1932 Willys Silver Streak

[Gojeep]

Setup the laser to get the exact distance from the rear axle centreline to the firewall on the Willys and the Chrysler 300 SRT8 donor. It was greater than I expected and was a 500mm-20" difference! This would mean that the Willys 113" wheelbase would grow to 133"! So something had to be done.

Notice the section cut out in front of the B pillar that is arrowed. That is 70mm-2.75" missing.

Then the whole B pillar was removed back to the original factory join at the seat riser.

Then all the leaning forward of the B pillar modifications was done on the other side with the same section cut free. I then removed all the calking from the end of the floor seam and marked forward from there 165mm-6.5" in different spots as the seam is not straight.

Then out came the cut off wheel and reciprocating saw to sever the shell in two!

Then I rolled them apart so the next stage could begin.

Before I cut it in two I made a support leg that bolted to the centre bearing mount inside the tunnel. It was at the exact height so alignment later would be easy and wouldn't fall after cutting the shell in two. The jack is at the balance point of the front section to make it easy to move the section around by myself.

The edge was de-burred and the assembly holes that had a dimpled edge were flattened out so they would sit flush on the seat riser flange.

Part of the excess was rough cut off, just in front of the seat riser flange, avoiding anything that had to remain attached to it.

It was then easier to remove the remainder of the floor from the flange after the spot welds were ground through with the die grinder and the seam splitter used. The tunnel was also drilled for plug welds in the original spot weld locations. The rest of the floor can be reached using my portable spot welder that I bought from a business that used to make limousines and funeral cars.

First test fit showed the height difference now between the top of the tunnel and the rear seat pan. Not unexpected given the slope in the transmission tunnel.

Wedges were cut out of the tunnel and the edge shrunk to bring down the height. The flange on the seat pan was also cut so it could be bent up to meet at the same angle.

4 reference points were measured at floor level as well as from the top of the A pillars across to the top of the C pillars, both sides, before the cutting took place. Then exactly 165mm-6.5" were deducted from these to make sure everything was true and square when it came back together. By having upper and lower points of reference, it makes sure the shell is not bending in the middle. Having the rear and front halves supported at four points also stopped this from happening. Even after welding everything back together, they were all to the millimetre. The B pillar was then welded back in 70mm-2.75" further forward and the part behind the B pillar was step flanged at the factory seam after 95mm-3.75" was removed.

Seam sealer/adhesive was added between the layers before welding them together. The factory assembly holes were also welded shut along the seam.

Etched primed to stop any rust while the build continues. Will be zinc coated and a seam sealer used later on.

This brings the distance from the rear seat raiser to the back of the B pillar the same as our daily driver, a 2010 Jeep Cherokee. We could test access on that and knew we didn't want anything less, but this will still have a deeper and wider seat base than it giving more room once seated. The front opening is still greater, but will be shortened some more once the new A pillar is made just behind the existing one. This will be vertical rather than angled as now and will lengthen the firewall to cowl distance to suit the Willys.

 

[Gojeep]

Video update of the donor body getting cut in half.

 

[Gojeep]

Well a lot of time working things out with little to show! Thought there maybe some interest in what has been going on anyway, so waited until I had a little bit show. No exciting developments though, so skip the post if that is what you are looking for.

Sometimes I just need to see some progress instead of just spending a lot of time working through different options for the build. So one thing I did was address the moving back of the transmission mounts. The stock position is pictured on the right spot welded to the bottom the floor pan transmission tunnel junction. On the left I have mocked up the other mount after moving it back 140mm-5.5". This is how much I determined earlier I could move the engine closer to the firewall.

I cut the vertical parts of the mounts off and bolted it to the cast alloy transmission mount. It is actually made for either the 4 or 5 speed transmission options. The models of the transmission is cast into the piece with an arrow pointing which way to the front. You just turned it around and used the other set of bolt holes between the two models.

