restoration.

Pulley side view of engine

The float chamber for the dual mixers is attached at the bracket the paper tag is tied to, and lies beneath it.  This engine is in remarkably complete condition.  After finding a parts engine with a float chamber, the only thing I have to make is the oil line which runs between the wrist pin end and the crankpin end of one of the rods and the fuel lines to and from the float chamber.

Magneto side view of engine

This photo (above) shows the two push/pull rods (above the throttle rod).  Each cylinder has it's own push/pull rod.  Both valves of each cylinder are mechanically opened using one rod and one rocker arm.  Not many manufacturers used this method of operating the valves because there can be no overlap between the closing of the exhaust valve and the opening of the intake.  Also, it complicates the cams and followers somewhat.

Rocker arms

This photo shows the inside of the rocker cover.  The mixer throats and needle valves are at the bottom of the picture.  The mixer throats are made of pot-metal and are pressed into the head.  The rocker arm for the right-hand cylinder is operated by the upper push/pull rod.  The arm is pushed (moved toward the viewer) to open the inlet valve and pulled (moved away from the viewer) to open the exhaust valve of the right hand cylinder.   The left-hand cylinder is operated by the lower push/pull rod - pushing opens the exhaust valve and pulling opens the intake valve.  The shaft for the left rocker also supports the right rocker arm.

Top view of engine

The top view shows the location of the magneto, gas tank, oilers and the speed adjustment, located on the end of the crankshaft opposite the pulley.  Look Ma, TWO mufflers!

Flywheel and crankshaft detail

Another oddity of the Edwards engine is the location of the flywheel.......BETWEEN the crankshaft throws!  The drive for the cams, mag and the governor are at the left end of the crankshaft.  Starting is accomplished by using a strap with a short peg in the end of it.  The peg is inserted into the hole in the pulley and the strap is wrapped around the pulley.  Haul on the strap and, if you're lucky, the engine's off and running.

Crankshaft and flywheel arrangement

Take a gander at how the flywheel is incorporated into the crankshaft!  It ain't one piece, folks!  If you look closely, you will see that the flywheel is bolted to the crankshaft at the crank throws.  To hold the whole rotating mass in alignment, lead is poured into the space between the crankshaft web and the flywheel.

Float bowl as received

Tommy Turner located a parts engine that included a "not too perfect" float chamber.  It's better than nothin'

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 May 2006:

I got the float chamber apart today with a combination of LPS2 and a torch.  I think it's been in a fire!  There was nothing at all left of the float and some of the brass parts were semi-brazed together.  It would be really nice to find another one that's in better shape although, barring this, it is rebuildable. but will require remachining all of the steel and small brass parts.  We can save the float arm, the main housing and the cap for the float valve.  Everything else is pattern material.

Before we disassembled the engine, I made note of the timing marks.  The crankshaft gear to the cam is marked in an unusual place.  See the left-hand photo.  The crankshaft gear has a mark between two teeth that matches with a mark on the edge of one of the cam gear teeth.


Timing gear locations on engine before disassembly

The right-hand photo shows how the mag and crankshaft bevel gears match up.  From what little we can find out about Edwards engines, we were led to believe that the magneto gears were made of pot metal.  This one must have been replaced because it's made of brass.  We find that if the mag is bolted to the mount, the gears miss meshing properly by about a quarter inch (the mag's too far away from the crankshaft).  We figure that whoever made the gear and machined the keyed taper inside the bore of the gear got the taper diameter too large, making the gear fit too close to the mag body (therewas actually some friction between the back face of the gear and the mag body when the shaft nut was tightened moderately.  Frank's going to machine a bushing so the gear sets out the proper amount.

Piston sticking out of bore

The Edwards engine is just chock-full of surprises that make me say, "Now, what in the heck did they do THAT for!!??".  One example is the way the pistons protrude from the deck of the cylinder block at Top Dead Center.  YES, both pistons do the same thing.  The picture above illustrates what I'm saying.  The arrow points to one piston at TDC.  The bright gray line at the tip of the arrow is about half the width of the 1/4" wide top  piston ring that is exposed.  (I don't think that this particular engine is an oddity, either with the block machined wrong or too-long rods.)  At least, Edwards engines didn't have a problem getting a top ring ridge.  Other than that, there's no explanation as to why this was done.

