While browsing Ed Warren's "Home Made Steam Engines," Volume 2 -- The Mill Engines" (Camelback Press, 1999), I once more became intrigued with his beam engine, Peggy. Though simple enough, it was far too big for my lathe and mill, and its 1-1/4" bore and 1-1/2" throw probably exceeded the capabilities of my compressor. How would I test this monster?
So I decided to build "Mini-Peg," a half-sized clone. But a mostly depleted scrap bin precluded even that, so Mini-Peg has become "Pegasus," with enough design changes for me to call it my own. The project that follows is being built with the "By guess and by gosh" method, with a lot of reference to Mr. Warren's plans for concepts. Some of the parts have been and will be built with my CNC Taig mill -- not because they need to be, but because I need the practice with CNC, and if the project is successful I may want to make more for Christmas presents.
Here are some of the changes I have already made:
The column supporting the beam is 3/4" aluminum solid round, instead of the 1-1/2" steel tubing in Peggy. And the pivot is built into it, instead of a separate piece.
The beam is 1/4"x 5/8"x 9" white oak, and the pivot hole includes a brass bushing, for in this engine the axle is fixed and the beam rotates around it.
April 29, 2003
Base
The base was a CNC project, half-sized of Peggy as far as the machining is concerned, but machining centered on a 4"x12"x1/4" piece of aluminum. I may have to do it over if the other parts stray too far from Ed's design. I know I'll have to cut a slot for the flywheel, but that will have to wait until I build one so I'll know the required size and location of the slot.
| Here's a photo of the base as it is at present. Time: 6 hours drawing and proving the G-codes. 20 minutes drilling and machining. | |
Beam
The beam was cut from a piece of 1/4" white oak that I've been using as "test pieces" for doing CNC practice. I put some dark stain on it to protect in from dirt and grease.
 |
Note the 1/4" bronze bushing in the photo. |
Axle
| The axle is a 3/4" piece of 1/4" steel rod, drilled and tapped 6-32 at each end. I had hoped to tap through in one operation, but my tap was not long enough. Also, I was not able to tap as deep as I'd hoped for fear of breaking the tap. So I have threaded holes at each end, with only about 1/8" of threads. Not a problem, for there will be no stress on these screws, for they're there only to keep the axle from falling out. I chose 6-32 screws because their heads are slightly larger than the 1/4" hole, so no washers are needed. |  |
Column
I was lucky while building the column, for I cut it more than twice (probably) its intended length. This meant that when I screwed up the top I could just flip the piece and do it again. The second time was more to my liking.
 |
The right end is the reject. I haven't cut it off yet, because at this point I don't know how long it must be. Probably about 2/3 of its present length. |
Column/Axle/Beam Assembly
| Here's a photo of how the column and beam are attached with the axle. As you can see, the column has been in the scrap box for a long time. But it will clean up nicely. |  |
Time spent today: 12 hours. (What can I say? I got interested!)
May 31, 2003
I've been remiss in keeping up with this web page. I've made several parts in the past month. Not as many as I'd like to have, however. It's times like this I wish I were retired! But here we go on the parts finished so far:
Pillow Blocks (2)
This was both fun and educational -- fun, because machining and CAD usually are fun for me, and educational because I learned something new about Vector, the CAD-CAM program I'm using. I found I can draw a part with the dimensions shown in the original plans, and when I'm done just tell the program I want it half-size (or any other size). The program reduces everything to the size commanded. I did this some time in mid May, and as I recall it took about 8 hours to draw and mill the parts.
 |
In this case I drew the part according to Ed Warren's design, except I extended the base by an extra 1/4 inch to make room for the mounting holes, and then reduced the whole drawing to half size. The part was drilled and milled without a hitch. |
Crank Shaft
| This was also done in mid May -- several times. The crank throws were no problem, but welding them was tough. I could not get enough heat to weld steel throws on 1/4" steel round. Finally I cut new throws from 1/4" brass, and these welded up ok. It took a full day to do this (8 hrs) because of the welding problem. |  |
Eccentric
 |
May 24: No problem at all with the eccentric. I drew it up according to Ed Warren's design, reduced it by 1/2 and turned it from cold rolled steel on my MicroMark 7x14 lathe. First try was a charm. Time: 2 hours. |
Eccentric Strap
| May 24: This part mounts on the eccentric, and in conjunction with some linkages will operate the valve routing steam (or air) to the cylinder. This also is Ed's design, drawn in Vector and reduced to 1/2 size. Milled from 1/8" aluminum. Time to draw and mill: 2 hours. |  |
Connecting Rod
| This is a 3-part assembly, and a departure from Ed's design made possible because of the smaller size. The parts are Rod, Beam Coupler, and Crank Coupler. |  |
 |
Rod. 1/4" dia. cold-rolled steel (CRS) turned down at both ends and threaded 6-32. |
| Beam Coupler. This is milled from 3/8"x3/8"x 1" brass. The slot is 9/32" by 1/2" deep, with a 1/8" hole drilled through to pin the coupler to the beam. (All of these dimensions will be spelled out in the drawings when I put them up.) This was a 2-hour part, milled by guess-and-by- gosh. Guess I'll have to draw up a plan for it. |  |
 |
Crank Coupler. This is my design, milled from 1/4" brass. It's important to separate the two pieces and bolt them back together before drilling the 1/4" hole. If you drill the hole first it will be distorted by the width of the cut, and you'll have to drill it again. You may even have to mill the part again. |
Assembly
| Here's the assembly so far, as of May 31, 2003. There's a lot yet to do, but at least this end is beginning to look like a steam engine. |  |
June 7, 2003
Into everyone's life a little rain must fall ...
