When I’m welding aluminum I often get a lot of little gas bubbles that, for the most part, dont appear on the surface but only when I grind off some of the weld (buildup welding)
It happens when repairing cracks in automotive wheels and on thin 1.5mm sheet, but I suppose that has to do with oxigen being drawn in from the backside of the weld.
Last time I had to do some buildup welding on a relative thick gearbox adapter plate, which then got machined down to fit a sensor. When grinding down the weld I noticed it was filled with tiny bubbles although not present on the surface. During welding I also noticed some bubbles float to the surface. I switched to different travel speeds with the same results. I also let the part frequently cool and wire brush in between passes.
I cleaned the part with acetone/ss brush/acetone prior to welding, and cleaned filler wire with scotchbrite and acetone.
Setup:
Thermal Arc 202 AC/DC
CK flex head 26 air cooled torch
#8 cup with gas lens
10L/min gas flow 100% argon
3/32 2% lantanaded tungsten (clean)
1/8 4043 filler wire
Welder set to 200amps but when part heated up I used about 170amp with foot pedal
AC balance switched between 25%, 30% and 35% (same results)
Frequency switched between 60, 100 and 150hz (same results)
Postflow 8 seconds
Arc length was about the same as tungsten diameter
stickout less than cup diameter
No abnormal torch angle
Clamped the part to aluminum backing plate for heatsink
Base metal AW-6029 (never heard of it, various filler wire charts don’t even specify it, so that made me a bit skeptical)
Getting pretty frustrated about it because, as far as my knowlege goes, I ruled out every possible problem. It mosly happens when doing buildup work. when welding automotive intercoolers, aluminum plumbing etc. with single passes only, I never get any problems. To me it looks like some kind of hydrogen contamination but I have no idea how to prevent it besides the things listed above.
I’d suspect the base metal has some zinc as part of if it’s composition and the bubbles are from that metal. They look far too fine to be porosity or “operator flaws”. And you setups look to be appropriate.
Auto ice wheels are often an unknown composition of junk alloys. And drivetrain housings are often cast materials that are not very refined on metallurgical quality, just efficient processing.
Auto ice wheels are often an unknown composition of junk alloys. And drivetrain housings are often cast materials that are not very refined on metallurgical quality, just efficient processing.
I’ve also had the same problems when welding 5754 base metal while using 5356 filler wire, everything else was about the same as the setup above.
It was a 1/8” thick undertray of a car and I had to fill up some bolt holes (drilled them out to eleminate any junk), and 2 small cracks (used a jigsaw to cut the cracks for cleaning and stress releif, no paint or oil residues on the blade)
I welded them with a clean stainless steel backing plate, and then cleaned the backside of the welds with a dedicated aluminum carbide burr and acetone, and welded the backside.
Both sides looked flawless when I was done, but when I started grinding the welds to a flat finish, the same bubbles started showing up
It was a 1/8” thick undertray of a car and I had to fill up some bolt holes (drilled them out to eleminate any junk), and 2 small cracks (used a jigsaw to cut the cracks for cleaning and stress releif, no paint or oil residues on the blade)
I welded them with a clean stainless steel backing plate, and then cleaned the backside of the welds with a dedicated aluminum carbide burr and acetone, and welded the backside.
Both sides looked flawless when I was done, but when I started grinding the welds to a flat finish, the same bubbles started showing up
nice picsBartNL wrote: ↑Sat Jun 03, 2023 4:27 am When I’m welding aluminum I often get a lot of little gas bubbles that, for the most part, dont appear on the surface but only when I grind off some of the weld (buildup welding)
It happens when repairing cracks in automotive wheels and on thin 1.5mm sheet, but I suppose that has to do with oxigen being drawn in from the backside of the weld.
Last time I had to do some buildup welding on a relative thick gearbox adapter plate, which then got machined down to fit a sensor. When grinding down the weld I noticed it was filled with tiny bubbles although not present on the surface. During welding I also noticed some bubbles float to the surface. I switched to different travel speeds with the same results. I also let the part frequently cool and wire brush in between passes.
I cleaned the part with acetone/ss brush/acetone prior to welding, and cleaned filler wire with scotchbrite and acetone.
