Monday, April 5, 2021

Building the firewall

Oka, finally time to do the last assembly on the firewall. Here's a hole, prepared for the installation of a bucked rivet. That a 100 degree countersink. Aircraft guys have extra special tools.
Here's what the hole looks like with a rivet dropped into it, and another rivet beside it to show you what it looks like before it gets bucked.
And here are the tools - a pneumatic rivet gun and a block of steel to act as the "buck" - the anvil you hold on the back side. You can see the end of the rivet sticking up into the countersunk hole; the countersink actually acts as the form for this end of the rivet. You just hold the buck against the other side, and use the pneumatic hammer to smash this end until it's flush.
Like so. Yeah, it's not perfect, and there are some apprentice marks on it, but this is the back side. I did it that way on purpose.
This is why I am making the whole thing removable - so I can service the fuel tank down the road, so to speak.

Here's the firewall pulled out again for today's work - making those little notches at the corners, to clear some weld bead on the frame, and then to mark and drill for the row of screws which will go along the top edge to secure it to the frame.

Wednesday, March 17, 2021

Big post today!

Today, I continued working on the mounting "toes" for the firewall. First, I used my 3D printer to make some spacer blocks. This let me provide a consistent offset between the face of the fuel tank and the firewall.
Here they are on the floor of the car, just aft of where the driver's seat will go.
And here they are looking down between the firewall and the tank. This let me get the angle brackets I had made all drilled and riveted up, so the lower portion of the firewall is all ready to bolt into the car.
To get the upper part ready, however, I will need to change tactics. Here are two rivets - on the right is the "pop" type that I have been using for most of the panel attachments throughout the car. The disadvantage of this type of rivet is that it is not flush on either side of the joint; on one end, it leaves a small domed surface, and on the other end, it leaves a protrusion which is even larger.

On the left is a bucked style of rivet, which is set with a rivet squeezer; it actually deforms into countersunk a countersunk hole on the back of the work to provide a flush finish on both sides.

These are aircraft rivets, used in applications where drag reduction is paramount, so no protrusions can be allowed.

Here's my first experiment is using these bucked rivets. This is the back side of a joint I made in a couple of pieces of scrap. I took two pieces of 1/8" aluminum, drilled holes and made countersinks on both sides, and used a rivet gun to deform the "shop end" of the rivet to fill the countersink.

Safe to say these pieces are securely fastened. I did a few more just to do it, and they all worked very well. I'm convinced.

Last but not least, here's the upper section of the firewall. It's all Clecoed up to a piece of angle, ready for the rivets to go in. I haven't shown it, but the holes are already countersunk on both sides, and I just have to buck the rivets.

That's for next weekend, though.

Tuesday, March 16, 2021

An overdue update

I haven't been good about updating in the last couple of weeks. I don't feel like I've had a lot of time to work on the care lately, given other obligations - but here's what I have been up to.

Here's the fuel tank in the car. It sits back flush to the vertical place formed by some frame members. It's a bit hard to see from this picture, but there are a couple of things that protrude through  that plane, and therefore through the firewall skin I am putting there - the filler neck and the vent fitting.

So, with the firewall skin mounted in the car, I marked and cut the holes for those items. The vent hole will probably need to be enlarged later, but the filler hole is just about perfect. 

Tubes from both of these will go up and outboard (left in this picture) to the fuel filler on the fender.

Here's a detail shot of that same location from inside the car.
Because I am not using the supplied fuel tank from the kit, I also have been working on building a firewall to replace the kit firewall between the fuel tank and the passenger area.

Here, I have three panels mocked up to form the lower and middle sections. There's also going to be a vertical upper section lying tight against the upper aft firewall skin (this gets important later), and a small lower center section to cover most of the gap (but leaving a space for shift cables and wiring to pass through).

And here, I am working on the "toes" of the firewall, which will be bolted to the floor of the car. The tank itself also has mounting toes as well as some tabs at the top which will be bolted to / through the frame, so the tank is really going nowhere.

