The Lambo certainly has more parts! It is pretty amazing, the design work that went into it. Even has functioning paddle shifters that work the sequential shifting tranny.
I 3D printed some grommets for the hoses that pass through the front firewall for the A/C and heat. I used TPU which is a flexible material and it worked out fairly well.
When I had them installed I realized that my design of the grommets was more complex than it had to be, but that's what engineers do; we design the simplicity out of things. Two heater hoses, 2 A/C hoses, and an evaporator drain line.
And I used this on all the A/C O-rings. Read about it somewhere and it is interesting stuff. Like a sticky, liquidy rubber. Supposed to work very well at preventing leaks at connections.
I had the rear firewall panels off to run the shifter cables and now it is all buttoned up again. I added some sound-deadener to the floor panels; thicker foam will go over that. The wiring is somewhat tidied up now, up to the dash area anyways.
So it is time to work on the center console. I have some ideas on what I'm going to do there to have access to the fuse box and other relays that will be under the console panels. More on that when I see how it works out.
The Marr connector that you see is from the mod described by Mechie3 for getting rid of the starter-cut relay. I later soldered them properly and saved the connector for my basement wiring.
Now with all the wires traced and removed (and some of them are fairly thick so it is a lot of bulk removed), there is only one relay being used. So out comes the handy saw again to cut the box down, and I’m left with this:
Rick
Rick, do you mind showing me where the starter-cut relay mod is you mention from Mechie? I was looking through his wire dieting thread and I did not see it nor his PDF doc. I could have overlooked it, but it would be super helpful. Thanks!
You are right, he does not mention about getting rid of that relay. I've been hunting around trying to find what I actually referenced and can't find it. I'll keep looking and see what I can find.
You are right, he does not mention about getting rid of that relay. I've been hunting around trying to find what I actually referenced and can't find it. I'll keep looking and see what I can find.
Rick
That would be awesome thanks! I have been using your dieting write up which has been helpful! I want to delete the secondary air pump because my engine doesn’t have it, but the harness does. I have removed the A/C already and just went ahead and cut out F127 the secondary air fuse/relay little box. After doing some more digging I see the relays are tied to the ECM, injectors, cam sensors, and wastegate solenoid. So I spliced the relays back into my harness. How did you remove this from your harness? How do you remove it but still keep the critical engine functions like the injectors? The ECM pin in question is C30. E87A5D4E-FF4D-40B4-8431-B983FEE9549D.jpeg
The wiring diagram you are showing is a little different than the one I used. Pretty sure mine is for the 2006. Not sure if that matters or not. I removed all of those relays and wiring. From what I can see, C30 from the ECM powers the relays to open or close them. Does it care if it is triggering something that is not there? You should be able to pull all of those relays out (all six of them in that relay box) and still have the engine start and run. The rad fan won't work, but that won't matter for this test. The connection you are describing for the other functions is the one that also takes power from the C30 trigger, so trace it and leave it connected to the junction where it goes to the injectors and other systems. It looks like that is "E/G(TB)-11" in my diagram....correct?
I am referencing the 06-07 Subaru FSM so I believe they are the same in principal. I know you removed your rad fans, but I am going to leave them so I will have to see how to keep them intact. I will also look at the C30 line some more to see if I can find the junction to remove the relays, but keep the E/G(TB)-11 functions. Thanks!
First thing, I did some additional machining to clean up the shifter assembly I made. Also made a shift knob and then powder coated all the various parts. I’m really happy with the way it came out. The switch at the rear is for the electronic parking brake.
What to do with the center console panels? Since there is a lot of stuff in there that I may need to get at (shift linkages, wiring, fuse box, relays) I wanted it to be easily accessible. So, the stock aluminum panelling was out. Besides, with my shifter I would need to modify the stock panels anyways. I didn’t want small removable panels since that would most likely mean seams, joints, exposed fasteners, and a lot of work. The easiest thing would be to have it all removable. How about removable in a minute or two? How about removable without any tools? How about no exposed fasteners?
A quick mock-up on the side panels with cardboard:
Something like this, obviously with some sort of panels on top. The forward one would run up under the dash and have a place for the A/C and heat controls. I could make these panels from thin wood, covered with foam and vinyl.
While looking at the cardboard pieces it came to me how to meet all the goals for the console panel attachment.
