The Homebrew Engine

Part Seven

Making Improvements And Corrections


Making It Ready For Showing

You can click on any of the photos to enlarge them.

27 July 2008:

I got energetic today and  made a set of counterweights.  Taking dimensions, I figured approximately what the off-balance cubic dimension was.  This included the crank cheeks, the big end, crankpin and about 3/4 of the rod.  

I then divided this figure by two to get the cubic dimension each of the weights had to be.  A bit of fiddling with dimensions for the weights came up with something close.  I never said I'd get it exact and, anyway, it's a one lunger.  It'll shake some no matter how much I work on it unless I want to use something sophisticated like a balance shaft.  I ain't THAT much of a nitpicker.  Whatever I do here's got to be better than no balance at all.


Turning down the blank.                          Cutting off the blank.                           Facing the blank.      

For the counterweights, I used a piece of the 5.5" shafting I used for the cylinder head.  I turned the diameter to 4.250" then cut off the blank a little over 0.750", the thickness of the crankshaft cheeks.  It went back into the lathe and was faced to the finish dimension.


Cutting the blank in half.                           Milling the cut edge.                              Milling the slot.     

Then, I cut the blank into two pieces.  The two halves were laid together in the mill vise and the cut edge was cleaned up to make both weights exactly the same size.  Then, without removing the weights from the vise, the slot for the crank cheeks was milled just a shade smaller than the crank cheeks.

The counterbalanced crankshaft laid in place to check for fit.

I used my big vise to press the weights onto the crank cheeks.  In the next session, I'll drill the weights and drill and tap the crank cheeks for mounting bolts.


28 July 2008:

Today, I got the counterweights bolted to the crankshaft.


Tap drill weight and crank cheek.                 Clearance drill weight.                   Crankshaft ready to install.

I clamped the crankshaft with the weights pressed on into the mill vise and milled small flats on the weights.  Then I marked the bolt holes on the weights and drilled them with a #3 bit for tapping it 1/4-28.  I re-drilled the clearance holes down just to the crank cheeks (note my masking tape depth gauge).  The clearance holes were used to guide the tap.  

I drilled the holes deep enough to go into the main pins about 1/4 inch in order to more solidly lock the pins to the cheeks.  I had some grade 9 (I think) 1/4-28 machine bolts that, with lockwashers were just long enough to work.  The way they are dimensioned, the bolts thread about 3/16 of an inch into the pins.  The bolts were liberally coated with Locktite and the heck* was tightened out of them.

*: "Heck" - just a little tighter than "The Crap".

Running at about 600 RPM.

After reassembling the crank and rod, I did a little work on the governor linkage to free it up.  Since the joints kept working loose and causing air leaks in the fuel line, I soldered the fittings between the check valve and the needle valve body.  I also made a screen for the fuel pick up tube in the tank.

Originally, I had some rubber feet on the skid.  They didn't work very well and were old enough to have become gummy so they were deep sixed and a couple of 2X2 end rails were made.  The whole works is a lot steadier now.

It only took a few flips of the flywheel to get it to start.  As it breaks in, it is getting easier to start.  I have learned that I have to hold a towel over the air intake to choke it every time I start it.  After it gets running, it does okay.

I pronounce the counterweights a success.  I didn't put a tach on it but I think it only starts shaking at about 700 RPM.  Any slower than that and it is smooth except for a little "piston jiggle".

The governor now works fine although it responds a little slow and may be a little too sensitive because it hunts a little at some speeds.  I used a "rag dynamometer" against a flywheel and it tries to hold speed.  When I made the throttle bellcrank on the mixer, I made it longer than I thought it needed to be and drilled a string of holes in it to hook the throttle rod in.  I have the throttle rod in the closest hole to the shaft and moving it one or two holes away should de-sensitize it enough for it to be stable.

Also, I'm happy with the power it makes.  No telling what the horsepower is but it could be in the 1/4 HP range.

I had to stop the test run after about 15 minutes when the carbon monoxide detector went off.


5 August 2008:

Fiddling with the fuel mixture, flutter choke and governor linkage didn't keep the engine from hunting at low crankshaft speeds.  I finally figured out that, at low speeds and advanced ignition, the throttle plate would close too much to draw fuel from the jet.  When this happened, the engine would falter and miss a couple of licks until the governor opened the throttle and increased flow enough for the jet to start flowing again.  The engine would speed back up until the governor closed the throttle and the process repeated itself.

