Thursday, December 31, 2015

Front floor support removal and installation

After reviewing all the damage on the front end of the car and building a suitable jig to support it, it's time to start actually cutting metal off the car. This is collision damage I'm repairing here, so it's more complex than just stabilizing the car and swapping rusty old parts for new metal, which is already a pretty involved procedure. 

All the frame pieces have to be replaced in a particular order, and my frame jig was built with this adjustability in mind. I need to be able to remove a part, and then place the new part correctly so it can be properly placed to support the new frame rail. Given my particular issues, the order for removing and replacing my frame parts is (1) Front right floor support, (2) Right side shock tower/frame rail assembly, (3) Radiator support. 

In this post, I'm replacing the floor support, shown below. The rear of the floor support is still in the right place but the front is bent where it meets the frame rail that was shoved rearward. The new part has to be put back in the correct place which may not be the same place it was when it was removed. Without correcting for the displacement, I'd just be 'building in' the misalignment I was trying to correct.


Where the floor support lives on a 67 Mustang. No Torque Box in the way in 1967 for Fastbacks.

Shiny new floor support, primed in SPI Black epoxy.

While researching how to do this, I noticed that most folks who are replacing a floor support were also replacing the actual floor pan as well, seen here and here as examples (thanks to Alex and Mike). Typically the floor pan would get cut out first, and the exposed and easily accessible floor support is removed and replaced while it just hangs there in free space. I'm in the unusual spot of not having a rotted-out floor pan in my car, so I have no need to cut out and replace the floor. I like that the floor is original, plus I've already primed it. Seems like a shame to just cut it out and throw it away. That's fine, except I couldn't find any examples of anyone replacing the floor support with the floor in the way.

Not one to be deterred by a bad idea, I decided to try to replace the floor support while the floor is still in place. I think it worked out OK. I used sheet metal screws to hold these pieces in place while lining all the new metal up. Once I was happy with the placement and the measurements, I burned it in (shown in a later post).

A new floor support is about $30. I burned up a $5 spot weld removal bit as well. The overall removal and replacement took about 12 hours, including priming. But the research and planning took much longer. Hopefully someone can benefit from my work and save some time.

Critical floor support dimensions on the 67 Mustang frame drawing in the factory service manual. 

Blue shows known good measurements/dimensions on my car, Green shows which dimensions I'm using to line up the new part. Red is the bent part of the car, so the dimensions are not useable from there. Dashed lines are my derived measurements, solid lines are factory dimensions, highlighted in the blue boxes. Since the jig is level and the car is leveled to the jig, I simply have to get the height of the new floor support to match the other side. Remember, I need at least two measurements in each of the three axis of motion to make sure the new part is lined up correctly.


On convertibles in '67 and all later cars, there's a torque box in this spot to tie the floor support to the rocker panel. With no torque box in the way on my car (yet), I can use this measurement when replacing the new floor support. I also have the new torque box primed and on the shelf for installation later, so I also used it as part of the fit-up for the new floor support.

There are 52 spot welds holding the floor support to the car. I know this because I used the factory weld and sealant manual to figure out the location and quantity of spot welds that were holding all this together. If you're doing any type of metal replacement on a car like this, I cannot overstate how handy this is. I'm a big fan of RTFM [Reading The Factory Manual ;) ].  Most of the major parts houses carry these. I have the whole set. In this manual, Ford is telling you exactly how they put this car together. In almost every case, if this manual says there are 10 spot welds holding piece A to piece B, I found 10 spot welds, and placed more or less as per the manual. $20 to make my life easier? Take my money!

(Check out this nifty history nugget. Granted these are Fairlanes, but same methods at work - the first 2 minutes show humans spot welding car bodies together back in the day.)



Spot welds that hold the floor support to the frame rail are visible here on the outside of the support.



More on the bottom. On the left of this image you can see all the wrinkling of the floor support from a 'minor fender' bender'. 



