My riff on the Upshur Twin

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Well-Known Member
Oct 6, 2008
Reaction score
I got the first run on one of six engines I have been building based on Hamilton Upshur's Twin published in Model Engine Builder in 2006.

The engine ran briefly on a prime yesterday before I hooked up the fuel tank, and pretty much fired right up today. That's a testament to Hamilton's design, careful building, and the S/S Machine & Engineering CDI ignitions.

I departed significantly from the magazine plans. Differences include:

(1) The crankcase is hogged from solid rather than built up from plates.
(2) The cam gear cover is integral with the crankcase cover.
(3) The crankshaft is machined from solid with counterweights.
(4) The camshaft is machined from solid, including the gear.
(5) The cylinders are cast iron with shrunk-on aluminum cooling fins.
(6) The heads are larger, and the rocker post attachment is beefed up considerably.
(7) The carburetor is a 3/4-scale version of Jerry Howell's V-twin carburetor.
(8) The flywheel is hogged from solid cast iron, including integral fan blades.
(9) The crankcase breather is based on Jerry Howell's V-twin plans.
(10) The intake manifold is a fancy fabrication instead of hoses per the plans.

One issue that arose is that one of the aluminum fin assemblies did not have enough shrink interference, and moved relative to the iron cylinder liner when the engine got hot. That's a problem, because the head is attached to the fin assembly, so the head drifted away from the engine centerline, carrying the valves and rockers with it. I wondered why the valve lash kept getting looser and looser on that one cylinder. I need to figure out how to re-secure the fin to the liner, perhaps with high-temperature lock-tite, but more likely a few radial pins or "dutch" setscrews.

I have five more of these in nearly the same state of building. The cost of sparkplugs at $25 apiece and ignition modules at $85 apiece is slowing me down! There are two ignition modules per engine.
Here is the first run of the second engine. The high-speed carb jet is not right, as the engine runs poorly at wide open throttle, but ok with the throttle mostly shut. The first engine does not do this.
By the way, I cured the problem of the cooling fins slipping by running long studs from the crankcase to the head. Now the whole assembly is more like an aircraft or Porsche engine. It was a challenge to drill the holes for the studs since they were half-holes at every cooling fin space. I used a super-long endmill, because a drill would just walk off line.
You would not need two ignition modules by using a wasted spark method.
No distributor needed but the HV coil must have both leads available.
That is the way I set up mine.
I tried connecting my S&S module to both plugs, but it only fires one. Once that one starts to conduct via its arc, there is not enough voltage to form an arc on the second plug. It appears I need a module designed internally to fire two plugs.
I found that RCEXL twin waste-spark modules work great. So now I only need one module and one Hall sensor per engine, which reduces the cost and effort considerably.

The video below shows the sixth of six engines mechanically complete. All six run. A couple need some carburetor work, and all six need final cleanup and some polishing, plus mounting on display stands. I am really chuffed, because I had zero issues with compression, valve seating, alignment, binding, or any of the troubles one often hears about. The engines pretty much fired right up straight out of the gate. One issue was the creep of the cooling fins alluded in the first post and which I cured by running long studs all the way from the heads, through the fins, and into the crankcase. The other issue was slippage of one of the main bearing bushings under thrust from the starter against the flywheel. I replaced the bushings on all six engines with ones having flanges to prevent that from happening again.

Here are some more photos of the six engines complete. I have an interesting story about solving a mystery that arose with one engine that would run beautifully on one cylinder, but extremely intermittently on the other. No guessing the answer until I pose the problem!
Rklopp--I think it's a big deal when I make one engine. I can't imagine building six simultaneously. Beautiful work.---Brian
I bought an old CNC Deckel mill and judged that it would be as easy to build six engines as it would be to build one. I think I misunderestimated :D! There was a pile a lathe work, and I don't have a CNC lathe. Soft soldering, silver brazing, polishing, and heat treating are also not exactly helped by CNC.

I plan to mount each on a phenolic base under an acrylic cover and give one to each of my four children, display one in my office at work, and one at home.
Here’s a minute-long video showing off one engine running well. It’s a bit smoky while accelerating. I am running 40:1 two-cycle fuel.
Congratulations on the completion of the entire project. That video shows a running engine that reminds me of a two-cylinder Maytag engine.

I hope you put an identifying tag on each one so that future family members will have the history and the family connection.

Do you have the next project in mind?

I have a few projects in my queue. One is to get a Satra/Morton M-1 that I built, running properly. I think the cam, which came pre-ground, is not right - there is zero overlap between exhaust closing and intake opening. I also have an M-5 radial kit that I have not started, other than to have the fixtures from building the M-1 cylinders available for the M-5. I am also itching to build two of Dave Watkins' Wild Rose III 16-mm narrow gauge (for O-gauge track) live steam locomotives. I built a Cracker, which runs beautifully, and two of Watkins' Idris, which don't run due to lack of boiler heating area. I want to recover from the Idris debacle. Children adore the Cracker running.
What a factory! When you did make the crank case out of one block. how did you fix the cylinders?
Is the crank case out of Aluminium?

Nice fotos! Greetings.
What a factory! When you did make the crank case out of one block. how did you fix the cylinders?
Is the crank case out of Aluminium?

Nice fotos! Greetings.
Cylinders were initially fixed to the crankcase via screws through a flange at the bottom. The cylinder heads were attached to the cooling fins shrunk onto the cast iron cylinders. This arrangement didn't work, because the aluminum cooling fins would lose their shrink fit on the iron as the engine got hot, and the heads and fins would move outward from the crankcase! I ended up retrofitting long studs extending all the way through the heads, fins, fins, and crankcase tying everything together a la old Porsches. I hogged the crankcases out of solid blocks of 7075-T6 aluminum. That material is stronger and machines more nicely than 6061-T6, but also tends to pit if water-based mist coolant is not dried off quickly, as I found out the hard way.