First, the cardboard pieces transferred to wood, then covered with thin foam:
I didn’t take a picture of them covered with the vinyl at that stage. But I did figure I needed to cut the side panels into 2 parts, since with the seats in, it would be very difficult to get them in and out if it was one long panel from the front to the rear. One panel will end at the front of the shifter and the other one will go from there to the front firewall.
So how to hold them in place without any exposed fasteners and no plastic door panel clips (you know, the kind that break off most of the time or else tear out of the wood panel)?
I 3D printed these plastic retainers. They are 6” long each:
Now, how to hold them at the top? Magnets!! I’ve used the small rare earth magnets a number of times recently in wood working and metal work. I was originally going to use them to hold the side panels in place at the top and bottom, but then thought that the plastic retainers would be a better way. Then I was going to use them just for the top of the side panel, but thought of another way.
I made a top piece from slightly thicker wood. The sides of it go down over the side panels to hold them in place.
This is the bottom of it with the magnets in place. The magnets are in small cups that apparently increase the magnetic force of the magnet. The cups are epoxied into flat-bottom bores in the wood. You can see how the sides go down to overlap with the top of the side panels to hold them in place.
The slots in it are for small storage pockets I’m 3D printing. There isn’t a lot of room for them since the shifter cables run through the area in the middle, but it will be useful for a phone or other small things.
When the printer finishes, I’ll put it all in place with the side panels and take a few pictures. I tried the top panel to see how the magnets hold and no issue there……..they really hold!
I’m really pleased with how this is coming out. I can pop the top panel off which then releases the side ones and then slide them up out of the retainers. Even with the seats in place it shouldn’t be an issue to remove them. No tools, no exposed fasteners.
Now I’m working my way forward with the panels. I’ll use the same concept there.
Very nice. My console was made with similar goals. Three screws hold it in (the storage part from back up to the front of the shifter). And it was to be able to handle pushing off of it with one hand to heft my bulk out of the car, without flexing, etc. Nice work you got going on there. Looks cool too.
Besides my 818 project I have another roadster resto-rod in process, It is a 1929 neo-classic built in 1968. The body has been heavily altered and the original interior is redone. The interior trim panels were vinyl / foam over luan wood. With a soft top and side curtains, the wood is at risk of water damage. In one of my day-jobs I GC'd the reconstruction of a facility for food product manufacturing. A requirement was washable walls, with "dairy board". That is white thin polyester panels 2-3mm thick. Some are smooth on both sides, some a smooth side and beaded side. I bought a 4 X 8 sheet at Home Depot, cheaper than plywood. I cut my panels to fit and took them to my trim shop for vinyl covers and beads. Stitching can be predone, but they could not "sew" through the board. Should be durable in bad weather, thinner than luan, and more finished edges. Actually glue and double faced tape also stick better than to wood. One could laminate dairy board for thicker parts,
When I was considering what to use for the interior panels, I did some google searches. Found a lot of the usual; some people that had been in the industry for decades that used nothing but luan wood without a problem, and others that said to never use wood. For what I was using it for, I thought plastic might be a good choice. Not that it should ever get wet, and traditional British cars had wood, and even cardboard, interior panels. So why did I end up using wood? The pandemic.
Stores are shutdown here in Ontario. They are open for "curb side pickup" only. So you can order on-line and then go and pick it up. If the stores were open I would have gone and looked at the various plastic panels at Home Depot. But I didn't want to order plastic panels without seeing them first to be sure I would be happy working with them. Since I've used wood before for interior panels, that's what I went with. But the next time I'm at Home Depot (when they are open) I'll take a look at the plastic panels they have.
I made the front panels similar to the rear ones; thin wood, covered with foam and vinyl. I glued a piece of wood on the back that sits on top of the ¾” frame rail. Then I printed some plastic brackets that screwed on, and they trap the panel on the frame rail.
The bottom of the panel is retained the same as the rear ones; it slips into the plastic retaining pieces on the floor. The plastic clips on the panel slip over the square tube so the panel can’t move in or out.
At the rear, the top piece is retained by magnets, and that holds the side panels in place. For the front, there is no structure along the top edge of the panels for magnets to hold on. So, what to do.
My wife told me there has to be a cup holder. She also said there has to be a glove compartment, storage for extra keys, phone, sunglasses, etc. I said no. I said if someone wants all that they can buy an SUV. She knows I don’t keep cars long and is thinking what a potential buyer might want. I compromised with the cup holder since that worked out as a good retainer for the panels.
I 3D printed a cup holder, with places in the bottom for magnets.