There's also spot welds holding the floor support to the floor on all three sides.



And a couple holding the transmission tunnel support brace to the floor support, marked in yellow. These suck because they complicate the assembly sequence and there's very little clearance to get in between the jig and the floor support.



Can't tell by looking, but there's 10 more spot welds hiding on the inboard edge of the floor support as well.




Easier to see are the 16 spot welds holding the floor support to the floor.



Preparing to remove the spot welds with the Communist Freight $5 spot weld cutters.



For removing the spot welds on the bottom, I had a little clearance issue. I built the jig to match the factory datum line, which means there's only about 7 inches between the bottom of the floor support and the jig - too narrow for a standard drill with a spot welder removal bit. So, I picked up a cordless Milwaukee right angle drill like this one and it worked beautifully for this job. 



Cutting welds out with wild abandon. Since I'm going to cut the frame rail out altogether later on, I didn't think I'd have to worry about stopping the cut on the first layer of metal. Then I realized it was a valuable  practicing opportunity, so I starting cutting the rest out the right way by cutting one layer only.



Wish I'd thought of that before I started though...



Since I'm keeping the floor pan, I intentionally drilled all the way through the floor and floor support in order to create holes in the floor for my new support to weld to. When I install the new part later on, I should see new metal though all these holes. Then I'll weld the floor to the support from this side, which is way easier than if I'd tried to weld it from underneath.



Here's where the decision to keep the floor really sucks. The transmission tunnel brace is also welded to the floor support (per the weld manual). There are two welds on the bottom and four on the INSIDE of the floor support! So, to get to the welds so I could cut them out and preserve the transmission tunnel brace, I cut an access door on the side of the old floor support and then drilled out the welds. When installing the new part, I'll have to weld these to the new part by coming from above and welding the transmission tunnel brace tabs to the floor support. Again, that factory weld manual is pretty handy.



The last 2 welds are inside the floor support. Here's the outline for the access hole in the floor so I can cut the transmission tunnel brace welds from the floor support.



Zing! A new hold in the floor. You can't see the ground because what you're looking at is the inside of the floor support and the transmission tunnel brace. This hole actually has to be larger to get the welding gun in there later on.



Once all the welds are located and cut, the support came out pretty easily. I guess I could have just done the floor pan as well and saved some trouble, but part of the reason for doing just the floor support and keeping the floor pan is that I couldn't find anyone else who'd done it this way. Come to think of it, there may be a reason for this...







The new floor support all pre-drilled for plug welds to simulate spot welds (per the factory weld manual). Primer is stripped away from both sides where welding will be done so the welds won't get contaminated. These will be structural welds, so they need to be clean, hot an well-penetrated.  



Comparing old support to new, a few things jump out. The old part is 16 gauge metal, whereas the new one is 14 gauge - much thicker. This will make welding a little trickier. Also, new piece is a little longer, so the hole in the bottom is in a different place relative to the rear of the part. Just keep this in mind when locating the new part relative to all those dimensions at the top of the post.



And since the frame rail here is bent due to the collision, I had to cut it completely out of the way so I could location the floor support properly. When it comes time to install the new shock tower/frame rail assembly, it should just slide right into the new floor support and be properly located.



Cut out the old part of the frame rail where it was welded to the floor support.



Blamo! After much measuring and checking, I installed the new floor support. It's held in with 'many' sheet metal screws to the floor plus the floor jack on the frame jig is applying a little upward tension.



Inside view, looking aft.The vertical gap between the transmission tunnel brace and the new floor support is too big here, but will get closed up better later on.


The view from inside. The sheet metal screws are installed facing up so now I have a bunch of sharp screws facing up hoping for some human flesh. Also notice the access hole I cut earlier has been opened up. 

And the floor pan is still in place, which was the point of this exercise.

No welding yet, that comes later. But the first replacement piece of structural steel in in place.



