Sunday, January 30, 2011

Cowl Panel Replacement - Episode 5: First and goal to go, but there’s a penalty on the play

Last time, I left off with the cowl install essentially complete with the remaining work involving the prep and install of the cowl end enclosures, apron brackets and rear fender mounts.  In the last few weeks, I have managed to tackle the majority of those tasks along with embarking on a rather extensive research project targeted toward replacing my old, ailing MIG welder (which I will cover in a separate posting).
To begin, I decided to protect the inside of the cowl kick panels with a rust preventative coating.  The product I chose to use is called Zero Rust based on its ease of use, good laboratory performance reports, its ability to be easily sprayed and the many positive end-user reviews and durability examples.  Since all of the places I envisioned using this product are out of sight when the car is completed, I chose the standard black color, although the range of colors is quite extensive depending on your particular requirements.
The most obvious thing I noticed about Zero Rust black was how heavy a single quart of this material is, indicating (to me anyway) a very high solids content.  It also has a rather heavy body to it and that makes brush application a challenge on many surfaces.  I started my application of the product with a foam brush (lasted about 5 strokes) and quickly moved to a conventional disposable bristle brush to paint the kick panels.   Note to self:  Home Depot foam brushes SUCK! 
 I used two coats on all surfaces but was rather frustrated at the thickness of the coats required to cover properly.  After a bit of research, it seems the black colored Zero Rust may be slightly thicker than the other colors (like red oxide) and could stand a little bit of thinning to ease application (more on this in a bit).  Indeed, some of my other blogger friends have used the red oxide colored Zero Rust and successfully brushed it on with apparently little hassle, so there must be something to this.
I let the inner kick panels dry for several days with the help of a heat lamp and then set off on the hateful task of sealing the entire cowl panel and ends.  I spent a lot of time evaluating the various iterations of available seam sealers on the market and decided that I wanted nothing to do with those that could be formed with a wet finger and cleaned up with water.  Generally speaking, these are limestone-based products and are usually referred to as “general purpose” seam sealers.  Naturally, a paintable, flexible seam sealer is desirable in most applications involving structural panels like this.  After much research, I settled on 3M #08405 Flexiclear Body Seam Sealer for most of my sealing requirements.  This stuff is kinda like household silicone sealant on acid.  Truth is, I think this stuff could glue together greased, day-old mayonnaise and not break a sweat.  The down side is this sh*t sticks extremely well to EVERYTHING including human hair, skin, eyelids, t-shirts, zippers, shoe strings, cat fur, carpet, light switches, toilet seats, coffee tables, light bulbs, tennis shoes and chocolate chip cookies (ask me how I know).  This stuff is bad-ass expensive and just plain bad-ass.  In the immortal words of Forest Gump……”that’s all I have to say about that”.
Anyway, with the inner kick panels fully sealed and protected, I switched my focus to the rear fender brackets which were thankfully included with my new cowl.  These brackets are an exact duplicate of the original parts and fit the cowl very nicely right out of the box.  After a few seconds on the wire wheel to remove the e-coat where I needed to weld, I was able to position each one exactly based on the previous measurements I made from the old pieces before their removal.  A few minutes with my (now) very fussy MIG and they were welded in place and solid.
Next, I focused on getting the rear inner fender aprons remarried to the new cowl via new Dynacorn apron brackets I bought from NPD.  As you might expect, I wasn’t thrilled with them at all due to their significantly thinner material and everything-but-fit condition.  Not to be accused of failing to give it a shot, I massaged these damn brackets for several hours until the fit was almost perfect and they looked quite nice.  Then after one more sanity check with them clamped into place, I took them off, threw them in my pile of “what-not-to-do” parts and exhumed my original brackets and set to work refurbishing them for re-installation.  I just couldn’t bring myself to accept the thinner gauge material after all.
Fortunately, my original brackets were in decent shape and I was able to remove them with little drama early on.  I did have to weld up a few stress fractures in the corners and tidy them up in the blast cabinet, but once I was done, I had a pair of brackets original to the car and fitting perfectly by most accounts.
With the rear apron brackets ready to install, I carefully marked the location of each required plug weld using a paint pen.  I carefully ground off the e-coat under each weld and dusted them with SEM Copperweld primer and welded them in, dealing my old MIG the coup de grace.  At this point, I had decided it was a mercy killing long overdue…..  I protected the back side of each plug weld with Zero Rust paint while they were easy to access and called this phase of the project done.
With my immediate welding tasks out of the way for the time being, I moved to the preparation of the cowl end enclosures.  As I mentioned in earlier entries, these panels were in astonishingly good shape when I removed them.  In fact, following several minutes work in the blast cabinet, they looked exceptional and ready for a coat of phosphoric acid to prep them for paint.  The following day, I decided to experiment with thinning the Zero Rust paint before application.  As it turns out, I really liked the performance of the paint much better with a slight thinning and will experiment more with the technique in the future.  Fortunately, I discovered that Zero Rust can be thinned with my standard PPG DT-870 reducer with fantastic results  (BONUS!).  As if that weren’t enough, I found I can also use PPG K-201 hardener with Zero Rust to speed the hardening process in sprayed application at a ratio of 5:1:1 (5 parts Zero Rust to 1 part DT-870 reducer to 1 part K-201 hardener).  These two bonus features alone make this product much friendlier to use than the comparable product POR-15.
With a slightly thinned Zero Rust mixture ready to go, I applied three medium coats to the inside of the cowl end enclosures and let them dry overnight with the aid of a heat lamp.  Next, I trial fit the enclosures to the cowl and adjusted the fit until they were as close to perfect as I could make them.  Then, I carefully marked the top cowl flange where each plug weld needed to go and center punched them to make them easy to drill with the spot weld cutter.  Once the plug weld holes were drilled, I went around each flange with my angle grinder and sanded off the e-coat and Zero Rust everywhere there was a weld location.  With the flanges in good order, I clamped everything back into place to check the fit one more time.  Now, if I only had a welder…….
Zero Rust paint applied to the passenger side cowl box.  I ain't bad with a spray gun, but this shot proves I STINK with a brush!

