Another Knucklehead Build

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.


Project of the Month Winner!!!
Project of the Month Winner
HMEM Supporting Member
Oct 24, 2007
Reaction score
Central Texas
After completing the Quarter Scale Merlin, I took several months off to work on a number of projects that had piled up around the home and shop. High on my to-do list was assembling a number of backup XP computers while the parts were still readily available. I've had an ongoing concern that my ten year old homemade shop computers as well as those running my wife's embroidery machines have been living on borrowed time. I could, if necessary, convert my Tormach to Linux-based PathPilot, but the hardware associated with my Wabeco lathe is still tied to Mach3. I also built up a couple Windows 7 machines so I could have at least one foot inside the modern world. I tried migrating to Windows 7 entirely, but I wasn't able to get some of my ancient CAD/CAM software nor my wife's embroidery software running on their 64-bit operating systems even in their so-called compatibility mode. Replacing all that software was pretty much off the table for me.

Committing to a new long term engine project involved a lot of procrastination and eventually came down to a decision between Ron Colonna's 270 Offy and Draw-Tech's Knucklehead. In order to shake the bugs out of the new shop computers, I modeled the Offy's crankcase as well as the Knucklehead's cylinder head assemblies in SolidWorks. I felt the Offy would probably be of wider interest to others since I'm not aware of any detailed published builds for it. In the end though I felt like I needed more time to consider some alternate approaches to the Offy's one-piece crankcase, and so for now I chose the Knucklehead.

I really liked the looks of Draw-Tech's CAD rendered Knucklehead but wasn't even aware of its existence until I came across Steve (Driller1432)'s HMEM thread:

His successful build validated the plan set and proved the model could be made to run using the original Harley timing. So I decided to do a thread on its build and, along the way, fill in some of the machining steps that Steve left out to perhaps encourage others to build one of their own. There was so much effort put into that engine's drawings that it seems a shame to allow them to languish on the forum's download site.

Even though it has only two cylinders, this engine isn't a beginner's project, though. It's considerably more complex than a Hoglet or even Jerry Howell's V-twin, but the finished result will be more reminiscent of an actual full-size engine.

I decided to begin the build by machining the exterior components of the head assemblies which I had already modeled. This included the heads, cam brackets, valve boxes, and rocker arm boxes. At first glance, the head assemblies appear to be the most complex parts of the engine, and their individual parts must fit precisely together.

My first step was to get hard copies of the pertinent downloaded pdfs since I've never been comfortable with working directly from drawings on a computer screen.

Because some of the key drawings were intended for E-size sheets, I dropped a flash drive off at our local copier store so they could print them out for me on their huge cut sheet printer while I ran some errands. When I returned, though, I was informed that the store's policy was to not copy or print out copyrighted material. They pointed out the title blocks in the lower right hand corner of the drawings that contained words to the effect that the drawings were not to be reproduced without written consent from the original owner. No amount of common sense reasoning could get me past the clerk I was dealing with. Instead of coming back later when someone a little less literal might be on shift, I printed the large size drawings out in poster board mode on my home printer and then carefully taped them together to create the large sheets.

It'll feel good to be making chips again, but with only two cylinders to deal with this time there won't be as many of them. -Terry
Both heads were machined from a chunk of 6061 from my scrap pile. Each head requires several operations, and their order is important to accommodate the work holding of the complex shape as it evolves. The head's 1/32" thick cooling fins are set on 3/32" centers and will wind up too fragile to be reliably gripped in a vise, and so their machining will be left to the very end. Most of the fins are deep enough that they will have to be cut with a slitting saw rather than be slot-milled with an end mill. In order to avoid folding any of the fins over, their external contours will be fully machined before performing any of the slitting operations.

The first operation included machining the outer perimeter of each head as well as its bottom surface. The bottom surfaces were finish machined right up front so they could provide references for the top-side finishing operations. Each bottom-side finishing operation included a hemispherical combustion chamber and a .010" deep mounting recess for a head gasket. In retrospect, I could have increased the depth of the recess somewhat to help locate the cylinder to the head.

