TIG welding a T-joint

OK I am simply not understanding your technique. If you have your amperage dialed in correctly, and you have brought your tungsten in close enough to get a puddle at that amperage, then you should have one narrow bead between the tubes.

First on T-joints you should divide the weld into 4 quadrants. If you were to place marks at 90 degree intervals around the vertical tube, with the first mark starting where the 2 tubes are flat to eachother, then you will end up with 2 marks on the flats and 2 marks dead center on the fillets. You should start and end each weld quadrant half way between each of these marks. So this is where you tack the tubes together.

4 tacks, 45 degrees off of the flats and fillets. This will give you 4 weld areas. 2 are almost flat butt welds, and 2 are almost regular fillets. You should dial in your amperage for the fillets, so 18 ga Mild steel would be: 18 ga = 0.047" = 47 amps + 30% = about 63 amps.

So that is where your amperage setting goes.

1/16" or 3/32" tungsten. 0.045" ER70S-2 filler rod. (never use filler thicker than your base metal)

Tack by melting a small ball off the end of the filler rod so it rests in the V of the joint touching both parts. Heat the ball directly until it flows out to both parts.

Now just play connect-the-dots.

Sorry but that is the best I can do without actually being there.

I don't think he is talking about tubes or pipes. I think he is asking about flat stock T joints. I have the same problem as he is having especially on thin Al. I have a little better luck using higher HZ and a little more stick out. The real problem seems to be from air sucking in from the backside of the joint. Try some foil backing and see if that helps....

You are correct, I should have made that clearer I suppose... I actually don't have as much of a problem with tubes, honestly. It's just doing sheets of metal that baffles me. The puddle seems to always be on one sheet or the other (or both), but never at the actual joint until I start glopping lots of filler in there. Then I've got this oversized C-shaped puddle that seems to either grow overly large or just sit there and lag behind the torch movement. I'm pretty certain my tungstens are ground correctly (I do quite a nice job on them, the instructor was convinced that I was grinding them with a jig of some kind because he couldn't imagine how I would be able to freehand something that nice) so I'd think the arc would go where I point it more or less, but it just never seems to want to go right to the joint. I'm guessing it's going to ultimately just be a "practice more" thing, but I'm hoping there's a useful tip or two that can help me on my way. Ernie's suggestion about using a tiny ball of filler to get the tack weld going sounds like it'd help a lot... I'll give that one a shot.

I do have the tungsten out pretty far so I can get it down into the corner, but I'm using a gas lens so I'd assume that's not a problem... a smaller cup (I've got a 7 right now) seems like it might help me get in closer, but with the gas lens it seems like it shouldn't matter since I'm not exceeding the stickout that can be done with it (I'm less than 6 tungsten diameters, which I recall reading was the limit with a gas lens). Would maybe a different tip shape help? Something blunter, or maybe more pointed? I can see that maybe mine's a bit too pointy right now... it almost seems like I get more heat from the sides of the grind than from the point itself. I'm not in class again until Monday so I can't really play with it until then, so hopefully I can build up a list of stuff to try before I get there.

Bryan

Are you "grinding" your tungstens on any type of oxide wheel or belt. This can lead to contamination, and causes your arc to want to wander erratically. Also try to make sure you point your tungstens parallel to your grinding surface so that all the "sratches" go to the point.

When I weld a Tee joint I point my tungsten at a very long angle, and stick it out as far as possible. The shape of the joint will inherently help keep the shield gas where you need it. Then just as Ernie stated, melt off a small ball of filler, give it a quick swirl to puddle on both pieces and run with it.

The only trick I can tell you is to use your gas cup as a guide. put it at a 45 degree angle in your joint, resting against both pieces. Then adjust your tungsten out as far as you can get it into the joint without touching. You can then just slide your torch along, keeping the cup touching as a guide.

I do this alot when fabricating with Stainless for food contact aplications. I can make robot perfect welds in appearance, but they are not very strong. Wrench

I had a tough time starting as well. Then I downsized my tungsten and using Ernie's balling technique I was able to start really well. What really made my beads really look beautiful was to move my tip up and down through the junction. This dissipated the heat to the vertical piece and then to the horizontal piece. Just keep getting that rod in there and moving down the line in a zig zag fashion (up and down). I was absolutely amazed at how improved the beads were.