Driver side cowl box coated with Zero Rust.

3M Flexiclear seam sealer is awesome stuff.  It takes on the color of the base material and seals extremely well.  Once dry, it is paintable in any color you choose.

Passenger rear fender mount welded in place.  Welds are looking progressively worse with my ailing  MIG welder.

Driver side rear fender mount welded in.  How much longer will this 'ol welder live?  Or perhaps, how much more can I take?

After wrestling with new Dynacorn rear fender apron brackets, I ended up refurbishing my original brackets.  Here I have marked the locations of each plug weld on the cowl with a paint pen.

Rear fender apron brackets welded in place and welds ground smooth.  This will be the final job for my old MIG as it's just not doing the job anymore.

Fresh out of the blast cabinet, the cowl end enclosures look great and ready for phosphoric acid treatment and Zero Rust paint.

I decided to thin the Zero Rust slightly for the end enclosures and really liked the way the paint responded.  I will be using this technique more as the project moves forward.

With the en enclosures fitted, I marked the top flanges with a paint pen to locate each plug weld location for drilling.

I center punch each weld location to allow the spot weld cutter to accurately center and run true.  The Blair rotabroach cutter makes perfect plug weld holes every time and can easily be controlled to go no deeper than the top layer of material.

With all of the weld holes in place, I sand off the e-coat and paint around each weld area to ensure a good weld surface.

With all of the prep work complete, the end enclosure is clamped back into place one last time.

Fit along every flange is checked for fit.  Looks good!

Top flange fit is exceptional.  Now, where's my new welder........