I bolted a small strip of metal across the console tubes, and the magnets on the bottom of the cup holder grab onto that, which holds the top piece in place.
I wanted the 4-way flasher switch mounted somewhere, but it is a pretty big switch and sticks out too far for my liking. I printed a bezel that the switch goes into and it sits recessed into where ever it is mounted. I mounted it in the side of the front console piece. If you look carefully at the previous picture, you’ll see that hole in the diver’s side.
Needs a little finishing up here and there, but I’m pretty happy with it. The cup holder panel overlaps the sides, and the cup holder goes in and magnets down to hold it and the panels in place. The other piece ahead of that also overlaps the top of the side panels, and it is held down with a large screw-knob inside it. That is accessed by reaching in around the front of that piece. It only has to be loosened and moved up and then the side panels can be lifted and removed.
One other small thing. There is obviously a gap between the front and rear side panels, so I again 3D printed a small trim strip for that, that slides in from the top.
I made a lot of use of the 3D printer for this. It is really handy for making small retaining brackets, holders, clips, and pieces like that. I don’t use it for anything that needs a lot of strength (though the plastics can be quite strong).
So maybe this will help someone with ideas on what to do with the interior. At least it is clean, easy work; no rust, grease, oil...
I can start to tidy up the rest of the interior now; insulation, carpet, seats.
818S/C : Chassis #25 with 06 WRX 2.5 turbo, ABS, cruise, PS, A/C, Apple CarPlay, rear camera, power windows & locks, leather & other complexities. Sold 10/19 with 5,800 miles.
Mk3 Roadster #6228 4.6L, T45, IRS, PS, PB, ABS, Cruise, Koni's, 17" Halibrands, red w/ silver - 9K miles then sold @ Barrett-Jackson Jan 2011 (got back cash spent).
When I made my exhaust, I knew it came close to a few areas that I wanted protected from the heat. It wasn't too close, and there would be air circulation in that area, but the plan was always to wrap it to keep the heat in.
When I took the original cat off the exhaust, and pulled the shield off it, I saw that it had some sort of insulation on it. I temporarily wired that piece onto my exhaust near the CV boot. Please ignore the crappy welding in all the pics.
After running the engine for a while I put my hand on that piece of metal and it was just warm.......very good insulation!
So I thought I would use that along with the exhaust wrap. I cut it into smaller pieces and wired it in place. If you are doing this, have good ventilation and/or wear a mask; the fibres in that insulation are nasty.
I haven't run it yet, but have read that it smells and maybe smokes for the first few heat cycles, so I'll back the car onto the driveway to run it. I overlapped it more than the instructions say, which is why I ran out and had to buy more; hence the different colors. The "titanium" one was supposed to be wrapped dry, which I did. The black one was supposed to be wrapped wet (well, not supposed to be, but recommended to make it easier to conform to the curves). I ended up doing that one mostly dry as well since they said to wet it but not to soak it, and it seemed difficult to do one without the other.
Running the engine will also let me see how the cooling system is doing and if there is air in it. I know there is, but will it come out with some heat/cool cycles is the question. Once I'm sure there are no leaks, I can put the body panels back on and get on with the build.
I wanted to get the roof put on to see how it fit and what additional work it would need. Seemed like a good idea to take the front fenders off first, and do the aluminum panels in that area. As was suggested in another thread, I drilled them all now and used clecos to hold them in place. The 2 panels with the arrows were riveted in since they would not need any work once the fenders are back on.
The panels with clecos will need trimming to fit along the fender later, so they were removed after all the holes for rivets were drilled. I'll need a bunch of small patch panels here and there to seal things, but that will come when the fender is back on.
The roof was a decent fit, but of course needs some work and adjustments.
The rear corners need to be pulled down onto the rear side panels.
The proper way to do this is to flip the roof over and bond some studs on the underside of the roof panel just in from the corner. A small piece of metal can then bridge the side piece with the rear piece and have a hole that the bonded stud goes through. Put a nut on the stud and tighten it to pull the roof panel down.
I got a bit lazy once the roof was on (not realizing that is was going to come back off many times) and took a different approach. I basically drilled counter-sunk holes and ran small bolts with counter-sunk heads to pull the panel down. I needed a few across the back and some down the passenger side as that was sticking up too. Doesn't look like it in this picture, but the roof panel is more or less flush with the surrounding ones now.
