Introducing ... the "Steel Webster"

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.
In aluminum, I have broken 3 flute 6-32 taps. Ended my misery in a hurry using 2 flute since. Maybe do the same in steel!
Thanks for the detailed build Awake. The Webster will probably be my next engine as soon as I finish my Kitchen Sink Engine.

I wish I’d have picked the Webster for my first IC engine instead of the Henry Ford replica. That leaky low compression S.O.B. Is causing me to pull out what remains of my hair at a frightening pace!

I read every one of your posts with great interest knowing that your work-arounds and improvements will make my project easier, and your pics are great!

John W
Great job. Excellent build log. I have a couple of engines under my belt, but I still have the Webster on my bucket list!!!
Thank you all!

All of the broken taps have been 2-flute spiral-point taps, which is actually part of what got me into trouble. I am used to using this style of type in larger sizes, where one can power-tap quite easily and safely. Not so much in 6-32 size, as it turns out. :(

The other tap I broke during this project was an 8-32, one of the two through the flywheel to secure it to the keyed shaft. The problem there was a matter of awkward setup - the configuration of the flywheel called for the set screws to be put in at an angle, matching the "draft" angle of the flywheel hub - and in fact, there would not have been any way to get into the spot without having that angle to help clear the rim. But I did not have a very secure setup for holding the flywheel at the proper angle, and the results were predictable. That one was a particular bear to remove, especially because of the limited clearance. At least one 2mm carbide endmill was sacrificed to the machining gods in the process ...
Sure! Here are the two that I've put up on YouTube - the first was the night that I first got it running, with a rats nest of wires and the gas tank just propped to the side:

The second was after I completed the "ignition box" that contains the 12v battery (of the type used in uninterruptible power supplies) and the coil, along with the plywood mount for the engine with the connectors to connect to the ignition box:

Part 10 of the build log includes the valves, valve guides, and springs:

Screenshot from 2020-04-10 20-47-09.png

I began by working on the guides. First I machined the inside of the guides, including the valve seat (though that doesn't show up well in this picture:


I thought I had a great idea for how I would machine the outside of the guides to be perfectly concentric with the insides - I machined a jig with a body over which the valve guide fit snugly, including a .094" stem that ended in a 2mm thread:


This let me set the valve guide in place and secure it with a 2mm nut:


Well, it seemed like a good idea at the time ... but in order to hold the valve guide securely, I had to screw down the nut rather firmly ... and the tiny 2mm thread broke off. :( So I machined the .094" stem away, leaving just the .219" section. Note that all this time I had not removed the jig from the lathe, so it was still perfectly true. I loctited one of the guides in place, let it set, and then machined the outside profile:


Still without removing the jig, I used my heat gun to break the bond and remove the finished guide:


Then I loctited the other guide in place and repeated the procedure. I wound up with two guides with which I was pretty happy. Here is one of the two, along with the part of the valve block into which it was then loctited in place:


Part 10 continues below ...


  • 10-Valves_and_Guides.pdf
    218.1 KB · Views: 273
Part 10 of the build log, the valves/guides/springs, continued:

Next I made the valves. I elected to do this with the stem secured by a live center in the tail stock, with the compound adjusted so that I could cut the 45° head of the valve in the same setup as turning the stem:


To let me cut right up to the live center but also be able to cut the head of the valve, I needed to be able to cut both LH and RH, so I used a round-nosed cutting tool:


Of course, I left the stem extra-long and made use of emery paper to get the stems to exact size (well, within .0002" or so), testing using a valve guide. (I kept the valve guide and the valve as a matched set, to allow for any variations between the valve guides.)

I chose not to try to part the valves off, but rather cut them off using the bandsaw, then trimmed the bottoms of the valves in my 7x14 lathe:


Finally I had two valves that fit smoothly in two valve guides ...


... or not. I did not have a very good .040" drill bit to drill the cross hole that I originally planned to hold the spring and its keeper in place, and I mucked up one of the valves. Sigh ... I remade the valves, this time according to the revised plans shown in the previous post.