The top surfaces were roughed in and then finish machined using a couple angle block setups in the mill vise. Care was taken to get these angles correct since they will affect the fits of the valve boxes to the rocker box. I made a few minor changes to the original drawings in order to provide some additional stock around the mounting screws in the rocker and valve box covers. I also machined the heads to use long-reach Viper spark plugs since I had a couple spares left over from the Merlin project. The original plan set assumes the use of more economical scratch-built plugs for which drawings are supplied. In any event, it's important to machine the plug's mounting surface and to drill and tap its hole in the same setup in order to obtain the best seal around the plug. I've also learned from bitter experience to not bevel the edge of the mounting surface around the top thread of the spark plug hole in order to clean up the burr that sometimes ends up there. Even with a spark plug washer, that extra bit of metal around the hole's top thread helps to seal the plug.

I made a couple minor changes to the vertical fin design to improve the cosmetics of their intersections with the valve cover boxes that will be machined later. When it came time to machine the heads for the valve guides and seats, however, I couldn't find the drawings for the guides nor the upper and lower valve spring perches in the documentation I had downloaded. Steve was kind enough to email me a copy of the drawing that he had used. After some deliberation I opted to combine the guide and seat into a unified cage to improve my chances of obtaining a concentric valve/seat combination. This was very easy to do, but the next head machining step required the cages to be machined and finally installed before the ports for the intake and exhaust tubes could be drilled and reamed.

The cages themselves were machined from C544 which is a free machining phosphor bronze alloy that I've used for the seats and guides in nearly all of my engines. I didn't have any stock of the proper diameter on hand, and so I ordered a short length from McMaster-Carr. I used to buy this material from Enco and then later from MSC (at much higher cost) after they acquired and killed off Enco. MSC, however sells their rod stock only in six foot lengths, and for this project I needed only six inches. What I received from McMaster Carr was clearly marked C544, but its outward physical appearance didn't match the other C544 material I had in my collection. Its spiral-colored raw surface looked more like 632 or 932 rod stock. I've included a photo comparing it with the piece of C544 that I eventually had to purchase from MSC.

The o.d.'s of the cages turned easily enough, but I couldn't control the chatter of the 7/16" carbide ball mill used to turn the i.d.'s. I managed to dig up some scrap cages from an earlier project, and their interiors were smooth with no signs of chatter. I thought I'd better cut some test seats before installing the cages in the heads. Sure enough, the material was so hard that even my carbide seat cutter left unacceptable chatter marks that would have ruined any chance of obtaining a usable seal. I made five trial seats and could not achieve an acceptable result using any of my manual piloted seat cutters no matter what I tried.

I reluctantly ordered the C544 from MSC, and what arrived appeared to be identical to the smaller diameter material I was used to working with. The cages machined beautifully with no internal chatter, and the test seats cut buttery smooth. The cages were sized for a light press fit in the aluminum heads, and high temperature Loctite retaining compound was used to insure they stay in place. A 9/16" standard end mill was used to plunge cut the major bore in the head for the cage, and its actual measured i.d. was used to set the target o.d. for the turned cages.

Two final head machining steps remain. These include drilling the ports for the intake and exhaust tubes as well as cutting the various cooling fins. - Terry















Looking forward to the build Terry.

OnlineMetals has different flavors of bronze. I bought a stick of their SAE660 C932 as well as C544 Phosphor Bronze for different purposes. They were different for sure & OLM packages the vendor spec sheets with the order. Or at least they do that on mine maybe because I'm across the border.

Speedy Metals has 932(SAE 660) & 954 (aluminum bronze) in round, not C544 phosphor bronze.
Terry, another one I will be following.
I know what you mean about the older computers, I run one running XP Pro as I have 3 film scanners($2500 each) that only have software for XP, I also still run AutoCAD 14, way to expensive to replace as I only use it about 20 hrs a year now. I have 2 motherboards, processers and ram put away in case it dies.

