Topic: Welding Sheet Metal

[Angela]

Note to moderators: I'm not sure where to post a thread concerning welding; I apologize if you feel the need to move this post.

I'm looking for some direction concerning welding sheet metal. I have an excellent MIG welder and am reasonably good at welding metal at 1/8" or thicker. However, when I'm faced with welding 22gauge sheet metal, I'm not so good. In fact, I'm terrible. I really struggle with warping, burn-through and getting the right penetration and bead height.

Yes, I can search through the millions of youtube videos looking for tips. However, for the purpose of restoring classic mustangs, I'm wondering if there are specific articles, books, videos etc that you guys would recommend?

I have a few specific questions:
(1) Should I be using an electrode positive or negative setup? I thought that electrode negative would inject less heat into the sheet metal, yet I noticed a comment on Miller's site that states electrode *positive* should be used for sheet metal -and that puzzles me.
(2) When welding autobody metal, should one use a straight or zig-zag pass with the electrode (gun)? I see reference to both techniques for autobody -yet I cannot imagine how both are correct. IMHO, the zig-zag pass (technique) would keep the electrode in a small area longer, which would increase heat and the chances for brun-through and warping. However, when I try straight passes with the gun I don't like the look of the weld...it's too narrow and tends to bead up too tall on the surface of the metal.
(3) What techniques do you guys use to reduce warping panels when replacing sections? I'll upload some examples of sections on my '70 which have been replaced and are merely tacked together around the perimeter. I think (but am not certain) that I now must fully weld the entire perimeter of the panel (VS using some sort of sealer) and am not sure how to complete the weld without warping the heck out of each panel. I've tried the "1 &6" approach wherein you weld 1". move 6inches and then weld another 1" length. But this still warps the test panels I'm experimenting with.

Here's an example of a replaced panel that is spot welded along the perimeter. I'm not sure what to call the joint...it appears that the two adjacent panels are slightly overlapped and spot welded from both sides (alternating).

[JKWilson]

While I'm no professional welder, and certainly not an "expert", I can offer a few things that may help. Is the Miller you're using a 220v or 110v unit? I have welded sheet metal on both of my cars (listed in my signature). On the fastback I used a 220v Miller unit, though it was in the 1990's so I don't recall the exact model and it wasn't mine. On the coupe I use a Millermatic 135 which is a 110v unit (which I own). The reason I ask about voltage of your machine is because I found I had to turn down the 220v unit much more. Once dialed in though, both machines function almost identical.

Using Miller's recommended positive electrode set up is the only way I've welded steel, sheet metal or other. Are you setting up your machine on scrap pieces before you begin? This allows you to get the best amperage/wire speed for good penetration. Be sure you're using scraps of what you'll be welding. I found that if I used two scraps of new metal for set up, but then welded new to old that it welded differently. Does your machine have variable adjustments for wire speed AND amperage? Some have a selector switch for the amperage and then variable adjustment for wire speed. Mine has both so I have a little more "tune-ability". I was taught the "standard" method of setting up the machine for the material. There's a little "formula" to get you close, but I have never used it because the machines I've used always had the set up charts inside the door. They are pretty much based on said formula, but why reinvent the wheel?! Once I had the basic adjustment dialed in I'd then fine tune for the material. This can be a little tedious but once there you're all set. One thing I've seen and was a quick and easy way to set up your machine was to set amperage, either at the selector switch or dial. Obviously lower amperage settings are for thin material. Once you've done that, then leave it alone. Now you do all of your tuning with the wire speed knob. You just lay out your material you'll be welding (using scraps), and orient it how it will be welded (horizontal, vertical, whatever). Pull the trigger and then adjust the wire feed speed as you listen to the weld. Once you have that nice consistent sizzle sound then you're all set! This method is really quick and easy. It also applies all MIG welders no matter what type of controls they have. I use this method now whenever I have to readjust my machine for material I haven't already been welding on. It sounds from your posts you don;t find Youtube videos helpful. I can say there's one which is GREAT for demonstrating this method. You don't really even need to watch it, just listen and you can hear it!



The only thing I've welded any continuous runs on was 16ga and thicker. Even then it wasn't a bead longer than an inch or two. You need to keep your runs short on thin metal and space them out as far as possible. This helps minimize warping. You just have to go back and forth, splitting the distances until you have the full run of the joint welded up. Don't be afraid to keep an air gun handy and blow air on the weld either! I have welded all of my plug (or rosette), welds in a single shot. You still need to alternate spots to maintain distance. Warping isn't as prevalent, but can occur.

