Homer's Project, and how it began.

This web page will be updated as work on the project progresses. There may be periods of inactivity since the demands of my business often take time away from my fun stuff.

This 1933 Franklin Airman is where it all started.  Today it is a very rare car, but my father bought it in 1933, and it was our family car until 1962. I learned to drive on this car, which was the beginning of my love for automobiles. Unfortunately I was unable to find the time or the money for a restoration. Franklin parts are either rare, or must be manufactured at a huge cost. This is an equation for spending lots of big bucks.

This picture was taken in 2003 just before it was sold. So with a little money in my pocket I decided to find a more reasonable car project.

 

I vaguely remembered a slick looking sporty car featured in a magazine many years ago. It had a roadster-style body made out of wood. I remembered it as being made a bit like a strip-hull canoe, which I had built back in the 1970s. I couldn’t really remember the name of the magazine so I started browsing the web in hopes of finding the plans. That’s when I ran across “The Book.”

I called a local bookstore, but they hadn’t heard of it. Since it was a Haynes publication, I went to their web site and ordered one. It was a little expensive having to pay shipping from England, but it arrived right on time. A few days later I saw the same book on the shelf at a local technical bookstore. Guess I tried the wrong one first. Oh well.

It seems everyone has a picture at this stage of development. Probably because getting the top of the frame married to the bottom seems like a real milestone. In retrospect, this really isn’t much.
Another milestone is building the back portion of the cockpit. Still a long way from a finished car. Looking back, I wonder why I thought this was picture-worthy. There is so much yet to do.

The frame is pretty well underway at this point. Here we have just finished checking front suspension geometry to assure proper camber shift with the body roll of an extreme corner.

Notice the position of the steering gear. For some dumb reason I thought it was too high and lowered it. As it turns out, I should have left it where it was. I must stop second-guessing myself.
Now this is what I call a real milestone – a rolling chassis. Look closely at the lowered rack-and-pinion. The resulting tie-rod angle will produce huge amounts of bump steer (bump-toe is more accurate). Clearly I should have left it as before. It will be relocated so as to be more horizontal at normal ride height.

Here is a close-up of the front left corner. The upper spring perch is not done the way Colin Chapman did it. It is more reminiscent of a street rod, but was necessary by a slightly longer coil-over spring/shock. 

Looking down the side of the frame, you will see a good deal more triangulation. On paper, this should make a much stiffer frame. The added cost in weight is less than I gain or lose from week to week.
Independent rear suspension from a Mazda Miata. It took a little time to work out a suspension geometry that would produce the correct camber change. If anything, it may slightly overcompensate body roll. It should really stick in the corners.
I had some work to do before fitting the engine. The 2.3 liter ford has a very deep sump, which makes the overall engine height way too tall. This is about to be given a little surgery to remove the ponderous bottom.
A little careful work with a saber saw yields a very shallow, but very leaky pan. Unfortunately, I didn’t take a picture of the wide “box” that was later welded to the opening. It formed what is normally called a winged sump, and still holds the 5 quarts of the original. To minimize oil sloshing, the “box” was given baffles to hinder side-to-side and for-and-aft oil sloshing.
Here it is with the engine and transmission installed. Not under its own power yet due to a missing drive shaft, gas tank, a half-mile of wire, and a few other minor things like seats. Still, it feels good to have this much finished.
I bought some Lotus 7 clamshell fenders that didn’t fit because of the added width of the Miata suspension.  To match the rear track, the front control arms had to be lengthened, which should give great cornering, but unfortunately, the fenders won’t work. The supplier agreed to swap for cycle fenders, but the packing carton was so large the shipping cost was more than buying the cycle fenders.

I need to disassemble the rolling chassis before giving it a coat of paint. But first I need to finish welding joints that are still just tack welded and attach lots of little things like Dzus fasteners that hold the nosecone to the frame, and brackets for the fixed ends of brake lines and cables.

Before I could determine the mounting points for the nosecone fasteners, I had to have a nosecone. These blocks of Styrofoam are where it started.

Next it was time to glue the rough-cut sections to one another to get the approximate nose cone shape. This is tricky because some types of adhesive dissolve the Styrofoam, and others don’t dry well. Resorcinol glue dries well because it doesn’t rely on evaporation, but it becomes hard and brittle. Carpenter’s glue takes a long time to dry, but it shapes better than resorcinol.