I removed the mixer and added a second very small jet in the throat of the mixer just upstream of where the throttle plate closed in the idle position.  A second check valve, a small tee for the fuel tubing and a needle to adjust the fuel flow completed the modification.

After several attempts, I got the mixer to behave and the engine now runs fairly smoothly through a range of speeds.  Although the idle needle needs to be worked on to get rid of touchiness in adjusting, it is a good addition.


10 August 2008:

Since I had the DC motor hooked up to "motor" the engine during work, I decided put it back on and make the engine useful by using the motor as a starter/generator.  The motor has a permanent magnet field and when running it with the engine, it will put out as much as 20 volts, far too much to charge a 12 volt gel cell battery.  

The regulator panel showing the engine pulling generator for about 8 amps worth of lamps.

The switch on the left is the master switch that disconnects the battery from everything.  The white button is the "Start" button and the switch on the right is the output switch to the binding posts.

The generator and the back of the control panel.

The only way to regulate the voltage out of the generator is to do it the dumb way and use a voltage comparator and a big 'ol power transistor.  The comparator switches the output of the generator off when it reaches 14 volts.  When the battery is low or the system is under heavy electrical load, the transistor is on all the time and all the output of the generator goes to the battery or the load or both (about 10 amps).  As the battery voltage approaches the regulation point (14 volts), the comparator starts switching the generator output off for longer and longer periods of time, causing the average charging current to fall.  (Note that shorter and longer times are in the range of a half a millisecond to a handful of microseconds.)

This scheme works fine and I have used it several times for float charging lead-acid batteries.  The junkbox provided all the parts I needed, including an ammeter (20-0-20 range) from some long gone car.

You can see a black plate on the back of the ammeter.  This is to shield the ammeter from the powerful field magnets in the motor/generator.  Without the shield, the ammeter will read about 5 amps charge with nothing turned on.

The pewter paint I had on the cooling tank was being dissolved by the oil that is thrown off the engine so I repainted it with barbeque paint.  I also added a cover to the cooling tank and a sight gauge.  The knob on top of the cover is one of those thingies that is used to keep a lamp shade in place on a table lamp...........another serendipitous junque box find.

I may have to do something to the cooling system.  After about a half hour running at load with the lights on, it started boiling and sloshing out cooling water.  Turning off the lights didn't completely stop the boiling so I may be "improving" it.  I probably didn't do the cooling any favors when I coated the inside of the "tank" with gas tank sealer to keep it from rusting.


24 August 2008:

I've been playing with the engine and have made some more improvements since the last update.

Since my back doesn't like a lot of strain, I mounted the engine on a roller stand from a long-forgotten project (Never throw anything away - at least until the garage door won't close or your wife threatens to call a divorce lawyer!).


The rig on the stand.                                    The billboard with the marquis lights.        

Also, as you can see, I have mounted a billboard with rotating marquis lights (left over from my Solid-State Ignition days).   I'll make up a placard that describes the main features of the engine and mount it so it is framed by the amber lights.

I also made up a light fixture that holds a couple of 12 volt auto taillight bulbs.  The lights and the billboard will give the rig something to do.  With everything on, the current draw is about 6 amps.  This is about all the motor/'generator can do but the engine can pull the load easily.

Then, I got tired of water slopping out of the tank when it expanded as the engine warmed up so I added an extension to the top of it.  The problem is now solved and the engine can be run continuously without losing water.

When I was running the engine a couple of days ago, it suddenly started laboring and stopped.  I took a look and found that the piston skirt was a little galled.  After pouring on a lot of oil, it eased up and was okay for a while but is soon started laboring again so I shut it off.

The oiler with spacer to contol the compression of the O-ring gasket.

What I found was the little vellum (gasket paper) gasket that I used to seal the joint between the oiler tube from the cylinder and the bottom of the oiler had come apart.  Crud from the gasket plus the resulting coolant leakage is what caused the galling.  

