Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / General Welding Discussion / Tig root porosity...please help me kill the gremlins
- - By devo (***) Date 12-03-2012 17:40 Edited 12-03-2012 20:26
Yes I searched the forum already.......the search engine leaves a lot to be desired...if I say "Tig" i don't want to see posts with the word "Tight"

So I am trying to learn how to do a root pass (GTAW) on pipe.  60 deg. included angle, .25" wall thickness, 6" carbon steel pipe.  Tiny (<.030") land, gap between 3/32" and 1/8" (sucky bevel prep that I just have to deal with).  Filler 1/8" for the bigger gap, and 3/32" for the smaller.  Amps around 80 for the smaller gap/rod, and around 105 for the bigger.  3/32" red tungsten, #10 cup.  I've watched the vids on weldingtipsandtricks, and I have done quite a bit of steel and aluminum GTAW, so while I'm no expert, I don't suck.  I've learned how to deal with the aluminum gremlins so far, and steel rarely gives me trouble, but this is driving me nuts. 
  What is happening is when I practice cover passes, walking the cup on the outside of the pipe, all is well.  Clean, quiet puddle, no problem.  When I try to put in the root (lay wire), if the gap is a little tight with the 1/8" filler, I get a root flush to a bit concave, but clean...a nice calm puddle.  I have even tried keyholing it and dipping the filler.  Still can't build up the inside much, but the puddle stays calm.  As soon as I try with the 3/32" filler, it turns to ****.  The puddle starts bubbling, the green aura of doom surrounds my arc, and soon I am making another trip to the tungsten grinder.  I've done all the standard troubleshooting approaches to isolate the cause of contamination.  No draft...bead on plate with 3/32" filler is good...autogenous welds on land of pipe are just fine...but I can't figure out why the puddle is getting dirty.  I've tried starting way up on the bevel, getting the puddle established, and slowly working it down to the edge, and before I can even bridge the gap, I get the bubblies.
  What's strange is that with .25" plate, I've done three 6" welds in a row no problem.  Root has around .063" reinforcement, weld face is clean, minimal silicon islands in the toes, fill passes and cap go in beautifully.  But something is happening with the pipe. 
  It seems like I should be trying to bridge the gap without penetrating all the way to the root, but then once the gap is bridged, let the puddle soak in all the way, and then start adding more filler.  Does this sound right??
Parent - - By OBEWAN (***) Date 12-03-2012 17:47 Edited 12-03-2012 20:39
post deleted.
Parent - - By devo (***) Date 12-03-2012 20:24 Edited 12-03-2012 20:27
""If you want the search engine to work you need to input more than just "TIG""........no s**t.  I was using that as an example of the shortcomings of the search function.  I've searched with every iteration of the words "Tig root pipe porosity "  and a few others to no avail, hence my post.  Booleans don't seem to work, have you tried it?
  As I said, I've gotten good repeatable results on plate with the same joint configuration, but on the pipe, not so much.  Could be coincidence, but I don't think so.  i can answer all of your questions with this.... Read my post first, all of it, and I have already eliminated dirty argon, breeze, adequate flow, clean base metal and filler, leaky lines, ID purge.  It's carbon steel pipe so no purge needed.  Maybe one of your robots can tell me how to make good Tig roots?
Parent - By OBEWAN (***) Date 12-03-2012 20:45 Edited 12-03-2012 21:10
What code are you working to?  I have worked to pipe specs in the past that do require ID purges for carbon steel depending on the thickness.  But, don't listen to a simple "robot programmer" like me. It sounds like you've got all the bases covered then.   Keep up the "good work".
Parent - By OBEWAN (***) Date 12-05-2012 12:09
It looks like this thread is a classic case that demonstrates the "art" vs. "science" in welding technique.  With the exception of watching the filler wire spec, it looks like "art" is winning.

If I ever need to learn how to pipe weld this will be a good one to remember.  And it would be hard to teach a robot to use the same technique.
- - By yojimbo (***) Date 12-03-2012 20:50
Regarding bevel prep- if you are hand beveling and ending up with uneven root opening, consider the following:  After marking the pipe squarely using a wraparound, and establishing your bevel, put a new, dedicated 7" wheel on a grinder to establish a flat surface on the bevel.  Best to use a 1/8" X 7" wheel as there is some flex to it and seems to work best.  Using that wheel, turn your ginder 90 degrees to the pipe and looking down, assuming your coupon is mounted in a tripod, maintain a 90 degree angle both vertically and horizontally and "face" the knife edge of the bevel to true it up.  Go lightly and maintain squareness to the pipe.  All you are trying to do is "break" the knife edge to a minimal land all around.  Once this is accomplished, your bevel should theoretically be flat and will eliminate the variations in root opening which are problematic when learning tho tig the root.  After you've established flatness, deburr the inside of the pipe to remove the material that will be left over from facing the bevel so you can read the true Inside diameter.  Not deburring will mislead your read of the ID of the pipe and when you go to grinding in the finished bevel you will go too deep, this re-creating the low spot on the bevel with their accompanying variations in root opening.  It takes some practice, but it is possible to hand produce a machined bevel.  Important to use a dedicated wheel for facing.  A 1/4" wheel will not ride the pipe correctly and cause you problems.