Tuesday, January 18, 2011

Cowl Panel Replacement - Episode 4: Divide and Conquer

After weeks of planning and prep work, I felt confident I had a plan put together to get the new cowl installed with professional results.  With the inner kick panels repaired and the flanges and saddles straightened and tidy, I spent several days getting the new cowl to fit as perfectly as I could manage it.  I venture I had the cowl in and out for trial fitting at least a dozen times over the course of things, and every time the fit got better and better.  Once I was satisfied that the parent sheet metal was in as good a shape as it could get, I treated everything with another dose of phosphoric acid and shifted my attention to a number a small issues that required attention on the cowl itself.
As I have come to learn quite clearly through the course of this restoration, even the best of panels is rarely ready to go out of the box, and this cowl was no exception.  However, while I have had many reasons to rail against several of the Dynacorn repair panels, this cowl was indeed very good from the off.  All told, only minor tweaks were required to get the cowl into shape for final fitting and involved little more than a bit of grinding here and there.  Particularly, the passenger front corner had a flange intrude into the upper flange area where the cowl end enclosure mates to the upper cowl panel.  This only became evident during the final part of the fitting process when I made sure the end enclosures would fit well with the new cowl.  I had a devil of a time trying to understand why the passenger end enclosure simply would not fit properly no matter what I did.  After a bit of noodling on the matter, I discovered the intrusion of this small flange and after a very quick swipe with my new Makita 4” grinder, the issue was completely resolved and the end enclosure suddenly fit absolutely perfectly!
With everything now exceeding my expectations for fit and finish, I prepped the flanges of the new cowl by drilling the plug weld holes with my Blair rotabroach and deburred everything with my trusty angle grinder.  With that completed, I taped off the cowl and firewall flanges and sprayed them with SEM #40783 Copperweld weld-thru primer.  I have found I like this product much better than the 3M weld-thru primer I have been using.  SEM Copperweld covers better and seems to reduce weld spatter better than the 3M product at a very competitive price.  And best of all, my local paint supply store, GNE Paint Centers in Lake Orion, MI, has it on the shelf.
Once the Copperweld primer dried overnight, we carefully placed the cowl in place, and using a pair of large drifts to realign the panel in its precise location, I buckled the whole works down using my high clamp force clecos and a variety of vice grip pliers and welding clamps, exactly as I had done when doing all of the fit work and preparation steps before.  Fortunately, I was rewarded with exceptional repeatability of the fit I had worked so hard to achieve and I was ready to weld the cowl in for good.
Welding in the new cowl required a bit of strategy to ensure the whole works didn’t move during installation.  First, I decided it was best to anchor the panel at each end and then move to the middle section of the front firewall flange to finish the job.  Although this meant that the most difficult welding operations were first on the menu, it ensured the fit was precise and no warping of the panel could occur.  I started by carefully welding the saddles in place from the bottom (an absolute B*TCH of a job).  Recall in my earlier ramblings that this cowl, as a one-piece assembly, could not be installed as it was from the factory and therefore the saddles must be welded from the bottom to ensure proper installation.  For the approximately 20-odd welds the saddles required, I had to invest about an hour per side contorting myself into all manner of unspeakable positions to get these saddles ready for primetime.
Next, I tackled the upper flanges at the ends of the cowl where they meet the A-pillar bases.  At this point, my ancient MIG welder decided to go menopausal and fight me most of the way.  After a few choice words and LOTS of manipulating of settings, I managed to get both ends welded up, albeit with more metal deposition than I like.  Oh well…….more friggin grinding.
At this point, the cowl is very solid in the car and the front firewall flange looks better than it ever has……and I hadn’t even welded it up yet!  Whooohooo!!!  With a second wind in my sails, I marched across the front of the cowl flange starting in the middle and working my way out to each end.  I bracketed each weld with welding clamps to ensure the flange fit was extra tight and was rewarded with an absolutely fantastic looking flange when done.  Finally, I removed all of the clecos and clamps and went back over each cleco hole with the Blair rotabroach and opened up plug weld holes.  After a few minutes welding, the cowl was completely welded in and looking tight and right.  The following day, I ground every weld down as I had done so many times before.  After a few hours, the flanges looked superb and the fit was truly exceptional.
Next on the agenda, I will coat the inner cowl boxes with an application of Zero Rust coating, blast the cowl end enclosures and prepare them for welding back in, fit and weld the cowl to apron braces and weld in the fender mount tabs to complete the job.  Oh yes……..and prepare my tax return so I can invest in a new inverter-style MIG welder!  A lot of work left before this cowl adventure is complete to be sure, but what remains is comparatively a walk in the park.  Off I go…….
After final fitting of the cowl end enclosures, everything was swabbed in phosphoric acid again to ensure a good bonding surface.

Firewall flange was repaired and straightened as part of the fitting process.  Here, it has been coated in phosphoric acid in preparation for weld-thru primer.

Inner kick panels were repaired and prepped for final cowl installation.  Note the nice, straight cowl saddle.  These outer kick panel surfaces will be coated in Zero Rust coating before being buttoned up with the outer cowl enclosures.

