Roof patch causing warp / oil canning

[nsb29]

I don’t think anyone is better at this than Robert and what he is telling you is 100% correct. In your situation and neither one of us having the skills that Robert has I still think metal bond is the way to go. I don’t think it matters whether it’s a round hole or a square hole just keep it to a minimum, if you use C clamps to hold your backer in place just don’t make them so tight you squeeze all of the adhesive out and then do the same with the patch it should equal out. The one thing I believe in 100% is to use a polyester high build like slick sand which has zero shrinkage as opposed to a 2K primer which will minimize if not completely stop any ghosting

[ItWillBeSlow]

Originally Posted by nsb29

I don’t think anyone is better at this than Robert and what he is telling you is 100% correct. In your situation and neither one of us having the skills that Robert has I still think metal bond is the way to go. I don’t think it matters whether it’s a round hole or a square hole just keep it to a minimum, if you use C clamps to hold your backer in place just don’t make them so tight you squeeze all of the adhesive out and then do the same with the patch it should equal out. The one thing I believe in 100% is to use a polyester high build like slick sand which has zero shrinkage as opposed to a 2K primer which will minimize if not completely stop any ghosting

Thanks for the response and I do agree the more I learn. As stated earlier by someone else, if a small ghost line in this area is my worst issue here, I will consider this more than successful.

When I think about the best way to do this, I believe it would be best to space the backer to allow the metal bond some room to fill and not completely squeeze out. I made a diagram of this below.

Please keep the comments and experience coming, I plan to tackle this within the next week or so and would like to iron out a couple of these details. I think the backer strips will be about 1/3" thick to properly attach the roof panel and provide a lip for the patch. If there are other thoughts please share.

[HAULIN' IT]

I'll add a bit more to the discussion.
First a question for Robert...Was the tailgate you showed in the video for sure mig welded solid? Ive seen some things like that before (glued, Lenco spotwelded along the edge & some tacked every couple inches that you could see EVERY weld & gap), but never a fully welded seam to that degree of ghosting. What I have seen several times (which if it was done on a big long seam like that, I can imagine it looking very similar) is the weld has a side (or both sides) with a real bad "undercut" left along the weld. An example would be a guy closes up the gas filler hole in the truck cab corner by laying a patch in behind & welding around the hole. Once finished, the INSIDE edge of the weld shows the "step" after some time in the sun because of the drop-off...the outside does not because it was able to be easily ground smooth with the original metal. One very good reason for your insistence of the butt-welding panels. Like you mention, you go to a car event on a hot day &/or as the sun is going down & dew is starting to set in...you see some interesting things!

To add to what Robert said about the question about how/why the ghosting...at least with using panel bond, I think there is another factor, but really all the same reason Robert stated & that is, simply the difference in the makeup of the items heating, cooling, ect at different paces & react differently to heat, cool cycles...we have a sharp cut metal edge of the original panel, we have a sharp cut metal edge of the patch...in between/around the edge is a very good insulating item (somewhat like a plastic material). See where Im going?

So what to do to minimize the ghosting? I have found a few things that go the right direction to minimizing the chances. First, you taper off the upper square edge of the original metal & the patch...similar to what doing drywall would look like. Second, As I mentioned earlier, I prefer the Fusor 208. It dries harder & is more solid than the "panel adhesive" products like 108, 110 ect. which is really more a metal glue intended for flanges which stays slightly "gummy" & can never really be feathered out.
This part is going to make Robert go take a blood pressure pill, but as I stated earlier...Im a realist. Body filler is really a friend to nearly everyone until they hone skills in like Robert has. So here it goes...Don't try to make it "too good". Yep I said it! What Im getting at is this. I promise you if you work real hard & make your pieces real close to flush not needing/having much room for "glop on top of it" it WILL ghost line on you. If you have room for a uniform coat of hard base filler like Duraglas, all-metal, something of "insulation" on top of your patch, the odds are much reduced you see a line. Just like nsb stated, a polyester primer is also my choice (Clausen Sandy or Rust Defender is my pick). I distinctly remember one of the Roofs I did years ago on an Impala. The local department had a car & wanted to make it an undercover detective's car. It started out Black & White with lights, antennas, ect. & ended up all White with pinstripes, mudflaps & full wheel covers...looked just like grand paps ride! I grabbed a pick hammer, tapped down the mess the last guy made drilling & mounting the stuff & glued plates over a couple holes on the roof, one on the trunk & an oval on the post from the spot light. That car was back in the shop for a deer hit & someone backed into it years later...couldn't find any of those spots, not a flaw to be found years later. It sat outside everyday, but it was White also. What Im getting at is: "The Roof was ruined by the guy with the drill", I made it better even if it would have shown a ghostline...no one but a trained eye was going to find one if it was there. One of the neat things of over 30 years "practicing" auto body repair...you get to try/see how things work out or dont work out. Lorne