Merle

Are you grinding the tungsten so all the scratch lines are in-line with the tungsten?

If the arc wanders from side to side, please check that:

-the tungsten is properly ground, with grinding lines lengthwise or

-the workpieces are not dirty, with no oxide on them.

At least, this is what my limited knowledge of tig allowed me to understand. I am also beginning...

I am. In fact I'll post a link to a close-up of one of my tungstens as soon as I can get my scanner hooked back up... I'm really quite proud of them. All those years of hand sharpening lathe tools and deburring shortened bolts on a bench grinder, I suppose. But yes, my tungstens are ground "by the book." They are, however, pretty steep as far as the angle goes... maybe even a little steeper than might be appropriate, since I've heard that a blunter point gives you less spread with the arc. In fact I think I'll probably try grinding a blunted point in class tonight and seeing if it helps, but not until I've tried all the suggestions made in this thread so far. I think I might go to a smaller tungsten, too, to see if that helps... although other folks in the class are getting good beads on their T-joints using the stock suggested size for the class. As I'm learning, though, TIG is a highly personalized process... there's more than one way to do it in every case, and ultimately you just have to find your own technique.

Bryan

The main rule in TIG is dial your machine into the proper amperage and then move the tungsten in until you get your puddle.

1 amp per 0.001" of thickness for steel and aluminum. Subtract 30% for outside fillets Add 30% for inside fillets.

Subtract 30% for stainless steel Subtract 50% for bronze. Add 100% for copper.

As an instructor I can tell you that most students start out with 2 standard mistakes. Their amperage is too high and the tungsten is too far away.

The lower your amperage the closer to the materiel you have to be.

How about when the materials being joined are different thicknesses, like say, 0.022" and 0.125"? Do you use an average setting, or the optimal setting for one thickness or the other?

Bert

Yes you have to find a compromise amperage, and cheat a little more of the heat towards the thicker of the 2.

Hi - abs. novice (attempting to) contribute here:

by T-joint you mean a fillet weld, between a flat plate and one standing vertically off it?

As ever a novice is helped by setting amps so "foot flat to the floor" gives intended welding power, this is it, for a TIG fillet-weld. You can get the amps way, way down and have a small almost viscous weld pool and no melted back-surface on either plate (sheet).

"foot to the floor" is probably risible for the experts here, but you have enough on your plate as a novice controlling torch angle and height, fill-rate, etc.

Melt pool on flat and vertical plate but they do not "run together":

You need to get that arc into the corner. I think you need to get the filler in there to first get a fillet to form bridging correctly across both pieces. Again, "foot flat to the floor" technique, you can come in quickly, get the plate surfaces to start melting and get the rod in to start the fillet, all very quickly. After first getting the fillet at the start-point, you can run it along the weld.

With a steady small constant-power arc you can have the arc getting its power right in at the corner and you won't have any prob.

Anyway, I'm being driven *&^%$' mental by another issue with fillet welds which I suspect is to do with rod size, so will post that question.

Richard Smith

(me, prev.)

Question:

You cannot tack-weld a fillet-joint without filler-rod -- as you get precisely the prob. discussed here - only ever end-up with melting the lower and upper plates but the melt-pools don't touch and "jump-together"???

Richard Smith

Ok, so a few more class sessions have come and gone and I think I'm starting to get the hang of it. Plus I have returned with a few images to show, and some other unrelated questions. My technique can't be too bad at this point as I even managed to weld a piece of 18 gauge steel sheet at a right angle to a 1/4" thick steel plate and not burn through the 18 gauge piece at all across a four inch long bead... the weld didn't fail when I hammered the crap out of it, either, so I think I'm getting somewhere!