I'll fill the small holes where the screws are and basically bond the screws in at the same time; grind off a little of the gel coat and use a thin piece of fibreglass cloth with some epoxy. I won't do that until I'm sure the roof is on for the final time so there shouldn't be much stress or movement around the bolt holes. Laziness now, more work later........oh well. But there will be other areas needing fibreglass/epoxy work as well.
The main area of concern is the A pillar and how it meets up with the front of the door. First thing was, I took off the thick piece of steel that is the mirror mount. No idea how you are supposed to get a seal between the door and the A pillar with this piece there. I'll figure another way to mount the mirror. So with that piece gone, I could lower the roof a bit more to minimize the gap between it and the door. BUT...
You still have the screw heads that hold the door panel in place. I machined the heads of some screws to make them a low profile.
I was OK with that until I started trying to minimize the gap even further. I decided to counter sink the holes in the door panel and machine some screws so they would sit flush with the panel.
I'll machine off a little more and then fill over the heads......or maybe not. When the door is closed you don't see these screws at all, so maybe I don't have to be too fussy with finishing them. I'm planning on having the car wrapped which is a little more forgiving with respect to surface defects than painting; not much more, but a little.
So now that I have nothing protruding above the top of the door skin, I can lower the roof down to get a nice minimal gap. Or so I thought. I had to grind out a fair bit of metal from this area of the A pillar to allow the roof to sit down lower without hitting the front edge of the door panel. And that was with the door panel skin trimmed a bit at the front. Funny thing is, it still does not allow the front piece of the roof to touch the frame in front of the windshield. Apparently you are supposed to rivet that part of the roof to the frame. Even with my roof being lower than a "stock" one, those areas are not in contact. Don't think it really matters though.
I think the only way to get a decent gap here is to build up the bottom of the A pillar to bring it down closer to the door. As it is now, the gap isn't any where near even front-to-back on either door.
And another issue with that area and thinking of how it is supposed to seal with the door. The top of the door panels slope up as you go from the outside of the door to the inside. With the door closed there is about a 3/8" gap:
So how do you seal that area? Anything that lightly seals the top of the door to the bottom of the A pillar with the door closed will be heavily rubbing as the door opens.
I have a way that I think will work and is very simple. Too simple to work? I'll do some more fooling around with that area and see.
I decided to take a break from that area and put in the trunk kit I have. Well, well, you can't get that in with the roof in place. One more reason to take the roof back off.
Not obvious in that picture, but the gap mostly evens out front-to-back and outer door area to inner. The mirror mount goes in front of that and is roughly even with the top of that inner door panel piece. A rubber seal on the bottom of the A pillar might seal it.
My wife had a look at what I was up to and when I showed her she said it looked very strange with the inner door panel piece extending to the outside of the car where it will be exposed to the elements. Not sure how many elements it will be exposed to, but she has a point. Also, I'm not sure how this will work with the door windows. Looks to me like the window will be just outside this piece, but may cut into it. Also...
These are supposed to line up? You can see that the inner panel can't be pushed back any further. The inner panel fits inside the door skin all around expect for the top, where it transitions from inside to outside (very top of the rear of the door in the picture above). This needs more thought.
Woha that's quite a big gap you got there! I used screws on some areas where I had gaps but none were as big as yours, I don't think just using screws would make it conform "ok".
Wha'bout using some rubber seals all around the inner panel between it and the door, would that work? It would also seal up the door on the side pod which prevents quite a lot of air (and dirt) flying in from that area.
AZPete has great pictures of the D-shape rubber he used. I think it was D-shaped.
Frank
818 chassis #181 powered by a '93 VW VR6 Turbo GT3582R Go-karted Aug 5, 2016 - Then May 19+21, 2017
Tracked May 27/July 26, 2017 Build time before being driveable on Sep 27, 2019: over 6000h Build Completed Winter 2021
Woha that's quite a big gap you got there! I used screws on some areas where I had gaps but none were as big as yours, I don't think just using screws would make it conform "ok".
In the manual they mention using a heat gun to shape that inner door panel at the top. I could see about heating it and bending it out to match the door skin. I'm not too concerned, if I have to use some foam strips I will, just seems strange to be that far off.
I thought being 3 pieces I could put the trunk in after the roof was on....nope.
Not a huge deal to take the roof off to do it. Pretty sure FFR made this for the topless 818 and didn't update it for the 818C. Or maybe I just have an old trunk kit. It works fine with the 818C, just needs a little extra work.
The pieces fit fine but there is no support other than along the back.