Once you get you machine set up properly there's one other thing to be aware of with thinner metals. Square corners burn through easily. When I do any patches I always round the corners slightly on the patch and opening to help minimize this. If you're butt welding (pieces placed edge to edge), you want a gap approximately the distance of your welding wire. Keep in mind that when you reach the ends you'll have square corners!

What size wire are you using? For sheet metal I try to stick with .023. I have used .030 when I've had it loaded into my machine and was too lazy to swap it. It burns a little hotter so you need to be mindful of that. I also use a little shorter stick out on the wire with .023 than I do with .030. Again, this affects the heat. As far as pattern goes I normally do a very small "C", and I do mean small! Imagine a line passing through the center of a "C" horizontally to represent the gap. I start at one end of the "C" and follow it around and then back again as I work down the bead. Some may refer to it as a zig-zag I guess. Either it or the straight are correct welds. I've tried both and settled on the one I use.

The final advice I can give is the same thing the violinist was given when he asked how to get to Carnegie Hall.....practice, practice, practice!

p.s. The type of weld joint in your your picture is called a lap joint.

[69GT350H]

Also not a pro, but Pos vs neg on my welder is determined if I am using Gas for Flux. I learned, use gas! And with gas, neg on the tip.

My apologies if this was a given.

[JKWilson]

Also not a pro, but Pos vs neg on my welder is determined if I am using Gas for Flux. I learned, use gas! And with gas, neg on the tip.

My apologies if this was a given.

Not to be rude or argumentative, but standard GMAW welding (MIG), is done with a positive electrode/negative ground. You can set up either way for specific results in the weld, but as stated the electrode is normally set for positive polarity (DCRP). To clarify a little more, I am speaking specifically about shielded (gas), electrode and not flux core, non-shielded welding.

If your method works for you stick with it by all means. From the questions the OP posted though they appears to trying to learn the basics. Swapping polarities outside of the norm is a little more advanced and can cause some confusion.

[69GT350H]

Not to be rude or argumentative, but standard GMAW welding (MIG), is done with a positive electrode/negative ground.

Not at all, as I said, on my Welder....   There is a sticker that states that flux core is one way, and gas the other. Its a 220v 35-110 amp Chicago Electric from Harbor Freight, and maybe that is the reason...   But definitely follow your welders instructions.

[Angela]

Thanks JKWilson for your excellent response. It must have taken some time to write all of that down, and I appreciate it.

I checked my welder and it does indicate that when using shielding gas (which I am), one should use electrode positive. I just could have sworn I'd been told that electrode negative was better for very thin metal. It would appear I misunderstood...

Since you asked, my welder is a Miller 210, which cn be used with either 220V or 110V. I've been using it on 110V, yet I certainly could use it with 220V and see if that helps when welding 22G sheet metal. From your response, it appears that switching to use of 220V will NOT help.

I practiced some more, using actual (scrap) metal from the car in question. I still am not able to weld together two sections of scrap 22G metal without warping the heck out of it.

So, this leads me back to my original question: based on the welding in the photo I attached above, should I continue pursuit of fully welding these seems, -or- should I use a product such as "All Metal" http://uschem.com/index.cfm?page=productDetail&id=54  to fill in the seam ?   I personally know people who have used the All Metal product for this exact purpose. However, it makes me nervous to do so, because I find it difficult to convince myself that the All Metal will keep moisture out of the seam. Feedback?

[69GT350H]

And with gas, neg on the tip.

And I stand corrected, as I went back out in the light of day and looked at my MIG again, it is set to tip pos for gas. Sorry for any confusion.

[JKWilson]

Thanks JKWilson for your excellent response. It must have taken some time to write all of that down, and I appreciate it.

I checked my welder and it does indicate that when using shielding gas (which I am), one should use electrode positive. I just could have sworn I'd been told that electrode negative was better for very thin metal. It would appear I misunderstood...

Since you asked, my welder is a Miller 210, which cn be used with either 220V or 110V. I've been using it on 110V, yet I certainly could use it with 220V and see if that helps when welding 22G sheet metal. From your response, it appears that switching to use of 220V will NOT help.

I practiced some more, using actual (scrap) metal from the car in question. I still am not able to weld together two sections of scrap 22G metal without warping the heck out of it.

So, this leads me back to my original question: based on the welding in the photo I attached above, should I continue pursuit of fully welding these seems, -or- should I use a product such as "All Metal" http://uschem.com/index.cfm?page=productDetail&id=54  to fill in the seam ?   I personally know people who have used the All Metal product for this exact purpose. However, it makes me nervous to do so, because I find it difficult to convince myself that the All Metal will keep moisture out of the seam. Feedback?