Today is the 2nd day of a 3-day weekend for me. I had planned to spend all three days making parts. Yesterday, on the very first cut of the very first part, the drive belt on my mill parted company. I had a spare belt, but misplaced it during the move into my new house. After hours of fruitless searching, I ordered a couple of new belts from Taig. But the weekend's shot as far as building parts is concerned. So I've been spending my time at the computer, updating this diary of the project and making more drawings. Hopefully next weekend I'll be back in business.
In general, the project's coming together nicely, with some exceptions: for example, the eccentric strap was originally designed as two pieces -- a short piece that rides on the eccentric, and a longer piece that bolts to it to extend its length. After making and assembling the pieces I realized that the full-sized engine required a total length of about 15 inches and that is why Ed Warren made it in two pieces. My half-size engine requires only 7", well within the capabilities of my mill, and a single-piece strap looks a lot better. I was making this single-piece strap when my mill went belly-up. So that's the first project for next weekend.
Here's a couple of photos of the two-piece strap:
The new strap will look like the lower photo, except it will be only one piece. And that is why my projects take so long.
June 13, 2003
My new belts arrived mid-week, and today I made the new strap, as shown below. A much better solution than the 2-piece assembly.
June 21, 2003
Today I made the piston, the piston rod packing gland, the piston rod packing nut, and one end of the piston rod. I'll turn the other end of the piston rod as the last step of assembly, when I know precisely the length required.
| Piston: This was super-easy. Strictly a turning job, with a hole through the center, tapped 6-32. The oil groove was made with the same cutoff tool used to cut off the part. The piston is turned just a skosh (I love these technical terms) less than 5/8", for a sliding fit in the cylinder. It's 1/4" thick, with the groove 1/16" deep by 3/32" wide. |  |
 |
Cylinder Head with Piston Rod Packing: Oops! I forgot to take a picture of the head and packing gland before pressing them together. The gland was turned down to 3/8", for a press fit into the hole in the cylinder head. I drilled it through 1/8" dia., and then followed with a 1/4" hole 1/4" deep, tapped 1/4-20 to accept the packing nut. The nut was turned from 1/4" brass hex, drilled through 1/8". I turned 1/4" of it down to accept a 1/4-20 die. I used LokTite in the pressed-fit for insurance. |
| Assembled (sort of): The piston fits vertically into the cylinder, with the head bolted to the top of the cylinder. Between the packing gland and the packing nut will be some lubricated packing material. Cotton? I don't know at this point -- I've never done this before. The packing nut will squeeze the material to make an air tight but sliding fit with the piston rod. Hopefully, that will prevent leakage of steam through the cylinder head. |  |
Time spent today: About 6 hours.
June 22, 2003
| Crosshead Assembly: The components are shown on the right. The coupling is made of 1/4" brass hex. |  |
 |
The piece on the left in the picture above is similar to the beam coupler pictured earlier, with a 1/4" piece of 1/8 steel, threaded 6-32". The middle piece screws into it. The middle piece and the piece on the right are pinned together, making a flexible coupling.
The assembled crosshead is shown on the left.
Time today: About 4 hours. |
June 27, 2003
Today was a laid-back day, in that it was a no-brainer. I cut a notch in the base for the flywheel to ride in so that I could snug it up against the rest of the engine rather than leaving it hanging out beyond the base, and I made four feet to get the base up high enough to allow the flywheel clearance from the table. Though the feet will be included in the drawing package, a builder could do just about anything he/she wants with them -- even no feet at all.
| Shown is one of four feet, made out of 1/2" steel hex, turned at its top end to 5/16". The foot is 1" long, which gives the flywheel 1/4" clearance. The screw is 5/16" x 1/2". This was strictly "freehand" on the lathe -- at least the first one was. Once I had what I wanted, I had to make three more just like it. |  |
 |
The flywheel shown is not the one I will use, although it's the right size. This one came out of my scrap box, one of my early attempts. Its center hole is slightly off, making the flywheel wobble as it turns. But it's ok as a mockup. |
Time today: 3 hours. (Told you I was slow.)