[ItWillBeSlow]

Originally Posted by HAULIN' IT

I'll add a bit more to the discussion.
First a question for Robert...Was the tailgate you showed in the video for sure mig welded solid? Ive seen some things like that before (glued, Lenco spotwelded along the edge & some tacked every couple inches that you could see EVERY weld & gap), but never a fully welded seam to that degree of ghosting. What I have seen several times (which if it was done on a big long seam like that, I can imagine it looking very similar) is the weld has a side (or both sides) with a real bad "undercut" left along the weld. An example would be a guy closes up the gas filler hole in the truck cab corner by laying a patch in behind & welding around the hole. Once finished, the INSIDE edge of the weld shows the "step" after some time in the sun because of the drop-off...the outside does not because it was able to be easily ground smooth with the original metal. One very good reason for your insistence of the butt-welding panels. Like you mention, you go to a car event on a hot day &/or as the sun is going down & dew is starting to set in...you see some interesting things!

To add to what Robert said about the question about how/why the ghosting...at least with using panel bond, I think there is another factor, but really all the same reason Robert stated & that is, simply the difference in the makeup of the items heating, cooling, ect at different paces & react differently to heat, cool cycles...we have a sharp cut metal edge of the original panel, we have a sharp cut metal edge of the patch...in between/around the edge is a very good insulating item (somewhat like a plastic material). See where Im going?

So what to do to minimize the ghosting? I have found a few things that go the right direction to minimizing the chances. First, you taper off the upper square edge of the original metal & the patch...similar to what doing drywall would look like. Second, As I mentioned earlier, I prefer the Fusor 208. It dries harder & is more solid than the "panel adhesive" products like 108, 110 ect. which is really more a metal glue intended for flanges which stays slightly "gummy" & can never really be feathered out.
This part is going to make Robert go take a blood pressure pill, but as I stated earlier...Im a realist. Body filler is really a friend to nearly everyone until they hone skills in like Robert has. So here it goes...Don't try to make it "too good". Yep I said it! What Im getting at is this. I promise you if you work real hard & make your pieces real close to flush not needing/having much room for "glop on top of it" it WILL ghost line on you. If you have room for a uniform coat of hard base filler like Duraglas, all-metal, something of "insulation" on top of your patch, the odds are much reduced you see a line. Just like nsb stated, a polyester primer is also my choice (Clausen Sandy or Rust Defender is my pick). I distinctly remember one of the Roofs I did years ago on an Impala. The local department had a car & wanted to make it an undercover detective's car. It started out Black & White with lights, antennas, ect. & ended up all White with pinstripes, mudflaps & full wheel covers...looked just like grand paps ride! I grabbed a pick hammer, tapped down the mess the last guy made drilling & mounting the stuff & glued plates over a couple holes on the roof, one on the trunk & an oval on the post from the spot light. That car was back in the shop for a deer hit & someone backed into it years later...couldn't find any of those spots, not a flaw to be found years later. It sat outside everyday, but it was White also. What Im getting at is: "The Roof was ruined by the guy with the drill", I made it better even if it would have shown a ghostline...no one but a trained eye was going to find one if it was there. One of the neat things of over 30 years "practicing" auto body repair...you get to try/see how things work out or dont work out. Lorne

Thanks. I am definitely not chasing perfection with this repair, and I am already making the truck 3x nicer than I ever intended to. I am honestly more concerned of a future crack or failure with the metal bond, than having a ghost line. This is an area that's hard to see on a truck anyway, and I would likely forget about it in no time. However, I do want to make sure I am following all reasonable steps to make sure its a durable repair.

Do you have any comments to my sketches above with the backer? My core concern is a small backer strip leaves little surface area for the original roof skin and the patch to all mate together. Do these strips need to be wider? Should I space it down to allow more room for the bond material to fill the void and not squeeze out?