I'll try to address your comments/questions one at a time.

Regarding the settings for polarity there are two answers. The standard set up is for a positive electrode. You can use a negative electrode to weld (with gas). It can be a more difficult task, but doable. Why? Because a negative electrode produces a cooler weld and can be more useful on thin metal. The problem with it is a less stable arc, greater spatter and lower penetration. I didn't go into this because you had mentioned having issues with penetration and bead height. To respond outside of the "norm" would kinda be like trying to help you to run before you can walk (no insult intended, just the best analogy I could come up with).

Many "newer" machines now offer dual voltage inputs (and I wish I had one!). When I welded the sheet metal on my fastback in the '90's I used a straight 220v machine. It can be done with a little practice and fine tuning. Since you have to option of using 110v I'd strongly suggest you use it. It will make your job easier IMHO. There is NOTHING on the car you can't properly weld with 110v. If you're building roll cages and such, well, that's a different story!

You're going to get warping..period! How much is what you want try to control. If you try to weld the entire length of a quarter panel and go at it non stop (even using proper spacing), you'll warp it. You can't rush it, you have to let things cool before moving on. You can also dolly down your welds while they're still warm to take out some warping. Doing this allows you to use the heat to your advantage. Be aware that MIG welds are "harder" than TIG welds. That's one reason why customizers like to TIG their work. The beads require little grinding, there's very minimal warping because of greater heat control and they dolly down easily. Location of your seam/joint also impacts how much warping you get. If you try to join a patch across the middle of a door skin, quarter panel, anywhere there is a lot of real estate, you'll warp the crap out of it. If you HAVE to seam there, plan on a LOT of body work/leveling to get it back smooth! This is where planning your joints/seams comes in. Try to seam in an area with the greatest support (body panel lines, edges, on bends, etc). Not matter where you weld, don't heat soak it. Minimal heat for penetration and minimal dwell time. If your machine is set up properly for good penetration without blowing through, you may be lingering on it too long (hence the build up of your beads). As long as you're pulling the trigger the machine will be feeding wire which translates into more material in the weld (bead). You may want to try dialing your wire speed down a little too.

If the picture you posted is of the actual seam you're welding I'd have to ask, where is it on the body? It appears to be a floor pan, but kinda hard to tell. As such, without knowing this it's kinda hard to offer any kind of informed opinion/recommendation on filling or welding. If it is the floor pan (and right where the floor turns up to the trans tunnel), I'd weld it. It is NOT 22ga though. It should be 20ga at the thinnest, but good quality pans are 18ga. IIRC.  They can take much more heat (relatively speaking!).

There's a pic of my coupe's floor pans at the bottom of the post which I seamed (butt welded), along the trans tunnel. There is no filler and if you look close you can still see some remnants of the beads. It's not a concours car so I wasn't too concerne about inside the car. Underneath is smooth though. You will also not see any extreme warping:



One thing about the filler (no matter where it's applied), I assume you'd be coating over it, correct? A good epoxy primer will be moisture resistant to protect whatever is underneath it. If you're really concerned about it though, you can use one of the fiberglass fillers. It IS moisture resistant.

Always glad to help out when I can. Otherwise, I learn a ton off of the other folks on here!

[Angela]

WOW, nice welding on your floor pans!

Yes, the photo I uploaded in my original post is a section of the floor pan. Using the metal thickness gauge that came with my Miller welder, I measured a section of the removed metal and it measured 22G. That said, I have the same seems to mess with on lower sections of the quarter panels as well.

I was mistaken, fyi, my welder is a Miller 211.

Your comment about fully welding the seams if they're at the floor pan.... is that because you're hinting that filling the seams with ALL METAL might cause paint cracking when the pan flexes? I'm really worried about warping the heck out out of the car while trying to weld these seems. I'm terrible with a body hammer and even worse on "oil canning" situations. On the flip side, I sure don't want the filler to crack after there's five grand worth of fresh paint on it.

[JKWilson]

Thanks for the compliment, but they're really nothing special to be quit honest. They look good under the car and they're just okay inside. It didn't concern me much because it's not a concours car (if I didn't say that already), and it'll be covered with padding and carpet. Hope that gives you some confidence in what you can do! 