Other things jump out:

Why did they use two mufflers?  There is a large chamber in the head beneath each muffler with only a thin partition separating the chambers.  There is no water jacket that area so there's no reason the two chambers could not have been made into one and just one muffler used.  I can think of only one reason for this.  The engines had a separate mixer for each cylinder and maybe there are two exhausts so the operator could see if one cylinder was running too rich or too lean by observing the exhausts.

Why didn't Edwards use a wet crankcase instead of the oddball oiling arrangement they ended up with?  There's a separate drip oiler for each cylinder.  The pistons have longitudinal grooves cut into the skirt that match up with the oiler holes in the cylinders (see the photos below).  At the end of the groove, there is a hole through the skirt with a copper pipe pressed into it which leads to near a cup on the rod close to the wrist pin.  The cup meets with a hole in the rod.  At one end of the hole is the wrist pin bushing and the other end goes to a copper tube that runs the length of the rod to the big end bearing.  The theory is that one oiler per cylinder can oil the bore, wrist pin and big end bearing.  This is all well and good if not for lubrication of the mains, timing gears and governor.  To oil these parts, you have to take off the "hogshead" and manually squirt oil into the main bearing oil cups and onto the timing gears and governor.  I'll bet most folks didn't do that near often enough and the engines wore out fast.

The wrist pins and the piston boss are drilled and a piece of hard wire with a 90 degree bend in it is put into the hole then twisted so it snaps into a hole in the lower piston skirt.  They surely could have thought of a better way to lock the pins in place.


                       Oiling slot in piston                         Oil line in rod and hole in lower skirt for wrist pin lock wire


Piston and rod details.  Note the bent wire wrist pin lock

The big question about oiling is that Edwards opted for a complicated and expensive way to do it poorly.  A wet crankcase would have made the engines last longer, made them substantially cheaper to manufacture and made them simpler for the operator to maintain.  The only remaining manual oiling required would have been the valve gear.

Why the oddball cylinder designations?   One piston is marked "A" and the other is marked "B".  That would make it a "B cylinder" engine rather than a "2 cylinder" engine, wouldn't it?.

Why make the engine run on either one or both cylinders?  The advertisments say that it can be run on one lung for lower power applications.  This is a throttle governed engine!  If it needs less power, the governor will simply open the throttle plates less, making the engine produce less power.  If you cut out one cylinder by simply shutting off the fuel to it (and NOT relieve the compression on the cut-out cylinder), the remaining cylinder still has to pull the dead one over compression.  I could make another bet that there is such a small amount of gasoline saved from cutting out one jug, payback in fuel cost would be exceedingly small for any load.

- Then there's the complication of two separate mixers.  One mixer would have worked just fine had it not been for the one-cylinder, two- cylinder deal.

What's the deal with the complicated fuel system?  The float bowl and some plumbing could have been eliminated by simply having check valves in each mixer and simple lines running down to the fuel tank, (yet another common farm engine practice).  I'll wager that the engines would have run just as well with the advantage of not having to mess with the problems encountered with overflowing float bowls caused by dirty fuel (there's no strainer I can see in the tank or line to the float bowl.)

Why did Edwards engines have such a complicated valve arrangement?  Stoddard-Dayton, an earlier car manufacturer (among others) tried this expensive and complicated method of operating both valves with one rocker arm and rod years earlier and dropped it in favor of the generally accepted practice of using a separate cam and pushrod for each valve.  The valve actuating system used here requires a fairly complex cam profile and two roller followers (one to push the rod and the other to pull the rod).  In addition, there's no way to machine a cam profile that allows any performance enhancing overlap at all.

I'll bet that, if the Edwards engine had been designed with atmospheric intake valves and mechanically operated exhaust valves (like virtually every other manufacturer of farm engines did), the Edwards engine would produce exactly the same power, would be a lot cheaper to  build and less prone to maintenance problems.

Why did they make the entire base of the engines into a water reservoir?  Most of the cooling water for these engines is BELOW the level of the cylinders.  Sure, more water means more cooling (sorta) but, if the water boils to below the cylinders, that extra water does nothing.  Looks to me like all that space makes is a nice home for an extended family of mice with their attendant nest material and droppings (which Frank's engine has!)

