[SMFRyan]

Greetings all,
So I have a job here that I am hoping to get some advise on. These are aluminum filter cans, the tubes are appx. 5.563" diameter 1/4" wall with a square cut on the end. The caps,are appx. 13/16" thick and profiled to form a single bevel weld groove.

Im using a miller synchrowave 250,
red 1/8" tungsten balled using DC+
#6 gas lens
100% argon ( can mix in some helium if needed)
4043 filler
positioner

Going to preheat a little, clean it with a stainless wire brush and acetone then do 1 pass to fuse it then without stopping a second pass to add filler. These always come back from testing with leaks, at least 50% of them, whereas the shop our customer was using before us (out of business now) never had any leaks.
I havent tried any yet, I am hoping to get some advise on technique.
How should the torch be aimed at the groove?
Would it help if the can was beveled as well to create a v groove?
Should the tube be laying on its side or standing like in the pics?
Any advise at all is greatly appreciated.

[Bill Beauregard]

You haven't said the pressure it'll hold. I would do it in one pass with 1/8 or 3/32 4943. Use enough heat to get penetration, dab filler as it sags.

[LtBadd]

Greetings all,

Im using a miller synchrowave 250,
red 1/8" tungsten balled using DC+
#6 gas lens
100% argon ( can mix in some helium if needed)
4043 filler
positioner

Did you mean to write that you're using DC- (straight polarity)?

If so you should be using 100% helium and a sharpened tungsten

[SMFRyan]

Greetings all,

Im using a miller synchrowave 250,
red 1/8" tungsten balled using DC+
#6 gas lens
100% argon ( can mix in some helium if needed)
4043 filler
positioner

Did you mean to write that you're using DC- (straight polarity)?

If so you should be using 100% helium and a sharpened tungsten

No sorry, AC 100% argon.
I am just using DC+ to ball the electrode.
Also I just noticed, I am using a standard #6 cup not a gas lens, all of my lenses are 3/32".

I will give one pass a try, would be nice if that works.

How much pre-heat would you guys recommend?


And sorry for the huge pics

[Bill Beauregard]

300 degrees F

[Oscar]

How is your tungsten sharpened (picture?)
Where are you aiming your tungsten (picture?)
Are you able to fuse the root first, or are you bridge'ing the adjacent walls first?

[SMFRyan]

Thanks Bill.

Oscar, yes I can do a root pass. That way seems to be working better for me than 1 pass.
I did 4 this way and they all passed the leak test.
I am trying to keep the torch around 11 to 12 o'clock on the tube and angle slightly towards the tube.

[Oscar]

Looks good, but with a balled-tungsten that size, I would be thinking arc-wander might have caused incomplete root fusion. Are you using pure tungsten? I ask because if you can avoid it, you should. 2% lanthanated can hold a blunt end/tiny "ball" much better to get fusion straight "down" the axis/centerline of the tungsten, but only if your welder has the necessary adjustability. Keep in mind I'm just a hobbyist in welding/fabricating, but I do a lot of "testing" and research into welding and arc dynamics, so to speak. The larger the tungsten, the better the tip "holds" without erosion, generally speaking.

I would be aiming for a grind kinda like the 2nd from left-to-right in this pic, courtesy of weld.com that I found on google Images:




For reference in case you haven't seen it:

[LtBadd]

2% lanthanated tungsten will hold its shape better then 2% thoriated, thoriated is typically best suited for DC welding

tungsten chart.jpg (71.16 KiB) Viewed 3286 times

[SMFRyan]

Oscar your first pic where you are showing the tungsten grind isnt showing up for me.
We are about to get hit with some awesome winter weather so Monday I will make a trip to get some Lanthinated electrodes and 4943 filler.

Really appreciate the help everyone!

[Oscar]

Let me attach it here again (you might have to clear your browser's cache, as my pic is hosted from Google and I've heard sometimes this happens to others who are trying to view it)



Here is the direct link to open in a separate browser window.

So you didn't reveal exactly what tungsten you are currently using? Was my guess of pure tungsten correct?

[SMFRyan]

It worked that time! Oddly its showing up the first time you posted it as well now lol.

Using 2% thoriated, Started out gorund like the 2nd from the right in the pic (30 degree) and balled slightly, it completely balled itself after welding 1 housing so I just ran with it.

So I sent 4 of these for testing and none of them leaked! And I have no idea what kind of pressure they have to hold, I will ask next time I talk to the customer. I would imagine at least 150 psi. and I would assume they need to survive higher psi then that during the test, however they may just be checking for leaks and putting faith in the welder as far as strength is concerned. I have about 150 of these things to do, aiming for zero failures

[Oscar]

It worked that time! Oddly its showing up the first time you posted it as well now lol.