After a little head scratching, I've come up with what I think is a good fix.  What I think is happening is that the oiler tube to the cylinder is expanding at a different rate than the water jacket and oiler parts are.  This causes the compression joint to squeeze the gasket as the engine heats up.  It squeezes the gasket enough that it can't expand again so it leaks after cooling off.  After a few heating and cooling cycles, the gasket gets ground up and fails.

First, I removed all of the gasket and substituted a small O-ring.  Screwing the oiler into the ball fitting would compress the O-ring but I couldn't tighten the oiler much because that would squash the O-ring until it closed off the oiler pipe.  Leaving the oiler a little loose didn't seem like such a good idea so I made the little brass washer that you see between the oiler and the ball.  This acts as a spacer and allows the oiler to be tightened down solid to the ball allowing just enough compression of the O-ring for a good seal.  Since the O-ring is made of rubber, it will take up the movement while maintaining a good seal.

I ran it for over an hour (didn't burn all of a 7 ounce tank of fuel) today.  So far, so good.

Ignition system

Another thing that bugged me was having to hook up an ignition system whenever I wanted to run the engine.  Since I had a battery and charging system already on the rig, I just made-up an abbreviated version of the Solid-State Ignition module without the built-in charger.  

The coil is one of the last of a bunch of magneto coils I purchased when I was in the ignition business.  Don't ask me what the coils are for........  I was told by the manufacturer that they were designed for some kind of 2-banger military engine.  They did a production run of the coils then the government cancelled the project.  I found out about them after they'd been on the shelf for several years.  I got a deal on some really nice coils!  They never would tell me how many they had on hand but they didn't run out while I was buying them.

There's an LED on the cover of the electronics box that flashes to indicate ignition and I've also incorporated a "Rev Limiter" circuit that interrupts the ignition whenever the engine exceeds a pre-set RPM.  I figured that this was a good idea because, for some reason the setscrew on one of the governor gears worked loose during an earlier run and the engine tried to run away and started hopping all over the bench before I could kill the ignition.  I don't think the governor will repeat that trick but.........just to be safe........

If anyone's interested in any of the electronics on this, I can work-up CAD schematics of the ignition and the voltage regulator circuitry and post them here.  It ain't rocket science but you will need some electronics expertise to build thems.

There are still a few issues that need to be addressed.  I may design and build a complete new mixer.  This one is touchy about fuel level in the tank.  If I make an overflow mixer, it will stay adjusted from full to empty tank.

Although the engine isn't noisy, I'd also like to experiment with a "Maxim Silencer" type of muffler.


27 August 2008:

I made the muffler today.  Again, it was made from scraps.


The inlet end.                                                                        More of the parts.

The intake end of the muffler is made from a grinding wheel clamping washer.  This came off of a $20.00 "1/2 HP" Chinese bench grinder I was foolish enough to think would be worth having.  Months later, I scrapped it and bought a decent ($150) Taiwanese grinder that is a LOT better.

The inlet pipe is cut from a piece of EMT and the body of the muffler is a piece of rusty black iron pipe that was rescued from something that had seen better days.  Note my lumpy welding!  Not pretty but serviceable.


        The templates for the baffle and outlet plates.                      Templates being glued to the baffle blanks.         

The baffle plate and outlet plate are made from a piece of 1/8" steel plate.  The hole size and number was calculated to be a little more than equal to the square inch area of the exhaust stub.  The total volume of the muffler is approximately the displacement of the engine.

I used my CAD system to lay out the holes then plotted multiples of the drawing 1:1.

I was going to make a Maxim type silencer for it but the scrap pile didn't come up with what I needed.


                Center punched.                                              Baffle and outlet plate ready for welding.

One copy of the layout was cemented to each of the blanks and center punched.  After drillling, the center baffle was whittled until it was a hammer fit into the muffler body.  It was knocked half way down into the body.


Baffle plate welded into place.                                                         Cleaning up O.D.          

I then tack welded the baffle plate into the muffler.  After I welded the outlet plate onto the body, I put the muffler in the lathe, squared it up as well as I could and drilled another #38 hole in the center of the outlet plate for use with the live center.  

It took some time but I managed to clean the lumpy welds off and make it look a little more presentable.

Completed muffler test fitted before painting.