This little bugger took me a few days to figure out.  The part of this small flange (painted yellow) on the passenger side front of the cowl saddle prevented the cowl end enclosure from fitting properly.

A few seconds with a body grinder and a clean-up with the sanding disk and the end enclosure fit absolutely perfectly!

I marked each plug weld hole with a paint pen to make it easier to identify where I needed to drill.  Notice the plug welds are placed between the factory spot welds.  While this may not be concours correct, it ensures the integrity of the panel and minimizes any structural problems with the assembly once it's welded in the car.  Since the plug welds will be invisible and the spot welds will remain, it will be virtually impossible to tell the cowl is not installed with factory spot welds.

A quick shot of the driver cowl saddle drilled for plug welds.  This part of the cowl must be welded from the bottom due to the design constraints of a fully assembled cowl.

SEM #40783 Copperweld has become my favorite weld-thru primer.  Excellent stuff, but pricey.

Driver side cowl saddle primed with SEM Copperweld

Passenger side saddle primed.

Here, you can see the cowl panel end has been taped and primed with SEM Copperweld after the flanges have been drilled and deburred for plug welding.

Final fitting begins with a pair of large drifts inserted into the cowl brace bolt holes.  Once this is done, a series of clecos and vice grip pliers and welding clamps are used to secure the cowl for welding.

The ends of the cowl are welded first to lock everything down.

The outer ends of the firewall flange are plug welded and tie everything together such that I could move on to the center firewall flange.

With the firewall flange completely welded, the cowl is now an integral part of the chassis structure once again.

Welds are ground smooth and brushed in preparation for a sealing coat of primer.  Here is the driver side......

and here is the passenger side after grinding.

The fit was absolutely superb.  Here is the passenger side top corner showing the fit between the cowl on top, the firewall and the inner kick panel at the lower saddle flange.

Here is the passenger side upper saddle flange fit viewed from the inside.

The driver side upper saddle flange area as viewed from the inside.

And finally, the passenger side top front corner showing the cowl on top, the firewall and lower cowl saddle area.
While there is still a lot of work that remains before the cowl project is complete, the results look good and the rest of the work in this area is comparatively easy.

Wednesday, January 5, 2011

Cowl Panel Replacement - Episode 3: From frustration to solution in a thousand easy steps