After completing the Quarter Scale Merlin, I took several months off to work on a number of projects that had piled up around the home and shop. High on my to-do list was assembling a number of backup XP computers while the parts were still readily available. I've had an ongoing concern that my ten year old homemade shop computers as well as those running my wife's embroidery machines have been living on borrowed time. I could, if necessary, convert my Tormach to Linux-based PathPilot, but the hardware associated with my Wabeco lathe is still tied to Mach3. I also built up a couple Windows 7 machines so I could have at least one foot inside the modern world. I tried migrating to Windows 7 entirely, but I wasn't able to get some of my ancient CAD/CAM software nor my wife's embroidery software running on their 64-bit operating systems even in their so-called compatibility mode. Replacing all that software was pretty much off the table for me.


This one is going to be interesting. Loved the Merlin Build, and I just downloaded the plans for this one from your link.

RE: old XP machines. Unless you need physical connections which don't come on modern machines (like a 25 pin printer port), you should be able to run XP as a virtual machine on a more modern computer. I have a friend who works for a company that still runs applications on Digital VMS based machines, and they use Windows-based servers running Virtual VMS machines. Something you may want to consider next time some of your hardware fails.
I was driving home from work today and was thinking, not heard much from Terry for a while after the merlin build. Glad to have you doing another build, I will be watching with interest.

Oh terry we r thinking alike, must be the Texas air:thumbup:
I too was in the debate of what to build next, and my choices were the knucklehead and the 270 offy also. And the main reason why I decided to make the leap above my pay grade was from watching you build your Merlin engine, i thought thru your thread, man this guy built this in the same amount of time I built my ma deuce,. So naturally I feel compelled to push myself again.
Though my first choice would be the offy to see you build as there are none it seems finished threads out there, I too understand about the offy crankcase after looking thru the plans , but none the less I will be watching this post closely as I can so I know where I'm going on my knucklehead motor.

I found the most difficult step during both modeling and machining the heads to be the drilling of the ports for the intake and exhaust tubes. It's the locations of these two corner features that make the front head different from the rear head. Some difficulty arises because neither hole is parallel nor perpendicular to any axis or feature on the head. The drawings contain the angles needed to locate the two components of each hole's axis so it will properly intersect the area behind the valve seat. It's also cosmetically desirable for the tubes to wind up centered between a pair of radial fins that will be machined later. Touch-off points for the actual drilling operations, though, aren't provided nor are they easily specified because of a lack of suitable reference features in the machining plane. I ended up using some shaky geometry involving a couple corner vertices that I indicated under a spindle microscope.

The heads were supported in a two axis swivel vise which, in turn, was mounted on a heavy duty sine plate. This setup provided the compound angles needed for all four machining operations with just a single reversal of the part in the swivel vise. The ports were through-drilled between the appropriate corners on the heads and their adjacent valve cages using a quarter inch drill. Counter-bores were plunged for the intake/exhaust tubes using a 3/8" end mill. Pucker factor was in the red during all four drilling operations since it would have been easy to ruin the heads which were so close to the finish line. An assembly drawing (lots of these are provided) shows these tubes being held in place with setscrews which I'll likely augment with Loctite. I like to test my engines' valve seals by pulling vacuums through the ports behind them. Any clearances between the port tubes and the heads are potential leaks.

In order to machine the cooling fins I found it convenient to first machine an arbor on which to support the heads using their already drilled and tapped head bolt holes. The slots between most of the fins were cut with a 1/16" thick slitting saw. The 2-3/4" diameter saw used in my 1-1/4" diameter arbor was barely able to cut the vertical slots to their full depths because of clearance issues between the arbor and various features on the head. I switched to a 1-1/8" diameter (1/16" thick) Woodruff cutter on a 1/2" arbor for the three vertical slots in the center of the head. This was necessary to avoid gouging the valve guides which, at this point, had been permanently installed. Had I followed the drawings and used separate guides and seats, the guides could have been temporarily removed for this operation and allow the same slitting saw to be used for all the vertical slots.