First off, thanks to everyone for their suggestions and advice. What really got me on the right track was turning up the current... I was a bit skeptical at first since I already had enough current to make a seriously wide bead... take it from me, one should never doubt the words of Ernie! The trick really is to hit the joint hard and fast to get a tight little bead at the point where both pieces meet, and then to move quickly across the piece. In fact it seems like you move a great deal faster than normal for things like stringers and butt joints... but, it does appear to do the trick. It has also helped that I'm working with thicker steel now... it's darned near impossible to learn on thin metal with something like this. Once I started getting the hang of it on thicker material I was able to move over to thinner materials, hence the aforementioned 18 gauge welded to a 1/4" thick plate.

I never did figure out the trick of melting off a small ball of filler to create the tack weld, though. The ball always stayed stuck to the rod, and would just creep up the rod and away from the arc. I did figure out a technique that worked for me, and it's been working well. I start an arc and just heat up the two pieces to be joined without actually melting them. Then I point the tungsten at the gap between the two pieces... the arc will naturally jump from piece to piece, or maybe even favor one piece over the other. Then you stomp on the pedal and give it a lot of current. Suddenly a puddle forms that flows between the two pieces. As soon as that happens, you release the pedal and you're left with a tiny, uniform tack weld with fairly minimal penetration but definitely enough to get the job done. I don't know how well that will work for others, but it's been working wonderfully for me on all types of joints and materials. Plus as I'm not using any filler rod I can use my free hand to steady my torch hand and get a very precise aiming as my hands naturally shake quite a bit.

Now then, some pictures and some questions. I'm not one to normally submit my work for public review due to a glaring lack of self-esteem, but I figure this will help me improve. First picture is of one of my lap joint beads that I think turned out quite well except for an odd dot in the middle of it... my guess is that it's a slag inclusion or something similar given the fact that the metal I was welding had an awful lot of stubborn mill scale on it, but maybe someone with more experience can tell me. I'm pretty sure I didn't dunk the tungsten there or anything like that, especially since I would have noticed the arc cutting out and the tungsten wouldn't look as uncontaminated as it does. The image can be viewed here:

Feel free to rag on me, though constructive comments are preferred. Next, a couple images of a T-joint I made. There's some rust on the bead now because it has been sitting around in some bad conditions, but the bead itself is still visible (except that the pictures themselves are a little crummy). Images are here:

I apologize for the poor image quality, in fact the second one is even a bit misleading as a result. The penetration is actually quite good and the welds are properly concave fillets without undercutting, but it's not enitrely evident from the images.

And last, three tungstens of mine that I'm proud to say I ground by hand, no holding fixture of any kind except my fingers:

The two on the left are lanthanated and the one on the right is zirconated. The middle tungsten was the one I used to create the above pictured joints. The left tungsten is actually not quite ground the way I like... I was in a rush grinding that one, the tip's a little bit dinged up. I've been pointing my zircon tungstens before using them on AC but blunting the tips a bit more than with DC use tungstens... it seems like I get a nice ball on the end that way that isn't so large as to cause the arc to get unwieldy, but I'd welcome any other point grinding suggestions anyone may have.

So that's my current progress, comments are invited! It's taken me something like ten weeks of class totalling maybe four to six hours a week of just TIG welding to get where I am, so hopefully I'm getting someplace worth being. Thanks!

This can work, but if you have even a tiny gap between the pieces you will just blast a huge hole.

Looks like a good start. Hard to tell how the bead size compares to the plate thickness though. I would like to see a slightly faster pattern.

Looks decent. Good profile overall, but you still need to practice a while to get consistency down pat.

I recommend turning off the amperage control on the foot pedal so the pedal is simply and on/off switch. Just switch amperage control from "remote" to "panel". Keep the contactor control set to "remote". It will force you to set the machine correctly for optimal amperage, and makes it easier to maintain a consistent bead width.

A little pointier would be good. Take a freshly sharpened pencil and copy the angle.

For the Zirconiums used on AC, try balling the end of the tungsten on DC Electrode Positive. Point the torch straight down at a clean piece of aluminum or better a block of copper. Bring up the amperage very gently until the tungsten tip melts into a perfect ball. Then back off the amperage and let cool before moving, and set the machine for AC.

You will get a much better ball on DC EP than you can get by forming it on AC.

Overall I think you are starting to get it.

Pointier tungsten, add filler a bit faster, and try turning amperage control to panel control.