The valve spring keeper was next, and it began by machining a bit of CRS to the outside diameter and drilling the .094" hole through the center:


I machined the shallow cavity that holds the c-clip:


Using a ground HSS parting tool, I machined the .188" diameter section over which the spring fits:


I used the same parting tool to part it off, and made the second keeper in the same way. Voila! Two valve spring keepers:


Of course, I should have made three ... because down the road, when I was getting ready for the final assembly, I lost one, and had to make another. :(

Part 10 of the build log concludes in the next post ...
Part 10 of the build log, the valves/guides/springs, final continuation:

Since I redesigned the valves to use c-clips, I had to make the c-clips. The first bit was easy enough - machine the OD, drill a 2mm (0.079") hole, and part off to the desired thickness:


That, of course, gave me an "O" rather than a "C"; the hard part was how to cut out the opening. I set a piece of scrap in the vise on the mill and milled a slot just wide enough to hold the O's that I had parted off:


I used medium strength (blue) loctite to secure the blanks in the slot; in retrospect, I should have used superglue or red loctite. However, this time I had made an extra blank just in case:


I used a 2mm endmill, taking very light cuts, to cut out the openings:


Good thing I made the extra blank - as I noted above, the blue loctite was not quite adequate, and one of the 3 blanks shifted loose, preventing me from completing the machining on it. Fortunately, the others stayed in place, allowing me to complete the 2 c-clips I needed:


The last bit in this part of the build was to make the valve springs. This was the first time I have attempted to make springs, and I have to say that I was pleased by how it went.

Before making springs, I had to make two accessories. First I made a "gripper" - not a very elegant piece, and I don't have any pictures or design to show - just a couple of pieces of scrap steel with a small slot between, such that I could clamp down using the tool holder screws to get the desired grip on the wire. Second, I made an arbor of the appropriate size - or at least, what I hoped was the appropriate size; I had trouble finding a table in Machinery's Handbook that quite gave me what I needed, so I had to extrapolate. (It seemed to work.) I drilled a .040" cross hole though it as a way to hold the start of the music wire.

Finally I could make the springs. I calculated and set the thread pitch that would give me the desired number of turns over the length called for in the plans, put the wire through the gripper and through the hole in the arbor, and adjusted the grip tension:


On the slowest speed of my lathe (30 rpm), I let it build a few starter turns, then engaged the half nuts to create the body of the spring, then disengaged and let it make a couple of finishing turns:


After snipping the wire, it sprang open a bit, just as the book said it would:


I trimmed down the start and finish turns, and tried it out - and it worked beautifully. My first spring (and the second that followed it) were a success!

This concludes part 10 of the build log. Next up will be a relatively simple part, the rocker arm.
Thank you all!

All of the broken taps have been 2-flute spiral-point taps, which is actually part of what got me into trouble. I am used to using this style of type in larger sizes, where one can power-tap quite easily and safely. Not so much in 6-32 size, as it turns out.
One thing to remember as well, if you're breaking taps, is that if you are going 2 diameters deep in mild steel and about three in aluminum, you can use a 60% thread depth, going to 75 - 77% which is where most of the thread charts go, increases the torque on the tap exponentially.
Last edited:
Part 11 of the build log, the rocker arm:

Rocker arm.png

Only took a few pictures of this part - it was relatively quick and simple. I cut a blank out, milled it to size, and started drilling the holes, first for the the M3 screw that holds the follower bearing in place and for the .219" hole for the bronze bushings (shown as .250" on the plan - I changed it when I built it, but haven't yet updated the plans):


I turned the piece 90° so that I could drill a hole to locate the end of the slot for the bearing, and then milled the slot until it fit the bearing:




Finally I milled to size the end where the tappet adjustment screw goes:


And that's all the pictures I have of this part. Not shown was tapping for the M3 screw, drilling and tapping for the 10-32 tappet adjustment screw, drilling the oiling hole, and filing to finish the slot so that the bearing could move freely. I also failed to take any pictures of making the bushings which get loctited into the rocker arm, or of making the pin which gets loctited into the frame and on which the rocker arm bushings ride. Those bits were probably more exciting than the part I've shown above ... oh, well.


  • 11-Rocker_Arm.pdf
    179.6 KB · Views: 223
Part 12 of the build log, the gears:


In my design, the exhaust cam is part of the larger gear, and the smaller gear has a hub section on each side, to allow for spacing of the flywheel and a place for the set screw. Both gears begin with slightly oversize blanks, with the critical dimensions being the 13mm (.512") ID of the larger gear (sized for a slight press fit of a pair of F686 bearings on which this gear will ride) and the 10mm (.394") ID of the smaller gear (for a close sliding fit on the crank shaft):



Along with the blanks, I made arbors to run between centers, one for each blank:



The blanks were loctited to the arbors, and once set, I machined the OD of each blank; shown here is the machining of the smaller gear:



Continued below ...