Fresh off the lathe, I did a test fit and started the engine.  It works about as well as expected and reduced the bark-bark sound to a nice quiet putt-putt sound.

Am I having fun yet?  You betcha!


5 September 2008:

Oh, yes - I've replaced the crummy little silent movie that used to be here with a sound movie that was shot on 5 September 2008.  The reason is that a neighbor gave me an old broken camcorder that I fixed.  Yes, Martha - I've cheated the junk man again!


29 September 2008:

PLEDGE:  I will not come back here and modify this post if, after I finish and this idea isn't successful, I have to go back to the mixer as it originally was built.

While I was making the new movie, I had a hard time keeping the mixture set so the engine misfired a lot.  I think that the reason was that the fuel had to be drawn from the tank with engine suction and, as the tank level fell, the mixture would slowly go lean.

"What the heck", I thought.  "Since it's running fairly well, I might as well see if I can make it run worse.".  To keep the fuel at the jets at a constant level, an overflow bowl arrangement would be a peachy idea.  Of course, I'd also have to come up with a fuel pump but I'll worry about that until after I've screwed-up the mixer.

Rooting through the 'ol junk box, here's what I came up with.


Junk automatic heating radiator air release valve                                       Valve cleaned up                      

I found this steam radiator air release valve in the brass junk box and cleaned it up.  These thingies were used on single pipe low pressure steam heating systems, in which the condensate water travels back to the boiler in the same pipe as the steam arrives in.  They let a little air (steam) out of the radiator so there will be steam flow to the radiator.  

I've pictured and am using it upside down from it's position when used on a steam radiator.  It seems to be about the right size and not too beat-up so I removed the guts and started cutting, drilling and soldering.

Mounting strap soldered to the bowl body before soldering to the air intake ring of the mixer.


Overflow bowl soldered to the mixer inlet and plumbed to the jets.

Fuel going into the bowl enters through the top casting and the overflow tube (a standpipe) runs up through the center of the bowl, entering where the little air release cap is located.  The stand pipe is sealed with a bushing made of a short piece of rubber tubing, compressed by the little cap.   The height of the standpipe can be adjusted by loosening the cap and moving the tube up or down.

Two short pieces of 1/8" copper tubing have been soldered to the bowl below the fuel level.


30 September 2008:

Today, I make the fuel pump.  I had several ideas and spent way too much time mulling over what to do so I finally went with what I'd dragged to the bench.

The idea is to make a diaphragm pump that runs off the exhaust cam.


Parts of the fuel pump before assembly                                               Completed fuel pump              

The only "new" materials used here were the diaphragm (left over from the Stover DV2 fuel pump rebuild), the pushrod and the bushing for the pushrod.  Because they aren't needed on the mixer now, the inlet and outlet valves were taken from the suction mixer.   

It took a while to get the fit I wanted on the pushrod.  After soldering the blank brass bushing to the plate (made from an old radio dial), I chucked it up in the lathe and drilled and reamed the hole.  Then I made the pushrod from lead alloy steel.  The last "tweaks" on the size were done with a file, then worn out 400 emery paper to get a nice almost zero clearance fit.  This fit is needed because there will be side thrust from the top of the pushrod as the cam moves over it.

Before quitting for the day, I tested the pump with mineral spirits and, surprisingly, it works.


1 October 2008:

It's now running on the new mixer setup.  I suppose you should call it a carburetor because it has all the elements of a carburetor now.  It has a constant level fuel chamber, an idle circuit and a high speed circuit.


As you can see, it ended up being a sort of "spaghetti junction" with the lines from the tank to the pump, from the pump to the fuel bowl, from the bowl overflow back to the tank and one each from the bowl to the high speed and idle mixers.

I did have to do a little tweaking.  First, I forgot to drill a little vent hole in the cap of the bowl and that made the engine surge and quit whenever a plug of fuel overflowed back into the tank.  

Second, I need to put a lightweight spring in the outlet valve of the fuel pump, as it has to push the ball up to seal off the outlet and this causes it to move fuel back and forth a bit in the line.

Third, I have to revise the venturi in the mixer so the high speed jet works correctly.

There are a couple of seepage leaks in my fittings.  This is easily fixed because I didn't use any thread sealer when I assembled the plumbing.