[MP&C]

First I'd like to say thanks for all the kind words, but I fear you guys overestimate my skills or abilities. I've had quite a bit of metal thrown into the scrap pile, serving as quite the lesson of what not to do. Have also had to strip down a paint job due to overeager application of Ospho by the owner into pitted areas, and me spraying epoxy primer directly over top of it (having no prior experience with Ospho). Sunlight over the course of the first year of car shows reactivated the deposited acid in the pits where it was now outgassing and we had small 1/8 and smaller circles where the epoxy was delaminating above these pits. ..

But I am a rather fast learner, and tend to offer opinions to help others capitalize on my learning curve. So when people start singing praises of using Ospho and other such snake oils, I am quick to offer my insight. Many of the proponents of these products simply are amazed by the minimal amount of work needed (ie: shortcuts) and have recommended product based on this and not much else, as most were still sitting in bare metal in project stage.. Lorne, this is what really get my blood pressure going .. People giving half ass recommendations that haven't finished the job to know the full story yet. Hey, I get it. We humans are lazy, and looking for an easier way, so that's why so many of these short cut products are sold. But that, folks, is why Robert is so damned opinionated..

Originally Posted by HAULIN' IT

…...First a question for Robert...Was the tailgate you showed in the video for sure mig welded solid?

Here's the answer from the horse's mouth. Yes, I asked him to clarify once before and this was his response. Names changed to protect the innocent/unknowing. He also spoke of his experience with adhesives, and I am leaving that as well, perhaps you can speak to that based on your experiences....

Xxxxxx, question for you if I may.... The video I took of your tailgate I have posted in a couple places in hopes that someone can see the results up front and perhaps choose another method. I do have someone asking more particular questions, as it appears he is one that swears by the flanging and panel adhesive method. So if I could, a couple questions.

Did you attach your flanged repair with a panel bond adhesive or by welding?

If it was welding, was it plug/spot welded and about how far apart, or was it a continuous weld from end to end?

I'm always trying to learn something, hope this finds you in good health.
Thanks, Robert

Robert, good to hear from you.

As for the flange repair there was no panel adhesive used and the lip was sand blasted and surface prepped before tacking and then drilled and plug welding approximately every 4 inches,

We went back and moved from one side to the other and did a continuous weld then ground it down.

Having never had any previous knowledge of how to properly make and weld panels this was the only way I was aware of, I know better now.

Thankfully the section still looks the same as when you saw it.

On another note, we used a cheap HF English wheel to make a insert for a 34 Chevy coupe and it turned out pretty good for not having any previous experience with rolling metal, we also had to make and replace all the reinforcement that had been cut out to make a dirt track car we were returning to the street.

The reason I am tell you this is we used the panel adhesive thinking it would minimize vibration between the filled top and the braces, the next weekend when we were ready to start sanding we noticed the panel adhesive drew the top and you could actually see the pattern of the bracing.

We ended up using piano wire to saw back and forth to cut the panel adhesive loose, It took us several weekends to correct the problem but the car was eventually painted black cherry and it turned out perfect.

Hope I answered your questions, that wagon you have been building has set the bar high! Xxxxxx

Originally Posted by HAULIN' IT

…...One of the neat things of over 30 years "practicing" auto body repair...you get to try/see how things work out or dont work out. Lorne

Lorne, thanks for adding your experiences here with the adhesives, I agree fully that sometimes another method may be in order, and that is the decision the owner must make in each circumstance. Sometimes you get people that don't want to share with others what didn't work as they don't want to admit "failure", or whatever the reason. I see those as a learning moment, and would feel remiss if I didn't share, especially given the costs today of paint and supplies.

Originally Posted by ItWillBeSlow

Should I space it down to allow more room for the bond material to fill the void and not squeeze out?

I do not have personal experience (disclaimer) but what I have read is that any seam should be clamped tightly. I would interpret that as meaning the joint may be stronger as a thin film between the two panels, and not used as a "filler". It may be, based on the roof repair on the 34 Chevy written above, that too much of a gap and the adhesive does it's own shrinking which pulls the area downward.. My thoughts are that the spacer is going to weaken the joint. Perhaps Lorne can share his thoughts on this as well..





.