  • 12-Gears.pdf
    239.9 KB · Views: 248
Continuation of part 12 of the build log, the gears:

Since the gear blanks are mounted on arbors, it is easy to transfer them over to the dividing head on the mill to cut the teeth. Initially I cut the gears using my module 1 "semi-hob":


This is a home-made cutter, but unlike a true hob, the teeth are just horizontal, rather than being a continuous helix. To get the best results with this sort of cutter requires making multiple passes per tooth, with each pass slightly rotated and offset. In other words, it can be tedious ... and even more tedious when someone (who will not be identified) messes up the math and winds up having to cut the smaller gear twice.

When I finished cutting the teeth on larger gear, I also switched over to an endmill and began cutting the cam in stages:


After a bit of smoothing with a file and emery paper, I had a very nice cam gear with cam (attachment 3):


All that remained was to remove each of the blanks from the arbors using my trusty heat gun:


Naturally, I immediately mounted the gears on the crankshaft and on the pin in the frame on which the cam gear rides ... at which point I discovered that, when one designs for a 24 and 48 tooth gear set, and then winds up cutting a 20 and 40 tooth gear set, for some reason the gears do not mesh. (I failed to take a picture ... it was the most deflated feeling to see these beautiful gears sitting with a good quarter inch of air in between.)

So, I had to make new blanks, and cut both gears again ... and at that point I decided to break down and buy a couple of M1 gear cutters. They were inexpensive imports, but seemed to do an adequate job:


With the CORRECT gears cut, I still needed to put the key way in the smaller gear, which I did with my trusty 7" shaper:



I also needed to drill and tap for the set screw that bears on the key. To position it correctly, I used a long key in the keyway to "hang" it over the jaws of the vice, then tightened up the vice and removed the key. Then I drilled and and tapped the set screw:


This concludes the building of the gears; next up will be the ignition.
I like the idea of cutting the key way with the cutter inverted. No need to tie the clapper down!

Did you do this one with a single width cut also?

John W
Can I sidetrack here for a moment ?
I've always liked the idea of one cutter cuts all for a given module .
Altough tempted , I never tried it myself .

If I understand correctly , the cutter is basicly a rack to wich each gear , regardless of the nr of teeth must mesh .
The only parameter needed to make the cutter is the pressure angle .
Sofar I perfectly understand .

But then , after the first pass wich cuts each tooth to its maximum depth , the Z axis must be lowered or raised a certain amount , and the gear neads to be indexed again with 1/2 or even 1/4 of a division added or substracted .
That iteration produces a perfcect involute in theory .

Only I'v never been able to figure how to do it .

How do you calculate the Z amount .
Haven't got a clue really .

For the indexing .
I suppose a 20T gear could be set up as an 80T , where you would use 1 , 5 , 9 , 13 for the primary pass
2 , 6 , 10 and 4,8,12,16 for the +/-1/4 and 3 ,7, 11 , 15 for the 0.5 pass and so on .
Even with a computerized dividing head that would take considerable time and concentration .

Does that make any sence :)
John, I thought it would be a good idea to cut inverted for just that reason ... but I felt like it was a better experience cutting the other way, with the clapper free to move. After cutting a few key ways on the shaper, I am wondering exactly what the logic is for immobilizing the clapper. Of course, this probably means I am doing it all wrong!

No, the cutter is only .085" wide, so once I got to depth I had to shave off the sides. This did help in terms of getting the key way lined up correctly, since despite my best measurements with the mark-one eyeball gauge, I was a bit off to one side at first. :( But that just meant I mostly shaved the other side, and voila! I won't presume to say it is perfect, but definitely it is a nicely functional key way.
Stragen (or Mitsuko?), that makes perfect sense. I'm attaching a .pdf of an article that I have been (slowly) working on for submission to the HSM magazine that is mostly about how I designed and used the "straight rack" cutter shown above. It illustrates (or at least, attempts to illustrate) what is going on when using a cutter like this.

I'd love to get feedback on whether it makes sense - I had to do a lot of condensing to squeeze it into an appropriate length. See if it helps, and if not, that will help me to try to make the article more clear!


  • With Pictures.pdf
    7.6 MB · Views: 322
For the 'straight rack', I imagine it could make quite a difference in precision for something like change gears for a lathe, but for non-critical applications like cam gears for a model, the same type of cutter but only doing 1 pass at full depth produces a decent facsimile of the involute shape. It is made up of facets of course, but a few minutes of running together with oil seems to knock off the high spots and they quieten down considerably. I've never had a set fail or wear out yet, although my engines don't see huge amounts of running. I will say the straight rack method is a lot easier when you only have to do 1 revolution + 3 teeth of the blank to get a decent gear.
Cogsy, it may have been a post of yours some time back that talked about using only one pass, and I mentally marked it at the time as something I needed to consider. As always, theory is an excellent guide to experience, but ultimately, experience trumps theory!

Latest posts