I asked about them being floor pans because minor warping isn't a real critical concern IMHO. Also because floor pans are thicker metal than body panels. This means they're a little more forgiving to heat. You might want to check your metal thickness with another gauge or for more accuracy, a set of calipers, micrometer, etc. The following are close : 22ga = .0299; 20ga = .0359; 18ga = .0478; and 16ga = .0598. I have seen the "fingers" get spread metal gauges causing errors in them. Also forcing the metal in can actually shave off a minute layer causing false readings. The metal should slip in with a hint of contact, just like a feeler gauge. Which, BTW, you can use to check your metal thickness gauge for accuracy!

Your quarter panel patches are thinner and warping is a concern. Just cut out as little as possible and stay as close to support areas (places where it welds to another panel or support, e.g pinch welds). You also REALLY want to do butt welds. Unless you flange one of the panels (patch or original), to meet the other, you'll never get the contour correct. I'd recommend doing the same for you floor pan. That's how the floor pans were replaced in the photo.

Work on your floor pans first to get the hang of it before tackling the quarters. This will build your skills and confidence.  BTW, the easiest and most accurate way to get a good even gap on a panel to butt weld is to overlap your panels then cut the pair with a body saw. The width of the blade determines the gap and it's a perfect gap for welding. Just make sure the rest of the panel is securely in position prior to doing this. If you need me to elaborate on this further I'll have to get a better explanation together, probably with an illustration.

I recall our conversation about your oil canning issues. You'll get the hang of it, just don't give up.  It also shouldn't be an issue here because you shouldn't be stretching the metal. Good luck with it!

p.s. I realized I didn't address your All Metal question. No, not a fear of it cracking. Just wanted to confirm what/where you were replacing the panel(s). Remember though, it's intended to fill, not seal. Most people use it to replace lead in leaded joints (high stress areas), like the window pillar joint, quarter panel to sail panel joint, upper trunk panel, etc...

[Angela]

Thanks again. Sure, I'll double check metal thickness with a pair of calipers.

All of the metal that needed replacing has already been replaced. However, I had help and the very experienced person who helped me spot welded all of the replaced sections around their perimeter. This included floor pan sections (the tunnel was left intact), a lower quarter panel section approx 6"x10" and the tail light panel. In some areas the person helping did go back and more fully weld along the perimeter of the repaired section. However, in 75% of the areas touched, the repair remains spot-welded. I know must determine how to proceed through completion. I am "OK" for most structural type welding and pretty darn good at body work and paint..... yet I'm not so good at sheet metal welding and I haven't a clue how to repair oil canning issue which I posted in a separate thread.

For sure, the oil canning issue in my quarter panel is kicking me in the rear. I've been trying to practice on an old "oily rag container" I picked up at a swap meet, which just happened to have an oil canning issue of its own. Let's just say that my efforts were only mildly successful in tightening the metal, and I made a horrible mess of the contour of the thing in the process.
I've bought a couple books on meat shrinking, but they don't describe the process by which one knows where to begin heating and hammering on an oil-canned area. I also have searched you tube with the same result....there exist a few videos of people repairing oil canning in cars, yet they don't mention where or why they heat specific areas. Until I understand where and why I should hold a torch to my quarter panel, I am convinced I'd only make things worse.

As far as my welding question, I agree that I should start with more welding on the floor pan seems. Should I be attempting to "simply" continue adding more and more spot-welds? Or, should I start laying down 1" long beads until the seam is fully welded?

Back to the ALL METAL question; I was told to use that product, instead of welding, to fully close the seams at my floor pan patches, quarter panel patches and where the roof lead was removed.  I've seem cars done this way, yet I wonder how they'll hold up after a year of use. Perhaps I'll call the manufacturer and see what they have to say. I suspect they won't want to answer the question.

[WT8095]

Note that the "spot welds" shown in your photo are not made the same way as the factory spot welds. The factory used a resistance welder, which runs a large electrical current through both pieces of metal, melting the metal in a spot where the two pieces were pinched between the electrodes. If you don't have this type of equipment available, the spot welds can be effectively recreated by plug welding. In plug welding, a hole is drilled in the top layer of metal, and a weld bead is placed inside the hole.

This video has a nice closeup view of the technique for plug welding:

[JKWilson]

Note that the "spot welds" shown in your photo are not made the same way as the factory spot welds. The factory used a resistance welder, which runs a large electrical current through both pieces of metal, melting the metal in a spot where the two pieces were pinched between the electrodes. If you don't have this type of equipment available, the spot welds can be effectively recreated by plug welding. In plug welding, a hole is drilled in the top layer of metal, and a weld bead is placed inside the hole.

This video has a nice closeup view of the technique for plug welding:

While commonly referred to as "spot welds", you are correct that the welds in the OP's picture are not. They are however plug welds (or more correctly, rosette welds), the same thing you were describing.