Why did they have the main bearing studs offset?  See the photo below.  I wonder how many main bearing caps were broken when tightening the caps to squash the paper shims and get the correct drag?

Main bearing cap

Maybe the above questions are reasons to wonder how the Edwards company survived even six years!  In any case, these are interesting engines and I'm looking forward to getting this one restored and running.

7 January 2007:

Last summer and fall, Frank and I made some of the parts we need to get this jewel back together and running.  

Once I got the float chamber apart, I found that it had been in a fire!  The float itself was totally gone and just about everything else was almost fused together but the brass body of the chamber was good.  The nut above the float needle had partially melted (you can see the little dimple in it where it melted a little) so Frank is making another of those puppies.  Because the top plate and central shaft for the chamber were so rusted and pitted, Frank decided to make new parts.  The float needle was also literally almost "toast" so Frank used his considerable skills to make another one.  I made a new float out of a piece of balsa wood and sealed it with some gas tank sealer.


Float bowl parts

When the magneto was set in place on the engine; the gear, a brass recasting, missed meshing with the crankshaft gear by about 1/4".  It was found that someone machined the tapered bore in the gear too big.  Frank made a sleeve and soft soldered it in the bore of the gear and it now fits correctly.  We also found that there was a lot of slop in the ball bearings on the mag so I took it apart we found that one end of the housing needed to have a little taken off of it so Frank machined some material off one of the end caps to get a slight amount of pre-load on them.

The magneto setup takes a little head scratching to figure out.  The magneto runs at crankshaft speed, has two lobes on the point cam and the rotor also runs at crankshaft speed, firing the plugs alternately every half revolution.  If you think about it, it makes sense.  The crankshaft is 180 degrees (a-la John Deere) and each plug is timed to fire at it's TDC position.  In brief, each cylinder acts like a Briggs-type one lunger in that it uses one spark and wastes one.

When Frank was having fun getting the biggest chunks of dirt, grease and other stuff out of the hopper, he found a mouse condo with the usual acrutements of bedding, acorn shells, and "processed acorns" (see photo).  He also found some cast iron tear-shaped lumps of cast iron (goobers?) that were a real challenge to remove since they just barely fit through the available openings.  I've heard that they were used along with tumbling to break up any remaining sand after casting.  Can anyone confirm this?

Mouse condos, "processed acorns" and casting goobers

The engine has rested quietly in boxes and cans but, yesterday, I made something for the Edwards that never was shipped with the engines but should have been.  From the factory, the mufflers are mounted directly to the head and the exhaust blows all over the hopper, making it a blackened mess.  I went to the local hardware store and got a two foot section of 1-1/2" galvanized pipe, cut it in half, turned and bored the ends and now have extensions that will allow the hopper to remain pristine.  If, in the future, someone decides that it's just too gauche to have those non-original pipes on it, all they've gotta do is take 'em off and put the mufflers where they were originally.

Today I finally got a real start on the final cleanup and fitting.  There was a little of the original paint left on the block so I polished it up and tried to get as close a match as I could manage.  After spending a couple of hours mixing and testing, I think I've got a combination of RustOleum colors that are close enough for Guvmint work.

Then I decided to clean the head up and get it ready for paint.  When I took the valves out, I found that one of the pressed-in cast iron valve guides was broken off flush with the head, leaving only about 3/4" of bearing.  I tapped what was left of the guide and used a bolt and socket to pull the pieces out (see the photo below).  Apparently, someone in the past worked on the engine.  The valve was stuck and, in trying to remove it, the guide was broken.  A new valve was made but it was put back together with the broken guide and never run.


Removing old valve guide

I didn't have any cast iron or bronze to make a guide out of so I looked in the 'ol stock bin and found a hunk of 3/4" leadloy (lead/steel alloy) round bar stock.  Leadloy is noted for being easily machinable and I figured that the lead content would add a little to it's bearing properties.


New valve guide

I machined the small end of the guide to 0.5005" diameter with a slight taper toward the end so it could be pressed (with a hammer and block) into the head.  I made the big end 0.600" in diameter so the spring would fit over it.  Since my bar stock was 3/4" and the spring seat part of the guides is about .900", I just whittled a washer to fit snugly over the small end of the guide, making allowance for the thickness of the washer.  I lapped the valve with the new guide and it was dead on center after some rust was ground away.  The valve stem diameter is 0.312" and I had a reamer that size so the fit is a bit snug but I figure if the valve doesn't stick in the first hour or so of running, it will loosen a bit and be fine.  Time will tell.  

