tig overhead

Date 09-04-2007 01:21

Hello reflex_ak, I have not specifically seen the exact thing that you are describing, however I did have a pipewelder come in to buy some time to practice up for a test that he was preparing to take. In his case he was fitting the pipe with a slightly wider gap on the top and feeding the filler in from the top as he welded the initial root pass into the bottom from approximately the 4 o'clock to the 8 o'clock position. After completing this portion he switched back to the more standard approach to welding the joint out. This method was what was required for the particular test that he was trying to qualify for. Haven't seen this method since.
It is possible that they are telling you that you can't do it in the manner that you have described because of the likelihood that the particular work that you are testing for would not allow this to be done due to physical limitations of the particular welds(not having access to the backside of the joint). Just my $.02. Regards, aevald

By swsweld (****)

Date 09-04-2007 02:59

Not that I have ever taken THAT test but I will try to help. I don't know your skill level on tig so please don't be offended if I say things too basic. What degree is your bevels? Where or how is it tacked up?
If it is 45 degrees I suggest a larger cup, 37 degrees a smaller cup. Gas lens if possible. Tungsten sharp, ground longitudinal not radial.
Do you have a run off tab to start and stop your weld? I know that they will not let you twist your wire(nice try though) but can you feed from the top? Allans point about not having access to the back is most likely the reason you will have to feed from below.
I would suggest 1/8" wire,(less feeding) feed wire almost parallel to plate in line with groove. Hold wire to the top side of the plates, this will allow you to get root reinforcement(penetration), you are fighting gravity so wire entry is important. Also keep a tight tungsten to wire gap. Do not long arc. This also helps the penetration. Your tungsten can not have too much stickout or you will not get the penetration. Don't know if you can walk (rock) the cup or if you will freehand. Prefer to walk it. If you loose the wire (keyhole) momentarily don't worry, with tig you can fix it fairly easily on the spot. You can feather the stop and reconsume it ensuring that you fill the keyhole. Use moderate heat, too low and you will fight it too hot and it will be keyhole city unless you are very good at feeding wire. I'm only guessing about 115-120 amps. That's if you use 1/8" wire. Also if your bevels are 45 use lower heat, 37 use the higher heat. Set your argon on 20CFH or slightly higher. An extra few cfh might help push the root in. No more that 25CFH. I said might. I also might get called out on that one.
After a few inches of welding the gap may shrink and the 1/8" wire will not fall through anymore. If that happens just keep light pressure on the wire and walk the cup it should go in very well like that.
Is it tig out or tig for 2 passes then stick? For tig hot pass turn heat up 10-15 amps keep tungsten directed to the walls moving quickly over the middle so you do not reconsume the root and suck it back. Pause briefly at the walls.
Good luck hope others provide some better tips. Let us know how it turn out.

With former employer at a nuclear outage, I had a welder repairing an excavation at six oclock. He wanted to knock it out so he twisted his wire. The RT showed the twisted wire unconsumed on the film. That's how we found out. Dig it out and fire up the preheat and do it again.
As I was told a long time ago "How come we don't have time to do it right the first time but we always have time to do it right the second time?"

By reflex_ak Date 09-04-2007 15:09

Thanks so much for the tips, This site is really awsome. Sorry my post wasnt detailed enough. I am using 8" plates at 37 bevel that way I have an inch of runoff each way.tig root and hp then lo-hy. I am walking the cup and it seems to help if I have the cup almost 90 from the plate. They will let me feed it from the top and that seems to work the best but because I dont have alot of experience with this process I seem to keep sticking the rod to the sides of the keyhole. thats why it worked with the twist. I could just walk right under the bundle. I will try the 1/8 rod laying more flat and it seems from what Tim says....I should just keep that end of the rod in the middle with a little pressure instead of trying to find the hole evertime. I like the idea of keeping a short gap between the rod and tungsten. I wasnt thinking of that but it makes alot of sense. I guess I should angle the cup more to get the tungsten in more? Dan

By Shane Feder (****)

Date 09-05-2007 08:43

Dan,
There is not a lot of difference between an overhead plate and the bottom of a pipe.
Feed that wire and keep on feeding it as fast as you can.
Gravity will make your weld want to drop so that is why you have to keep shovelling the wire so you don't get root concavity.
You don't need to come in at 30 degrees with your wire, rest it on your tack (above it) and feed it in,that way it will be a steadier feed.
If as mentioned you can't access the back of your root to feed your wire try putting a small arc in your tig wire so you can feed into your root gap from below but still have the tip of the wire in the puddle on top.
Very hard to explain in words but hope you get the idea.
Regards,
Shane

By makeithot (***)

Date 09-15-2007 16:18

aevald, I am wondering about this 5/32" root opening is this not excessive for a tig root. Most welds I have done were fit tight with a feathered edge and those that had a root opening were 3/32" with a land of the same demension.

By aevald

Date 09-15-2007 19:35

Hello makeithot, when it comes to gaps, lands, bevels and other things having to do with fits on pipe or anything else, I typically observe these and take note of the result of these set-ups before I decide if I agree with the method. If a WPS can be qualified with these parameters then that's all that really matters. I would agree with you that I typically would not select this particular fit for welding myself, however that doesn't make it wrong. I would agree with your description for the fits that I have generally seen, knife edge, 3/32" gap, in my case no land, 1/8" filler, yet as I said everyone likely has their favorite set-up and if they can qualify it then they should be good to go. Great to hear from you, Regard, Allan

By medicinehawk01 (**)

Date 09-06-2007 08:16

Hello reflex_ak,When I weld pipe using GTAW (whatever wall thickness) I prefer 5/32" gap using 1/8" rod to root. Usually, (referring to pipe now) the gap closes up so IF you are lucky, you wind up with a "strong" 1/8" gap which is what I like, especially between 4 o'clock and 7 o'clock. Keeping the tungsten stick-out as much as desirable (so the arc is as close to the root opening as possible), you try to keep the filler rod above the internal area of the root opening. In other words , you keep the wire above the inside of the joint so that fusion of the weld metal fuses to the base metal and you end up with "more than flush" internal penetration. Welding plate, I imagine the same condition would be desirable so as to produce a root which will be more than flush and thus harder to suck-back. I "may have" used your wire twisting technique because the fit-up demanded that I do something drastic to make the weld, but I know it is not sound practice. You do not have to result to such extremes IF the fit-up is good. Ofcourse the edge preparation should have a knife edge and you would need to minimize torch ocillation (welding overhead) to put in a root which is "more-than-flush" so you have less chance of sucking back the root pass which I imagine is more likely given gravity is wanting to prevent a weld where the back side (root side) penetration is more than flush. A back feeding technique is basically the same thing but the root opening has to be larger than the welding rod so as to alway keep the filler metal as much as possible to the back side of the open root. What needs to happen is that you can make a sound root pass feeding the rod from the bevel side (using plate) once you have mastered that technique I mentioned earlier. "Back feeding" which seems to me is what you are describing and cannot be done on a test because access to the back side of the joint is not possible IF were an actual production weld because it is assumed the base metal would be in the way and not allowing you to back feed. When welding pipe though, it is possible to back feed nearly 90 percent of the joint assuming the root opening is large enough to pass the rod thru the pipe joint.