With the vertical flanges cut off I was able to slide the mount back on the narrowing tunnel the distance needed. Also making sure that I had it positioned exactly the right distance left to right as well.

A bit hard to see in this view, but made up some folded angle pieces that were curved to suit the tunnel shape so it was still supported on the sides of the tunnel and the floor as before. You can see by how much the flanges are now further away from the dropped section.

Notice here how much further away the dropped section is now from the tunnel edge. The vertical flange used to go straight down the side of it.

I used all the drilled out spot weld holes to plug weld the mounts back in, plus added extra for the area towards the rear. This is now ready for when I later start building the engine mounts after the Willys frame is in place.

I made some templates, both positive and negative ones, of the vertical door profile to help make the door frames etc easier. Also to make sure the shape stayed consistent down the length of the car. So I cut them using the front door A pillar steel plate that the hinges screw too as a template. Then I discovered that the door plate to take the other side of the same hinge, was different! Why the factory stamped these out differently, even though they are right next to each other, is beyond me. You might be able to make out the pencil line showing that the steeper radius of the door compared to the A pillar.

Then I checked the side of the grille and it was shallower again, so I modified the template to follow this instead.

I like the shape of the grille so going to keep that profile from one end of the car to the other.

The other thing I like is that it tilts the top of the window frame in 0.5 degrees rather than out 0.5-1.0 degrees that is stock. I prefer the look when the windscreen/windshield is narrower at the top than the bottom. This makes the roof narrower than the beltline at the window sill level as well which looks better to me.

This will also allow me to get as close to the side of the dash as possible to help reduce the extra width across the front of the body shell. The gap left is the narrowed door frame width plus trim on the side of the dash and door card.

To better show the difference, I have altered one end of the door skin in the background to that of the new profile. The door closer in view is still stock.

The difference is 23mm-7/8". Not a lot but that is doubled when taken over both sides. Every little bit counts when trying to hide the extra width of the donor shell.

 

[Gojeep]

Continued:

Next thing I wanted to check was the sweep, radius or curvature of the body horizontally. To know what I have stock, I raised the doors up in the middle until a metal ruler would lay of the two doors without a gap in the middle underneath it. This is to make sure I don't end up with a 'double hump' when looking down the side of the car. I lot of cars have this straight from the factory in the early days until maybe the 70's? Now I could place my large 1220mm-4' profile gauge on them and transfer it to some left over ply from the truck bed floor.

I made both a negative and positive sweep template, refining the curve using an adjustable body file set to the matching profile. I put the front of the template as close as I could to where the outside of the door profile will sit from the body. It looked like it has too much curvature for the extended body length I have now. Also the amount you see sticking out from the body shell is how much it will be covering the running board. I want more of the running board to show.

The plumb bob, beautiful old one brass from my late father, shows the minimum I must clear to leave enough structure around the seat belt reel mount. Really need even more than this as want to keep the rear parcel shelf structure intact too. The end of the template also shows that the body tucks in way too early, taking out most of the rear of wheel arch.

So the decision was made to re-profile the whole body to a shallower radius! Going to mean a lot of extra work, but I'd rather do that so I can give this re-imagined body a fighting chance to look half decent in the end. So I had to come up with a way to draw a new giant radius from scratch. The adjustable profile gauge is no good at doing this at all if there is nothing long enough, with a consistent radius, to lay it on. I tried different types of string pulling from a pivot point, but even nylon was too elastic over this distance to be accurate. I ended up using plastic coated wire tried around the bottom of the pencil. The pencil line shows the new radius compared to the stock one.

The radius matches the upper part of the door and thought that was a good starting point.

It took most of my garage driveway to get a pivot point far enough away to use. The end of the wire loops around the gate drop bolt and the pencil at the other end. The wire was kept taught while keeping the pencil as vertical as possible while drawing the line from one side of the template to the other. The new radius was 7100mm-23'4"

Was curious to see what the stock radius was to see how much it has changed. Was surprised to see it was only 5200mm-17'. So I had a big difference of 1900mm or 6'4".