I was looking at what to weld or bolt to give it support and then came up with what I think is pretty easy. There are mounting pads that aren't used (for the trunk pins or something on the non C version). I bought a piece of aluminum angle, and found a some large square aluminum tube. I bolted the tube pieces to the trunk pin mounting pads, riveted the angle piece to that, and that lines up nicely with the front of the trunk piece and raises it up just the right amount.
I may put a couple of screws to hold the front in place. The side pieces don't hit the body exactly, and there are those cutouts at the front corners of the trunk. I'll seal those up when I put the roof and hatch back on and see what else needs to be sealed.
With the hatch piece in place, you can see how it sits with the trunk.
I'll make a dividing wall to seal the front of the trunk up to the hatch glass; only has to be about 3" or so. And to hide the trunk contents from the outside of the car, I could black out the glass for the rear few inches. Overall, I'm happy with how it came out.
Now I'm working on the power windows. More of a challenge there than the trunk, but I have something in mind that may work. Stay tuned.
OK, on to the power windows. I'll go into my usual amount of detail here, so grab a coffee.
There have been many complaints about these and a lot of threads on how to modify them to get them to work better....not well, but better. So what exactly is wrong with the design?
The most basic thing is that when the window is up, there is only about 3" of glass left in the door. That is not enough to give good support to the window. BUT, if there were good tracks inside the door at the front and rear of the glass, that would help a lot. And why there isn't, and can't be, leads to other issues with the design.
This next part can get a little confusing, but may help someone who, like myself, is thinking how the window system can be improved.
The rear of the glass is just about straight, from an up-and-down viewpoint when looking at the glass from flat-on. So that will move up and down along the channel that is moulded into the roof. So far, so good. But the front of the window, roughly straight as well, is not parallel to the rear. So you can't have a track at the front since it wouldn't be parallel to the rear track and that simply can't work. I made a template that simulates the glass, and another one that simulates the roof channels.
A bit hard to see, but the dashed line is about 1/4" away from the "glass" and simulates the roof channel. When the glass moves down, it would slide along the rear of the glass/channel, and here's what happens at the front:
The front of the glass was lined up with the solid line. With the glass moved down only about 3", it has moved rearward and no longer has any support at the front. Again, this is with the rear of it moving down along the channel in the roof. So you can have a guide channel inside the door at the rear, but one at the front would not do any good. The glass will move out of the channel as soon as the window starts to lower. If the front of the glass was parallel to the rear, then this would work fine. But it isn't and can't be modified to make it like that. At least I'm pretty sure you can't cut tempered glass.
Assuming there is a way around that, there is another issue. The rear of the glass is curved in-and-out, when looking at the glass flat-on. This matches the curve in the roof channel. So when the window moves up and down, it goes through an arc. Nothing wrong with this, pretty much every car does it like this. Except the very old ones with flat glass, and giving it some thought, Factory Five should have just used flat glass and it all would have worked out better....but I digress.
The problem is, the front of the glass does not have a curve, it is basically flat. So think about the rear going up and down and moving in an arc (in and out from the side of the car) and the front is an 8" long section with no curve. So you can't have a track in the front since a long flat section of flat glass is trying to move in a curved track. Now some people have added a length of track to the front inside the door, and this works. It works since the non-parallel part of the glass at the front moves out of the track, but the nose of the glass right at the bottom will stay in the track. So there is support with a track like that, but only at the very bottom of the glass.
Now if you have followed all this....good. So we can add a track at the rear (inside the door) and another track at the front (that will only contain the bottom corner of the glass) and that will give decent support for the window. But it will not give much support when the window is up, and that is when it is really needed.
I played around with my templates for a long time trying to come up with a way to support the glass when it is all the way up and still have it move properly. There is only one way I could see to do that. Add a complete channel around the window, attached to the door frame and opening with the door. This used to be a common way to do it before frame-less side windows became more popular. The good news is that there is a fairly large channel moulded into the roof that can accommodate a feature like this. The bad news...well....it is a lot of finicky work. At this point I think it is going to work, but there is still a compromise in the design due to the way the glass is made (the flat at the front not matching the curve at the rear).
I couldn't find any steel channel with the correct dimensions I needed. I did find hollow rectangular tubing that would work after splitting it in half.
This is 1 1/2" x 3/4" material, which gives a 3/4"x3/4" channel which is just about right. It has a wall thickness of just over .080" which was a little thicker than I wanted. But I soon realized that bending this to the shape I needed was not easy and it tended to go a bit pretzel-like, so having some extra wall thickness to smooth out later is a good thing.