The same saw was also used to cut the slots between the radial fins around the lower perimeter of the heads. The depths of these slots depend upon the particular side of the head, but all are contoured to follow the rounded corners of the heads' exterior perimeters.

Somehow I managed to complete both heads without any significant screw-ups. These little workpieces ended up with a lot of machining, and there'll be several more parts just like them to come. It's the detail added by all those operations, though, that will add the interest and realism to the finished result. - Terry












Hi Mayhugh
Jack here, It's so good to see members interested in my model. All should learn from your detailed description of every thing. Awesome Looking forward to your build, and will follow. I am now designing the Flathead next.
Great Work
Each head assembly will include a pair of valve boxes, a rocker box, and a rocker shaft support bracket. All the boxes, and especially those covering the valves, will require quite a bit of machining in a number of different setups.

The easiest to make parts in the head assemblies are the support brackets which will be bolted in extra wide slots between the vertical cooling fins. They were machined in cookie sheet fashion, and only a couple additional operations were needed to drill and counterbore mounting holes for their 2-56 cap screws.

The valve boxes are considerably more complex, and the left and right boxes on each head are mirror images of each other. The plans includes drawings for both-handed parts as well as a number of other mirrored pairs. In my experience, this is fairly unusual with model engine plans and is very much appreciated. One of my struggles with the Howell V-twin involved machining its fairly complex mirror-image carbs from the single-handed drawing provided.

The interiors and outer perimeters of all four boxes were roughed out in the first mill setup in which eight 1-72 mounting holes were also drilled and tapped. There isn't much space on the floors of these boxes for their 2-56 mounting cap screws, but the heads have to be counterbored so they won't interfere later with the valve spring mounting pads. The first secondary operation included the removal of a fillet left on the rear of the box by the first operation in an area where a sharp inside corner is required. A third setup was needed to finish machine the boxes' interiors and provide the clearances needed around the rocker shafts and arms. The boxes were trial fitted to the heads, and the hole locations for the rocker shafts were transferred from the installed support brackets before drilling and reaming the holes through the rear walls of the boxes.

The machining for the valve box covers included drilling and counterboring the holes for the 1-72 cap screws. The covers were flipped over, and clearance channels for the rocker arm assemblies were machined. A nice touch is that the top surfaces of the covers are not flat but are designed with slightly contoured surfaces on a large diameter radius. In order to machine these contours, the covers were bolted to the valve boxes, and the boxes were held in a vise for machining. While still in the vise, a flange was machined into the nose of each assembly for a close fit into the rear of the rocker box. - Terry















That's some really nice machining going on here, nice work & set-ups.

I really like the bead blasted finish, what are you using for media & at what air pressure?

The media is glass bead from a local Harbor Freight. I'm not sure of the grit - it was the only thing they sold about 5 years ago when I bought it. There's no pressure regulator on the air supply feeding the cabinet and so it's running 90-120 psi. - Terry
The rocker boxes protect the rocker arm hardware and enclose the top-end lubrication system. In addition to the large bolts that protrude through the side covers, it was the shape of the rocker boxes that resulted in the term 'knucklehead' becoming associated with this particular engine.

Whenever I make parts like these on my Tormach, I can't help but feel a healthy respect for the skill and concentration that would be required to manually produce them on a rotary table. In my case, the rear and peripheral surfaces of both boxes were machined at the same time before flipping the individual workpieces over to hog out their interiors. I took a few liberties with the original interior shape but paid close attention to the clearances required around the heads of the 8-32 button fasteners that will be used to mount the boxes to the heads. I initially left some extra stock on the bottoms of the boxes so the mounting feet could be trimmed to their final lengths in a separate operation that included trial fitting them to the valve boxes on the heads.

