2024

JANUARY

Did I say a third try making the Kamm cove?!?! I DID!

So, while making the cardboard contour template to fit inside the Kamm cove foam piece, I realized that I was also making the contours to create another Kamm cove. Only this one would be more accurate, having more contours. To then create a finished piece, I could spray insulation foam between the contours and sand down to shape. And it would begin from the accurate cutout of the Kamm cove outline, which when all is done, I could then simply insert into the tail of the foam body. This struck me as a more accurate, easier to be accurate, and easier to mate with the foam body approach. It was well worth trying at this point. So I glued the cove contours onto the Kamm outline cutout.

That done, the next step is to make the taillight cones to glue down on the Kamm outline. I laid those out, having plotted my cone shape from my 3D drawings, and laid out the shape on some thick, glossy card sheet. I made it long enough to wrap it twice around. Then I cut out three diameters to glue inside the rolled up shape at the top, middle, and base.
That done, I cut out the placement positions from the contours and inserted the cones, and stood back and studied what I made. It did not look right. The outer cone look too tall and too narrow at top, while the inner cone looked too short and too wide at the top.

Back to my drawings and refining my 3D points and using my photos of the car to derive the shapes, heights, and placements. Slowly, painfully slowly, I am settling on final dimensions for the taillights. It is frustrating and I have to admit that I am stuck in this muck. But I must come to conclusions I can live with.

FEBRUARY

Back I went to my drawings and my photos. I found three photos that approximated orthographic rear, top, and side views. I attempted to determine the vanishing lines to establish the correct position in perspective for features of interest. I then compared what I got from the drawings (making tracings) to the Photomodeler drawings.

Vanishing lines were hard to work out. I tried to find measureing points in the same planes of interest. After comparing the tracings to the Photomodeler master drawing, I decided that the Photomodeler drawing was the most reliable drawing to work with. Still, I adjusted its dimensions for the taillight cones to the nearest 1/8 inch, and the differences looked to be minor or not observable. So, once again, I created patterns to make the taillight housings. They will be very close to my first set, but more trustworthy.

MARCH

With the contours glued down to the backing board and the tailcones placed, it is time to fill it in with the spray insulation foam, large-gap formula. It is not as controlled coming out as I expected. Controlling the amount was difficult and not exact. I ended up spraying down too much, which resulted in air pockets under the surface when I came back over an area. And, the large-gap foam cannot be easily built up. It comes out too much and therefore too heavy to hold itself up. So I needed three sessions to get enough to mostly cover the contours. And each session needed 2 to 4 days to cure. And even then, when I sanded into an air pocket, I found uncured areas that did not have enough air exposure.

Carving began by using a hand saw to trim the excess off the outside and the inside. Then I discovered my large-grit, 40-grit sanding disc did a nice job of quickly sanding it down and leaving a smooth surface. But angling it, I could reach most areas. I used a Dremel with a wire brush to clean off the cones without damaging them. I needed to add more foam, this time with regular formula spray foam, to fill the air pockets I uncovered, and to build up the outside edge. Two rounds later, I had a decent-looking Kamm cove foam form. The last step was to do the ends. I cut off enough foam to glue on a block of foam on each end. I then used a template to mark the outline as seen from the back, and then the outline as seen from the side. That quickly shaped it into completion.

Finally, it was time to surface it with Bondo. This was not a easy as I imagined. First lesson - don't use a metal putty knife. One, Bondo sticks to it like, well, like to metal. Two - the knife is not flexible enough for a complex curving surface such as the cove. Second lesson - Bondo dries HARD! It is very time-consuming to sand by hand and sanding block. I switched to a 40-grit sanding pad that was not as aggressive as what I used on the foam. This, too, could be angled to reach nearly all of the cove. But it had to be held securely to run it along a surface.

Meanwhile, while things were drying, I turned to the engine vents. My first idea was too aggressive for the foam and too hard to control. My second idea took a scuffing disc and shape it to the contour of the vent cross-section. Then, to push or pull it thru the foam, I mounted the hand drill on a cardboard cradle. I then pushed the scuffing disc along my marked lines. One thing I had to do was cut and pull out the blue glue pieces. While I was at it, I also cut along the marked lines hopefully to give a clean line when done. Removing the glue saved the contour shape of the scuffing disc from being worn down by the rubbery glue.