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

10 January 2007:

Yesterday, I transferred some more dirt and rust from the engine parts to me and painted the large castings.  In the process, I found out that one of the solid babbitt main bearings was loose in the block.  They look like they were made to fit the block and not poured in place.


Making a new locating tab

The cylindrical tab that locates the shell used to be there but somehow got broken off and lost.  There wasn't a trace of it in the locating hole.  Since the bearing looks fine to reuse, I measured the position of the tab and drilled and tapped the shell 8-32.  I then got a brass 8-32 screw and turned the head to .250", the diameter of the hole in the block.  A flat was filed where the screw seats.  JB Weld secured the screw in the shell and a little grinding with a burr put the brass screw below the bearing surface.  Fits nice!

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

12 January 2007:

Today, I got most of the rest of the painting done.  I miked the cylinders and find that the bore is just a shade over what I assume was the standard 3.000".  Actual measurements are in the 3.004" +-0.002" with very little taper and about 0.001 out of round, good enough to go with.  checking the ring fit in the cylinders finds that the gaps are over 0.040".  Worse is the 0.014" groove to ring clearance of the top rings on the pistons.  The other two grooves in each piston are only about 0.0015" over the 0.250" nominal ring width.  Dave Reed is advising as to whether or not to get new rings.

I set up the mains and found that I only had to take one thin (paper!) shim out of the cam end bearing.  And, yes, I did grit my teeth when tightening the nuts, not wanting to hear the snap of breaking cast iron.

Last summer, I soldered a new neck on the tank and Frank made a brass sleeve to replace the rusted-out 1/8" pipe threads in the tank outlet.  Today, when cleaning up the gas tank and preparing it for paint, I found a bunch of pinholes.  It got the Lee Pedersen gas tank sealer treatment.  While the old gunk was coming off the outside of the tank, I now question what color it was originally.  It could either be engine color or aluminum.  Does anybody know what color it's supposed to be?

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

20 January 2007:

I consulted with piston ring expert Dave Reed and he suggested going to 5/16" wide  rings in the top grooves and machining the grooves to 5/16" plus a thousandth or so.  Since the lower two grooves in both pistons were only worn between 0.0015" and 0.002", Dave said we could stay with 1/4" rings there.  The new rings along with some plug wire and hardware were ordered.  

One of the pistons set-up in my lathe with the top groove turned.  Note the plugged locating pin hole in the bottom groove

And, yes Martha, I do have a right-hand carbide tool in another station on the tool mount!  Interesting is the plugged ring locating pin holes in the pistons.  You can see one of the plugged holes in the lower groove.  I suppose Edwards ordered a bunch of pistons early on with the pins in them.  Later, they must have decided not to use them.

Edwards DID get it right with the wrist pins (but not the locking arrangement).  Before taking the pins out and figuring that the pins were, like the F-M pins, not really hard, I got my automatic center punch to mark them.  When the punch snapped, sparks came off the sharp end of the punch, leaving a flat-ended punch.  Those pins are HARD!

Other stuff is getting done.  I've decided that the gas tank is going to get a coat of Krylon Chrome (Chrome is the color of choice because I have a can of it sitting on the shelf!).  I know this is not the original galvanized finish but it is moderately close.  If, at some time in the future, someone comes up with a DECENT galvanized finish, it can be stripped and repainted.

The rod bearings are in pretty good shape but the loose shims were beat to death.  One sheet each of brass 0.002" and 0.005" shim stock was ordered.  

The governor is cleaned and back in place.  I suppose the timing gears and cam can go back next time I visit Mr. Edwards.

The American Bosch two banger mag works but I'd like someone to let me know what the point gap is so I can make it as good as it will get.  It is a Model FX-2-ED1.  

(Added, 27 January:  I got a reply to my post on Smokstak from Rudy, who owns Adrian's Magneto Service in Manitoba, Canada that the gap is supposed to be 0.014".  THANKS!  I reset it from 0.012").

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

26 January 2007:

Before doing the final assembly of the float bowl, I wire wheeled the corrosion off.  The gas tank was painted, fittings were cleaned up and fuel lines were made.  