Setup:
Thermal Arc 202 AC/DC
CK flex head 26 air cooled torch
#8 cup with gas lens
10L/min gas flow 100% argon
3/32 2% lantanaded tungsten (clean)
1/8 4043 filler wire
Welder set to 200amps but when part heated up I used about 170amp with foot pedal
AC balance switched between 25%, 30% and 35% (same results)
Frequency switched between 60, 100 and 150hz (same results)
Postflow 8 seconds
Arc length was about the same as tungsten diameter
stickout less than cup diameter
No abnormal torch angle
Clamped the part to aluminum backing plate for heatsink
Base metal AW-6029 (never heard of it, various filler wire charts don’t even specify it, so that made me a bit skeptical)
Getting pretty frustrated about it because, as far as my knowlege goes, I ruled out every possible problem. It mosly happens when doing buildup work. when welding automotive intercoolers, aluminum plumbing etc. with single passes only, I never get any problems. To me it looks like some kind of hydrogen contamination but I have no idea how to prevent it besides the things listed above.
that looks more filler related than base metal. cheap junk filler?
my thoughts here is to drop cup size down to a 5. no gas lens if possible.
drop filler size down to 3/32, that should slow down the puddle freezing a tad which may help to let gas out and make less ripples which makes it better for the next pass.
also a long shot but check torch for gas leaks. sometimes you get it where it acts as venturi effect and sucks air into the gas flow.
the other is contamination. aluminium soaks up crap, gearboxes are always bad. but that should show up in the first pass. if its only showing up in the build up i doubt its a base metal problem.
tweak it until it breaks
4043 is not the best filler for 6000 series parent. I'd don't like it the best of times because of porosity and peppering.
5356 is indicated for 6000 series parent. If you suspect porosity is from parent alloy contamination, put a buttering layer of 5356 down first, grind it back to just above flush and then fill out the joint/repair. That should eliminate the contamination, however I suspect the problem is the 4043. Also if you are using 1/8" filler, make sure you are putting plenty of heat in to flow the filler. If you don't have enough grunt, drop back to 3/32".
5356 is indicated for 6000 series parent. If you suspect porosity is from parent alloy contamination, put a buttering layer of 5356 down first, grind it back to just above flush and then fill out the joint/repair. That should eliminate the contamination, however I suspect the problem is the 4043. Also if you are using 1/8" filler, make sure you are putting plenty of heat in to flow the filler. If you don't have enough grunt, drop back to 3/32".
Flat out like a lizard drinkin'
just to add, its hard to tell from pics but i would be overlapping the weld a lot more. flatter beads, closer together, quicker dabs. you will have to do more beads.
part of that is all the high points is wasted as it all gets machined off,
but also you want to avoid getting oxide stuck down in the toes where it doesn't always get cleaned out by the arc (same problem can happen with butt welds or cracks). oxides in the weld will cause porosity.
a lot of those bubbles look inline with the edges of your beads.
but its harder to tell on screen than in real life.
part of that is all the high points is wasted as it all gets machined off,
but also you want to avoid getting oxide stuck down in the toes where it doesn't always get cleaned out by the arc (same problem can happen with butt welds or cracks). oxides in the weld will cause porosity.
a lot of those bubbles look inline with the edges of your beads.
but its harder to tell on screen than in real life.
tweak it until it breaks
AJ Clayton
- AJ Clayton
-
New Member
-
Posts:
-
Joined:Mon Jun 19, 2023 10:51 am
Aluminum can be a problem child in so many cases. I’ve had similar problems. Had better results using 5356 than 4043 in so many jobs.
Clean with ss brush pushing in only one direction then go to a carbide burr and grind area down. (No abrasives of any kind.) Thick material not cracked go down deep as first bead pass. Pre heat material to 500+ degrees or so.
Pre heating helps with having to use less amps and gets out a lot of moisture and gasses.
In most of my cases if it’s going to have porosity it can be seen in weld puddle or during process. If so get back on it with the burr then not later. Grind out weld and start over.
Second try is always better 90% of the time.
Clean with ss brush pushing in only one direction then go to a carbide burr and grind area down. (No abrasives of any kind.) Thick material not cracked go down deep as first bead pass. Pre heat material to 500+ degrees or so.
Pre heating helps with having to use less amps and gets out a lot of moisture and gasses.
In most of my cases if it’s going to have porosity it can be seen in weld puddle or during process. If so get back on it with the burr then not later. Grind out weld and start over.
Second try is always better 90% of the time.
Thanks for the replies, its appreciated!
@Tweake:
I’ve also had good results with the exact same filler wire, I dont know if there is a possibility that good and bad wires are in the same package but yeah….ive seen crazier things so why not.
my idea with the #8 cup was that because of everything getting so damm hot, maybe its better to keep my last dab shielded a little longer to prevent contamination, but I’ll try a standard collet next time.
