The Stovebolt Site Forums The Shop Area Paint and Body Shop Fixing oil canning on the cab roof

When I got my truck, the roof was partially caved in, with a few creases. I managed to get it mostly back to the right contour, but it has a serious oil-can toward the rear just right of center. I sent a PM to Robert (MPandC) for some advice, seeing as he's the guru for all things metal. He was kind enough to call me and talk me through how to fix it after I send him a few photos and videos. His advice was to try and define the extent of the oil can, mark the perimeter of it (where it doesn't move) and then push around the perimeter with a thumb to find spots that were "locked" (not able to make the oil can move). He also advised using a donut dolly to shrink the numerous lumps (showing as dark spots where the primer is sanded off) I made I pushing it back into shape, saying that I had probably stretched the metal. He also talked me thru his trick of using a magnet to locate spots on the opposite side of a panel so the dolly can be positioned correctly.
Robert, if I messed up translating what you told me, feel free to correct me.

Anyway, HERE are some photos and videos that I pointed Robert at. I laid a strip of sheet metal across the roof, first near the front, then moving it back a bit for more photos. I also used the strip laid front to back on the right of center. That highlights the low spots in the roof. The low spot just behind center can be pushed upward with very little pressure from underneath, but the high spot right and rear of center takes some pressure to cycle the oil can.

If anyone else has suggestions, please weigh in. I plan on updating this as I work through it, hopefully getting it straight so I can move forward with paint. I'm sure there are other cabs out there that have a caved-in roof, and if my efforts can help someone else, that would be a good thing.

Robert explained it correctly. I had the same issue with my roof. I found that when I found the edge of the oil can area, I could use a hammer and donut dolly and shrink that area to stop the oil canning.
A 2 inch pvc straight pipe fitting works also as a dolly.

Thanks Wayne.

I had no doubts that Robert was correct. That's why I hit him up. I've been a long time follower of his work. I have a "collection" of donut dollies: a 2" PVC pipe coupling, and a 1-1/2" one.

You can also use a iron pipe cap (2") as a hollow dolly and a shrinking hammer. Years ago I saw a tool a fellow from Vietnam had made. It looked like a flattened C or sideways U with a sharp end and a hollow end...hinged in the very center of the C. You could shrink metal like a champ as long as you could make the tool reach the damage. No matter what you did it would always hit the center of the hollow end.

Today I got around to defining the oil can area(s). Turns out there are two of them.
In the pics, the inner circles are where the bending takes place when cycling the oil can, and the outer marks are where when pushing along the line, the metal doesn't move. The straight line along the left side of the first pic is where the inner roof starts, so won't be able to use a donut dolly there for shrinking. May have to go with heat there. Pic 1 is taken from above the passenger door and Pic 2 is taken from above the driver's door.

Any thoughts where to start shrinking? I'm seeing some lumps in the lower left of Pic 1 that are probably good candidates. That's near where the metal is highest within the bigger circle. I'm thinking leave the smaller circle alone till the larger one is almost gone.

A roof skin on this truck, in its original form, has a crown in both the x and y axis. This crown is what provides the structural support of the roof skin. Kevin's pictures show that the roof has been neglected or to be more accurate, used as a ladder or something. Any dent in the roof is going to interrupt the crown where the skin will lose the support it once had.

Any dent in that roof skin, or any panel for that matter, will typically cause a stretch as it gets pushed inward. About 99% of roof dents are caused from the outside and push inward. When removing said dents, some shrinking needs to occur in conjunction with bumping the dent. Looking at the bare spots (dark circles) in image 5140 on your flicker page, these suggest that dents were hammered upward, likely into a hard dolly?, with little if any shrinking occurring. The fact that the paint is sanded off reinforces that these are stretched and when they were hammered upward, may have been stretched even more. Looking at the image 5140 again, the center of the panel is shown to be recessed in comparison to the flat strip of metal.

Please understand that oil cans are a chore in/of themselves when the panel is in front of you, much less looking at pictures. But what I perceive has occurred is that the extensive dents and stretching that went with them, combined with hammering the dents upward (possible more stretching), has a bit of excess metal looking for somewhere to go. This is normally seen in the form of a sinusoidal wave, and as there was less damage in the middle to lock things in place as occurred with the dents, the center was pushed downward. My thoughts are the large center oil can (low) is a side effect of all the stretching from dent formation and dent removal with not sufficient shrinking to counter those effects. I would start by leaving the side effect (large low oil can) in the center alone, and attempt to shrink each high spot to relieve some of the pressure pushing outward. Use donut dolly or shrinking tip on a dent puller, I'd shy away from a torch as I don't think you want excessive shrinking going on which may send you to the other end of the spectrum. Slow and steady shrinking to sneak up on the repair. Enough shrinking to either side of the center of the obvious and larger of the dents should show progress in the form of the oil can starts to loosen up slightly. Once some of this pressure is relieved, then start some off dolly bumping where you may see some highs and lows adjacent to each other. The other thing to keep in mind is that any localized low adjacent to high spots (bare) are themselves high spots when you look from the bottom. So they will likely need shrinking as well.