If I had it to do over, I would use urethane foam. It is easier to glue, and shape. Styrofoam has tiny beads that break off during shaping and float around the garage, getting into every nook and cranny. They also develop a static charge that makes them stick to everything. The weight on the top of this stack is there to hold things together while the glue dries.
After rough shaping the foam, I covered it with layer after layer of drywall compound. Sanding the “plaster” gave a smooth surface that could be formed into the finished shape. Here I am doing a little finish sanding with the foam and plaster being held to the frame with a cotton rope. A belt would have been better since the rope cracked the thin layer of plaster, which I had to patch and sand smooth again.

Note the scuttle frame in the background without it's skin. You will see that change in a few more photos.

Once the plaster was shaped, I painted it with primer filler. Then did some wet sanding with 400 grit wet-or-dry paper. I used two different colors of primer so it would be easy to see the high spots. Here, the white spots are the base plaster; the next coat was red primer. The gray spots are gray primer, which was sprayed on top of the red. It took several such coats before I was happy with the finish.
After the surface was smooth, it was waxed. I actually gave it three coats of wax. Then I sprayed liquid mold release on top of the wax. All this to make sure the fiberglass resin would not stick to the form. This shows the fist layer of polyester resin gel coat tinted black. I painted two coats of gel coat with cheap, disposable paintbrushes. Then two layers of fiberglass mat separated by cormat (a light-weight stiffener). Once the resin cured, the foam was removed to make a female mold into which I would lay up the finished product.

Since the form was designed with a ¼” lip around all edges, it would be impossible to remove the finished nose cone without splitting the form. This shows the result – a right and left side with a plywood flange laminated to the form.

This photo was taken right after the form was separated, and finished nose cone removed.
This photo was taken right after the finished product was removed from the mold and placed on the frame. It’s not sitting square on the frame because there is still a lot of trimming and finish sanding to be done. The grill opening is just rough-cut in this photo. In the next picture, the profile looks more normal since you can’t really see all the trimming that remains.

With the car in the background it’s hard to make out the dash and firewall (called a scuttle in England). The steering wheel is also hard to make out, but it’s there, and it actually turns the wheels. 

If the ground clearance seems a little low, that is because we haven’t made an attempt to adjust that yet.

Scuttle is still just a frame without skin. Be patient. That is all about to change.

This was taken a little later, after I had sanded the part-line and done a little trimming on the edges. In this picture, the grill just stuck in the opening. The grill was made from the front of a damaged shopping cart that had been taken out of services. Some day I’m going to adjust the ride height so the car sits level.
Before cutting the aluminum for the scuttle, I created a paper pattern which included "bend lines."  The cut lines and bend lines were then transferred to the aluminum. 

I thought bending to the correct radius would be difficult, but fortunately, the phone company provided a handy  tool. By inserting the sheet metal between the telephone pole and a PVC down-tube, the PVC pipe became a perfect mandrel to bend the sheet metal around..

This is what the scuttle (cowl) will look like once it is finally in place. In this view, it is only clamped in place with a c-clamp on each side. The sheet metal is aluminum with the white protective film still in place. It is bent to conform to the shape of the cowl frame, but won’t be fastened until the frame is disassembled for painting.
When it came time to put the skin on for real (and get rid of the c-clamps) I decided to use a technique used in aircraft construction. Notice the pop rivets are flat top, countersunk. The method used was to clamp everything in place and drill the aluminum and steel tube at the same time. Then unclamp and countersink the steel tubing. Then put the clamps back on, and dimple the aluminum into the countersunk holes in the tubing. The last step was to insert the pop rivets. 
It was a little  tricky getting the bends around the ends (for riveting) so the aluminum skin fit tightly against the scuttle frame. It took about a bazillion clamps, but the extra time and effort paid off. This is the result. Obviously, the dash has been temporarily installed for this photo.
Here is another view. I included it because of the interesting reflection of the picket fence in the background..

You'll notice the screws around the perimeter of the dash panel. They screw into holes drilled and tapped into the scuttle frame.

It was a nice change of pace to fabricate the exhaust header. Primary tubes are 1.75 inch exhaust pipe in the form of a Tri-Y (4-2-1). The small pipes all funnel into a 2.25 inch pipe. Eventually this bad boy will get ceramic coated.  It should last longer and keep the under-hood temperature down.

This photo also shows a change to the book frame. In the book, the panel where the exhaust exits has a single triangulation tube. I decided to add some stiffness here for two reasons. One, the 2.3 L engine is pretty heavy and develops a lot of torque. Two, the top and bottom tubes are long and unsupported which allows those tubes to flex. The added triangulation should add a lot of stiffness to the frontl.  