The arc was looking a lot better now but wanted to see if I could improve on it even more, or would it be not enough curvature? Only one way to find out.

I still had some 'wobbles' in the line which took a long time to even out after cutting, so needed a way to reduce that. So I clamped the pencil to a square with the lead only slightly deeper. It didn't have to be square to the surface to be accurate, just a consistent angle while I pulled on the wire tightly.

After a couple of new curves were drawn, cut out and tried, I ended up at 10 metres or 33 feet! I was up against the house on my carport driveway, as had run out of room on the other.

This give a good perspective of how long the new radius is. Coming up at nearly double that of the stock one.

I think this is the best radius as it doesn't cover as much of the running board and gets the stock wedge shape back again as it has moved the widest point of the body back towards where it is stock. This will make the transition to the grille better as well.

Next was to alter the negative template to match so I could see how the new arc continued into the rear of the wheel arch. It now leaves all the structure I was after, so pleased with the outcome.

So the bottom sweep template is the new radius with the stock one directly above it. Also the other radius tried as I got to the end result. It took a lot of time and then also perfecting the sweep so it was consistent from one end to the other. Now I finally know what shape the body will be made both in the vertical profile and horizontally.

I was wondering how this new arc compares to my 2010 Jeep Cherokee, which the Willys body will end up close to in size once done. Turns out over the distance of the doors, it still has a 20mm-3/4" gap, showing how much more it still has than this. It would take around another 2000mm-6.5' of radius to get near as shallow to match it, so confident the profile wont look too modern.

 

[Bubblehead617]

Dayuum.......

 

[TugboatBill]

Marcus,
I always knew you were at another level but I was wrong. You're another level beyond that.

 

[rocket]

Amazing work again as usual Marcus. I'm looking for inspiration on my Cummins Dependable Diesel Roadster, The Johnny Cash Special. My Wife calls it Chitty Chitty Bang Bang I built it around the Radiator shell.

The main body work is a 1962 Olds Cutlass that I narrowed and shortened. It still has the boxy rear end and Cutlass trunk that I've never quite liked. I've been considering cutting the back off the body and doing an Auburn style Boat-tail rear end. But I'm not sure exactly how to do it. Any input would be helpful from the Master.

 

[wfoguy]

Marcus, way more thinking than doing. My 1st thought was this would make the truck look simple. Maybe that was correct.

 

[Gojeep]

Amazing work again as usual Marcus. I'm looking for inspiration on my Cummins Dependable Diesel Roadster, The Johnny Cash Special. My Wife calls it Chitty Chitty Bang Bang I built it around the Radiator shell.
View attachment 183850
The main body work is a 1962 Olds Cutlass that I narrowed and shortened. It still has the boxy rear end and Cutlass trunk that I've never quite liked. I've been considering cutting the back off the body and doing an Auburn style Boat-tail rear end. But I'm not sure exactly how to do it. Any input would be helpful from the Master.
View attachment 183851
View attachment 183852

Being so wide, you would have to make sure you had enough length to create a boat tail without having too much rear overhang. Inspiration from pre-war coach builders is a good way to go and see how they went about it in design and shape. The Duesenberg's were all coach built and were a larger car. Also Studebaker, Lincoln, Cadillac etc of the late 20's early 30's all had large roadsters.
The top line really needs to roll down towards the bumper for this period and be rounded like the hood sides. This can be done having a large tube rolled and then quartered. Pretty sure 'Bad Chad' used this technique rather than the traditional way of shrinking the edge to roll it and making a seam through the middle of the curve.

 

[rocket]

Thanks for the input Marcus. When I built this thing I had 8x10 color glossy prints of late 20's to early 30's Roadster Royales posted all over the walls of my shop for inspiration.