Donut dollys... I like using things that go a bit slower and allow you to creep up on the end result where you can better watch what's happening as you go along. When I tend to use a faster method it's like sliding into an off ramp at 90 for me...


I would shy away from heat and try the donut dolly and see how that works.. Keep in mind, the optimal efficiency in using a donut dolly is when your hammer pushes the excess into the donut where the springback brings it back to flat. Too much hammer swing and you leave a dent, not enough and you're there forever. So it's a fine line to using that tool, but I find it does well for me, others haven't had the same luck but I think it's all to do with patience.

With a high crowned panel and wanting to bring a bulge down/inward, I would use a flat faced body hammer. The flat hammer works well here in pushing a convex crown inward and capturing any excess, making a smaller crown. With a fairly flat panel as you have with your roof skin there, a flat hammer on a flat panel doesn't bode well with enough panel pushed into the dolly for effective shrinking. The crowned body hammer (Snap-on BF618) on a flat panel allows you to push excess metal into the donut, just don't swing so hard you leave a dent. Start soft swings and work up as you see how it progresses. You should be able to watch any bulge decrease in height. Any stretched low (not the middle oil can) would be treated as a high from the bottom side, ie: hammer from the bottom into a donut dolly on the top. If you leave any dents you are swinging too hard. But at least make one dent so you have a benchmark to ease up from. If your patience grows thin, my next preference would be a heating tip on a dent puller. Does a better job of pin point focus than a torch or shrinking disc.

to clarify my rambling above, a flat hammer on a flat panel doesn't work well at all at pushing/capturing excess into the dolly. you'll be there forever with no results.

Hence the crowned hammer on a flat panel, to push excess into the donut. Just as I mentioned the pin point accuracy of a heating tip, with a donut dolly I prefer sized at the diameter of the hammer face or slightly larger. Usually with the crowned hammer I use same size diameter dolly as the hammer, as the edges don't really hit. This smaller size allows more precision in bringing down the high spots when it isn't a good consistent bulge. Like the ones you show, the sanding shows highs and lows, so a smaller donut would allow better precision in picking out only the high spots.

Got it, Robert

I just happen to have a BF618 and it looks like it matches up to the 1-1/2" PVC coupling that's one of my donut dollies. It leaves about 1/16-1/8" around the perimeter of the hammer head. Or should I be going with a smaller PVC coupling? I can try each to see what works best.
I'm not familiar with a shrinking tip on a dent puller. Can you elaborate on that? I may have to go buy another tool. [on edit] figured it out.

Thanks for your help.

Story time. A few years back, an older man (what am I saying here? ) who had quit driving had his car parked across the street, and I caught some neighbor kids jumping up and down in the roof. When I told them it wasn't any way to treat the car, they said the old man told them it was OK. I'm not sure I believed them, but they quit. I'd sure hate to try fixing that roof.

That pvc coupling should work fine. Here's the shrinking tip on my dent puller. also referred to as a heating tip. It has enough mass that it doesn't "weld" like the normal tip, and will heat up a pin point area for a localized shrink.


https://hosting.photobucket.com/albums/y167/rmccartney/Metalworking/.highres/Picture207-1.jpg


Here's some shrinks:


https://hosting.photobucket.com/albums/y167/rmccartney/Metalworking/.highres/Picture208-1.jpg

Is that heating tip a stud welder and your using it without the studs?

Mine does not weld the studs, it has the “throwing star” style consumable that you weld to the panel, pull, then twist to remove and use on the next one

For those of us that didn’t understand what Robert was talking about, I looked it up and included the [LINK].

Pretty neat system and now it makes sense to me.

Here's a picture of the "throwing star" in action. This gives you 3 sides to use for pulling dents. then once dent is pulled, a quick twist releases it, ready for the next one....


https://hosting.photobucket.com/albums/y167/rmccartney/Metalworking/.highres/Picture358.jpg


In addition to the heating tip I showed on page 1, they also have an interchangeable tip for welding on those studs for vinyl top molding clips.

After I posted last night, I found a video of that "throwing star" dent puller. Pretty slick. Easier than a stud welder and separate slide hammer like I borrowed and used on a few spots.

I actually have something that should work on heat shrinking. I attempted to build a one-sided spot welder from an old microwave that died. It didn't work for that, but should do the trick for spot heating. Has a nice big copper electrode that I can shape to somewhat of a point. Will give that a try.

An employee at a body shop I managed had a method of dealing with an "oil can" situation- - - -he would push the dent inward, glue a wooden paint stir stick to the back side of the panel with contact cement, and skim the top with Bondo. He didn't make it to collecting his second week paycheck!
Jerry

Update??