At this point, I’ve been unable to find narrow enough seats to fit inside the frame so decided make fiberglass buckets with  Kevlar reinforcement. This is a plywood and plaster form from which the seat was to be fabricated. 
This is the seat fresh from the mold. Unfortunately the headrest lacked stiffness and wasn't tilted far enough forward.  Although it was a nice fit in the cockpit, it couldn't be used. 

Rather than take the time remake the form and seat, I decided to buy aluminum dragster seats that would fit the narrow cockpit. They will require padding and upholstering, but then my hand made buckets would have as well.

Here are the dragster seats from the back, however I really took this photo to show the fuel tank in position. I welded this from mild steel instead of aluminum. After leak-testing I coated the inside to keep rust from forming. While steel is heaver, the weight difference wasn't great enough to justify paying someone to weld one since I haven't welded aluminum since school (a long time ago).
This photo shows the pedal assembly, master cylinders, and throttle cable in place. Eventually the frame for the pedal assemblies will be enclosed by aluminum panels that will form a tidy little box similar to that of a Series II. Naturally, it won't be exactly the same since these  master cylinders face forward. 

The throttle cable works a bell crank that actuates the throttle (left of picture).

The steering column is stabilized by the ball bearing pillow block right behind the upper U-joint. Even though it hasn't been on the road yet, steering feels very tight.

I pulled the engine as one of the tasks required to prepare the frame for final assembly. While sitting on a bench I fabricate mounts for a small Nippondenso alternator. There are several versions of this small, internal fan unit, but this came from an 89 Toyota pickup. Even smaller units come from forklifts, but this was a low cost item from a local wrecking yard.. 
The upper and lower mounts were fabricated from scratch. The lower one can't be seen but was welded from 1.5" by 3/16" thick steal, and provides a solid pivot point. The upper bracket was made from two spherical joints, one LH and one RH thread. Nuts welded to opposite ends of a small tube make a turnbuckle for easy belt tension adjustment. Jam nuts lock things in place. The string is a way to find the drive belt length -- 34".
  The original carburetor baseplate was cast with a mount for the EGR valve. As a result, it was very think, and caused the carburetor to stick up above the top of the valve cover. The added thickness of an air cleaner was dangerously close to the bonnet. Earlier I had made a block-off plate for the EGR port, but it looked ugly and did nothing to make a smaller power plant. My solution was a 0.50 plate of aluminum and hours of drilling holes and work with a file.
Here you can see the bright, new baseplate under the carb. It might not be clear to you, but I really notice how much lower the carb is now. 

Just behind the timing belt is an oil catch can from Jaz. It only holds a pint, but I'm sure I won't need more volume because there is a drain valve at the bottom. I don't plan to run PCV to the intake, but if I do, it would be a simple matter to block the can's breather port, and use the extra vacuum fitting (left side of can). 

I decided not to use the original monster cable clutch release in favor of a hydraulic one. This CNC brand is almost a dead ringer for a Willwood unit. They are identical in size and shape, and they both pull.  My selection of CNC was highly scientific - a local dune buggy shop had one in stock, I had to order a Willwood. As I recall price about the same.

The CNC unit came with a long, threaded stainless rod for the "fixed" end. I cut off 2.5" and rethreaded the end for the spherical joint. I cut off the cable end of the throughout fork and welded a grade 8 bold for the other spherical joint. I don't like how it turned out, and will probably re-fabricate something better. 

You can see the frame is now bare, in preparation for paint. However, before I could start painting I had to decide what to do about a roll bar. I opted for a bolt-on unit. This photo was taken mid-process. The rear struts haven't been added yet. Here, steel plates were bolted to the frame's shock towers. Then the roll bar was positioned and tacked in place. Next it was unbolted and final welds were done on the bench.

Since the book frame was modified for independent rear suspension, here are a few words about what you see here. Tubes RU1 and RU2 are normally at the edge of the frame's axle opening. With IRS, they interfere with the lower control arms. Notice they have been moved inboard to be the anchor for lower A-arm mounting brackets. The book calls for an angle brace (tube Z) between RU1 and B1. Since RU1 has been moved, the four tubes that form a diamond provide similar stiffness. The diamond wraps around the bottom of the differential. One tube needed a notch to clear the differential nose.

By the way, it takes a long time to paint all those tubes.

Now that the frame has been painted, I'm cutting and bending aluminum sheets. Here is a start at the pedal box. I can't wait to see the master cylinders sticking out from the front panel.

The back of the pedal box, facing away from the camera, will butt up against the firewall of the scuttle.

The white paper is a template of the shape required for the tray in the engine compartment. The template will be transferred to aluminum sheet. Then I'll be cutting up again. 