 

[Gojeep]

Just a video update on Elvira's build.
https://youtu.be/BbhH03FpSLo

 

[Bubblehead617]

Jeez Marcus, ya make my head hurt just trying to keep up with all your thought processes in this!

 

[TugboatBill]

It's incredible to see all the thoughts that go into a project like this. Amazing.

 

[Gojeep]

Thanks for the comments guys.
There was a mention of showing what thought processes go into a build like this, so there you have it!
Less than half of what I had as was starting to sounds like I was babbling too much and I couldn't even watch it myself!

 

[commando_willys]

Just a video update on Elvira's build.
https://youtu.be/BbhH03FpSLo

Enjoyed watching the video of your radius adjustment thought process. Appreciate you sharing it.

 

[Gojeep]

I needed to check to see if I could run straight from the side of the grille to the A pillar, the line that the hood/bonnet sides will take. So I placed an axle stand where the edge of the grille would be once widened and then projected a laser line to the outside of where the A pillar will end up. That line runs down the outside of the tape measure and shows how much of the firewall is too wide!

If I cut the firewall straight up from that line I wouldn't even be able to fit the A/C unit back in on the passenger side, or mount the drive by wire throttle pedal on the drivers. That is when I realised that only the bottom part has to be that far in, from just above the chassis rail, and from there up it just has to follow the shape of the door skin. Here you can see how it curves up now, getting wider at the top. I can now just get in the A/C unit in on this side.

Here you can see both the modified passenger side along with the drivers side marked out ready to be cut the same.

Here is how I went about it. First the firewall was cut free from the cowl side and then the section removed from the firewall. Then a relief cut was made where the lower part of the body steps in as this will need to be able to compress once the bending takes place. A ratchet strap pulling over some angle iron, to keep it even, was used to pull the cowl across to the narrowed firewall and then a bolster was hit from the inside along where I wanted it to bend form. The gap left is due to the firewall being curved across the front.

The cowl side was extended to meet the firewall once more using matching steel from the door sill that had been removed earlier. The hole for the wiring harness was moved across and the drive by wire throttle pedal still fits onto the inside of the firewall on the stock mount.

I set the grille up off centre so the passenger side is where it will be once widened later on to get more of a visual. The black line on the floor shows the line that the hood/bonnet side will take to the A pillar. I may run a slight curve to it yet, but that will be decided later on depending if I need to clear the side of the HEMI.

Narrowing the firewall reminded me just how much the firewall is curved. I had made a note of it that it was 84mm/3.3" closer to the grille in the middle than the sides. But did I remember to take that into account when calculating the engine bay length needed for the the engine and radiator/fan etc to fit in? I'm already moving it as close to the firewall as it can go, 140mm-5.5" closer than factory, but suddenly thought I was measuring from the side of the cowl and not the middle of the firewall! Only one way to find out and that was to sit the radiator, condenser and fan combination in front of the engine the same distance from it as it was installed in the donor engine bay. I remember it already being very tight but did have it all written down. Above shows it with the only difference being that I have turned the radiator 90 degrees so it will fit inside the Willys grille. This is no problem for the radiator to run that way as this was quite normal until cars had lower frontal area and had to run cross flow radiators instead. However you cannot do this with the A/C condenser as the tubes must remain horizontal for the gas to return back into a liquid state efficiently. So a taller and narrower condenser will replace the stock one later on.

Turns out that I did not have enough engine bay length after all, and I knew I was already a bit short even measuring from the side of the firewall. So looks like an extra 100mm/4" will be needed to fit everything in, and that is with the condenser/radiator/fan unit inside the grille shell as much as possible. Just as well I have already shortened the body as much as practical, otherwise the car would be awfully long! Least it wont look snub nosed now with the longer than stock body behind it. More in keeping with the big luxury cars of the era now too.

