Cracking issues on 1/4" aluminum tin welded butt joints - 5052 -

[Andrew J 649]

(can't seem to edit the post title but auto-correct changed "tig" to tin)
Hello All. Im fabricating another batch of planter boxes in 1/4" 5050, but I'm having major cracking issues this time around. Though the one I've attached a picture of is a funny shape, most are rectangles. All have a 32" long butt joint in the middle, joining two halves. The client wanted all bent corners so moving the joint to the middle seemed a good way to go. Last time however the joints were at the corners, and as some of you may remember, i did a series of shorter (2" or so) welds into beveled out areas.
Im first cutting a vee on both sides of both pieces (about a 1/3 of the depth) then butting the parts and clamping them down to the table. I make a series of short welds, about 1" long, about 5" apart. Then i flip the box and do the same on the other side, clamping bars across the surface to hold everything flat while it is off the table.
Settings and details:
Sync 250, #20 water cooled torch, 210 amp setting, ac polarity, ac balance 7-8 toward max penetration, 100% argon, 1/8" 2%lanthinated tungsten, 1/4 - 5/16" stick out, #8 and #10 alumina nozzles, medium gas lens, 15 second post flow, trying both 5256 and 4043 at 3/32"
The welds for the most part look good and go down smooth and shiny. The tungsten is staying well shaped and silvery. But when i flip the part and weld the other side, i crack the welds on the first side. Ive managed to get one of these together by doing very short welds, and grinding out cracks (with a carbide burr) and re welding, but there likely is some way to make longer welds and not get cracking, no? It seems like the longer my welds are, the hotter the part gets (obviously), the more the other side cracks.
Any advice is greatly appreciated.

[H80N]

Hello All. Im fabricating another batch of planter boxes in 1/4" 5050, but I'm having major cracking issues this time around. Though the one I've attached a picture of is a funny shape, most are rectangles. All have a 32" long butt joint in the middle, joining two halves. The client wanted all bent corners so moving the joint to the middle seemed a good way to go. Last time however the joints were at the corners, and as some of you may remember, i did a series of shorter (2" or so) welds into beveled out areas.
Im first cutting a vee on both sides of both pieces (about a 1/3 of the depth) then butting the parts and clamping them down to the table. I make a series of short welds, about 1" long, about 5" apart. Then i flip the box and do the same on the other side, clamping bars across the surface to hold everything flat while it is off the table.
Settings and details:
Sync 250, #20 water cooled torch, 210 amp setting, ac polarity, ac balance 7-8 toward max penetration, 100% argon, 1/8" 2%lanthinated tungsten, 1/4 - 5/16" stick out, #8 and #10 alumina nozzles, medium gas lens, 15 second post flow, trying both 5256 and 4043 at 3/32"
The welds for the most part look good and go down smooth and shiny. The tungsten is staying well shaped and silvery. But when i flip the part and weld the other side, i crack the welds on the first side. Ive managed to get one of these together by doing very short welds, and grinding out cracks (with a carbide burr) and re welding, but there likely is some way to make longer welds and not get cracking, no? It seems like the longer my welds are, the hotter the part gets (obviously), the more the other side cracks.
Any advice is greatly appreciated.

.

Here is an excerpt from a Fabricator article Best practices for welding aluminum

What Causes Cracking in Aluminum Welding?

Hot cracking and stress cracking can occur during aluminum gas metal arc welding (GMAW) and gas tungsten arc welding (GTAW) processes. Both types of cracks, even when small, can prevent welds from meeting code requirements and can eventually lead to weld failure. Hot cracking is predominantly a matter of chemistry, while stress cracking is the result of mechanical stresses.

Three main factors increase the probability that hot cracking will occur during aluminum welding. The first factor is how susceptible the base material is to cracking. For example, some alloys like the 6000 series are more prone to cracking than others. The second factor is which filler metal you use. Third is joint design—some joint designs restrict the addition of filler metal.