Wednesday, December 23, 2015

End of year review and current status


Merry Christmas! As 2015 closes out, I'm looking over all the progress so far, and I'm stuck by two things: I've gotten a lot of good work done, and I have so much more to do. I keep reminding myself this is a long term project ;)

Let's get the basics out of the way.

Yes, I'm still working on it.
No, it's not done yet.
No, it's not down on its own wheels yet.

I haven't posted much the past month as we're currently in the midst of a kitchen remodel that will likely drag until Spring, and this takes up some of my garage time on the weekends. Between this and the Christmas season, it's hard to get time to post much lately. And, if I'm perfectly honest, it's cold in the garage, and I'm basically a wuss when cold weather shows up. (Seriously, it's not even that cold - like 50 degrees.)

But, fear not, for I have time during the last week in December, so I'll get some posts together and published as January plows on. I have some cool stuff to show on the front frame repairs, which has lots of welding, cutting and general chaos.

Astute viewers will note that the blog isn't' really in 'real-time' yet - Posts are still catching up to the current State of the Project. But I'm getting close,which means I'll be posting events as they happen. As such, you might be surprised to learn what actually happened in 2015. Even more astute viewers (which I assume is anyone actually reading the text) will note that any posts previous to what's in here is the 2012-2014 story.

Let's review what got done this year, what we've learned, and what's coming up next.

I discovered how many layers of paint are on my car (2).

I finally realized that media blasting in my driveway is just not worth it.

I tackled my first exposed sheet metal patch on the drivers side fender. Not Terrible, but for 50+ hours, it better be Not Terrible.

Finally took on the notorious Fender Apron Overlapped patches.

Demonstrated once again that Rust Never Sleeps. 

Enjoyed finally getting an item completely checked off the 'to-do' list.

I removed my first set of coil springs using the right tools and didn't almost die in the process.

Removed the front brakes and learned that 4-piston calipers from 1967 are still pretty awesome pieces of hardware, and that 30 years between brake fluid flushes will give you something that looks and moves like maple syrup. Pro tip: Change your brake fluid at least once per decade. Maybe more.

We removed the front suspension (almost all original) and decided to keep as much as possible for the rebuild.

While removing the steering system, I figured out how much fun a pitman arm puller is for breaking those old joints. Turn, turn, turn, BANG! Don't worry, the hammer is for the mice in the garage, not the car.

Steering system out (less steering box). Finally decided to keep and refurbish the power steering setup and use the manual adapter. 



I rediscovered the joys of taking things apart. The metal work is rewarding in a different way, but few things compare to a disassembly montage.

I learned that roll pins can be your Worst Nightmare if you don't locate and remove them properly. Look closely for the little jerk poking up in between the clamp ends. Still haven't resolved this yet.

The rest of the floor needs a good sanding and shot with a couple coats of epoxy primer. It looks OK from here, but actually the floor has more rust spots than a red-headed lifeguard has freckles.

I finally designed and built a frame jig for the Major Metal Work. 

And got the car up on it for the actual Major Metal Work.

So that's where  we got to on the blog this year. My plans for the next year, like most of us, are Grand in scale and cost. In particular, I'm hoping to get some major decisions worked out for the car:
  1. Brakes - I'm staying with the power-boosted 4-piston factory caliper set-up; for the calipers, I need to figure out if I'm rebuilding, outsourcing a rebuild, or just buying new parts.
  2. Steering - I've settled on the factory-like power steering setup but using a manual adapter - more effort but more road-feel and fewer (no) hydraulic issues.
  3. Suspension - the standard coil spring/strut rod setup will stay, with roller bearings where it makes sense and quality shocks. The stance will be level and the springs will be GT+ rates.
Once all of this is sorted and installed, I have a rusty quarter panel/wheel housing to deal with next. This part of the project promises to be one of the hardest. I have yet to decide on how to actually do the repair.

With any luck, once the metal work and paint stripping is done, we'll be ready to get the car out for bodywork and paint. 