Haven't had a chance to get to it yet. Have had guests up from the lower 48 (and a fishing trip ) that kept me away from the project.

I've been thinking about getting a steel 1-1/2" coupling to use as a donut dolly (has a bit more mass than the PVC one I have), AND backing that with a big magnet that I sourced from a speaker. Thinking that would allow me to locate the spot to hit on top "on the fly" for shrinking. I stopped by the plumbing shop yesterday, but it was closed.

Good idea with the coupling. To add a bit more mass, set the coupling on a flat metal table top, melt doen a few wheel weights, and pour an inch or so of lead into the bottom of it. The cane I carry has about a pound of bullet casting lead poured into the tip- - - -sort of an "attitude adjuster" for somebody who thinks they're picking on a crippled-up old man!
Jerry

I recently used a tool I made for installing a harmonic balancer as a shrinking dolly. Worked great as it has allot of mass.

Check out this guy on y tube .Carter auto restyling easy repair on oil canning on classic cars and trucks. The video is o a 52 chev truck.

Thanks Kevinski,
I viewed that video, but he didn't do any shrinking on that particular roof. He did explain why I have some lumps in my roof as I got a little to energetic bumping it back up.

Today I had a chance to do a little shrinking with the PVC coupling as a donut dolly. The smaller oil can toward the driver's side rear got much tighter (harder to cycle). Going very lightly and checking the lumps with my hand to see how they're flattening out. The larger oil-can is starting to change as well, but still about the same size, and the depressed area in the center front is still depressed and cycles pretty easily.
I've been eyeballing where to place the dolly reference to the hammer. I tried welding a handle on a high strength magnet (has a steel backer plate) that I got from a computer hard drive to be able to get it off the metal, but soon discovered that it became detached from the backer plate AND WAS NO LONGER MAGNETIC! I guess the arc welding current messed it up. So backed up and used JB Weld to glue the handle on another one. Waiting for that to cure to use it to locate spots on the opposite side of the metal.

Kevin,

It's more likely the magnet lost its magnetism from heat. You can kill most magnets with as little as 300 degrees Fahrenheit.

Spent a couple more hours today with the donut dolly shrinking metal on the roof. The big oil can is now VERY hard to cycle, the depressed area on the forward part of the driver's side is much flatter, and now there are two overlapping (kind of in a figure 8) oil can spots toward the back and straddling the center. The one toward the driver's side rear is basically gone, as it takes a LOT of pressure to depress that area.

Making progress, but it's slow going. Here's another video. Link

Back at work on the roof. Did some shrinking on the two figure 8 oil cans, then sanded all the marks off and started over. I have 4 distinct oil can spots now (see the picture - looking from back to front). The ones numbered 1 and 2 in the photo were the two figure 8 ones. They've gotten a bit smaller, but there's still a large overlapping one (#3), and another smaller one has reappeared to the left rear (#4). It kind of feels like I'm chasing these around. Still working with the PVC coupling and convex hammer, which does seem to work well in shrinkinig. Have not gone to heat.

Anyone have any suggestions? I'm having trouble identifying the spots around the perimeter that are locked per Robert's explanation. Should i be shrinking in a line (curve) around the perimeter? That's what I've been doing in basically parallel rows of hammer blows. I'd sure like to get this knocked out soon.

Can you take another set of pictures with the profile gauge(s) so we can better see what you have.

Thanks for weighing in Robert.

First pic with the front edge of the strip resting on the crown at front center (I believe that's propping it up a bit)
Second about 6" further back
Third another 6" back
4th pic center fore and aft
5th about 6" toward the driver side
6th about 6" toward the passenger side

The worst gap is less than 1/8", so getting close. Only oil can #3from the previous post is a "hard" oil can (takes quite a bit of pressure to cycle it. They all pop back to the upper position.

Phak 1 The lead body man I was around 52 years ago would heat a red place about like a quarter then back it with a flat dolly then pound it with a FLAT round body hammer about 1 1/2 inch in diameter. A few of those and the "oil can " would disappear. was not high strength body steel,he hated that sorry era !!

When I went to Auto Body trade school (a mere 54 years ago, I was taught to heat up a spot cherry red about the size if a dime and hammer and dolly around the spot working the metal into the center, then quench it with a rag dipped in water. That was for extreme stretches, and in Kevin case, I think that would do more harm than good.

Kevin, can you also do gauge front to back in a few different spots?

Robert, the last 3 photos in my last post are front to back (centered, left, and right of center.)

Got a call from Robert.Thanks

He said it looked pretty close and suggested propping some thin plywood underneath and using a slapper on the remaining high spots (previously marked with a pencil squiggle.)

Tried that and am getting closer with a little more judicious shrinking with the donut dolly. Now have just two light oil cans that are smaller than the previous. It's been a slow process. Kind of hard to see the marks on the photos and to get the 2x4 propped in the right spot underneath.