The last thing I need to do to the frame is weld a mount for a torque reaction arm I fabricated for the differential. In the Miata, reaction torque is handled by a formed channel from the differential nose to the transmission tailshaft housing. 

In this photo, the torque arm is the diagonal tube that runs from the right side of the differential to a cross-brace at the top of the driveline tunnel. It is rubber mounted where it joins the frame to dampen vibration. The torque arm is heavy wall tubing in the shape of a Y. The Y end brackets the stock torque arm mount. It is a little like a ladder bar with the big end fastened with two huge through bolts between the top and bottom of the differential nose. 

This is a cool tool for cutting sheet metal. The cutter produces a coil of metal as seen here. It's easy to use, and leaves the edges clean. I figured it was a worthwhile investment since there would be a lot of aluminum cuts.
The first major piece of aluminum to be fastened to the frame. Seemed worthy of a photo at the time. Attaching this piece showed me where the book left out a lot of little frame details. For example, there needs to be a frame member to fasten the edges of each panel to. Riveting to air just isn't secure. In this case, I had the forethought to bend flanges on this panel where it fits around the driveline tunnel. That gave me something to rivet the tunnel skin to.

I came up with a quick way to assure the rivets were evenly spaced, and in a reasonably straight line. Drilling holes at each inch mark on a yard stick worked very well. So well that I only used this wooden template once. I immediately made a template using a metal three foot rule.
Here I started putting the floor panels on with the frame up-side-down, on jack stands. Notice the metal drilling template made by drilling holes at each one inch mark of a metal three foot rule. Since the side skin will bend over the bottom and be riveted to the same frame rail, I skipped every other hole. When I fasten the side skin, I will put those rivets in between. The end result will fasten the floor every inch, and the side every two inches. 
I didn't break a single drill bit while drilling 127 holes through the aluminum and the steel tube.. All perfectly aligned thanks to 10 Cleco clamps that made sure the aluminum didn't shift during the process.

After the holes were drilled, the aluminum skin was removed and deburred. Then every hole in the frame was countersunk. In this photo the top three holes have been countersunk, the nearest ones have not. 

When the skin was ready to be fastened for good, each hole was dimpled into the countersink. That allowed the use of flush rivets. The Cherry Q brand of rivets were used which can be found at aircraft supply stores. They have higher tensile and shear strength than standard pop rivets, and the mandrel remains in the hole. That way they are self sealing (to an extent). The mandrel also lends to greater shear strength. 

Wearing gloves while drilling  makes it easy to brush shavings away -- almost a requirement since the shavings from one hole make it hard to see the center punch for the next. Without gloves it's easy to get nasty metal slivers.
This is the result from fastening just one side of the floor pan. Lots of drill shavings, and 127 rivet mandrels.
3 pieces of Aluminum + 281 Rivets = 1 big smile. It really looks like progress.
I'm sure I took some photos of the assembly steps that took place between this picture and the last, but who knows where they are. I'll keep looking.

Here, I'm just finishing the rear aluminum skin.. The car is on jack stands so I have enough room to get underneath and "roll" the skin around the bottom frame tube. The top is now held in place with the screws that mount the snaps. I've heard of people using clamps to hold the metal in place while they work the bottom. Using the snap screws worked better for me -- no shifting, and no clamping marks.

This is one of the front fenders fresh from the mold. Making the mold was simple. I just bought a heavy steel trailer fender, bobbed it so it wouldn't cover so much of the wheel. Then painted the inside. After the paint was completely dry, it got a heavy coat of wax mold release, then a coat of PVA mold release. This is the untrimmed result.
Here they are after a little work. They've both had the excess fiberglass trimmed. The nearest one has been sanded with fine wet or dry. The other is just how it came from the mold. 

If I'd have been smart, I would have mixed the pigment a little darker so it would be a little closer to British Racing Green. Maybe next time.

What you see here are the wires being sorted for the general part of the car they will be run from the fuse block (already mounted). This photo was taken a few minutes before Jon Lexau dropped by to help with the wiring. Like most other things I estimate, wiring took a lot longer than I expected. Even thought I purchased a pre-fab wiring harness (and fuse block) it still took most of a day just making connections. That doesn't count the prep work of drilling holes, running wires to the proper part of the car, installing rubber grommets, and wrapping the bundles in plastic loom.  I'm sure Jon thought it would just be a couple of hours, I know that's what I thought. Thanks Jon. 
You couldn't see this bundle of wires in the previous view. This is the under-dash wiring -- instruments, ignition, horn, lights, turn signals, etc. This is what Jon did while I ran the wires to the engine, fan, horn, front lights, rear lights, and turn signals. Basically, everything outside the car. It would have been impossible for both Jon and I to have worked on the dash at the same time- just not enough space.