Another thing I have also established is that the engineer is happy with my plans to stitch weld the Willys original chassis rails top and bottom directly to the sides of the body making it into a unibody instead of hard bolting it to the chassis as was stock. He also inspected and approved using the donor unibody frame over the rear independent suspension, using all its mounting locations, before the Willys rails are then used the rest of the way again. They will also run all the way to the very front of the car as well holding the engine and front suspension in place.

The connection between the two will use the factory donor stiffening brace that runs up inside the rear frame pressing. This was not connected to the outer sill before. I have trimmed it back the thickness of the boxing plate I'm going to add so it will be flush with the rest of the inner door sill.

For the first time new steel is being added to the body that was not from the left over donor shell! I have folded this plate up with slots cut into it that line up with the ends of the inner brace. The inside was etch primed and weld through zinc coating added as it wont be able to get to later on.

It was fully welded around the sides and through the slots onto the inner brace to make it one unit. The black line shows where the top of the chassis rails will be once fitted leaving enough room to weld the top of the chassis to the plate.

The top of the ply, that I am using for the body curve template, is at the height that the chassis rails from the Willys will sit at.

 

[wfoguy]

 

[Gojeep]

I have mocked up the B pillar into position and notice how far out the angle of the top of it is due to angling the lower pillar forward until it is vertical. Another thing I was working out was angle of the window sill line. The stock Willys only has 0.5° of slope but the donor Chrysler had 2°. I had to raise the front and lower the rear at the parcel shelf just to get down to that number due to it becoming steeper after shortening the body length.

If I went for that extra 1.5° over stock, would there still be enough height to fit the grille in without cutting it down? Turns out there will be so that is not a problem.

But what would it look like? Above is the stock mock up with approximately the extra length I have to put in the doors to fit the shortened donor shell.

This is with the angle added and it is around 2° total.

Originally the thought was to add to the bottom of the doors as well to help the window height to lower door ratio.

But then once you add the extra angle to that, it looks like it might too much?

With the increased window sill angle, maybe only 50mm/2" extra would be all that is needed? I don't want it looking too heavy either.

What is not in the other mock ups is the fact that the whole rear wheel arch has to be raised 50mm/2" as well, bringing that distance between the top of the arch and the beltline back to what it was before the wedge was added. Above that has been done along with the extra 50mm/2" door extension from the photo before.

Back to working on the upper B pillar. Look how much the donor Chrysler leaned inwards! You can see why I chose to cut the B pillar off where I did, just before the bend. I'm still hoping I can use the plastic interior trim throughout right up to the window sill line which is where the orange string is.

This is the inner half of the steel pillar that I had saved after cutting it off. I thought it would be a good challenge for some reason to see if I could straighten it by mainly stretching.

I was successful, but it took way too much time. Then had to add length back to it after cutting off the angled top part as well as add some width to match the doors between the windows.

Then I added the top mount seat belt adjustor after straightening that out to fit as well. Then stood back and thought, I have never ever adjusted that in my life as it is always at the top! The pillar also narrows to give more headroom for the inward tilt that I have just removed as well!

So I scrapped it all and made a new piece from scratch that doesn't taper and has just a fixed upper belt mount. Oh well, it was good practice to see what could be reshaped if needed. The overall height is right to the top of the roof skin so will be cut down exactly later on. Rather have more than needed to work with.

 

[wfoguy]

Decisions, decisions. What is different from most projects is that we maintain the original body measurements and make everything fit inside. You are modifying to fit everything. Do you think if you were a solidworks operator and had 3d scanned everything it would be much simpler to make all of the adjustments before cutting?

 

[Gojeep]

Decisions, decisions. What is different from most projects is that we maintain the original body measurements and make everything fit inside. You are modifying to fit everything. Do you think if you were a solidworks operator and had 3d scanned everything it would be much simpler to make all of the adjustments before cutting?

For sure. Guys on the metal shaping forum are scanning whole cars and cnc cutting bucks to form panels too etc. Way quicker. Don't have the funds to do any of that or the room for a buck even.