The other day, I decided to put the timing gear/cam/push-pull rod assembly in.  The way it's built, you can't lay the assembly into the block without the pushrods hooked to the followers because you can't get to the lower pushrod pin in or out with the assembly in place.  I put the rods on and slipped the bottom one through the hole in the crankcase........DRAT!  The lower pushrod bangs into the skid.  Unbolt engine, lift the head end and prop it on a 4X4, put in the cam assembly, lower it and rebolt it to the skid.  Done!

Main bearing oil wick

Last week, I received some Edwards literature copies from Mike Camerer and Curt Andree.  MANY THANKS, Guys!!  The literature is extremely helpful!  I learned something else about the Edwards from the literature.  The main bearings are supposed to have wicks in them to draw oil from the depressions in the caps to the bearings.  Some felt does the trick!

I got shim stock and the other day I cut shims and fitted the rod bearings to the crankshaft.  

After I got done with the rods, I studied the literature again and found to my disgust that there is supposed to be a 1/8" pipe coming out of the end of the camshaft .  The end of the cam shaft was a little boogered-up but I had early-on decided to leave it that way.  NOW, I know what the booger was - it was where the pipe had broken off and someone had tried to get it out.  Figuring I'd have to do the "lift the engine" routine again, I thought "What the heck - I'll try using an Easy-Out on it.  If I booger it up worse, I'll still have to take it out."  Wonder of wonders!  I hammered the Easy-Out into the piece and gently turned and it came out!  I think that's about the third time in my life they've worked for me. 

Close, but no seegar!  Note the Alemite fitting The brass thingie to the left of the speed control knob).  I dredged it up as a temporary way to grease the cam shaft.  Frank has found a cup in his "junque" that we think is about the right size and is sending it.  When it gets here, I'll clean it up and swap with the Alemite.  

NOTE TO YOU YOUNG WHIPPERSNAPPERS:  "Alemite" grease fittings are the cylindrical ones with the pins that engage the grease gun with a quarter turn.  The ones you're used to seeing all the time are actually called "Zerk" fittings, probably because they were made by a company named Zerk.  They were so much easier to use than the Alemites, manufacturers changed over to them probably sometime in the early 30's.

I just had to lay the hogshead on it and see if the paint color was anywhere near the original

I'm getting close to hearing it run!  This is the hard part.  Waiting for the last batch of parts to arrive.  Today, I set it in time and now all I'm waiting for is the rings and the plug wires.  A couple more hours of work is about all it will take to finish it.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

26 January 2007:

Just as I finished making the gasket for the hogshead, the mailman arrived.  I got two presents.  One was the rings, plug wires and hardware from Dave Reed and the other was from Frank with the float needle cover, a couple of way-too-big grease cups and some other stuff.

I had to put one of the pistons back in the lathe and turn out about another half-thousandth from the top groove for the oversize ring to fit but I thought it was a little shy before.  THEN, the rings went on the pistons, the pistons into the engine, the head and valve gear on and the fuel lines connected.

TOMORROW, I think I just might hear it bark for it's first time in a long time.  I may need to get a helper to hang onto it because the new rings make a little hard to crank.


Note that it's nice and clean!  You can see the "slobber catcher" tin can on the left.  I couldn't help but polish all the copper and brass!

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

1 February 2007:

Well, I got it finished and worked up a good sweat trying to get it to fire.  Not a lick!  The mag's weak and won't even fire an 0.012" plug gap at cranking speed.  I took it off and subbed a capacitor - no better.  Tried several variations on the point gap - no better.  Then, I slipped the original magnet off and stuck on some rare earth magnets and made an armature.  It was only a little better and now throws about a 3/16" spark if you give it a good spin but is still pretty puny at cranking speed.  I've heard that these mags have weak coils and fail often.  That may be the problem here.

I've got the "improved" mag back on the engine but it's raining now.  Maybe tomorrow, I'll try some more exercise with the starting strap.

After conferring with Frank, we'll make a decision as to what to do with it.  In any case, I'm thinking of doing some shadetree engineering.......Take a VW (or similar) 4-banger distributor.  Grind off two of the point lobes.  Make a mount to fit the distributor to the mag base.  Taper the shaft for the Edwards gear.  Use one of my ignition modules on it.  I know it would work fine but fear that I may get smacked upside 'da haid by some purist.