I bent the sections and tack-welded them together. I learned later that this was not the way to do it. Because the part on the right has to be bent in-and-out (from looking at it flat-on) the tack welds soon came apart. My tack welds tend to be as strong as a final weld, or come apart as I am brushing off the weld. My welds tend to be functional and sometimes almost pretty. But I'm good with a grinder, so all is well at the end of the day.
Anyways, I started bending it using a small hydraulic press with a block in it to keep the channel from closing up.
I did this in small sections at a time to form a gentle bend. It worked fairly well for the top section. But the right section needed to be bent in the other direction on the channel, and this didn't go well. So I got out the oxy-acytelene torch to heat it and bend it. There are good and bad things about that.
To open up the channel it works well. Heat the sides red hot and open the channel. To close the channel didn't work so well. The sides tend to move out, then have to be heated and pulled in. To bend it "sideways" really doesn't work well at at; it bends, but in various directions.
Now I'll make it clear here that I am no expert at metal work; mill and lathe work, OK, but bending channel....not so much. I did get most of the frame done, but it was bent, straightened, re-bent, moved, pushed, pulled, until it was a pretty poor result. But with enough effort I was almost pleased with it. The car shows that modify and make body parts from scratch are not calling me yet.
In case I wasn't clear, this will be attached to the door frame with some adjustability. Hopefully the window will slide up and down in this track and be fully supported the whole way.
There is one thing I did learn about bending the channel. It is much easier to cut it almost all the way through in several places, close or open the bend, and then weld up the cuts. Yes, there is a lot of welding and clean up needed, but the bends end up in pretty much the direction you want without also going in other directions. Of course there is welding distortion, so this is very much an iterative process.
There are still a few issues that I'm working out. Like I said, the front of the glass is flat, unlike the rear that moves in an arc. And there is the issue of the front not being parallel to the rear so not having much area for support (other than the very bottom curved part of the glass). I'm working on that area now and think it will be OK.
This has been a huge amount of work. Not there yet, but I'm still hopeful.
Rick
Don't know why I sometimes get an extra attached picture at the bottom.
Very nice.
Time and effort well-spent, and FFR should take note (copy) your work.
Some time in the not-too-distant future, you will forget the pain, but you will have a better car "forever".
Here is a similar but not identical mod on a GTM, by far the best solution and perhaps will give you some additional references. Too bad I can’t employ it with a Targa top.
Having to take the door off to install or remove the outer door skin is a pain. The screws holding it on go up from the bottom and down from the top, so they are not accessible with the door installed. I don't like that.
With the outer door skin removable you can set the door latch and work on the power windows, get everything set up, then put the door panel on.
One way to do this would be to bond some brackets onto the door skin and use those to bolt to the frame. I almost did that, then came up with what I think is an easier way.
I modified some T-nuts (1/4" thread) by grinding off the prongs and shortening them a little.
Then enlarged the holes in the bottom of the door and pressed them in. I used a little instant glue on the other side to hold them in place, but not sure that was needed.
I enlarged the holes in the frame that the 1/4" bolts normally go through, enough for the body of the T-nuts to fit in. I cut some bolts short enough to engage the thread but not stick through the T-nut. There is enough clearance under the door to keep the panel contacting the rocker panel area to allow the shank of the T-nuts to go under the door frame, then the bolts pull it up into position. I put a rubber washer and a steel one so the protrusion of the T-nut shank above the door frame plate would not cause the bolts to bottom out without clamping the door skin.
I was going to do the same thing at the top, if I decide to not use the provided mirror mount. Then I had a better idea. I threaded the holes in the mirror mount that normally bolt to the door frame, through the skin. The holes are almost the correct size for a 5/16" coarse thread; just a quick pass with the correct size drill first.
Enlarging the holes in the frame and the door skin, and the bolts go from inside the door, up through the frame, the skin, and then thread into the mirror mount. Again, cut them to the correct length so they don't quite stick through the mirror mount.
If you want to get fancy you can use some sort of covers for the bolt holes in the mount. Even if you are not going to use the stock mirror mount, you can cut it down and use it for this scheme, as it helps fill the gap between the door and the A-pillar.