Thanks for following along. Here's some teaser shots of upcoming posts (use your inner movie-trailer voiceover for the full effect):



"In a world...where a Floor support is removed and replaced - while the floorpan is STILL INSTALLED..."

"Where an ordinary dude gets in way over his head..."

"Where metal yields to Massive Hydraulic Forces..."

I teach myself to do plug welds upside down.

And discover the secret method of removing a radiator support in 27 easy steps.


Keep the faith, friends. It can be done.



Monday, November 16, 2015

How to design and build a frame jig in only 6 months!

This one is long - but it's a big step, and one of the reasons for doing this blog is to show others doing this work how I did it.

Remember The Damage? It's time to start having hard conversations with ourselves. Specifically, how is someone (me) who has only done brake jobs, oil changes, and water pump replacements going to remove and replace an entire structural chunk of a classic American car without making a total shambles of it all? We're talking about pulling parts off the car that were really never supposed to come off - ever - and put them back exactly where they are supposed to be.

I need to put a new one of these on the car:

Dyncorn Shock tower/Frame Rail assembly from CJ Pony parts. Pretty!

No worries, I have an actual Ford Shop manual for my car. There's whole paragraphs on how the power steering control valve works. There's gotta be pages on this. I'll just open it up the manual and see what it can tell me...



And, no kidding, this is all that's in the manual. One drawing. This, on its own, is not very helpful to me. Right off the bat I'm sure I don't know enough about this.

So after a LOT of research, reading, pondering, and sketching ideas on napkins, I finally feel comfortable enough to give this a go. All told, I thought about this part of the project for about a year as I was doing other stuff. Again, I'm slow, I know.

Here's the deal: in order to do this sort of work, one needs to build or acquire a body jig to hold and stabilize the car, then place the car on said jig so that the only parts that move are the parts that are being removed and replaced. The big questions here are:

  1. what kind of jig? (body cart, fixed-base, or - my favorite - custom hack job)
  2. where to attach the jig to the car? (hint: not to parts that are going to be replaced)
  3. what should it be made of? (hint: not bricks or jack stands)
  4. what sort of music should one listen to while building a jig? (this one's a no-brainer: Will Smiths 1997 hit "Gettin' Jiggy With It" - lyrics [by Nas, who knew?] are interjected at appropriate points in the text.)
What is a body jig? Well, in short, it's a frame that supports the car while it's being worked on. It can be made of wood or metal, but it has to be sturdy enough to not deflect under the load of the car while parts are coming off and going back on. Basically, you're putting the car in traction. Note this is not the same as a rotisserie, which is great for general restoration work, but I wouldn't want to use one for structural work.

["Got to get jiggy with it..."]

Most of jigs I found online were for cars getting big pieces replaced due to rust damage. If you think about it, that's way better than collision damage - you just build the jig around the rusty, but un-bent car. Of course, it helps to have a frame shop confirm the frame is straight and true before you start, but in essence, that's it. But for collision damage like I have, I need to build the jig to support the car where the frame is undamaged or where it's the frame is supposed to be, This is harder for many reasons, one if which is that not-overly-helpful frame measurement drawing up above. I wish it had more, you know, measurements.

How many frame measurements do I really need?  I gave this a lot of thought. My first guess was "a lot more than are on that drawing". But that's not very specific or helpful. It turns out, that's not even the right question to ask. What's really needed here is a way to ensure the new frame rail/shock tower assembly goes into place on the car correctly. The right way to look at this is to ask "Which dimensions do I have to control in order to make sure the new assembly is located correctly?" As it turns out, the correct answer is "at least 6".