Stay tuned.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

1 February 2007:

Overnight, I had a brainstorm.  Knowing that the mag was sorry, I figured if I could get it up to some decent speed, it would run.  What I did was to make a temporary ignition for one cylinder.  See the photo below.


Temporary Ignition System

I used the solid-state ignition module (lower left) that I run my Hoyt-Clagwell with.  The trigger wire runs close to the right hand cylinder intake rocker and touches it all the time except when the intake valve is open.  When the rocker moves back toward the closed valve position for number two, the number one cylinder is at TDC,  The wire to the trigger input is grounded at this point and the number one plug fires.  It's difficult to get the timing advanced much with this arrangement but as long as it will run at all, it is enough to get up enough speed for the sorry mag to fire the #2 plug.

Before starting, the number one plug wire from the magneto is grounded and the solid-state plug wire is attached.  With this arrangement, I could stuff a rag into the #1 mixer and turn it over a time or two to get it primed.  Then, with the solid-state ignition on, an easy pull of the strap had it running on that cylinder.  Opening the #2 mixer got that jug firing.  I turned on the oilers and let it run for a few minutes with half new/half old ignition, trying unsuccessfully to get it to run smoothly. 

While the engine was running, I  turned off the solid-state ignition and removed it's wire from the plug and connected the still grounded magneto wire.  When I ungrounded the mag wire, it would only fire occasionally on that cylinder.  I swapped it back to the old/new setup and let it run for about fifteen minutes.  Then, I stopped the engine and removed the plugs, gapping both at 0.020" to try to get it to run better.  The new/old method was used again to start it.  When the plug wires were switched, it now ran on mag on both cylinders but was still VERY raggedy running.  No amount of mixer adjusting could get it running right.


Here it is, galumphing away and making steam.  Note the high-tech magneto magnets and armature.  

While I was fiddling with it and while it was running, I swapped it back to the new/old arrangement and swapped the old magnet for the new magnets and armature.  It would not hit a lick and made a very puny spark.  Could it be that the magnet just needs charging or is there something wrong with the coil.  The original magnet sure doesn't hold a keeper very well.  Once I swapped back the rare earth magnets and the bent steel armature, it ran on the mag (but badly).   I let it run like that for an hour and would have let it run longer to seat the rings but I was getting irritated listening to it miss, fart and pop so I put it to bed.

You Edwards guys are right.  It took no time at all to get the hopper cooking.  I wonder why they get to a boil so fast.

Now, I've got a question for you Edwards experts out there.  When one of these engines is running right, does it settle down to the typical chug-chug.....chug-chug......chug-chug routine or do they just naturally run sloppy?  If they can actually run smoothly, I think the mag's the cause.

Another question........Does anybody know someone who can successfully rebuild an American Bosch FX-2 magneto so it's reliable and throws a really hot spark?  If so, please let me know.  I'm afraid the coil may have to be rewound but that's not certain yet.  Email me if you have the answer.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

8 February 2007:

I think I'm going to try to take the magneto apart to get at the coil and condenser.  If I'm successful, I can get a new coil and condenser and put them in myself.  Details to follow.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

10 February 2007:

After having taken the mag apart and checking each component, I find that the coil is most likely good.  While it was scattered all over my bench, I sweated the cover off of the cunningly concealed condenser can and replaced the original wax paper insulated one with a modern 0.33 mfd, 640 Volt film capacitor.  

After I put the whole schmear back together, I think I've figured out what the two setscrews on either side of the point cam do.  My -theory- is that, since the cam assembly fits rather loose in the housing and the setscrews are pushing against the housing at each cam location, the set screws could be juggled to get the points to open the same amount on both of the cams.  I only had a little luck and got them to open 0.014" on one lobe and 0.015" on the other.  That's as close as I could get.

The result of all of this was to get - maybe - 1/16" more spark using the rare earth magnet assembly.  It's now almost 1/4" when spinning it by hand.  I've got it back on the engine now and the next time I feel like getting slapped in the face with the starting strap, I'll see if it starts on the mag.  If not, I guess a magnet recharge is next.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

19 August 2007:

After the last entry, I got it started again but it didn't run markedly better so I decided to build a solid-state ignition system for it to see how well it can be made to run.   After doing that, I -just might- put the mag back on.