Only the bottom corner of the glass is in the channel and if the glass does not move precisely, it comes out of that part of the track. One issue with it slipping out here is that the glass can then move as it goes up and not go into the track along the top. After a lot of puzzling on what to do, I found a simple solution. Instead of cutting the rectangular tubing down the middle to make the channels, I only cut one side of it off to leave a deep channel. I cut slots in it to bend it to match the travel of the glass, then welded the slots.
Since this is in the door, it doesn't show. This traps more of the front of the glass. Also, the track is "thicker" than the thickness of the glass by a fair bit, so the flat glass moving in a curved track should work OK.
Another issue I found is that the track at the rear puts the glass right in line with the door release tab in the latch mechanism. This probably depends on where the door frame ends up; it can vary on how close it is in-and-out depending how the latch is set up. I bent the tab at the bottom (where the loose bolt goes through to activate the door release) but that didn't solve the problem. I had to get more serious to move it out of the way.
So with that done, does the glass go up and down? I didn't want to motor it up and down since the glass is riding in the metal channel and I was afraid of breaking it. But I tried it in different positions to see how things looked.
But the good thing was that the glass stayed fully in the forward track and gave good support. The bottom of the glass is fully in the channel and the top of the front portion is still partially in.
I'll clean up the channel and paint it. I have to add the channel liner that the glass slides on, then I'll try it with the the glass in and the regulator hooked up. If it works, one more door to go.
While I'm far away from getting to the door window installation, I totally love what you did and will use it for inspiration when the time comes. Given all the challenges you had bending u-channel, and getting it to fit, what are your thoughts on possibly 3D printing a channel? I haven't done 3D printing yet, and would imagine I'd have to connect multiple sections, but the curved channel may be much easier to fit. Thoughts?
While I'm far away from getting to the door window installation, I totally love what you did and will use it for inspiration when the time comes. Given all the challenges you had bending u-channel, and getting it to fit, what are your thoughts on possibly 3D printing a channel? I haven't done 3D printing yet, and would imagine I'd have to connect multiple sections, but the curved channel may be much easier to fit. Thoughts?
Interesting, but I don't think it would be good. Assuming you are talking of 3D printing with plastic. I don't think the 3D printed channel would be strong enough, but I generally print with PLA, PETG, sometimes ABS. I know there are stronger materials now, but they generally take a higher temperature and I can't use them. So maybe with a channel designed to maximize strength it might work. Would be a challenge to 3D model the channel, but maybe a fun challenge.
I would be a little concerned with the strength when the window is down and someone uses the channel to close the door. The steel one I have is plenty strong for that. The good thing about 3D printing is that it is pretty cheap to try things. Model up a section of the channel, print it, and see how strong it seems.
I'm hoping what I learned with bending the channel for the first window will make the second one easier. Heating and bending is not the way to go. Cut slots, bend to open/close the slots, and weld. Maybe with a proper pattern of the window (with the correct curves in it) a fabrication place could make a nice channel with proper bending equipment? This is obviously a huge experiment for me, and there are likely better/easier ways to do it. A home-made channel roller?
I've had it working with power, but not happy with it yet. The tracks need some tweaking. This is all the way down (and maybe even a little more than I'll set it since the bottom of the window is just about hitting the door skin.
I put some paint on the frame but did a sloppy job with it. Need to smooth it out and do a better job.
The main thing I'm not happy with is how it mates up with the rubber window strip along the door. With the window down, there is no gap. But with the window up, there is a large gap around the middle of the window. I cut the curve in the door skin to clear the window in all positions. But I think I need to try and move the top of the window inwards when it is down so I can move the rubber strip closer to the glass and not have a gap when it is up. For those that have the windows done, do you have a fairly consistent space between the door cutout and the window in all positions, so the rubber is rubbing evenly all along?
I'm working on the second window now and I'll see if I can get it to fit better with the rubber strip. If so, I'll come back and tweak this one.
Man, that shifter and bell crank....noice.
I took a detour on my 818 that started with a 3d printer and has resulted in a cnc router of my own just because I want to make some of my own parts. I'm not too proud to say that I have trouble conceptualizing parts to do complicated things like this, but with all those fantastic pictures you took, I have a fighting chance. I think that makes your work DOUBLE impressive. I'm also really encouraged that you managed to shorten your cables. Well done all around!
Thank you! I started with a cnc router as well for wood projects. Then started doing aluminum on it. Works well as long as you keep the cutter lubricated so the aluminum doesn't stick to it. If it does, instant ruin of the cutter and the project. I bought a small cnc mill with spray coolant (the "mist" coolant system could be used on a cnc router just as well) and an enclosure. Pretty much all the parts I make are what is referred to as 2.5D, as they are basically flat plates with various cutouts and external shapes. And it is sometimes a challenge to get the design from what you picture in your head to something that you can actually make. But fun all around!