I believe the original design intent was that a pair of -014 o-rings would be used to seal the entries of the valve boxes to the rear of the rocker boxes. The parts drawings include the necessary clearances for them, but they don't appear on the head component assembly drawing that I downloaded. I do plan to add them because otherwise the rocker boxes will almost certainly leak since oil will be sprayed onto the rocker shafts right where they exit the rear of the rocker boxes. The assembly drawing does, however, show equally important o-rings underneath the heads of the rocker bolts which will seal them to the front cover.

As it turned out, I managed to make the machining of the relatively simple covers more complex than that of the boxes. The covers were initially milled out in typical cookie sheet fashion according to their drawing, but the generous fillets around their front faces left little material around their perimeters for any secondary work holding. Not only did the rears of the covers need to be faced to bring them to their final thickness, but I decided to add a snug fitting .020" high boss on the rear surface of each cover to positively locate it to the box. Since I had some 1 mm o-ring cord on hand, I also decided to mill an o-ring groove around the perimeter of the cover in order to seal it to the box.

I bolted each partially machined cover to a piece of MDF with its unfinished rear surface facing upward. When clamped in a vise, the workpiece was securely supported with its finished front surface facing down but perpendicular to the mill's spindle. However, the axes of the cover weren't perfectly aligned to the mill table, and the final fit of the cover to the box critically depended upon this alignment. Fortunately, Mach3's g-code library includes G68 which is capable of rotating the machine's coordinate system to that of a misaligned workpiece. It was easy to calculate the precise angle of the part in the vise by measuring its corner coordinates under a spindle microscope. This was my first experience with G68, and to my amazement the snug fitting covers turned out to be near perfect fits to the boxes.

With the covers in place on the boxes, the locations of the holes for the rocker bolts were transferred from the rear of the boxes to the covers, and the holes were drilled, reamed, and counterbored for the bolt heads.

The final operation on each rocker box was the drilling of a pair of holes through its bottom for the push rod tubes. The axes of these holes were, of course, at compound angles with respect to the surfaces of the box not unlike those for the intake/exhaust ports in the heads. Fortunately, the drawings supplied touch-off points for the centers of these holes which greatly simplified the drilling operations. Strangely, the holes didn't result in the rocker boxes becoming mirror images of one another as I might have guessed. Since I haven't and don't plan to model very much more of the engine's assembly, I hope this doesn't come back around and bite me later. - Terry















Nice work Terry.

I just finished doing some O-ring work for my radial. Fortunately while the part was still in the lathe, I was able to test fit its mating part only to determine the fit was going to be way too tight. So I was able to modify a bit on the fly. With only 1.5mm diameter cord, it was pretty teeny amount. For O-ring parts like your cover where you want it to have a certain fit or seal, do you dry run the groove width & depth on a piece of scrap first with your O-ring type/durometer in hand? I found some info online but it was a bit generic - mostly hydraulic seal related. I have one more sealed part sealed to make, so thought I'd pick your brain.
Normally, I just follow the published recommendations for the groove size, but in this case I cut several trial grooves in some scrap to determine the best height for compressing the o-ring. I ran a narrower groove than typically recommended because I wanted the o-ring to stay put so I wouldn't have to deal with it during what might later be some tricky assembly with the rocker arm components. This meant also making the groove deeper than normally recommended. I ended up with only .005" compression and a 100% filled groove. - Terry

p.s. I was following your project during my hiatus - nice work.
Hi Terry!
Congratulations for the wonderful building. This is really awesome work.

I realized you use a CNC machine.
I'm curious about what CAM software are you using to generate the G code.

I also have a CNC milling machine and today struggling to develop the Fusion 360 Post Processor for the milling CNC control. Not much Autodesk assistance here in Brazil.