APRIL

While all the above was happening, on the side I prepared drawings showing contours thru the windscreen and side glass around the cabin. My thought is to closely follow the shape of the windscreen and extend it upward as a complete windshield. The side close I hope to do the same and make it so it can crank up and down. The difficulty is the changing rake of the windscreen as it flows from the front crease around to the doors and back to the rear deck. The angle changes relative to the horizontal plane from almost 30 degrees at the front windscreen crease to nearly 50 degrees at the rear deck for the side glass. Most of the change occurs where the windscreen bends around to the door front cut, and is nearly constant on either side of that bend.

Just this quarter I discovered a program to allow me to create cuts thru my surfaced 3D Monza SS file. It is a freeware viewer from Solidworks called eDrawings. I created several cuts of the cabin area from the front view and the side view and the top view. I could then test what I would get if I extend the glass higher to where a top would be. It looks good to do so, maintaining the same shape wraparound windscreen as the original car. To have the side windows roll up and down requires a constant (unchanging) curvature side to side and up to down to be able to keep fitting the opening slit in the door for the glass. If I can keep the same shape for the front windscreen, then I might be able to make it so I can swap out the taller windshield for the 5-inch high windscreen like on the original car. That way I can have it both ways. When shaping the windshield/side glass/top, I will duplicate the outlines on my chassis to confirm the final actual dimensions of this critical piece. And if successful, I will had a removeable top that will stow up front. Otherwise, I will have to go with the heavier fold-up convertible top with all the complication of stowage behind the seats and the extra weight.

Back to the Kamm tail. And to do that, it was back to the Kamm tail drawings as they located the tail light cones. Using my concluded Kamm width of 43.5 inches, I confirmed the location of the cones, and the height of the cones based on my 3D plots. With the cones in place, I needed to hard-surface them, since they wwere juat made of thick paper. I used my epoxy gel from the gas tanks, but found it would smooth out but not even out the lumpiness of the semi-viscous gel/paste. Rather than take the time to sand them smooth, and likely end up with something more oblong than round (I knew because I tried), I decided to cut them out and make new ones from sheet styrene, which I did. That took another week.

While the new cones dried, I cut out a profile of the cabin top in 1/2 inch foam and placed it using my laser-leveling technique after leveling the chassis. Then I climbed into the driver's seat and placed a level on my head and had my wife snap some photos. Turns out my head is JUST under the top line of the roof, effectively even with it. That was better than I feared, but still meant I needed to raise the roof, say, 3 inches for minimal operating clearance. But before I design that, back to the Kamm.

The new cones looked good and sturdy and fit into the original openings. Then came the setback.

MAY

I like to doublecheck my measurements, so I consulted the one photo I had that showed the Kamm cove functionally as an othrographic look-down view. I had earlier worked out the vanishing points, so I dropped the drawing onto my 3D Kamm drawing. And lo and behold, it did not confirm the location of the taillights. . . Well . . . it DID match them, but not the ones I was working with.

You see, I only plotted the driver half of the car. So I then flipped that and aligned the two halves where I deduced the centerline should be. But I also did plot all four of the taillights. And when I flipped the halves, the left and right set of taillights did not overlay each other, i.e., they were offset from each other. And the plotted photo taillights aligned not to the driver side lights, but to the passenger side lights, which wanted to be farther apart.

I refined the photo vanishing point, which changed little, but confirmed what the photo showed me. I tried to scale the width to the width of the license plate bezel, but kept getting a wider Kamm than I could justify with the published width of the car. What to do?!?!

Well, . . . in any case, it meant that the distance between the inner taillights was too narrow. They had to be wider apart. And that meant I had to make the Kamm wider. And THAT meant I had to CUT APART the Kamm cove and add a big spacer of about 1-1/2 inch!!
So, I did.

Time to take a short vacation. Which I really did.