Stress cracking can occur when an aluminum weld cools and excessive shrinkage stresses are present during solidification. This could be due to a concave bead profile, a too slow travel speed, a highly restrained joint, or depression in the end of the weld (crater crack). How Do I Stop Cracks From Happening?

In some cases, preventing hot cracking can be as simple as choosing a filler metal with a weld metal chemistry with lower crack sensitivity. Each aluminum filler metal has an American Welding Society (AWS) classification that corresponds to the Aluminum Association registration number, and together the two identify the particular alloy chemistry.

Always reference a reputable filler metal selection guide to make the best choice because not all aluminum filler metals are suitable for every aluminum base material. Some filler metal guides give recommendations specifically for several weld characteristics, such as cracking, strength, ductility, corrosion resistance, elevated-temperature service, color match after anodizing, postweld heat treatment (PWHT), and toughness. If cracking is a concern, select the filler metal that has the highest rating in the cracking category.

Also, using an appropriate joint design can help prevent hot cracking. For example, a beveled groove joint is a good option because it allows for the addition of greater amounts of filler metal, which increases the amount of base metal dilution, making it less prone to cracking.
It is possible to prevent stress cracking by using a filler metal containing silicon. When allowable, this type of filler metal lowers shrinkage stresses, particularly in crack-sensitive areas like the beginning and end of the weld (or craters). Also, use an automated crater fill function or other approved methods of crater filling to minimize the opportunity for cracking to occur in the crater. Increasing your travel speed also can help decrease the opportunity for stress cracking on aluminum by narrowing the heat-affected zone (HAZ) and reducing how much the base metal melts.

Preheating is also an option to combat stress cracking because it minimizes the residual stress levels that are present in the base material during and after welding. Closely monitoring the heat input is key to making this work. Too much heat can lower the base material’s tensile strength in some alloys to unacceptable levels.

http://www.thefabricator.com/article/aluminumwelding/best-practices-for-welding-aluminum

[H80N]

You might also consider trying 4943 filler..... it has higher silicon content and has low hot cracking sensitivity....

http://www.hobartbrothers.com/uploads/pdf/datasheets/Hobart_Maxal_4943.pdf

http://www.hobartbrothers.com/produc...%26reg%3B+4943

[Olivero]

Never heard of that, I think you need more weld in there, looks almost fused which aluminum seems to crack very easily when fused without much filler.

Remember that once the puddle is molten, the metallurgy changes of the aluminum and so your filler helps make up for some of those short comings.

[ryanjones2150]

What filler metal are you using?

[lars66]

Leave a little gap between the plates so they don't pull into tension as the weld is cooling. might take a little trail and error to figure out how much gap.

[Andrew J 649]

Thanks all. Really good info in those links H80N (fabricator, hobart), and id love to try some of that 4943. Olivero, there is a pretty good amount of filler being laid down. The photo is a bit deceiving i think. Its 1/4" material and the vee is nearly 3/16" total width (about 3/32" bevel on each side) and when i weld I'm getting a joint that is proud by a heavy 1/16". Anyhow, might try going deeper, but at that point the bevels will be touching. Ryanjones2150, i started with 5356 which i tend to use when the base is 5052. I am also trying 4043, but getting the same result. Lars66, a gap i haven't tried and maybe will, but i don't have any spare pieces as big as the actual parts so not sure if a test on a smaller section will tell me much, but I'm going to try it. What about starting on a cold table? Is there a chance that heat is creating a steam/condensation on the back while i weld the front, trapping bad elements in the weld? I think ill try blocking up off the table 1/4" or so, if anyone thinks that could be an issue. Or warming the table, but that seems a bit nuts. Also, if welding a long butt joint like this, is it important to start from the middle and work outwards? Im not sure i did this, and i might be creating tension.

[aametalmaster]

I would have migged it. But that's just me...Bob

[ryanjones2150]

But that's because you're the aluminum-mig-master, Bob.