[JKWilson]

All of the metal that needed replacing has already been replaced. However, I had help and the very experienced person who helped me spot welded all of the replaced sections around their perimeter. This included floor pan sections (the tunnel was left intact), a lower quarter panel section approx 6"x10" and the tail light panel. In some areas the person helping did go back and more fully weld along the perimeter of the repaired section. However, in 75% of the areas touched, the repair remains spot-welded. I know must determine how to proceed through completion.

<snip>

As far as my welding question, I agree that I should start with more welding on the floor pan seems. Should I be attempting to "simply" continue adding more and more spot-welds? Or, should I start laying down 1" long beads until the seam is fully welded?

Back to the ALL METAL question; I was told to use that product, instead of welding, to fully close the seams at my floor pan patches, quarter panel patches and where the roof lead was removed.  I've seem cars done this way, yet I wonder how they'll hold up after a year of use. Perhaps I'll call the manufacturer and see what they have to say. I suspect they won't want to answer the question.

I'm kinda confused by your statement that all of the metal needing replacement has been replaced. Are you trying to say it's partially installed but needs finish welded? Also, based on the floor pan photo, if your "experienced person" helped/did this for you, in the future I'd inquire a little deeper as to what they're experienced in! I know that sounds pretty insulting, and well, it is...no way around it. If they're an experienced welder, so be it. The plug welds look pretty decent. If they're supposed to be experienced with body work/sheet metal replacement, well, they'd be considered a hack. Overlapping patch panels used to be somewhat common in the days before MIG welding was common in body shops. Panels were brazed in with a torch and the heat caused huge warping problems. Overlapping panels cut down on this..some. If the joint wasn't near an edge to run the overlap to, then they'd install the patch from the backside. My own father used to do it many moons ago. Some older folks continued to do it this way, even with a MIG, because "that's the way they've always done it". If nothing else, if the patch panel is an example of how your quarter patch is in, neither should have used the "spot welds". If you're gonna do it with a lap joint, they should have been tack welded at the seam. Your could have accomplished the same thing you have now by simply using self tapping screws to hold the panels together awaiting welding. All the "spot welds" did was create more work.

Would it be possible for you to post a few more pics of what you're trying to work with? From your statements it almost sounds like you're not fighting a true "oil can", but heat warping the panel from incorrect welding. I've made up an illustration to show the different types of joints found in sheet metal repair. They do not contain "spot welds", because where these are properly used are kinda out side what you're trying to do. They are used in full panel replacement (no seaming necessary there!). The picture is posted below. In the lap joint that's finished you can see how filler changes the contour of the panel. Not a good situation. It also illustrates the problems with either a lap or flange joint regarding water incursion points (specifically for your floor pan). If you want to eliminate this area for a lap joint, you'll need to weld the joint under the car...not fun! Trust me, it's MUCH easier to do all of your welding and grinding on the topside of the car if you can vice underneath it! Of course, with a rotisserie mounted car  the bottom can become the top easily.

Regarding the use of All Metal as a sealer, I can't stress enough...forget it. Drop the idea right now. It's a FILLER, not a sealer. Yes, it's waterproof. Yes you may get away with using it as a sealer, but why?! It's also THE hardest to work out of all of the fillers. When it sets plan on a LOT of elbow grease to level it out! If you are absolutely stuck with using a waterproof filler as a sealer, I'd suggest using a fiberglass filler like USC Duraglass or Evercoat Everglass. As far as cracking generally speaking none of them will crack if used in the proper application (correct areas), and applied properly (proper thickness). They will withstand flexing, but only to a very limited amount.

In closing, your concerns about fillers cracking after a few years under $5k worth of paint is a good line of thinking. EVERYONE is concerned about this...or should be! It appears you want to apply paint and filler properly so that it lasts. Why do you want to short cut the metal work by not properly welding and then making up for it with a filler? I'm not trying to insult, but just point out the realities of it. You certainly seem like you truly want to do the best job possible. For that..GOOD ON YA! I'm just like you, I want to do the best possible job. I HATE having to do things over because of poor or incorrect methods. It costs me time and money, not to mention the aggravation! I'm a firm believer in doing it right the first time!!

Here's the image I put together. See if it helps any to understand the points I'm trying to make and feel free to ask questions. Good luck!

[J_Speegle]

Would also like to mention that allot of restorers are no longer replacing sections or panels (unless there is no other choice) but rather replacing onlu what is needed even though this may result in 20 small 4x4" or slightly larger patch panels on their projects. All using but welds. In this way they maintain the factory panel to panel welds and date codes as much as possible