Lack of penetration on AL

[hotrodder]

I'd love to see a weld from somebody who knew what they are doing broken apart?.

It's softer sheet ally and nothing broke but FWIW http://www.mig-welding.co.uk/forum/media/albums/ally-weld-abuse.743/

Using brute force or a BFH to attack/break welds on stuff is fairly limited in usefullness... on something like the soft sheet example above it kinda works as a ductility test but on heavier stuff and/or things made from tube/pipe it's kinda useless because in real life things fail from fatigue unless whoever designed the part got things very wrong. This is why destructive weld testing typically involves cutting samples from a weld coupon to subject to nick break tests, bend tests and inspection of etched samples
Attack a sound weld coupon with a BFH and, after getting FUBAR'ed it'll usually break in the HAZ so won't highlight any porosity or small areas of LOF etc [/gross generalisation]

 

[Brad93]

He's in NZ

 

[Mid Wales welder]

For some time I have noticed I'm not getting full penetration on aluminum tube.

The settings;
200+Amps AC, 25% Cleaning (or 75, Kempii is backward here), 120Hz rounded squarewave.
12 LPM Argon, electrode 2% lanthanated 2.4mm. Filler Rod 4043, cleaned of course.
Aluminum is 6261 - slightly stronger version of 6061. Cleaned. Tube is 25 x 5mm.

The side faces were bevelled, the fillet joint not. I've tried gapped, bevelled, plain, plain gapped. All the same.

Yes - I did have a little oopsie and dipped the electrode. The electrode is running OK, tip kinda rounded, purple when cooled. The puddle looks OK, no cracks post weld. Some small bubbles in the weld after "disassembly".

The problem is that when I had to take some frames apart I noticed the penetration wasn't very deep. To get them apart you have to beat the daylights out of it ie. it's not really lacking strength. By day lights I mean big vise, 4lb hammer, many strikes.

Any thoughts ?

Hi Gavin, I see this all the time with Aluminium fillets. Getting the prep and setting right is obviously important, Brad and Hotrodder both have a lot of experience. But also you need to have a good technique, keep the tungsten as close to the arc as possible with not too much torch angle, then only add filler when you seen the front edge of the puddle melt down into the root. If you add filler before this you will just get lack of fusion.
Are you using a foot pedal or torch switch? I would be starting off on about 250amps as that's quite thick box.
Hope that helps.

 

[Hood]

Ok now bear in mind I am not an expert and just self taught so I do lots of things weirdly compared to convention. Normally I would probably be a bit lower on amperage, maybe 180ish.
Anyway a couple of pics, 5mm 5083 with 5356 wire and a quick wire brush as the Alu is a slightly grotty

 

[Hood]

Oh and was with a WP26 torch with standard collets etc and 8LPM

 

[Hood]

And just for a bit of fun I used the MIG on the same material and this was the result.
Now also I want to mention I am new to the EWM Pulse Mig so it will take a bit before I get fully into the swing of it, only had it a few weeks and not used it at all the past week and a bit so not much practice. That will change next week as I am back to fitting out the 10m Cheetah Catamaran

Anyway two pics, first the weld and lastly trying to break it.

 

[Richard.]

Looking at the pics the biggest fault I see is arc length which is causing your lack of pool control. Those fillets look to be twice the size of the material thickness which would suggest your arc length is too long and also causing your lack of fusion defects. This lack of fusion is known as tunnelling. Your two weld pools appear on both faces, you add filler which bridges the gap but you have a tunnel right in the corner of unfused material which is a big failure point on the joint. A. Marco test or nick break will pick this up better than a good ol fashion caning with a hammer.
I'd hit that with a light pre heat. Wouldn't need much but just enough to stop it sucking your heat out at start up because it's starting up that's key. Getting that pool going right in the corner from the off.
Your gas flow is too high and can have the same effects of too little by drawing in air from a turbulence effect.
I'd say your balance could go the other way personally. I weld at 80% DCEN quite often and it's a good way of utilising current limits because it will put more heat into the work and help with penetration. It also gives your tungsten an easier life and allows it to stay more in shape helping a little with arc focus. The compromise is oxide layer cleaning so be aware not to adjust beyond the point your not getting the cleaning needed. Play with it and see what you can get away with.
For a material thickness like that I would be bringing that frequency down. 120hz is where I'd be for something in the 2-3mm fillet range max. At those thicknesses I'd treat it like a thick butt joint and be around 70-80 hz.
Penetration with frequency is a terminology thing and I've often got the terminology wrong because the actual weld metal below the surface doesn't change with adjustment on freq. the thing the does change is the shape of pen. (DxW as Hr said.) so pen is narrow and deep with high and wider and shallower with low but heat getting in is more with low.
Once you lose pool control with an over sized arc length it doesn't matter what hz your running you are doomed with the LOF problem.

 

[Hood]

@Gavin Melville
Any progress updates?

 

[Gavin Melville]

Finally, an update, been "busy".

Getting there. Used a LOT of gas, and maybe learnt a few things.