Andrew, I have not experienced the problem you are having. Most of the aluminum fab work I do is with whatever aluminum I can salvage from the scrap yard. No telling what grade it is. The joint that's cracking...is it directly under the second joint you're welding? If it is, maybe you're melting into the first weldment a bit too much and sucking some of it back into the new weld. And is there any particular reason you're beveling this butt joint? Seems that if you're welding it from both sides that it would be unnecessary for a planter box unless you need 100% penetration for some reason.

The only time I can remember really having a situation similar to yours was a repair of some tool steel parts that I welding with 309. Both sides beveled, welds exactly opposing one another. Ever time the second side was welded, the first side would crack. I'd go back and let the heat soak in good and weld the first side again a little slower and that seemed to work. The stresses from the weld metal solidifying are quite extreme.

On the filler metal...I was doing some small parts for a job I have and I was alternating between the 4043, 5356 and 4943....have to say, I really like the 4943. Wets in nice, just seems to have an overall better workability in the weld puddle and it looks good when you're done.

[FusionKing]

You simply are not getting enough penetration.
Your cannot weld aluminum in the fashion you are and not go through what you are going through.
Throw out all the steel thinking, that people get away with.
The prep you are doing is simply not enough.
You need wider, deeper,and hotter.
The welds are not long enough. Skipping and then flipping would be how I would do it if I wanted to demonstrate how to make it crack.
If you make a gap, you will have to tack weld a lot. Then the tacks may or may not crack.
It doesn't appear you are getting enough parent metal dilution.
In time you will learn what you can and cannot get away with. It isn't the filler. Just simply prep and weld sequence, along with weld spacing in relation to the sequence, as well as how much heat you get into the part before you move to the next weld.
I'm not wanting to come off as a know it all. I just recognize the type of problems that are common with welding aluminum plate and just how fickle it can be. IMO you need to be way more generous with the beveling, amperage and filler.
2 welds for 1/8" don't equal a procedure for 1/4" on aluminum. In other words 1/2 on each side don't cut it. If you don't go deep then you need a very high crown and width.
BTW very nice work. It looks like you are making some really nice stuff. Hopefully you are making enough to make it all worth it.

[shovelon]

You simply are not getting enough penetration.

You need wider, deeper,and hotter.
The welds are not long enough. Skipping and then flipping would be how I would do it if I wanted to demonstrate how to make it crack.

And thicker. There just is not enough weld filler reinforcment to counter the pull of the base material. Like FusionKing says base metal dilution is key to strength if you use 4043. You won't have to worry about base metal dilution for 5356 or 4943 as they both don't need to pickup alloying agents the base, just thicker.

[SPW]

Preheat

[Andrew J 649]

Thanks everyone that weighed in on this. Big help. All seems to be going fine now. I can't say which of the changes made the critical difference, but i can tell you what i changed.
Pre-heat - I wasn't doing any before, and the shop is a bit cold now and there is no heat on. So I'm warming the parts up to what i'd guess is about 150f.
Spacing - Im now gapping the seam by about 1/32". This seems to make a big difference in how the weld goes down. I need to use more filler and fill faster but thats probably helping too.
Set up - originally i was starting with the parts clamped to the table. The table is cold and there is a lot of surface under these parts.. i could be wrong but i'd think there would be some condensation happening between hot aluminum parts and a cold steel table, and if so this could be getting into the weld. So I'm up off the table on aluminum square tube.
Amperage - I bumped up from 190 to about 210, and am now going full pedal through the whole weld and going fast.
Filler - Ive gone back to the 5356 and is working fine. 3/32" dia. So i think i can rule out the filler as the problem
Order of welds - Im now starting from the center and working outward. Tack on center, then to the left and right of that, then to the left and right of those, til I'm at the outer ends.

Im still flipping the parts as i was, doing a bit on each side of the seam, but much less so. Im laying longer tacks to start, closer to 2" now instead of 1". And then laying 3" - 4" welds after that.

Again, thanks everyone. Couldn't have moved forward without this input.