Back again and back into the garage. I tested working with spray insulation foam and sort of shaping it by covering it with wax paper and placing a weight over it. Yah, it sort of worked, but the center area where I made an indentation stayed liquid until I removed the waxed paper. More testing needed.

The Kamm cove shaping continues by adding foam to the outside edges to bring them to the 43.5 inch width I settled on. I was also able to shape it into the cove toward the outer taillight opening.

While the foam and Bondo dries, I began mocking up the removable top I hope to use in lieu of a convertible top. I start with 1/2 inch foam and make a centerline profile. I cut out an opening shape of the cabin from thick corrugated box cardboard. The piece with the hole went on the chassis and is positioned to correspond to its placement on the car. I then positioned the centerline profile and carefully climbed into the driver's seat and checked my head height against the profile with a level. My head would rub the roof, so I add three inches to the profile and try again. Yup, it looks like for sure I will have between 2 to 3 inches of head clearance. Good enough.

I cut another foam profile to set on the foam body and begin cutting and foam-gluing upright panels to it to represent the windscreen and side windows. Next step is to determine how to mate the two, differently-angled surfaces (windscree to side glass).

JUNE

Let's begin the month reviewing how we got here. The first June pix shows the different attempts I have gone thru making the Kamm cove. There are four foam versions and three major cardboard patterns I made trying to get this to look close to the actual tail cove. The third pattern is shown in use in the 2nd pix. Aligning it is a study in best-fit, as the 3rd and 4th images show looking along the top and bottom edges.

I continue throughout the month repeatedly checking the cove outside rim and adding Bondo repeatedly to repeatedly build it up to eventually sand it to shape. Waxed paper works great to make a flat, smooth surface. Just pull it off when the Bondo is firm but before it hardens. During drying times, I continue adjusting the window panels to determine how to achieve the final shape. The side windows must go up/down about 10 inches, but not all the way down. I intend them to stop at 5 inches of height to match the real car's side glazing. It is a straight-line section up and down but curved slightly front to back. Note that they must go up and down and when seen from the side, they are cut to a rake of 45 degrees, which on the rear rake will extend into the B-pillar when raised. And by quite a bit. The problem is merging the 25-degrees-from-horizontal of the windshield to the 62-ish-degrees-from-horizontal of the side windows with their 45-degree angle. But after much measuring and plotting, I think it will work and the hard top will detach and fit into the front trunk.

Continuing with the Kamm cove. I realized I needed to focus on tightly defining the lip, or edge, of the cove. Little by little, the rear-view outline was established by slowly building up all arond - especially at the major bends. Then, I could focus on building up the edge to the correct depth (front-to-back, center to outside wings). It took multiple sessions adding and shaping Bondo applications. When THAT was established correctly, THEN I could shape the edge into the cove area. I did a rough sanding on the top edge transition, got tired of working on it, and moved back to working on the top.

JULY

Building up the top was a truly step-by-step, trial-and-error process. I had my side profile determined and had started with the windshield at an aggressive rake of about 27 degrees. That was greater than the +/-22 degree rake of the original 5-inch high windscreen. The side windows were positioned but needed to be shaped for curvature and especially to blend into the windshield. Then the roof was added and roughly shaped; then the B-pillars; and finally the rear window surface.

Along the way I had some fun with it when I place a remainder cut-out piece of foam at the back of the roof. It sure looked like something Syd Mead might do on one of his USSteel rendering, or that Ed Roth might toss in. It was very 1950s, rocket-to-the-future looking. But then I began to think now it might be very practical . . .
The two wing extensions on the side could be used contain Cibie headlights on the front, and extra safety taillights on the back ! ! ! Naaawwwhh - That would be TOO wild.

By the last week and after much shaping on the driver's side, the basic roof shape was good enough to test . . . Would it be tall enough on the chassis?!?! And the answer is ------

YES! With an inch or two to spare.

BUT . . . The windshield rake prevents the steering wheel from flipping up more than a notch past the vertical position.

I think I will have to go with a swing-away like on the '61-'63 T-birds. Or else a removeable wheel.