It was very well worth getting the factory wiring harness. They were color coded, labeled, and sized properly. All of this made for a much cleaner, faster, and error free job. On previous projects, I did my own harnesses. I'll never do that again.

This is the instrument panel after the wiring job, but before I tucked some of the wires up, out of sight. I think it looks pretty nice. I tried to keep the layout in keeping with Colin Chapman's design for the original Lotus Seven. I've used more instruments, and different light and wiper switch, but it's reminiscent of the original.
Vroom. Vroom. It makes noise like a real car. Here I am adjusting the carburetor. Now it's really getting exciting. Note the little tiny red battery. Just trying to keep the weight down.
Once the wiring was done, and the engine went Vroom, I didn't want to take the time to build my own rear fenders. These were from Curtis Unlimited. They were inexpensive, but needed a good deal of work before they were ready for paint.

I decided to mount them while in gel-coat. Then they would be solid for the necessary body filler and sanding it took to make them smooth and ripple free. It might have been better to just pay the extra money for Caterham fenders.

Yes, that's a big smile. I'm actually going to drive it up onto a trailer, and take it to DMV for a VIN inspection. There's a major story about that whole process. Check it out after the last photo.
One of the early drives with my daughter. It's low, and loud, and fast, and you just can't help smile. 

Not long after this picture was taken, I lowered the seats. The bottom cushion was just too high. Seating is now 2-1/2 inches lower, and actually more comfortable. The only thing is that now I have to look up to see the top of the tires on passing trucks (that's me passing them, of course). 

After a few hundred spirited miles, the cam in the little 2.3 liter developed a flat exhaust lobe. I probable should have gone through it before putting it in the car, but it seemed to run good in the Pinto donor car.

My 14 year old nephew, Ryan and I replaced the cam with the engine still in the car. After pulling the old cam, I wanted to make sure we cleaned out as much of the worn metal that was certainly floating around in the lubrication system. We pulled the pan, and flushed solvent through the oil passages. We scrubbed the top, and inside the crank case. Then we replaced the rod and main bearings. This is really the hard way, but it was only two days before track day at Portland International Raceway.  

While working under the car, a drop of oil fell into Ryan's mouth. I told him that he now qualifies as a real grease monkey.

At Portland International Raceway with an engine that only had 250 miles on new bearings and new cam. What excitement. The engine performed like a dream, and the car  really did go fast, and cornered well. Best of all, it passed tech inspection so I could run. 

A lot of cars were faster then mine, but I can't imagine their drivers had as big a smile as I did. How's that song go -- I did it my way... from scratch. 

Here it is at the All British Field Meet the day after running at PIR. I thought it was kind of cool to be parked beside its cousin, a real seven. Notice how much taller my roll bar is. That's because I drew a line between the top of the front part of the frame and the top of my head. Then I extend the line to where the roll bar would be. That way if the car was resting upside down on the roll bar and the front of the frame, my head would still be above ground level. I did the measuring before  lowering my seats. That's why the roll bar is so tall. I'll fix that someday.

Now here's some irony. The car won People's Choice 1st place. I never expected it because the fenders still need paint, and I haven't finished the inside trim. Not only that, the day before I was on the track at PIR and the car still had track dirt stuck here and there. Oh well, if they want to give it to me, I'll take it.

Here's a passenger seat view taken while touring the Columbia Gorge Scenic Highway. The car in front is a Westfield 7 owned by Dale and Sue Fillmore. Good friends, but they had to buy their car instead of build it. 

I had to say that because Dale was going to build one with me, and even bought the aluminum for the body. Then he found a deal on the Westfield. Anyone what a bunch of 2024 T3 Clad aluminum?

Here we are with Club Lotus North West getting ready to start a summer drive from Grand Prix Imports, in Wilsonville, OR. We crossed the Willamette river by ferry, and later crossed a smaller river on a historic covered bridge. Ultimately we got to Silver Falls State Park for a short rest before heading back through a scenic mountain road. Lots of curves and lots of fun.
This was a short stop at the Cape Horn scenic outlook on the Washington side of the Columbia Gorge. Donna and I were on our way back from a Sunday brunch at Skamania Lodge. 

Fenders still need paint. Hood needs louvers and paint. Roll bar needs lowering, but it's still a blast to drive.

More photos to come. More words to come.