All these measurements really boil down to making sure the part in the right place relative to the rest of the car. So really, I just need to control:
  • fore-aft position, 
  • the left-right position, 
  • the height off the datum plane
and the rotation around each of these three axis:
    • roll, 
    • pitch, and
    • yaw 



    If I can ensure I have enough measurement points to control these 6 axis, I can be sure I've put the new part in the right place. Each axis requires a minimum of 2 measurements, and some will do double-duty for another axis. So now I know what I'm hunting for: measurements that help me control these 6 axis. I spent a lot of time online trying to find other measurements to use in addition to the 9 or so the shop manual had for the front of the car. As it turns out, some of the numbers from the '69-'70 Mustang manual are also valid on the '67-68's, so between that and some other numbers I found from experts online, I was able to cobble together a list of about 15 measurements I felt good enough to work with. Drop me a line if you want the specifics. By request, I've included the frame measurements I used at the bottom of this post.

    [..."I got the feeeeever for the flavor of a crowd-pleeeser...]

    I decided to build a jig that will hold most of the car steady using hard mount points on the car and include some adjustability with strategically-placed scissor jacks to allow me to dial in the locations of the new parts. This is done so I can compensate for the old, bent parts coming off the car, and put new, unbent parts on the car in the correct place, and still be able to crawl all over the car and not worry about knocking it over (bad) or having it fall on me (worse).

    Enough with the babbling. Let's build a jig. 

    The shop manual's frame drawing has a side view and a top view: 



    The side view has a line at the bottom called a 'baseline' - also called it a 'datum' line - and it's used as a reference. Essentially it's the zero line for height. The measurements on the side view are measured in inches up from that datum line. It could be anywhere actually, so you could move the datum line down a foot and just add 12 inches to each vertical measurement. I'm not going to complicate things (a first, I know), so my jig will use the datum as shown, which means I'll mount the car exactly as shown in the drawing - for example, the forward hole of the leaf springs at 7.98" over the datum line and the front bumper mounting holes are 12.94" over the datum line. 

    I chose to build mine out of 2"x2"x 0.125" square tube stock. It's easy to go all crazy here, some guys are using I-beams or 4-inch bars, some guys are using wood. That's cool, but I can't weld wood (yet), and if I can't weld my jig, well, what's the point? The math suggests the 2x2 square tubing is plenty strong enough for a nearly naked '67 Fastback, I just need to manage flexing of the jig over long runs. And in true beginner fashion, I'll weld the snot out of it, in keeping with the "overdoing it" methodology I've used thus far.

    ["...I makes it hot, gettin' jiggy with them...]

    I made a ladder frame out of two 10-foot sections crossed by three 42" sections. I'm using leveling feet at the corners for leveling and regular bolts at the other intersections of the cross-pieces for extra flex prevention. The pieces are all clamped and leveled in the driveway as it's built so I don't accidentally weld in a flexed piece by mistake.

    ["...You tryin' to flex on me? Don't be silly, gettin' jiggy with it...]



    Precision milk crates ensure the ladder frame is perpendicular in all the right places.

     Leveling foot screws into grade 8 nut welded over a hole in the rail. This provides the fine adjustment for leveling the jig on the floor.



    The datum plane is really just the imaginary horizontal plane formed by the tops of these long rails on the jig - that's my zero-level for all vertical measurements.



    I built the jig under the car to my specs and then lowered the car onto it. I used a laser level to make sure I get the jig's datum plane level.

    I borrowed Ted's Super Cool Laser Level. Includes plate for mounting on tripods or sharks.



    I used sections of the bar stock as a poor man's retro-reflector (yeah, I went there) for leveling the jig. I marked where the laser should be on each bar, placed one in each corner, and then set the laser off to the side of the car to hit all four corners. 

    "No, Mr. Bond, I expect you to be level within 1/16 of an inch over 10 feet!"

    Once the ladder frame was under the car, I built cross-pieces with vertical posts on them to attach to the car's hardpoints. 

    Crosspiece with vertical supports for the leaf spring mounting holes. 


    Crosspiece with vertical supports at Lower Control Arm (LCA) mounts.This one was assembled in place and then welded. I could get a grade 8 bolt through the driver's side LCA hole, but not the passenger side due to the bent frame rail.

    Passenger side LCA Mount. These holes in the jig should be exactly centered in the LCA Mount. When the new piece goes on, they will be.