AUGUST

The Kamm cove just does not want to be finished. It was primed and I thought it was ready to add the taillight cones when I caused a big problem. I placed it in the sun for an hour to thoroughly dry the primer. When I came back, there were cracks in the surface and the foam had expanded between the ribs. I gouged out those all enough to add new Bondo and started sanding it all again. I had to build up the rim in places and correct the overall outline and rim heights. And still, imperfections kept appearing, requiring touchups. Very frustrating.

While Bondo was setting, I worked on the poptop, continually shaping it. The window surfaces were most important to establish, especially where the windshield meets the side window, and what slanted angle the side window would take. Since the side window will roll up/down, it will need to be the same angle front and rear. (quick aside - It does not HAVE to, but in order to avoid an open gap between windshield and window, it will have to go up as the same angle as the windshield A-pillar. And the same issue occurs at the B-pillar.) But how much of a rake could I give it and still have a nice profile on the B-pillar? Remember, the windshield at the A-pillar has quite a rake to it in side view. I got it close to 60 degrees from vertical. Then I reshaped the roof and did a rough match on the driver side before removing the top from the body.

While more Bondo sets, I decided to get the chassis going. The engine had not been run in two years. I first needed to replace the starter. That turned out to be a simple job. Next I suspected the fuel pump was not doing its job. I checked in with the local VW mechanics and learned I could test it simply by pushing open the flap valve on the MAF. It worked! And so did the engine. This takes me into...

September

... But that revealed another pesky problem - an oil leak around the #3 pushrod tube. White smoke galore as it dripped onto the exhaust pipe. So I pulled it and added RTV to the seals and tried again. Smoke galore again. So I pulled it and cleaned off the RTV and replaced it with the last of my seals and simply oiled them and reinstalled the tube. I also checked behind the rocker arm cover on the other side, and discovered the washer for a rocker arm lock-down bolt had broken in half and left the rocker arm out of adjustment. I fixed that, then started up again. Well, first of all it ran rough, and then there was a LOUD metallic POP, and it ran well! I decided the push rod was slightly misaligned and had corrected itself. It ran well, so let barking dogs bay!
More Bondo. Time to adjust the shift linkage to be able to get it into gear. By now, I am into...

October

... and I can only get 2nd and reverse to work - and reverse only works when I put it into the 4th position! Something is very wrong. The best guess is two-fold: the T4 shifter has limited reach compared to a T1 shifter (this IS after all a T1 transmission), and/or... After talking with the builder of the transmission, who has done hundreds (yes, hundreds) of these transmissions, there must be something wrong with the shifter/shifting geometry. How else to explain that it is engaging reverse in the opposite position than it should be?

That has to set a bit. The body takes precedence. With the cove nearing completion, I need to address reinforcing the body with an internal frame. With the help of wonderful neighbors, we lift the body off the base and place it on its side, exposing the underside. I begin to cut holes to run two lengths of 2x3s, placing them at the level of the waistline. Actually, it is two pairs of 2x3s that are screwed together along their overlapping lengths. These are then foamed into place, while at the same time I foam gaps between panels and where panels abut but need more reinforcement. The 2x3s protrude at the rear such that I can attach a 1/2" plywood backboard that I will bolt the Kamm cove to. I also add leg-like extensions to keep the body in place on the base. I then clean up the surrounding frame that holds the base, center it relative to the garage, center the base within the frame, and add 2x4s front to back to make the base rigidly placed in the frame. Then I call back my neighbors and in a quick minute, we place the body back on the base. The next day I recenter and level the body on the base, and I am ready to go again. And it IS much stiffer and not wiggly anymore.

Back to the cove. It is time to add the fillets to the taillight cones. I decide against automotive clay. I would have to heat it to about 170 degrees to slather it around the cones. The cones are made of styrene, so my fear is that the heat will melt or distort the cones. I choose instead to use the Aves modeling compound I experimented with on the one-fifth scale model. It's like a stiff clay, so not as easy to shape, and must be sanded to shape after setting, but safer for the taillights.

Check back for more progress!