Having a Cowling Good Time!

The last really rusted spot on our little Mustang is the engine cowl. This is a notorious spot for major rust, and our car hasn't escaped the ravages of water and time. I knew that we would lose all of our hard work on the floorpans if we didn't give special attention to thoroughly repairing this area.
Before we attacked the problem, I did some research on the Web and learned that the design of the fresh air intake on the Mustang was poor to begin with, and Ford didn't help matters by assembling the cowl as a two-piece unit with no paint on the inside. This automatically ensured that rust would begin forming after the first rain or washing that the car endured. In addition, if the car was left outside, leaves and other debris would pile up inside the cowl and eventually block the drain holes. Once this happened, it was only a matter of time before the forces of nature would eat their way through the sheetmetal. Once that happened, the floorpans were doomed to the same fate. After reading all of this, I felt more than a little frustrated at Ford (and all American manufacturers, I'm sure) for basically ensuring obsolescence in their product. Is it any wonder that many Americans used to buy a new car every two or three years?
So, putting aside my frustrations, I picked up a grinder and

Opening up the drivers side of the cowl

cut out an access hole on each side of the cowl below the area where the fender mounts. This is a good place to work from to make small repairs. For our car, though, that small opening wasn't enough, so I took the plunge and cut out a nice-sized hand hole on the top of the cowl on each side. I made sure to make clean cuts, because these pieces were going to be welded back in.

Rust holes around the outside of the "hat"

Once opened sufficiently, I could see what our job consisted of, and it wasn't pretty. Both "hats", which are the shafts that funnel air into the car, were almost completely rusted off at their bases. In addition, holes big enough to fit my hand through were on either side of the hats, which I found after removing the aforementioned handfuls of leaves and dirt.
Since we're still trying to maintain somewhat of a budget, and the project is dragging somewhat, I decided to try and fix the holes without removing the top half of the cowl. This is a very labor-intensive process that requires the removal of around 200 spot welds. I knew that if we started down that path, Daniel would become completely demoralized!

Note the big hole in the lower right

Instead, I decided to clean out the cowl through my openings as best I could with scraper and vacuum. Once satisfied that I had all of the loose stuff out, Daniel
gave each side a liberal coating of Rust Seal, the super hard epoxy paint that we're using on the underside and floorpans. My thinking was that if I sealed the metal from future corrosion, I could also create a good bonding surface for fiberglass matting in there and form it around the hats, effectively fiberglassing them in place.

Applying fiberglass

This worked better than I anticipated! Even the biggest holes were spanned with the fiberglass matte, and in not time we had created a new base of the cowl and bonded the hats back in place. After our glass work cured, we then applied a liberal amount of seam sealer all around each of the hats, and all over the surface of the cowl. When I say liberal, I mean we used about a tube and a half between the two sides. Again, my thinking was to create a durable surface that would
both bond to the old metal and seal out any future water contact to the old rusted areas.

Plenty of seal sealer should do the trick

After a few skinned knuckles and gobs of sealer under the fingernails from daubing it with my hand, we then did a "bucket test". I not only wanted to recreate the water flow that might normally come through the cowl vent from rain or washing, but I wanted to flood the area to know for sure that every possible leak point was sealed. Fully confident in my work product, I positioned myself under the dash with a drop light while Daniel poured about two gallons of water into the cowl. I was met immediately with a flood of water on the inside!
Undeterred, I took stock of the situation. I quickly identified (what I thought was) all of the remaining pin holes and gaps. More caulk was applied. More water was poured in. More water flooded into the car! More resolve was made to stop the leaks!
We did this routine a total of five times.
I could tell Daniel was getting a little tired of the process, but I tried to impress upon him the importance of this repair, probably

The first bucket test...or the second?

more so than any of our other repairs. If we worked this hard to get the car right, only to ignore or overlook a small leak, then all of our work would be for naught after a couple of years. We forged onward.
After the first and second tests, I spent more time under the dash trying to figure out exactly where the leaks were that evaded my repairs. I finally determined that the front seam of the cowl assembly, which is only spot welded, was letting water through to the interior. We both agreed that the only reason that water was getting through there was due to the volume of water that we were pouring in at one time, which was probably more than anything that would occur in normal use. However, since I had figured out the hidden problems, it made sense to keep caulking.

Closing the patient back up!

We did the fifth and final test this morning. With fingers crossed, I again got under the dash, while Daniel poured yet another bucket of water in. And....success! I came out as dry as when I went under the dash.
For final insurance, I applied another liberal coat of Rust Seal over the seam sealer, to fill in any possible pin holes or little spots where water would accumulate.
Finally satisfied with our efforts, it was time to weld up our holes.I feel pretty pleased with the end result. Now we can move on to the final bit of sandblasting in the engine bay, and start some reassembly!