March 24, 2001

My Radical Radial was a successful project, in that the design resulted in an engine that worked the way I had hoped. But it was full of false starts, mistakes that made me start over, and compromises. This version will be considerably streamlined, and hopefully will run better on less air.

The first change is in the cylinder blocks. Rad-1 had a cylinder block with valve chest attached. Rad-2's cylinder block has the valve chest built in, resulting in a valve much closer to the cylinder it services. This allows a much more compact drive chain. Rad-1's cylinders were drilled through the blocks, with the bottom of the cylinder capped with a piece of angle aluminum which also served as a mount to the frame. Rad-2's cylinders bottom out short of drilling through, eliminating the need for the angle aluminum. The cylinder block will mount directly to the frame, further reducing the profile of the engine.

I will be making as-built plans as I finish each part, so that when the project is done the plans will be available for anyone interested. A link to the plans in .pdf is at the bottom of this page. The finished cylinder blocks are shown below. Time so far: 8 hours.

I suggest you wait until the project is complete and tested before building according to these plans. Otherwise, if I have made a mistake (heaven forbid!) then you would be doomed to repeat my mistake! (Well, maybe "doomed" is too strong a word. But you wouldn't like me very much if you had to throw away parts because of my mistake.)

March 25, 2001

I made the four pistons today. Slipped up on the last one (far right in picture). The slots took multiple passes with the slitting saw, and the last pass on that one made the slot too wide. I'd have redone that piston, but I'm out of suitable material. If I get more aluminum round, I'll make a new one. But this is a cosmetic problem, and probably will not affect operation. Time: 2 hours. Added the piston drawing to the plans.

April 1, 2001

I made the frame today, and drilled the hole in it for the crankshaft. This is the same design frame as in Rad-1, except for two things. First, it's made of 1/2" aluminum instead of 3/8", because that is what I had. Second, the hole in Rad-1 was for access to the crankshaft, to allow screwing on a retainer for the piston rods. The hole in Rad-2 is for the crankshaft itself. It's drilled 5/16", with a 3/16" bronze bushing inserted. Notice in the picture that the bottom legs of the frame are longer than the top legs. This is to allow the bottom legs to fit into holes in the base. The cylinders are positioned equidistant from the crankshaft. At present the cylinders are just placed there with the "eyeball" method, and are not yet attached.

Time: 4 hours.

As-built plans for the frame have been added to "Rad-2 Plans."

May 13, 2001

Two parts made today -- the crank assembly and the valve eccentric.

Crank Assembly: I turned a 1-1/8" aluminum round to 1", drilled and reamed a 3/16" hole in the center for the crankshaft, and parted it to 1/4" thick. This would be the crank disk. I moved the disk to the mill, and drilled/reamed another 3/16" hole through, offset from center by 3/8". This hole, with the crank inserted, will give the engine a stroke of 3/4".

Then I drilled/tapped 6-32 from the perimeter of the of the disk to the center hole for a 6-32 x 3/8" setscrew. I used a long setscrew on the theory that the more threads in contact the less chance of them stripping when tightened. Probably the disk should have been steel, but I didn't have any of appropriate diameter.

I cut a 3/16" steel round to 3/4", and drilled through and tapped it for a 6-32 screw. This retaining screw will prevent the piston rods from walking off the crank in operation.

I inserted the crank in the crank disk and sealed it with Loctite, taking care not to get any in the threaded hole. I screwed in a 6-32 x 1/4" screw as shown, and the assembly was done.

Note: Redesign of the crank shaft eliminated the need for the 6-32 screw and allowed shortening the crank to 1/2". I left it as-is on mine because I had already completed this assembly and sealed it with Loctite. The plans do not show this change, because they are "as built."

Valve eccentric: I did have a piece of steel round large enough for this piece. I turned 1/4" of a 3/4" round to just a skosh under 1/2" to provide a slip fit into the 1/2" holes of the valve rods. ("Just a skosh" -- I love these technical terms!) I cut off the piece, leaving 1/32" of it at its original 3/4" diameter to act as a retainer for the valve rods.

All the valved engines I have seen have been tuned by rotating the eccentric to the proper angle and then anchoring it with a setscrew. This engine cannot be tuned that way, because a setscrew would be covered by the valve rods.