One of my biggest concerns while tackling the cowl replacement was an area in the driver side kick panel wall that was completely rusted away as a result of a large amount of debris that was trapped in the cowl box and allowed water to collect and rust the area away.  As soon as I was able to properly survey the damage, I set off to finding a replacement panel or patch for the area, fully expecting that something would be available from Dynacorn given they offer complete 69-70 bodies.  I scoured every Mustang parts supplier catalog I have and found absolutely nothing available.  At this point, I figured it probably a case that the part was too new and was just a phone call away.  Yeah……right.  No joy on that idea no matter where I turned.  So with my hope waning, I gave Dynacorn a call for the straight skinny directly from the source.  Yep…….another disappointment.  The word I got from the gentleman I spoke to was that Dynacorn withholds certain parts to maintain exclusivity for a period of time (a moratorium of sorts) and that there were no immediate plans to offer the inner kick panel as a replacement part.  So, with that disappointing thread of information set deeply in my brain, I set out to see what was available in the used parts market. 
After several days of phone calls and emails with most of the major used mustang/classic car parts yards around the country, I simply couldn’t justify the cost of buying a used part cut out of a wreck when the total cost to my door would start approaching the cost of a new floor!  So naturally, I did the only thing left to do:  make my own panel die and hand-form a patch.
I will have to admit up front, I really wanted to avoid making a patch from scratch.  Not because it is beyond my capabilities, but because I really don’t like my MIG welder for doing delicate work.  In fact, it really sucks and at some point I will replace it with a Miller when the means to do so arrive.  But, I have to dance with the date that I came with, and suffer through it as best I can.
The first task involved in this repair was to make several cardboard and Mylar patterns of the kick panel opening to allow accurate layout of the profiles.  These templates must be as accurate as possible and they must be able to be indexed to the body so that their absolute position can be established every time they are transferred from the body to the work and back.  I have to give the template credit to my dad as he is an absolute soldier on these things while I’m away at work or generally jerking around doing something else.  Thanks Dad!
Once the templates were made and the fit verified, I transferred the marks I used to define the general repair area on the kick panel to the templates so I knew exactly what area to duplicate in the repair die.  Then, after some rummaging through the scrap steel bin, I gathered up the material necessary to fabricate a forming die that I could use to accurately fabricate a decent patch panel.  I decided to use a piece of ½” steel plate as the base for the forming die as I knew I would be hammering on it quite a bit and needed a good solid base to work with.  The remainder of the die was fabricated from ¼” hot rolled steel strap as this duplicated the step in the panel perfectly.  Using the cardboard and Mylar templates, I transferred the proper contours to the steel and set off to grinding, filing and sanding them to match the shape exactly.
Once satisfied we had the contours correct, I welded the step forms to the base plate using the templates to verify alignment then ground the welds smooth and polished them with a flap wheel to remove any burs and smooth the surface.  At this point, I could now reproduce the inner curvature of the patch to allow a lip to be formed that would duplicate the detail of the original panel, so back to the templates we go.  Once the curvature was established and transferred to the metal, I carefully cut out the rough shape with my flame wrench (a.k.a. my oxy/acetylene cutting torch) and finished the surface on the grinder, belt sander and flap wheel.  I completed the die by gently rounding all of the inside corners to reproduce the proper bend radii and de-burred everything once more with the flap wheel.
Satisfied I had a good forming die to work with, I cut a piece of 20 gauge galvannealed steel I had in my stock to be slightly larger than the shape I required and clamped it into the die.  Galvannealed steel is a rather specialized sheet metal material that is coated in a durable form of galvanizing for great corrosion resistance and is annealed to allow it to be formed and shaped with tools rather easily.  Some of the higher-end restoration supply houses still carry it, but specialty metals suppliers around the ‘net show it available in small quantities as well.  It’s good stuff and I highly recommend it for this type of work.  Of course, plain annealed mild steel will work equally well, but will lack the extra corrosion protection.
While the tools required to do this type of work are relatively simple, it is common to have to make a tool to do a certain job or achieve a certain shape, and this job was no exception.  Fortunately, I have another little restoration secret I’ll pass along to anyone who finds themselves in a similar situation.  Often times, you will need to make a forming tool that needs to be struck with a hammer to get the results you need.  However, you must be very careful to not assume just any old steel is suitable for the task, as striking many types of steel with a hammer can often result in splintering or fracturing of the tool resulting in very dangerous flying debris.  Fortunately, there is an inexpensive and plentiful supply of form tool “blanks” available right from your home improvement store in the form of masonry chisels.  That’s right, those oddball chisels used to split bricks and concrete make excellent and very customizable forming tools at very low cost.  These chisels are designed for heavy striking forces and are almost always made of good quality forged steel.  Try to stick with the US-made tools as they are usually of much higher quality than the less expensive import competitors.  Often, these tools have a very broad range of widths allowing you to create almost any shape you need for complex forming jobs.  In my case, I cut back a 2” wide flat masonry chisel to give me the proper tool thickness I needed and then created a slightly curved “doming” tool on the belt sander.  This tool allowed me to form the inside corners of the patch with accuracy and quite neatly.
I started the forming process by slowly forming the outer step down to the flange surface starting with rubber mallets and progressing to my ball peen with the doming tool and plashing hammers.  As the panel started taking shape, the flange area started to get wavy as the material shrinks and stretches to fit the form.  This is a little disconcerting for first-timers, but a bit of patience and a little more work with the rubber mallets quickly remedied that condition and I was ready to form the inner lip.  This lip is intentionally formed last as it is the point of no return on the form and adds rigidity to the patch.  In similar fashion to which I started the forming process, I started rolling the inner lip with the mallets and finished up with the ball peen and planishing hammers until the lip fit the form tightly.
Now that the majority of the patch panel details were established, I began trial fitting it to determine what needed to be trimmed to get the fit just right.  Then I marked the patch and trimmed the outer edges using my rotary shear so the edge would be nice and square without the deformation that tin snips cause.  Another check against the body and I trimmed the inner lip to spec and fit it one final time.  Satisfied I had a patch that would work, I stained the repair area with spray machinist dye (an inadvertent mistake I’ll explain in a bit) and scribed the patch outline on the kick panel and drilled each corner with my Blair spot weld cutter to give the corners adequate clearance to allow me to cut the remainder of the damaged area out with my abrasive wheel.  As it turns out, I should have used brush-on machinists dye as the solvents in spray dye created an immediate flash-rusting problem I will need to address once the work is completed.
I carefully cut the damaged area out with the abrasive wheel just inside the scribed marks so I could ensure a good fit.  After test fitting the patch again, I slowly massaged the edges of the hole and the patch until I had about a .030” gap all the way around the patch to allow good weld penetration.  With the patch clamped in place with my trusty welding clamps, I tacked the patch in place in several locations around the periphery.  I tweaked up the fit just a touch with some light taps from my planishing hammer and went ahead and welded the patch in fully.
To finish the patch up, I ground all of the welds with my angle grinder and brushed the area with a wire wheel to tidy up any scale or bogies I missed.  While it’s not perfect, the patch looks a helluva lot better than the mess it replaced and once I can get the whole area sandblasted, I should be in good shape.  Now I will move over to the passenger side kick panel and repair some tiny pinholes and (hopefully) start preparing to install the cowl for good.  Still a lot of work ahead, but this was a good leap forward.
A rather depressing sight.  This is the rusted out section of the inner kick panel that requires repair before the cowl work can continue.