All in all, the engine runs substantially better because I don't have to chase the fuel mixture as the tank empties.

It looks like my first outing with The Homebrew Engine will be at the "Goat Days Festival" on Saturday, October 18th in Blountstown, Florida........... That is, if the world doesn't end before then.


2 October 2008:

I added the check valve spring to the fuel pump, fixed the fuel line leaks and re-worked the venturi of the carburetor.

The check valve spring allows the pump to work much better and I could reduce the stroke significantly and still have more than enough fuel pumped to the fuel bowl.

Old venturi (left) and raw material for the new venturi insert (right)

A couple of old air conditioning tubing fittings make up the raw materials for the new venturi insert.  These were turned to fit into the old venturi housing.  The new venturi ends-up even with the centerline of the main jet.


The new venturi insert in place                                                             The carb with venturi.      

The new venturi is a snug slip fit inside the old venturi housing and the former flutter choke housing with the fuel bowl is used to hold it in place.  Note the ring where the two air conditioning fittings were soldered together.

If I feel like it in the future, I may remove the venturi and insert and bore both of them out about 0.100".  This is because the main jet is hard to adjust to get the engine to run properly at partial load without overfueling at full load.  As it is now, the engine runs significantly better than it did with both the suction/flutter choke system or the fuel bowl system with the old venturi.  It ran even richer before I turned off the web of the flutter choke to open up the air inlet.

Some day, I might rig up some kind of prony brake on one of the flywheels and use a digital bathroom scale to measure torque and then calculate horsepower.  This little engine is not a wimp by any means.  The generator will make 10 amps at 14 volts which calculates to be about 140 watts.  Since I think that the generator is most likely about 50% efficient (including the belt losses), the actual engine load is probably somewhere in the neighborhood of 300 to 400 watts.  This is about a half horsepower and the little engine isn't even breathing hard to do that amount of work!

I guess it will make around 1 to 1-1/2 horsepower if it were tweaked on a dyno and run full tilt.  Not too shabby for a pile of junk, eh?


26 July 2010:

Since my last posting here, The Homebrew Engine has been to a couple of shows but has been mostly resting underneath a towel.  Yesterday, since I'd been meaning to do it for some time, I made a radiator to replace the cooling tank that was inadequate if the engine was run under load for a while.

Radiator mounted.

And a fine radiator it is.  I test ran the engine today after fixing a pinhole leak and found that cooling is still inadequate.  After about a half hour, it started boiling until I turned off the lights and let it idle for a couple of minutes.  I temporarily mounted an old 12 volt computer muffin fan on the radiator core and it did lower the temperature about five degrees but, under load, it's still a bit iffy.

I decided to check top tank and bottom tank temperatures to see if the problem was flow.  I used an I.R. thermometer I bought from Harbor Freight and found that it is defective, reading about 80 degrees low.  My thermocouple meter with the thermocouple stuck in the top tank does, indeed, indicate 212 degrees when it's boiling.  I guess I'll have to see if I can get a replacement ...... Drat!

Once I find out if the temperature difference is between inlet and outlet is too great, I'll think about a water pump.


27 July 2010:

Because the price of naphtha (camp fuel) has risen to over $9.00 per gallon, I decided to see how the engine runs on pump fuel, the kind without alcohol in it.  It runs all right and seems to handle a little more spark advance but it's hard to tell any difference in power.  Then there's the matter of the smell of the exhaust.  With naphtha, the exhaust smells better.  

Today, while testing regular gasoline, I decided to hook a small thermostat to the fan and mount the thermostat on the top tank of the radiator.  The only thermostats I have are ones that open on rise, the opposite of what I need.  That particular problem was solved using a resistor, transistor and diode.  The fan turns on at about 170 degrees and turns off at about 160 degrees.  

Just after the fan started.

Thermostat and reversing circuit.

Before I put the cover back on it, I'll make a better looking fan mount.

By the way, using the thermocouple on the tanks of the radiator, I measure around 20 degrees (F) difference, not a whole lot.  Adding a water pump is iffy and I'll only do that if I get really bored and want another "re-invent the wheel" project.


Anyway, that's how it is as of this time.  I'll take it to some shows in the winter and hope to see you!

Questions or comments?  Email me at:

[email protected]