Here's what I've done lately:

I built a base and shaft assembly that mimics the original mag.  The original mag gear fits on the taper end of the shaft and meshes with the crankshaft gear.  The blob of epoxy with the wires coming out of its is a Hall-Effect transistor that is mounted on an arm that can be moved a few degrees radially around the shaft so I can adjust the timing.

A phenolic disk is attached to the shaft.  It has two magnets located 180 degrees across from each other.  The magnets cause the Hall-Effect transistor to trigger the ignition.  I've still got to work out the circuit to do this but it's not rocket science.

I've thought about how to make this work and have decided to use one of my surplus two-cylinder mag coils left over from the Solid-State Ignition project of a few  years ago.  It is seen at the top of the "mag".  

One plug is connected to each tower of the coil.  The coil fires once each half revolution of the engine.  To keep the plugs from firing at the wrong time (bottoms of the strokes), I have a rotating short connected to the end of the shaft opposite the gear.  I suppose you could call this an anti-distributor.  

When cylinder 1 (the plug of which is connected to the left-hand coil tower) is getting ready to fire, the shorting bar rotates to where it shorts the right-hand tower (plug) to ground.  In effect, this kills the spark that would occur at the bottom of the stroke for either cylinder.  There is an unused spark at the top of the stroke between exhaust and intake but, as in lawnmower engines, it does no harm.

Time will tell how well this arrangement works.  

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

21 August 2007:

Yesterday, I got energetic and designed and built a rat's nest electronics board for the "mag".


  "Mag" shown as mounted on the engine.                              "Mag" shown with the hogshead in place.

Since this isn't gonna be a production thing, looks aren't a biggie.  I intend to make a cover for it so it looks a little nicer.  There are two buttons shown.  One of them turns the ignition on and the other one turns it off.  I've also incorporated a "drop dead" timer into the electronics so if the engine quits, the ignition turns itself off after 30 seconds.  The LED flashes when the coil is triggered so I can set the timing.

The "float tickler"

To make starting the beast a bit easier and so I don't have to squirt gas into the inlets to prime it, I made a "float tickler".  You see it on the float bowl.  It's that little spring loaded brass button thingy just opposite the needle access cap.  When you press it, it pushes the float down a little in order to make the fuel system flood.  Works fine!  Press the button for a few seconds 'til gas runs out around the shaft of the button and it's got plenty of fuel loaded into the intakes to start.


Yes, dear.  The driveway WAS clean before I started playing.  Yes, dear.  I'll clean it up!

I took it out in the driveway in the nice warm (100+ heat index!) afternoon.  Tickled it and gave the strap a pull and off it went first pull.  I also whacked myself in the head with the strap!  I let it run for a while until an exhaust valve stuck (the one I made the guide for - it's a little snug but will wear-in).  Shut it off and pecked the valve loose.  Oiled the heck out of it and then started it again on the first pull.  Yes, I whacked myself in the head AGAIN!  I'm getting to really dislike that method of starting the thing.

And "Thing" it is!  Even though it starts easily, it is the poppin'est, fartin'est, misfirin'est, smokin'est engine I've ever fooled with.  Adjusting the fuel mixture every which way will only get it to run steady when I've got the timing retarded to 15 degrees (more or less) past TDC.

I may have gotten a hint while fooling with it this afternoon.  While it was running as well as it would, I noticed that there is a slight amount of fuel vapor being blown out of the inlet ports.  I'm going to check the valve timing.  If the intake is closing too late, it could cause the blowback.

More to come.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

22 August 2007:

Starting the Edwards (providing it WANTS to be started!)...... pretty straightforward except for the actual cranking.  First, some fuel must be squirted into the intake ports.  Then you crank.  In response to a numerous request, here is how the Edwards engines are cranked:


             A leather strap with a pin attached                       A hole in the pulley that fits the pin in the strap


The pin in the strap fits into the hole in the pulley                 Wrap the strap around the pulley and heave

Try not to get hit in the face when the strap flies off the pulley!