The solution was to put the setscrew in the crank disk, and lock the valve eccentric to the crankshaft. The engine will be tuned by rotating the crank disk to the appropriate position and locking it down with its setscrew.

Here's a shot of the crank assembly and the valve eccentric mounted on the crankshaft:

And here's another showing these parts in relation to the rest of the engine:

Time for these parts: 4 hours.

Drawings for these parts have been added to Rad-2 Plans.

May 20, 2001

Today I made the piston rods and attached the cylinders to the frame. Easy work, but time consuming.

Piston Rods: I made these out of 1/16" x 1/2" x 2-1/2" brass. I put a 1/8" reamed hole at one end for the piston's wrist pin, and a 3/16" reamed hole at the other for the crank. The two holes are 2-1/4" apart, center to center. I made four of these, and then rough-cut them to shape with a Dremel cut-off tool. Further work with a file, belt sander and sand paper contoured them as shown. Luckily, the ends of these will be semi-hidden - I'm not so good with handwork!

Attaching the cylinders: I first marked the frame at the intended location of each cylinder's open end, 1-13/16" from the center of the crankshaft. I thought I could attach a cylinder to the frame with Loctite, and then drill through the frame to mark the cylinder. That way, the holes would have no choice but to line up. Great idea, if it had worked! Loctite releases with the application of heat. Drilling generates heat. Need I say more?

I ended up locating then drilling the holes in the frame, and then clamping the cylinder to the frame. Then I ran the drill bit through the existing hole to mark the cylinder. Separated the two, then drilled the cylinder mounting holes and tapped for 6-32 to a depth of 1/4" Then back to the frame to enlarge those holes to 3/16", giving me some manuevering room when I mounted the cylinder.

I completed this operation for each cylinder before going on to the next.

Here's a shot of the engine so far, showing the crankshafts and cylinders in place.

And here's the flip side, showing the mounting screws for the cylinders. I may countersink those holes to allow the screws to lie flush with the frame. I haven't decided yet, because the bushing on the crankshaft will hold the flywheel far enough away to provide clearance. Countersinking would be purely cosmetic, and I'm not sure it would improve the looks.

Time for this session: 8 hours.

A drawing of the connecting rods has been added to Rad-2 Plans.

May 27, 2001

Valve Rods: I've made previous valve rods from .064" brass. I decided to make these out of .025" brass, less than half the thickness of rods on previous engines. This is really flimsy stuff. But there's no great strength required, since they only push and pull valves back and forth. A light load -- hopefully, light enough. I decided to go with the thin stuff so I could reduce the depth of the eccentric. If it holds together over the long run, I'll stick with that thickness for future engines.

Here's a shot of one of them. The large hole is 1/2" diameter, and was drilled with a stepping drill. A caution is in order here. Use a stepping bit with large holes in thin material. A regular bit will grab the workpiece and destroy it (and possibly a piece of you!)

The piece was first drilled, with the holes 1-1/2" apart center-to-center, and then contoured with a Dremel reenforced cut-off tool and a belt sander.

I'll do the valves next (today) and add them to this writeup when I'm finished with them. Also, I'll add the plans at that time to Rad-2 Plans.

Time spent today: 4 hours.

June 2, 2001

Piston Rods Revisited: The piston rods described earlier worked just fine, except for one problem; the four rods stacked on the crank pin threw the top and bottom rods out of alignment with their respective pistons, causing them to bind. I had thought of making new pistons, cutting each wrist-pin slot to match the offset of its piston rod.

Then I took the project to a SCRAP (Southern California Rarely Accurate Putzers) meeting and asked for ideas. One of the members suggested a Scotch Yoke design, in which each piston rod was connected to two pistons. That cut the stacking from four rods to two, and completely removed the binding. Here's one of my Scotch Yoke type piston rods:

And here they are mounted in the engine:

Now the engine turns smoothly, with so little friction that it will run nicely with air supplied to only one cylinder.

Time today: 4 hours

Note: I have redrawn the Rad-2 Plans in Delta Cad, a simple CAD program that allows me to present more accurate dimensions. Hopefully, these new drawings will be easier to read than what I had up originally.

June 15, 2001

Valves: I made the valves from 3/16" brass purchased from a hardware store. These proved to be slightly oversize, so they would not fit in their cylinders. I reduced their diameter to a slip fit by mounting them in the lathe and sanding them down with 420-grit cloth sandpaper, running the lathe at high speed. Buffed them with the cloth backside of the same sandpaper.