Using 2% thoriated, Started out gorund like the 2nd from the right in the pic (30 degree) and balled slightly, it completely balled itself after welding 1 housing so I just ran with it.

So I sent 4 of these for testing and none of them leaked! And I have no idea what kind of pressure they have to hold, I will ask next time I talk to the customer. I would imagine at least 150 psi. and I would assume they need to survive higher psi then that during the test, however they may just be checking for leaks and putting faith in the welder as far as strength is concerned. I have about 150 of these things to do, aiming for zero failures

My bad! I missed the "red" part in the original post! Sorry about that. Also, 2% ceriated works pretty good. I just got some in 5/32" diameter for my new TIG.

Also, did you wire brush the inner "parts" of where the tops meet the cans? Kinda like "behind the weld joint area..



Basically, what I'm getting at, is the cleaner the aluminum behind the puddle, the better it fuses:

Go to 11:45 in this video. See how clean the aluminum is inside the pipe and how it fuses nicely
svSi5_ZLFn4

Also, Jody did a similar video:
RJk0li0fbIo

[SMFRyan]

All good!
Yes sir I wire brush the inner parts plus the top of the caps and about 1" down both inside and out of the tubes. Basically anything thats anywhere near the weld gets cleaned. Surface finish is not a concern.
Ill check out the videos at lunch.

[tweake]

Using 2% thoriated, Started out gorund like the 2nd from the right in the pic (30 degree) and balled slightly, it completely balled itself after welding 1 housing so I just ran with it.

i would get better tungsten. 2% lathinated or a tri mix.
go up a tungsten size or reduce cleaning action and/or use some helium. basically try to reduce the amount of tungsten balling. you need a bit of balling for cleaning action to work properly but to much can make it very wide arc but you need to focus it more into the root.
also look at using high frequency AC.

[Oscar]

also look at using high frequency pulse.

On AC, a lot of machines limit how much pulse adjustability there is due it interfering with the wave shape's polarities. HTP limits there machines to 10pps, the Lincoln Aspect 375 limits it to (I think 25% of the main welding amperage). I think Miller doesn't have any caps on AC pulsing, but with the wrong pulse frequency it can change the wave amplitudes incorrectly.

[tweake]

also look at using high frequency pulse.

On AC, a lot of machines limit how much pulse adjustability there is due it interfering with the wave shape's polarities. HTP limits there machines to 10pps, the Lincoln Aspect 375 limits it to (I think 25% of the main welding amperage). I think Miller doesn't have any caps on AC pulsing, but with the wrong pulse frequency it can change the wave amplitudes incorrectly.

sorry that should be AC frequency rather than pulse. (wheres my coffee.........)

[Simclardy]

Can you slow down the positiner to get better penetration, and maybe tighten up the ripples a bit? Just a thought. Did you get to see the failed units? Crater cracks? Any clues?
Good luck

Sent from my SM-G930V using Tapatalk

[SMFRyan]

I am stuck with 60Hz square wave and no pulser unfortunately. Some day I will buy a good inverter machine with all the bells and whistles.
The positioner is already almost as slow as it can go and if I up the amperage any it boils the aluminum. I can however try to tighten the ripples a bit.
The ones with failures seem to be due to porosity. The welds looked fine but the there would be pinhole leaks without being able to see anything wrong. This is actually my first time welding these as my old welder quit so he could move closer his family. However he was standing these upright and basically bridging the joint by trying to fuse the beveled lip down to the tube. I dont think he was ever penetrating down to the root like I am so I think we are off to a better start this go around.

[SMFRyan]

Heres my setup, looks strange but it works good.

And disregard my messy shop lol, hasnt been much time for cleaing lately

[Oscar]

Also, did you experiment with the AC balance for more penetration (More EN)? So long as you don't have pepper in the puddle, more EN is typically better, generally speaking. I don't know how the sync250's knob is mapped out in terms of percentages, but it might be worth trying out on practice parts.

[SMFRyan]

As soon as I can get the blue electrodes im going to try dialing it back some, its about 50% right now

[SMFRyan]

All I can say is WOW!
The store didnt have lanthinated but they had 2% ceriated so I tried those and what a huge difference...

The new setup I tried
3/32" ceriated w/ #6 gas lens
using some helium
set the ac balance around 7 and I feel like I could take even more cleaning out of it.
Dropped amperage down to around 175

The 3/32 electrode balled up quite a bit so I will try the 1/8" tomorrow.

[SMFRyan]

This is the fusion pass. Still doing 2 passes it just seems to make it look better in the end
Once I do a few more of these they should start looking better

[LtBadd]

Excellent, good to see you're working through it!