Using cardboard and Mylar templates, the proper contours of the die can be established with accuracy.

Once the shapes are set, the die parts are checked once more to ensure a correct fit and then clamped securely to the base material for welding.

I welded the die form parts with short 3/4" stitches along the back and sides to minimize warpage.  No need to weld it all the way around.

The welds are then ground down with the angle grinder.

Once more, the templates are used to verify the shape of the forming parts and then the Mylar template is marked for the inner curvature of the base plate.

With the curve marked and the Mylar template cut to fit the curve required, the shape is transferred to the base with a marker.

Here is the base plate flange curve ready to be rough cut with the torch.

Here is the rough cut curve at about 1/16" oversize.  Not a bad cut if I say so myself.  Virtually no slag and a very narrow heat affected zone.

Here is the finished die, ready to go to work!

I used 20 gauge galvannealed steel sheet to make the patch panel blank.  Here it is clamped into the die and ready to form.  But wait.......
I determined I needed to make a special doming tool for this project and chose a 2" US-made masonry chisel to be the donor for the project.
I cut the blade back on my band saw to get the root thickness I was looking for.
After a few minutes on the grinder and belt sander, I had a nicely formed radius that was just right for forming the inner corners of the patch.  Quick, easy, safe and CHEAP!

Here, I have completed the first phase of forming the patch.  The outer step of the patch is complete and the slightly wavy flange that remains can be seen.  This will soon disappear with a little hammer work and the forming of the inner flange.

In this shot, I just finished the initial roll of the inner flange with a rubber mallet.  This established the crease line and makes further forming much easier.

The inner flange is almost formed here.

After a little more work with the ball peen and planishing hammers, the inner flange is formed with plenty of extra material to work with.

Here's a look at the back side of the patch.  Almost looks better than the front!

After trimming some of the excess material off the patch, I'm ready to start fitting it to the kick panel.

The test fit looks really good and I can now mark the back edge of the patch for trimming to size.

You can see a light pencil line around the back edge of the flange step where I will trim the patch of excess material.

I darkened the trim line using a sharp marker to make it easier to see in the shear.

Using my trusty rotary shear, I trimmed the excess material off the back side of the patch.

And there it is!  A completed, hand-made patch panel ready to weld in place.

If you look closely, you can see the scribe lines in the machinist dye where the rotten part needs to be cut away.  Also note the flash rust areas that came up immediately when the spray dye was applied.  I should have used brush-on dye instead.

Before cutting the bad section out, I drilled two holes in the corners of the repair area to allow the abrasive wheel to complete the cuts without snarfing up the corners of the repair.  First the top.....

Then the bottom.

With the rotten section cut out I was able to fit the patch to the kick panel quite closely.  A little massaging of the patch and panel edges and the patch was clamped in place for welding.

The patch was tacked in place and everything was checked once more to make sure it was in order before welding.  Looks good!

After a few tense moments, the patch is fully welded in place.

With welds ground smooth and the surfaces brushed clean, the patched area looks quite serviceable, albeit not perfect.  However, once prepped and painted, it will be completely invisible.