June 28, 2003
Valve Bearing Block
I did the two bearing blocks in CAD, according to Ed Warren's design, and found a couple of problems caused by size reduction. First, the parts were so small that I would not be able to secure them with the screws I had on hand, and second, even if I got around the first problem, any alignment error, however so small, would cause binding. So I redesigned the part as shown:
I milled this out of 3/4" x 3/4" x 1-1/2" brass. After squaring the ends, I drilled through and reamed 1/8" for the shaft, then milled out the center. This ensured alignment for the shaft. Alignment of the mounting holes to the base was a challenge for my old eyes. So I cheated. First, I located and center-punched the holes in the base of the block. Then I mounted the block on the base and secured it with Loctite, and drilled through the block and the base. I knocked the two pieces apart, and then enlarged and counter-sunk the holes in the bearing block. Finally, I tapped the holes in the base 6-32. |
 |
 |
Here it is, mounted on the base. Note the hole near the bottom of the picture. It is one of the four holes originally drilled in the base to mount the original blocks. If I can patch them well enough to conceal them when I powder-coat the base, then everything is copocetic; If not, I'll have to make a new base. I won't make that decision until the engine is complete. Who knows, there may be other holes to fill. Actually, I kind of anticipated problems of this kind, for I'm making changes to the design as I go along. In fact, there are holes under the not-yet-secured cylinder that will have to be filled as well, and new holes to be drilled for cylinder mounting. Time for this part: 8 hours, including the two original bearing blocks that I decided not to use. |
June 29, 2003
Valve Pivot
Today was mostly mounting existing parts, rather than making new ones. One I did make, however, is the valve pivot. This piece will transform the horizontal movement of the eccentric arm to vertical movement driving the valve. Two pictures are shown: the pivot silver-soldered to its drive shaft, and the assembly mounted on the valve bearing block.
 |
 |
Time: 3 hours, including miscellaneous assembly.
July 4, 2003
Valve Cross Frame
Ed Warren made this as a three-piece assembly -- probably because of the size of his engine. Since Pegasus is only half the size of his Peggy, I was able to do it easily as one piece.
Time for this piece: 1-1/2 hours, including drawing, cutting and assembly.
 |
This was another chance to practice CNC, though it could have been done easily on a manual mill. The piece is 1/8" aluminum, 1-1/2" wide by 1-3/4" high. The three holes are tapped 2-56. |
And here is how it will look mounted in place. The 2-56 mounting screws will be secured with Loctite in the threaded aluminum, and cut off flush with the brass pivots, during final assembly. The 2-56 hole in the top of the frame will accept the valve rod assembly, which will look a lot like the piston rod assembly shown to the right of the frame. Clearance problems prevented me from mounting the frame "perzackly" over the valve rod assembly's intended location, so I will have to make a small extension to the top hole, in effect moving that hole about 3/8" to the right. I won't make that extension until the valve rod assembly is built and installed. Vertical travel of this assembly is 1/4", which should be adequate to operate the valve. |
 |
July 12, 2003
We're getting close to the moment of truth. I have one more assembly to build before I can test this engine. That assembly is the valve rod packing gland. Without that gland there would be too much leakage around the valve rod, preventing pressure buildup in the cylinder. Maybe I'll get that done tomorrow. Meantime, I've built two minor parts today: the input nipple and the valve rod frame extender.
| Input Nipple: Mostly I run my engines on compressed air, off a hose whose inside diameter is slightly less than 1/4 inch. I made this nipple out of 1/4" brass, drilled through 1/8", slightly rounded on one end to facilitate pushing the hose on, and threaded 10-32 x 1/8" to screw into the valve chest. |  |
 |
Valve Rod Extender: This piece has two functions. First, as part of the Valve Cross Frame above, in effect it positions the frame directly over the valve rod. Second, the 1/8" x 3/8" slot gives the frame "wiggle- room." Without that slot the frame would try to pull the valve rod sideways, causing a bind. The slot is about twice as long as it needs to be |
| Here are the two pieces mounted in place The nipple is about an inch long, allowing plenty of surface for the hose to grab. The valve rod frame extender screws on to the valve rod, which is tapped 6-32. The slot is attached to the valve rod frame via a 2-56 x 3/4" screw. That screw, among others, will be cut to fit during final assembly. When possible, I wait until final assembly to cut screws to fit, as the smaller the screw the harder it is to handle. Also, when cut to fit they are harder to identify during the many times the engine is disassembled and reassembled during construction. This way, they're interchangeable during construction. |  |
Time so far today: 4 hours.
September 21, 2003
Well, I finally got it running. This project was a real bear, with me making mistakes all over the place. Did the cylinder/piston assembly three times -- not because the design was wrong, but because I got careless. The last time through I put five hours into the assembly, and then misaligned the last hole. Maybe in a month or two I'll do that assembly one more time; but in the meantime, it runs ok. That last screw looks cockeyed though, so I think I'll cut it short and Loktite it in for looks; the other three screws will hold it well enough.
I powder-coated the aluminum parts -- red, yellow and blue. I'm getting better at that, and am pleased with the way it turned out.
Click here to see it run.
 |
 |
Copyright © 2001, 2002, 2003 All rights reserved. See Terms of Use.
Copyright © 2001, 2002. All rights reserved. See Terms of Use.