Why I Dislike Sunroofs (Part I)

In perusing car ads, you’ll sometimes see the term “slicktop,” which means that the car has no sunroof. While I loathe the car-salesman-vibe of the term almost as much as “restomod” or the phrase “all the hard work’s been done,” it’s gained traction because it accurately conveys the lack of an option as a positive. That is, if you say “no sunroof,” it makes it sound like the car is missing something that it should (or at least could) have, but in the opinion of many including myself, the opposite is true—the lack of a sunroof is actually plus.

Sunroofs are a pain. Most of them are rarely used, but that doesn’t prevent them from having or causing problems, and when they do, the hassle far outweighs their occasional sun-on-the-face benefit. While I don’t hate sunroofs the way I hate, say, decent rust-free cars that have been drift-modded, they can be more trouble than they’re worth. Here’s why.

The history of the modern sliding sunroof extends further back than I expected. Surprisingly, it appears to have begun a hundred years ago in a country not known for sunshine or fair weather—England. Richard Oliver Mobbs of London patented his sliding roof panel in Great Britain in 1925 and formed The Pytchley Autocar Company to build and sell it, with Morris and Austin adopting and incorporating the design in 1932. Pytchley, unfortunately, appears to have been one of the many casualties of World War II. Cadillac, Oldsmobile, Buick, and LaSalle apparently also offered a similar sliding roof panel just before the war. Like the Pytchley roof, their design also died on the vine.

The sunroof as we know it really took wing in the 1960s. The German manufacturer Golde produced both fabric and metal sliding sunroofs that became available on several German marques. It’s perhaps appropriate that “The People’s Car”—the VW Beetle—may have largely introduced the sunroof to the American public. Beetles were available with roll-back fabric sunroofs as far back as 1950, but a hand-cranked metal roof became an option in 1963. The first American car with a factory sliding sunroof was the 1960 Thunderbird with an available Golde roof. Unfortunately, there was no cranking mechanism; it needed to be slid by hand. And it was expensive. It was only available for a year. But in the late 1960s, the sunroof came back to Detroit, with the American Sunroof Company (ASC) supplying power sliding metal sunroofs to Ford for installation in Cougars and Thunderbirds.

In the mid-1970s, a confluence of fuel cost, emissions, and safety factors essentially killed off the production of convertibles in the United States for about a decade, leaving sunroofs and T-tops as the available options for those who wanted to see some sky while driving. In my humble opinion, neither of these comes close to giving you that wonderful whole-body-relaxation response of driving a fully-open car (Porche’s Targa top is a bit better), but the huge manufacturing advantage of the sunroof is that it maintains the car’s basic stiffening structure of a roof supported by pillars—the A and C-pillars at the front and rear, and the B-pillar behind the front door. As much as I love convertibles, with their roof and pillars gone, the body of the car needs to be further stiffened to compensate for the lack of a roof. My ‘99 BMW Z3 roadster feels very solid, but on older ‘verts, you still get a lot of body twist. I used to own a 1984 Alfa Spider, and it was so flexy that if I put the nose on ramps, I couldn’t open the doors.

Sliding sunroofs, and their cousins, tilting but non-sliding glass moonroofs, remain popular and are standard equipment on many higher-end vehicles. But here’s the problem with sunroofs: They all leak. They’re designed to leak. Or, rather, they’re designed to deal with the fact that they leak. The sunroof seal is mainly there to reduce wind noise. Water is expected to get past it. Don’t believe me? Every sunroof in every car sold today and likely sold for the last 50 years has a four-sided moat beneath it with drain ports at the corners. The front two corners drain into rubber or plastic hoses that go down the A-pillars, the rear ones out the C-pillars. On many vintage cars, manufacturers didn’t take into account the fact that these cars would still be loved and driven half a century later. On my beloved 1970s BMW 2002s, the front sunroof drain hose protrudes through the inside corner of the wheel well. So far so good, but the rear one stops short and dumps water into the inside corner of the wheel well and the rocker, invariably rotting it out from the inside.

And things can go wrong with the “moat” system. A few months ago I wrote that my 2008 Nissan Armada was dripping water from the right side of the sunroof. It turned out that the drain hose had become detached from the right front drain spigot. I was thrilled that that was the problem, as the Armada forum had reports of the moat being cracked, which would’ve been much more difficult to repair.

That gets to the second issue: Sunroofs are complicated. In addition to drainage, the mechanism for sliding involves not just a pair of tracks but also cables and an electric or and-cranked winding mechanism to pull them. Plus, the sunroof not only moves back and forth but also up and down. Watch it closely sometime. When you open it, the back edge first drops down so it can retract beneath the surface of the roof, sort of like one tectonic plate subducting beneath another. It reverses this action when you close it. That means there’s a mechanism that not only slides it back and forth, but also manages that upward and downward motion of the back edge. If any of this breaks, the sunroof becomes very difficult to close. It’s little wonder that many folks in the vintage car world open the sunroof when they buy the car to make sure it works, then leave it shut.