6261 ally is not the same as 6061.
When you think the arc is short enough, it needs to be shorter.

The image of the weld face on looks undercut. It isn't. The etching process got left in the bucket too long.

Thanks for your help here.

Kind Regards,
Gavin.

 

[Richard.]

Finally, an update, been "busy".

Getting there. Used a LOT of gas, and maybe learnt a few things.

6261 ally is not the same as 6061.
When you think the arc is short enough, it needs to be shorter.

The image of the weld face on looks undercut. It isn't. The etching process got left in the bucket too long.

Thanks for your help here.

Kind Regards,
Gavin.

View attachment 112365 View attachment 112366 View attachment 112367

A million times better Gav.

 

[zeppelin builder]

That rules me out then

That looks about 8 mm thick if it is you will need to bevel to a sharp edge tack up at the corners and use a 3 mm gap do a root pass then aditional passes to fill and cap pre set the straight tube as it will pull or straighten it while its got a bit of heat in

 

[Brad93]

Lots better. Good clean etch, nipped the root on the left hand side but just missed it on the right.
I imagine you ran the side on the right first then it had a bit more heat in it on the other side?

Its just practice getting the puddle to flow right down into the corner.

 

[Richard.]

That looks about 8 mm thick if it is you will need to bevel to a sharp edge tack up at the corners and use a 3 mm gap do a root pass then aditional passes to fill and cap pre set the straight tube as it will pull or straighten it while its got a bit of heat in

A 3mm gap!!
It's a fillet joint
It doesn't call for full penetration just sufficient penetration and fusion. Square edged, plenty of heat and direction will see that into a perfect fillet joint.
Bevels on Al fillet joints can make more trouble than they are worth
You have a much sharper angle to try and get fusion into the corner and it's simply not required on gage like that. Other than a little light on the fusion (right side) and if I'm really been picky the leg is a tad big. I would say the op has hit that very nicely.

 

[zeppelin builder]

looks heavy wall to me that tubing if it was full pen i would prep and gap it ,when doing fillets you must wash the metal thoroughly in before adding filler
a lot of people are using a lowish current which is creating problems shouldnt have to hang around forming a pool ; pre heat enough to get rid of any water as
when you start welding the plates will sweat and water gets trapped in joint , i prefer a pedal too as you can nail it at the start and reduce a bit as you get heat in there ,most of the alu capping runs i see on here look too cold , humped up in the middle and not washed in to the toes,, the ripples look too widely spaced as well ,like dab of wire every 10 mm but maybe im being critical ,remember your cap is your signature

 

[zeppelin builder]

Lots better. Good clean etch, nipped the root on the left hand side but just missed it on the right.
I imagine you ran the side on the right first then it had a bit more heat in it on the other side?

Its just practice getting the puddle to flow right down into the corner. View attachment 114152

good job mate credit where its due

 

[Brad93]

That box section looks to me to be a bit of 1" x ¼".

I wouldn't prep the two filleted sides. I would just put a well fused ¼" fillet in.

I would however prep the outsides where they are more of a butt joint to be around 37.5 degrees and just put one hot pass in there to sit just proud of flush.

 

[zeppelin builder]

try telling a lloyds inspector[ its a bit of a butt joint] either full pen or not

 

[Brad93]

try telling a lloyds inspector[ its a bit of a butt joint] either full pen or not

I've never worked to lloyds standard so who knows what is required. I used to work to aerospace BCAR. Some of the specs call for backing gas on Aluminium!

We've got the lloyds specs at work, I'll have a look tomorrow.

 

[Richard.]

try telling a lloyds inspector[ its a bit of a butt joint] either full pen or not

Zep I'm a qualified welding inspector. Multiple passes on that joint would not be required. It would likely make things worse with the size of the haz you would achieve. In your line of work yes an open root full pen, multiple passes to fill the joint. You can't take that approach to a 6mm lump of Alu box nor can you tell me a welding inspector will want to see that either.
Welded properly (and the op has done a good job of showing it done well) square edged prep plenty of heat at start up, nicely blended toes like you rightly said and there is no reason you cannot get sufficient penetration, fusion, fill in one pass on that.

 

[zeppelin builder]

I've never worked to lloyds standard so who knows what is required. I used to work to aerospace BCAR. Some of the specs call for backing gas on Aluminium!

We've got the lloyds specs at work, I'll have a look tomorrow.

I've never worked to lloyds standard so who knows what is required. I used to work to aerospace BCAR. Some of the specs call for backing gas on Aluminium!

We've got the lloyds specs at work, I'll have a look tomorrow.

it all boils down to what the client wants i have in the past held the caa licence for full pen butt alu welding so i would like to give my help to anyone that will listen i started doing alu at 16 years of age im 59 now so done a few tricky one off things with tig and o /a my mate did procedures for AWS alu heavy wall pipe with 6mm filler a complete animal so we know our stuff up here scotland always been miles ahead in welding skills as we were the pioneers in the oil industry
in the pre thatcher boom years