The Modular Tower Engine - a new, experimental design

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
HMEM Supporting Member
Sep 4, 2019
Reaction score
North Carolina
With some fear and trepidation, I am introducing a model engine that I have designed, and which I am about half-way through building. The fear, of course, is that I have no idea yet if it will work! Below are some CAD mockups of what it will look like. The first three provide different angles on the fully assembled engine; the last picture has the ignition components removed so that you can see the details of the crank and cam gearing.

It may not be obvious at this point, but this design was inspired by Longboy's Side Shafter and Super Tee - though I hasten to say that he is not responsible for the gaffes and mistakes that I am no doubt making! The key element of Longboy's designs are the connection of multiple cylinders through gears rather than through a shared crankshaft. You may be thinking, "but this "modular tower" is a single cylinder design." Quite right ... but you may note that the design includes the placement of holes and features that will allow a second cylinder to be placed at the other end of the tower - at which point, the tower will be turned 90 degrees in a boxer configuration, but the cranks will be connected to each other via gearing with the cam shaft. Now you know why I call it the "modular" tower engine!

One major feature of the design is also the primary source of my fear and uncertainty about whether it will actually work: The main shaft, i.e. the shaft on which the flywheel resides, is actually the cam shaft rather than either of the crank shafts. I've never seen this done ... but I can't think of a reason it won't work. That may simply reflect my ignorance and inexperience! One issue could be the piston having to complete all four cycles for one revolution of the flywheel ... but then again, this is also true of the various Atkinson designs. I know the "Atkinson differential" is a beast to get running, but from what I can tell the "Atkinson cycle" seems to be reasonably easy to get running.

I welcome your comments and feedback, disparaging or encouraging alike! Stay tuned for the build ... I will provide pictures and plans as they become available.


  • Screenshot from 2020-07-01 16-02-47.png
    Screenshot from 2020-07-01 16-02-47.png
    43.1 KB · Views: 338
  • Screenshot from 2020-07-01 16-03-49.png
    Screenshot from 2020-07-01 16-03-49.png
    35.3 KB · Views: 345
  • Screenshot from 2020-07-01 16-04-18.png
    Screenshot from 2020-07-01 16-04-18.png
    43.8 KB · Views: 349
  • Screenshot from 2020-07-01 16-03-35.png
    Screenshot from 2020-07-01 16-03-35.png
    34 KB · Views: 357
Interesting design. I suspect the engine will run, but two things come to mind. You may run into some torsionals between the crankshaft and the camshaft due to clearance in the gear teeth. Also with the flywheel running at half the crank speed, you'll have less rotating inertia to carry the piston through compression and over top center. I'm not saying this will be a big problem in a small model engine, they are just things that could give problems in a larger engine.

Good luck. It will be interesting to see how your design turns out.


Thanks, Chuck. Both of those have been on my mind, but especially the first - this thing may strip the teeth out of the gears pretty fast, if I can even get it to run. But maybe, maybe ... ? I decided to give it a go, and if it doesn't work, I'll re-purpose most of the parts into a more conventional design. And if it does work, I'll add the second cylinder and see if it works as a boxer. And if that works ... I have further plans. :)
Hi Andy, Guys,

I tend to agree with Chuck. Also for the same amount of effort to keep the piston moving will require a flywheel with twice the mass.

Though I cannot see any good reason why it shouldn't run.
Thanks, all, for the feedback.

Here's the first piece of the engine - the "tower," which functions as the foundation for everything else. Since there is nothing out of the ordinary in any of this, I did not take any pictures while machining - just the finished product.

A couple of notes. First, note that this is fully symmetrical - both ends, both sides, both faces are drilled/tapped/bored exactly the same way. This will allow the second cylinder to be added later one. Second, note that the bores are made to take the common-as-dirt 608 "skateboard" bearings - these will be used throughout to support the crank shaft and cam shaft ... which means that both shafts will be 8mm diameter.


  • IMG_8100.JPG
    66.9 KB · Views: 309
  • IMG_8104.JPG
    61.1 KB · Views: 333
  • 01-Tower.pdf
    190.7 KB · Views: 373
  • Screenshot from 2020-07-02 15-27-28.png
    Screenshot from 2020-07-02 15-27-28.png
    121.9 KB · Views: 339
Hi Andy, Guys,

I tend to agree with Chuck. Also for the same amount of effort to keep the piston moving will require a flywheel with twice the mass.

Though I cannot see any good reason why it shouldn't run.
I think you would need 4 times the mass of flywheel. The flywheel running at half speed has half the momentum and then it's geared to the crank, so the crank sees only half of that momentum torque.
I think you would need 4 times the mass of flywheel. The flywheel running at half speed has half the momentum and then it's geared to the crank, so the crank sees only half of that momentum torque.