I had the drivers side window working fairly well; not perfect, but not bad. So I decided to move on to the passenger side as that should be easy with everything I've leaned. Huh!
I had the basic frame done in a few hours and spent the next 6 days cutting/welding/bending/correcting...not easy at all. But I did learn more about how all this works. The main thing was that with the drivers side window I had the glass supported at the front and rear by tracks in the door, and the center of the glass by the power window motor assembly. That is too many supports fighting with each other. And as I said previously, because of the shape of the glass, the front support can't really be close fitting to the glass. So after days of trying to get the passenger side to work, I realized what the issue was. I cut out most of the front channel that is in the door to leave the glass supported by the rear channel and the center regulator (the motor assembly). It isn't perfect, but it works pretty well now. And the glass is nicely supported when up, down, and anywhere else in the travel.
Some tips for those that might want to try this. First thing....don't! Sort of kidding, but it is really frustrating when you realize that the shape of the glass is not designed to let the window go up and down properly. May sound strange but you'll find that out when you put some time into the windows, whether modifying them or not.
From the video, you'll see some of these details. The front of the glass comes out of the track when the window moves down. This works well since there are still 2 points of support. I also made the bottom of that track widen out so the glass finds its way back into it when on the way up. At the rear you'll see the top of the glass comes out of that track when the window moves down. I don't think the drivers side was doing that, but there is definitely a difference between the 2 sides. On the passenger side the window can go all the way down, so obviously the track is a different shape. The window moving out of the track at the top may be fixable with adjusting the regulator angle and glass support, but it isn't too much of a concern since it does work well, and it is still well supported by the glass below that point.
One other thing you'll see. The area of the door skin where the rubber strip contacts the glass looks all messed up. That's because it is. I cut that area out a bit too much and had to put some of the fibreglass back in. I'll smooth it down later. The tracks are lined with a felt-like material that is meant for this purpose. I bought a wide roll of it and glued it in place.
I haven't put the seal around the roof/door opening yet. When it is in place it will seal against the new window channel. And I bought stuff made for a 1/4" thick edge.
So now I'll go back to the drivers side and see if I can apply what I learned on the passenger side and get it working a bit better.
I've probably forgotten some details, so feel free to ask if you think you want to try this modification.
Looking good!
IMO, add body work to the front edge of the window track to blend in the recess of the A-pillar to the door skin (I did on mine), it will look so much better.
818SC chassis #206 EJ207 2.0L VF37 twin scroll || Cusco type RS 1.5 LSD || Wilwood pedal box (firewall attach) || Wilwood superlite front calipers BUILD Phase 1: 6/6/2014 car delivered || 5/24/2015 first start || 6/7/2015 go karted || 4/20/2016 hard-top-topped || 10/25/2016 registered || 11/18/2016 inspected & complete BUILD Phase 2: 3/8/2017 EJ207v8 || 5/29/2017 re-first re-start || 7/17/2017 re-assembled with race car bits
I've been watching this intently trying to determine what I want to do so that I can get a decent window solution for my targa top. Since I can't put tracks all the way up the side, I have to come up with something else. You have me leaning towards throwing away the curved glass and getting something that is flat since I'm making 3 of the 4 mating surfaces from scratch.
Looking good!
IMO, add body work to the front edge of the window track to blend in the recess of the A-pillar to the door skin (I did on mine), it will look so much better.
I saw what you did with yours. Very nice! When watching my window go up and down I did see that the channel I made is fairly well contained within the moulding along the roof, but that area doesn't look right (the triangle at the bottom of the A-pillar is recessed in a bit much).
I was going to mount the mirror somewhere else and reduce the gap between the A-pillar and the top of the door, but then decided not to. I'll probably make a new mirror mount piece that moves the mirror rearward a little; with that triangle area and the window channel the mirror might be blocked a little from the driver. I was thinking of making a piece to build up the triangle, but not have it attached to the door like you did. Maybe a 3D printed or CNC'd aluminum piece with a logo or something on it? Maybe the new mirror mount piece will extend the full length of that space and help seal it.
I took a look at what my drivers side window needs to get it working better. Then went back in the house.
Thanks: 
Likes: 
Reply With Quote
Originally Posted by FFRWRX