Gravely Model L tractors used a leather strap for starting but the guy who designed it must have had an Edwards and found a better way.  On the Gravelys, the pin was in the cast iron crankshaft pulley and the strap had the hole in it.  Also, on the Gravelys, the strap was wound so it came off at the bottom of the pulley, keeping you from punishing yourself (unless, of course, the 'ol Gravely backfired and jerked your hand into the pulley!).

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

30 October 2007:

I decided to do something about the mixer venturis.  Below, you can see the original pot metal ones.  Frank and I dissected what was left of the ignitor head and pulled the one remaining venturi.  That one was made of tin and shaped sorta like the ones I made today.


The venturis in Frank's engine have a main casting of pot metal with a couple of steel pieces screwed to the sides to neck down the airflow. 

As you can see, the pot metal pieces have a slot to direct the air across the needle valve.  We didn't think this was too good, because a lot of the air misses the needle and doesn't pick up spray.  On the new parts, the airflow is cylindrical when it meets the needle valve.  The cutout in the small end slips over the needle and seat, placing them squarely in the airflow.

I took a little license with the original ignitor venturi and made the final opening 0.5" instead of 0.6".  I think this will make the engine run better at light loads.  If 0.5" is a disaster, I can always pull 'em and hog 'em out to 0.6".  Also, I didn't have a bit that was 0.6" and my boring bar is just a tad too big to go from the 0.5" drill to 0.6".

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

20 December 2007:

Yesterday, I finished making the housing for the fake magneto.


The housing was made from a coffee can and some scraps from a junked air conditioning compressor unit.

The plug towers are made from some old phenolic insulators from, I think, a tuning coil from an old transmitter.  There are two buttons.  The red one is to turn the ignition on and the black one is to turn it off.  There is also a little LED that indicates when spark occurs.  The lever on the side is for adjusting the timing.

I also checked the cam timing and it is okay.  Before I run it again, I think I'll check the valve rocker adjustment.  Maybe the intake valves aren't closing completely.  It's gotta be something 'cause it's GOTTA run better'n it has!

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

21 December 2007:

Well, today, I installed the new venturi and found that one of the mixers isn't centered vertically in the port!  When I put the needle valve in, it bound in the venturi and wouldn't seat. I filed the needle shank a little to get it to run in nicely and clear the venturi (which, being pressed in, doesn't lend itself to being removed without lots of breakage and cussing).

Then I adjusted the valves (cut 'em a little more slack so I'm sure they close), rolled it outside and cranked it up.

Some Joy. It started right up. Only ran on one jug. The other jug was WAY rich. That jug is the one that has the non-centered mixer (natch). Fiddling with it, I found that the fixed part of the jet was sprung a little from being pushed by the venturi. Ran a tap size drill bit through the needle bore and re-faced the seat on center with the needle. Worked somewhat better - at least that jug would fire fairly regularly.

With the spark fully retarded (to a little after TDC), it could be made to run fairly smoothly but when advanced, it would lope. THEN the exhaust valve on the strong jug stuck. That's the one I made the new guide for. Shoulda made the guide outta cast iron instead of the steel bar stock I had in my junque or given it more clearance. It was a bit snug when I put it together.

Yes, Martha, Marvel Mystery Oil really works! After working the valve loose and sticking it a few more times, I added some Marvel to the gas and no more sticking.

Shorting one plug and adjusting mixture for the other cylinder gave steady running of both cylinders but only separately. I think the timing magnets in the magnetothingy are of different strengths and the timing is not the same for each cylinder. That means that one jug fires more advanced than the other and makes the engine run faster than the other jug wants to run - ergo misfiring on the retarded jug. It could also be that one of the throttle butterflies is not in time with the other ont.  When the original AB33 comes in, I'll try it and see if there's more Joy.

After a while, I got tired of futzing with it and shut it down and drained the rusty water out of the hopper. I think I'll use antifreeze in it because it really rusts a LOT. Almost couldn't get the drain plug out! Maybe the inhibitors in the antifreeze will keep the rust down.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -


It would be helpful to have a history of the Edwards company, serial number info for dating, etc.   If you have anything we can use, please contact me at the email address below.  

Information is trickling in but if you have brochures, etc. you 'd share, we'll put 'em on the page for everyone to share.


BOY!  This is FUN!

Stay tuned for further developments.  

Take care - Elden

[email protected]

Visits Since 8 February 2010

wordpress com
stats plugin