    Scissor jacks (Communist Freight) and U-bolts. Holes drilled in base plate for U-bolts. This just happens to fit perfectly over the 2 inch bar stock I'm using.


    Then the scissor jacks are mounted to the long rails and will be used to hold either the floor support (shown here) and a 2"x1" crossbar that the front frame rails will rest on. This it the adjustable part - since my frame is currently bent down and back, I'll use the jacks hold the old part for removal, and the raise the jacks to fine-tune to the height of the new assembly before screwing and welding it into place. The U-bolts can be loosened and the jack slid in and out of place as needed. Yeah, pretty frakin' awesome, right?


    The front crossmember is welded directly to the jig to make it more stable. That whole front piece is also going to be replaced well, so I'm fine to weld directly to it.

    Here's the jig supporting the front of the car. Three scissor jacks, two welded posts, and five bolted posts (there's one more at the idler arm mount). I left jack stands just under the rocker panels as 'just-in-case' backups if there was a problem, but they aren't actually touching the rocker. I'm also going to use the export brace and a Monte Carlo bar up top for locating these parts.



    This pic is taken once the jig finally came out from under the car much later. In fact, this was the first time I'd seen it on its own since I built it under the car. I'm just showing it to give an idea of the overall design. The vertical posts are for the forward leaf spring mounts and the LCA mounts. The castors aren't on when the car is being worked on - instead, the leveling feet and bolts are used to support the whole thing,

    Blamo. The Jig is up. The Car is mounted. The Parts are in. So let's get on with it.

    ["...Gettin' jiggy with it...<fade out>]


    Details on Frame measurements used for the jig:
    As mentioned above, I had to hunt and scrounge the web for measurements I could call 'real'. In the end I used the 67-68 frame drawing in the back of the manual (shown here), as well as some measurements from the 69-70 frame drawing. As it happens the front end between these two generations is pretty much the same, so a dimension on the 69-70 drawing will work on a 67 chassis. But I had to do a lot of research to figure out which would work.

    Additionally, and you have to be careful here, some measurements come from real restorers posting measurements online taken off good cars. This is tricky because you have to know the source is really good enough to use - this is hard on the interweb.

    So, here's what I used:
    From the 67-68 Shop Manual Frame Drawing :

    1. Red boxes indicate actual frame dimensions used (6).
    2. Blue line indicates measurement between frame rails just above the forward edge of the floor supports (27.62", ±0.125", from here). The relative location is the blue dot on the side view.
    3. Yellow line indicates the measurement between the rear aprons at the top where the hood hinges bolt on (40.0", ±0.25, from same link as #2 - good thread, and Rusty428CJ is one of those trusted sources.)
    4. The green lines are the diagonals from rear floor support hole to front frame rail forward hole, projected down on a parallel plane (like all plan view dimensions). Basic geometry and the assumption that the angle drawn from between the two holes at the rear of the floor support (29.88") forms a right angle with the line that is measured (as in, not shown here) from rear hole to forward hole. These two diagonals need to match. With my bent frame rail, they didn't. Good dimension on my car (70.00") was measured at hole edges. Your mileage may vary.





    From the 70 Shop Manual Frame Drawing :
    1. LCA mount, Left to Right (19.125")
    2. Forward frame hole, Left to Right (30.00") 

    [Update, 02/19/2016] - I forgot to mention two key pieces for this effort: the lower engine bay crossmember is a good reference point, as well as the export brace. I purchased a high quality aftermarket crossmember from Opentracker Racing because my factory-like repro fit so poorly in several areas. The new piece is designed to tie the frame rails together just behind the LCA mounts and it measures 29.5" bolt-to-bolt. This dimension doesn't appear on any frame drawing. The export brace is the "good" reproduction that is good for locating the tops of the shock towers to each other and the cowl.

    One last piece of advice: measure it all one more time.
    Good luck - hope this helps.