The reason why I chose 1/8 filler is also to keep the heat down a bit by cooling puddle more and not stay at the same spot for too long.
I checked for gas leaks and I’m aware of the venturi effect but everything is sealed off nicely.
The part came fresh off the CNC so no gearbox contact pre-welding.
I’m afraid I also can’t tell if the bubbles are on the edges or not
Next time I’ll stack the beads closer and see what happens.
The reason why I didn’t preheat the part was because I didn’t need full penetration welds, and trying to prevent the part from warping too much.
I did notice some bubbles popping up in my beads during welding, they appeared like little pimples so I stayed like a second or so longer on that spot until it pops open. I guess that was the sign to stop, grind, and switch to 5356???
By the way, pre-welding I zigzagged the base metal a couple of times without filler and moderate amps to let the cleaning action do its thing, warm up the part, and cooking out any possible surface junk, good or bad choice?
@Tweake:
I’ve also had good results with the exact same filler wire, I dont know if there is a possibility that good and bad wires are in the same package but yeah….ive seen crazier things so why not.
my idea with the #8 cup was that because of everything getting so damm hot, maybe its better to keep my last dab shielded a little longer to prevent contamination, but I’ll try a standard collet next time.
The reason why I chose 1/8 filler is also to keep the heat down a bit by cooling puddle more and not stay at the same spot for too long.
I checked for gas leaks and I’m aware of the venturi effect but everything is sealed off nicely.
The part came fresh off the CNC so no gearbox contact pre-welding.
I’m afraid I also can’t tell if the bubbles are on the edges or not
Next time I’ll stack the beads closer and see what happens.
The reason why I didn’t preheat the part was because I didn’t need full penetration welds, and trying to prevent the part from warping too much.
I did notice some bubbles popping up in my beads during welding, they appeared like little pimples so I stayed like a second or so longer on that spot until it pops open. I guess that was the sign to stop, grind, and switch to 5356???
By the way, pre-welding I zigzagged the base metal a couple of times without filler and moderate amps to let the cleaning action do its thing, warm up the part, and cooking out any possible surface junk, good or bad choice?
Good choice. You can also use a different balance setting (more cleaning) while zig-zagging. Then switch to less cleaning/more penetration once you begin welding. That way you can use less amps to weld and aide in controlling your distortion.
one thing to check is the CNC coolant. seen it a few times where they changed the band saw coolant and suddenly the welding goes to crap.BartNL wrote: ↑Thu Jun 22, 2023 1:12 pm Thanks for the replies, its appreciated!
@Tweake:
I’ve also had good results with the exact same filler wire, I dont know if there is a possibility that good and bad wires are in the same package but yeah….ive seen crazier things so why not.
my idea with the #8 cup was that because of everything getting so damm hot, maybe its better to keep my last dab shielded a little longer to prevent contamination, but I’ll try a standard collet next time.
The reason why I chose 1/8 filler is also to keep the heat down a bit by cooling puddle more and not stay at the same spot for too long.
I checked for gas leaks and I’m aware of the venturi effect but everything is sealed off nicely.
The part came fresh off the CNC so no gearbox contact pre-welding.
I’m afraid I also can’t tell if the bubbles are on the edges or not
Next time I’ll stack the beads closer and see what happens.
The reason why I didn’t preheat the part was because I didn’t need full penetration welds, and trying to prevent the part from warping too much.
I did notice some bubbles popping up in my beads during welding, they appeared like little pimples so I stayed like a second or so longer on that spot until it pops open. I guess that was the sign to stop, grind, and switch to 5356???
By the way, pre-welding I zigzagged the base metal a couple of times without filler and moderate amps to let the cleaning action do its thing, warm up the part, and cooking out any possible surface junk, good or bad choice?
take a coupon, heat it up and dunk one half in the coolant and let it cool down. then weld a bead across the length and see if you get bubbles on one half. that should show if its coolant related. (i can't believe i didn't think of that before, sorry).
if you havn't used a small cup on aluminum before you might be in for a surprise. it works really well, even where you would split the gas flow like outside corner.
gas coverage is not that critical, its going to oxidize the instant the gas is off it anyway and in the arc area the cleaning action breaks down the oxide. whats really interesting is standard tends to work better than gas lens.
its like everything is the complete opposite of stainless.
running the torch over before welding is usually a good idea especially on repairs.
tweak it until it breaks