I started the 1/16" valve notch with the point of a cutter in the lathe, just enough to guide a hacksaw. Finished the notch to1/32" deep by holding the hacksaw to it with the lathe running at medium speed. Here's a picture:

The notch and the following full-diameter section on the right are required. The narrow part and following wide part to the right end are purely cosmetic, allowing valve action to be seen from the back of the cylinder.

Time: 2 hours.

A drawing of this part has been added to the Rad-2 plans.

July 4, 2001

With more advice from S.C.R.A.P., I put all the plumbing inside the frame. I drilled a #36 hole from the end of each leg all the way to the point where they crossed, then milled a square channel into the center cutout of one of the two pieces. I notched the pieces as required to allow passage of air from the square channel to each leg. Then I bolted the two pieces together and drilled/tapped 10-32 for the nipple. The result is shown at right.

I tapped each hole at the 4 ends of the frame 6-32, and plugged them with set screws.

The picture shows another brainstorm. Rather than mount the engine as an X on the base, I decided to mount it vertically on one leg. No particular reason, except I already had Rad-1 mounted on two legs and I wanted a little variation. This required that I trim one of the long ends of the frame to match the short ends. The other long end is bolted to the base. Also, I found that the Rad-1 flywheel looked too small on Rad-2, so I bought a 7" pulley from the hardware store. It was built for a 1/2" shaft, so I turned a 1/2 brass sleeve with a 3/16" hole to fit it on this engine.

Fussing around time today: 10 hours. (I've never been accused of being fast -- or good at this, for that matter.)

Frame drawings have been modified to reflect these changes, and a pulley sleeve drawing has been added to Rad-2 plans.

July 15, 2001

It runs! Today I timed the engine on cylinder #1. With air supplied to that cylinder only, the engine had enough power to pull the "dead" cylinders through their cycles. In short, the engine ran on the one cylinder. Then I transferred power to cylinders 2, 3 and 4, in turn. Each cylinder ran strong enough to operate the engine. This verified that timing any one cylinder times them all. Finally, I plugged the "test port" on each cylinder, and put the air to the main input nipple that routes it through the hidden manifold to each cylinder. I turned on the air 50 times, and not once did it fail to start. I'm pretty happy with Rad-2 at this point.

There are still a few things to do before I can call this project complete. I need to replace the 6-32 set screws I'm using to plug the test ports, for they look pretty hokey sticking out of the frame. I'll thread some aluminum rod at 6-32, and screw them in with Loctite. Then I'll grind them off flush with the frame surface. Same with the holes at each end of the frame, where I drilled through to make the manifold

Also, I'm going to paint the aluminum frame and base black, after filling the deep nicks and scars of this scrap material. The pulley I'm using for a flywheel could also use a coat of paint. When all these things are done I'll post a picture of the finished engine, and a video clip of it in operation. Watching the two Scotch Yokes in operation fascinates me. (So does watching concrete harden.)

There was a small problem in initial testing that bears mentioning. The joint of the two frame members was too loose, and it leaked so much air the engine would not run. To solve this I made a gasket of card stock, and sandwiched it between the two frame members. Then after I bolted these two pieces together I saturated the seams with Loctite. Since it takes 500 degrees of heat to release the Loctite, running the engine on steam should not be a problem. This step is not mentioned in the Rad-2 plans. The gasket and the Loctite completely sealed this joint, and the engine runs smoothly.

Fussing time today: 6 hours. (3 of them just watching it run at various speeds)

July 21, 2001

So much for painting the base and frame black. My painting skills are not up to it. I did paint the flywheel, however -- yellow. This doesn't look too bad, and is an improvement over the bare metal of the pulley-made-flywheel. As for the base and frame, I spent several hours sanding these parts, improving the look of this scrap aluminum. Then I sanded with the cylinders in place, feathering the ends of the cylinders into the ends of the frame. This is how it looks today:

Click HERE for a short video of Rad-2 running. 3MB file.

Time today: 6 hours.

Estimated total time for construction: 58 hours.

Probably an equal amount of time was spent on the writeup and drawings. This hobby does keep one out of trouble, doesn't it?

Copyright © 2001, 2002. All rights reserved. See Terms of Use.-