Hi Peter, Guys,

Ah, yes ! I missed that ! A good reason not to drive a geared down flywheel.
I have a somewhat fuzzy recollection of having heard/seen of real life airplane engines doing quite that - running the prop on the camshaft to keep keep the revs under 3500 - anybody con confirm this?
I worked in a large stationary engine plant. One of the really big motors was made up of two smaller motors. The camshafts were two pice per each motor. These had an interlocking joint similar to hooking your left and righ5 hands together. Theis joint had a sleeve over i5 for a bearing surface. The double motor had a similar joint on the two motors. Pluse a larger crankshaft. I forget the hp rating but about 5,000 seems right. These motors were designed to run non stop for years. Most had automatic oil changers. At given time oil was transferred to barrels and fresh oil pumped in. Various alarms were in place. The customers just had to bring oil delivery trucks out occasionally to replentise oil. Most ran on natural gas so fuel was not an issue. These ran very large generators for community electrical power, Cylinders were designed so one cylinder with piston and rod could be replaced. The blocks had windows in the sides for access. One was in the world trad3 center basement. It was shut off just before the water got to the carb. In other words it was running nearly submerged. Jyou might want to look at this type of joint also.some model inline motors use this. A variant was used in the ‘60’s twin engine dragsters too.
I haven't been taking pictures of making most of the parts for the engine, thus far, but I did take pictures of making the flywheel, since this was a new approach for me - the last flywheel I made used a key, but this one uses a tapered, split hub that tightens into the flywheel, and tightens onto the shaft at the same time. It uses 4 screws to attach the hub to the flywheel, and has two set screws to help push it out after it is tightened (the taper is fine enough to lock). There is also an open hole in the hub that may be used for a starter mechanism - that is not yet worked out.

Here is the making of the hub, first as a slightly oversized blank with an 8mm / .315 hole bored through:


Next I mounted this blank on an 8mm (.314+) arbor that was made between centers, affixing it with Loctite:


I mounted this between centers in the lathe and finalized the size of the flange; then I put it in the mill vise, located the center, and drilled the various holes into the flange:


I also tapped the two 6-32 holes that will hold setscrews to use to push the hub out after the taper locks:


With the hub all drilled and tapped, I put the arbor back between centers, set the compound to 4°, and cut the taper:


With the hub finished (except for slitting the slot), I continued working on the flywheel. Earlier I had roughed it close to final size. Once the hub was made, I left the compound set at the 4°, mounted the flywheel, faced it off, and drilled it at 5/16":


Then I began boring, using the compound still set at the same 4°, until the hub just fit:


Unfortunately, there was a bit more spring in the boring bar than I thought, so after the final spring passes, the hub fit loosely. :(


Fortunately, there was a ready solution. I went ahead and cut the inset on the face of the flywheel, then faced off the middle hub area shorter until the taper fit snugly as desired:


Once the hub fit securely, I removed the flywheel and began to work on finishing the other side and the rim of the flywheel. I tried to do this with an arbor between centers, but to get access to the face, I would have needed a very long arbor ... and at 8mm diameter, a long arbor was too flexible. So instead I prepared an arbor using a piece of scrap stock in the lathe, cutting a 4° taper until I got the fit I wanted. I drilled and tapped for a central screw to secure it. I neglected to take pictures of this, but it worked well, allowing me to finish the turning of the flywheel. The final step was to put the flywheel in the mill vise, locate the center, and drill and tap for the four 6-32 screws that mount the hub to the flywheel:


I also neglected to take pictures of the final operation on the hub - with it still loctited to the arbor, I put mounted one end of the arbor in a spin indexer and secured the other end with a center. I positioned it to the right orientation and used a 1mm slitting saw to cut the slot through. Then I heated the hub and arbor to break the loctite bond, pressed out the arbor, and cleaned it all up.

The previous post shows the final results. I made the 8mm cam shaft and tried it out - the tapered hub tightens up and locks tightly. I spun it up to 2000 rpm in the lathe, and it seems to be free of vibration.

On to the next part ...
I think you would need 4 times the mass of flywheel. The flywheel running at half speed has half the momentum and then it's geared to the crank, so the crank sees only half of that momentum torque.
I'm going to take the other side of this argument! You can leave the mass of the crank as is. You have twice the torque input now from the reciprocating assembly per two revolutions of the flywheel.
Last edited:
Dave, I have thought it about that way as well. To be honest, I can see an argument to be made on both sides - more mass needed because it has to do twice the work per revolution, or less mass needed because it is getting a power input every revolution rather than every other revolution. I don't know enough to judge between these competing arguments!
I haven't found anything (yet) on the internet about a 4 stroke engine running the crankshaft at half the reciprocating items speed. That you are making it modular for another cylinder will let the project develop as a single to prove viability somewhat sooner.
I have run my 2-3-4cyl engines on one cylinder to isolate fresh engine combustion issues and the crank had no problem carrying the spark less dead cylinders along at a slower speed of course.......sometimes with the plugs in (2cyl) for compression.
Next up for your viewing pleasure, the cylinder. No pictures of the making process - nothing unusual involved - so just the finished pictures. I notice that it has got a bit of rust on it - I had the garage door open the other day when it was raining, and I've been paying for it ever since. :(


  • IMG_8114.JPG
    58.3 KB · Views: 298
  • IMG_8115.JPG
    51.5 KB · Views: 271
  • 03-Cylinder.pdf
    210.1 KB · Views: 260
  • 03-Cylinder.png
    107.3 KB · Views: 270
I've found that my digital camera somehow shows up rust that is almost invisible to the naked eye. And I have issues with damp in my shop too.

Latest posts