Tuesday, March 2, 2021

Firewall and fuel tank progress

Using one of those "stretched" angles I made last time, I started building a firewall to go forward of the fuel tank. Here, I have the angle attached with Clecos to a piece of sheet aluminum. There's a gap in the hole pattern because I need to cut out a window in the middle. This will form a passthrough for shift cables, wiring, and so on.
Here it is in the chassis, with the window cut out. This is going to be partly covered by a panel across the bottom, attached with screws so I can remove the firewall later to service the tank, should I need to.
I also did a little work on the fuel level sender that came with the tank. The sender is a 0-90 ohm unit - but the Subaru fuel gauge expects a 90-0 ohm input. So to avoid a reverse reading in the dash, I had to reverse the throw on the arm. And, because the sender unit is at one end of the tank, I then had to reverse the position of the actual potentiometer on the mount, so the float would still be to the right.
This is the reason I could not just rotate the sender 180 degrees - the mounting flange has an odd number of screws holding it down. I did try it at a 144 degree (two holes) rotation, but the arm fouled against the wall of the tank.

Next up is to get the tank actually bolted to the chassis. It does have mounting ears (you can see one of them in the previous picture), so I will drill through those and get some bolts into the frame and the floor. I am waiting on some rubber tape, though. Should show up today.

Tuesday, February 23, 2021

Don't underestimate 3D printed tools

Previously, in the build...

I'm using a third-party fuel tank in the car. This means I can't use the firewall that shipped with the kit, so I have to make my own.

I've elected to make it by riveting together flat panels with aluminum angles. So this is an angle here - nice and square. But I don't want it square, I want to open it up a bit, to about 112 degrees, to match the sloping front face of the fuel tank.

The first step is to anneal the aluminum. This is 6063-T52, which is heat treated to make it strong. But I want it soft.

Neat trick I learned - regular old Sharpie makes a pretty good heat indicator for this - just heat your aluminum until the squiggle disappears, and then let it air cool.
Then I just squeezed it in the vise between these 3D printed shaping dies. These aren't anything special, just PLA plastic. I got the idea by watching the YouTube channel "Stuff Made Here" - one of the things he does is to use 3D printed (much nicer and stronger than these) dies to stamp sheet metal parts. I figured it would be worth a try.

I am simply amazed at how well it did work, though. I put some grease on the dies, and just clamped it in the vise, loosened it, moved it along an inch, and repeat for the length of the angle.
Came out spot on, over the whole length. Way plenty good enough for my purposes.

The dies do show some deformation and wear after the one angle, so I am currently printing a new set with a slightly modified design. I figured the first set would be completely destroyed, though, since I only used 30% infill. The new set will be 100% infill, which more than triples the print time.

It's amazing when a half-assed plan comes together like this.

Tuesday, February 16, 2021

Worked on brakes this weekend

This was a weekend for brakes. Here in the front, I installed a third-party bracket that allows me to put these big Brembo calipers on the Subaru knuckles.
I'm using the same (massive) discs in the rear brakes, and the emergency/parking brake calipers don't quite open up far enough to squeeze them.
So I bought some .080" aluminum stock and made a couple of shims to space them out a little.
Success! Here is a rear knuckle with everything bolted up.

I also did a bunch of thinking and planning about the firewall at the rear of the cabin. I believe I will not be using the firewall sections provided with the kit, but I have not yet decided 100% what to replace them with. Options include .063" aluminum, .125" aluminum, mild steel, and carbon steel; and welded or riveted construction.

Tuesday, February 9, 2021

Installing the rear knuckles

So after I got some brackets and links set up the other day, today is the day I actually got them on the car. Here's the rear knuckle assembly on the driver's side, bolted up to the trailing arm and the lateral arms.

Note the spiffy adjustable forward lateral links - those should help dial in the rear alignment.

Here's both sides at once! Excitingly, the next section of the build manual is called "Install Driveline." I'm not sure if I'm actually going to drop the engine and transmission in next, but it's very, very tempting.
Here's a look at just one knuckle, with most of the connections faked up. There's also a coilover that goes in here, but I think it's going to be a while before that goes in. I certainly have to get the engine, transmission, and drive axles in before that.
On a side note, I finally took the time to organize all of the hardware that came with the kit. Not every cell in these totes is full, but a lot of them are. This should make finding particular fasteners easier, compared to my old method of rummaging through a cardboard box.