And so, with that, I can tell you about the sunroof in the FrankenThirty (my 1988 BMW E30 325is that was apparently rebuilt with pieces of two different cars). When I bought the car, there was no seal on the sunroof at all, its left rear corner was sagging badly, and it would neither open nor close though I could hear the motor trying to turn. There was obvious staining on the rug indicating a history of water intrusion. I needed the car to sit outside while my wife and I were on vacation (in fact, unless I sell something, the car will likely need to sit outside all winter), so for the short term I simply taped the sunroof shut with packing tape.

To be clear, I would’ve left the sunroof taped shut, but a friend of mine shamed me into fixing it. My goal was not to rebuild the sunroof so I could hit the switch and slide it back and forth with alacrity and without fear. It was instead to get it closed, leave it closed, make sure it didn’t leak, and not spend a lot of money or time doing so. While the FrankenThirty has surprised me with its ready-to-run happy-to-be-part-of-the-family vibe, it’s still a somewhat ratty car, and I had zero desire to have a working sunroof be an isolated pocket of restored flawless functionality.

Although I couldn’t move the sunroof with its electric motor, there’s a little access panel that exposes the underside of its main gear, in the center of which is an Allen key hole that allows you to use a ratchet wrench and an Allen bit to hand-crank the sunroof. This enabled me to close it. I could then slide back the headliner trim panel and reach into the tracks, from which I pulled out several pieces of plastic that were jamming it, along with a section of rubber hose that I believe a previous owner had placed to hold up the sagging left rear corner. I eventually realized the plastic pieces were broken sections of the part of the mechanism that lifts up the back of the sunroof when you close it. (Remember what I said about it being complicated?)

To assess what was wrong and how to fix it, the sunroof panel had to come off. And since the car would be sitting roof-less for an indeterminate amount of time, it needed to come into the garage.

The panel is held to the winding mechanism by six Torx-head screws, but to see them, you first need to unsnap the front of the sunroof’s headliner piece, then slide it backward into the same recess where the sunroof panel itself slides. It actually wasn’t that bad to sit inside the car with my head up against the underside of the sunroof panel, wield a Torx bit on a swivel ratchet extension along with a flashlight to help get it into position, and unscrew all six.

With the panel off, I could look down on the broken plastic lifting section and see where the pieces went. I also could see that, since the sunroof had been partially open for years, there was all manner of detritus underneath—pine needles, pieces of acorns, evidence of mice, etc. I made sure there were no more broken pieces, vacuumed up all the flotsam, and lubricated the track with spray white lithium grease. I also ran a wire down and blew compressed air through the two front drains (the rear ones are basically unreachable).

It wasn’t clear to me how much I needed to remove in order to replace the broken plastic piece, and it was difficult to identify the part in BMW’s diagrams. Looking on eBay, I could see it sold as part of a sunroof cable assembly that was no longer available new, but for which $250 New Old Stock (NOS) pieces and $100-ish used items were available.

With nothing to lose, I decided to have a go at epoxying it back together with J-B PlasticWeld. Because the broken plastic section was right at the point where the rod for the lifting mechanism met it, I elected to reinforce the epoxy with a piece of metal. I cut a section off a small hose clamp, bent it into shape, carefully laid the pieces and the PlasticWeld into position, let it dry overnight, then tested that it worked without self-destructing.

I looked into buying a new sunroof seal, but the $125 cost for what’s basically a piece of weatherstripping put me off. I have a variety of sizes of high-quality adhesive-backed closed-cell foam tape I bought years ago from McMaster-Carr for sealing the door flaps on heater boxes. A ½-inch wide piece of 1/8-inch thick foam looked to be a good fit if I cut it lengthwise. I cut four small pieces, put one on each of the four edges of the sunroof, and test-fit it in the opening. And remember—as I said, this is mainly a wind noise seal. It seemed to be the right size, so I cut one long section of adhesive-backed foam seal, cleaned the outer edge of the sunroof, and laid it in place. I then test-fit the panel and seal. It was astonishing how well it fit.

With that, I carefully re-mounted the sunroof panel, leaving the Torx bolts slightly loose to fine-position it. I gently hand-cranked it into position, gritting my teeth as the rod on the end of the plastic mechanism pushed the back edge upward. I expected it to fail at any moment.

It didn’t. It held.

I carefully tightened the Torx bolts, smiled at my masterwork, and unplugged the sunroof switch so I wouldn’t be tempted to open it. I then moved the car out of the garage and parked it in the driveway.

That night, it rained. In the morning, I was greeted with water dripping from the headliner and puddles on the seat and floor.

Okay, maybe I do hate sunroofs.

(Next week, I divulge the multiple layers of what went wrong.]

***

Rob’s latest book, The Best Of The Hack Mechanic™: 35 years of hacks, kluges, and assorted automotive mayhem, is available on Amazon here. His other seven books are available here on Amazon, or you can order personally inscribed copies from Rob’s website, www.robsiegel.com.