After establishing an even root opening I would suggest two things.  First- use a 1/8" tungsten.  It will not dirty up as quickly and carries the amperage better.  To gap your pipe I would suggest using a 5/32 opening when you tack up.  Build yourself some gapping tools by cleaning the flux off some 5/32 7018 and tig weld them into a "T" formation.  Those 3 points of the "T" will keep your pipe correctly alighned.  Keep the first tack small, 1/4" max.  The 1st tack will draw quite a bit and close your root opening less than a desired 1/8" if it's too big.  After the first tack, adjust the opposite side of the pipe to 1/8" gap as well [hammer and wedges] make the second tack your bottom tack and keep it to about 1/2", leaving your wedges in the root opening to prevent it from shrinking.  Use 1/8" wire for the root pass.  I suspect you are starving your puddle when you are putting the root in and not feeding enough wire.  I find this often causes a loss of control on the root and results in porosity.  I dont know why it does this, but I have seen it too many times no to know it will.  I took training at the Fluor Daniel school years ago to learn to use the wide gap tig root technique where we were taught with a 3/16" gap and to "backfeed" the root.  It took a lot of practice to get it to their standards but the results were worth the effort and enabled you to depost sufficient root reinforcement to prevent suck back and concavity. It is really the only reliable method I know of especially at the bottom of a 5 or 6 G weld.  The real trick of it is in the feed hand holding the wire.  Wire feed of the consumable will have the most effect on puddle/heat control assuming a scratch arc is being used, and if for any reason you are using a foot control, pleas, dissconnect it now and start using a scratch arc. 

The two primary techniques for root pass with tig is either "pressure feeding" or "back feeding".  Sounds like you are pressure feeding where the root opening is equal to the wire diameter and you are holding it into the bevel and just burning it in.  My opinion is that this will work well enough on roll out welds where gravity is your freind but will hinder you in positon welds where gravity it working against your melted puddle and pulling it down to create suck back or a flat root at best.  If there is anyone in your location that can demonstrate the backfeeding technique using a 3/16" root opening, it would be worth your time to  arrange a demonstration of that technique.  The welder who introduced me to it said, "Learn to weld like this and you will weld fearlessly".  I quit that job a week later and headed for Deerpark TX to attempt enrolment at the Fluor Daniel training center.

Last thing, start practicing in the 2G position until you get a better handle on it, you'll have less of a fight with gravity, then move to some roll out work before you sart on the 6G.  Body position is everyting on the 6G- you do not have to reverse your feed hand just get yourself in a position where you are stable, can see the puddle and still make the weld.  This probably won't help much, there's nothing like being exposed to proper training, but it has worked for me after a lot of practice.  Best fo luck.
Parent - - By devo (***) Date 12-03-2012 22:21 Edited 12-03-2012 23:49
Thank you for the reply....To address a few of your suggestions.
- I am trying this in 2G.  I've learned to trip and fall before I walk, and to walk before I run. 
-  Gapping tools..check
-  Deburring the bevel...check
-  1/8" tungsten for root on 1/4" wall pipe?  I'm running around 100 amps tops.
-  I am using foot control, but with lift start and I'm stomping the pedal the whole time, trying to replicate the way it's done in the world.
-  I can barely get the tacks to go in, let alone try backfeeding vs. pressure feeding.  But yes, pressure feeding is what I am trying
-  So I guess the main question is about tacking technique.  Do you back feed the tacks?  Or do you start high on the bevel, get the puddle going and move down into the root, bridge it, and soak it in.  Because that is what I'm trying, and what occasionally works.
-  More of the problem seems to happen when I am trying to use 3/32" filler.  I know 1/8" is typical, but will there be a greater chance of losing control with the smaller filler?
-  I live in Southern WV.  No chance of getting someone local to show me the technique.
-  I'm not sure if it is ever done out in the world, but how do you back feed on 24" pipe??Or is pressure feeding used on the larger diameters? Or is GTAW simply not used on the big stuff?(I thought it was)

  Thanks again for the advice, I'm headed back to the booth right now to try it out.
Parent - - By devo (***) Date 12-04-2012 01:35
Progress!!!
Nothing beautiful yet, but no porosity.  I ended up using a large gap(5/32") and the "backfeeding" technique.  In retrospect, I believe part of my problem with the 3/32" filler is that I don't know what kind of rod it is.  It could be S-6, S-2, or maybe even R45.  I'm using some filler from the 70's I found in the dungeon of our storage room.  Seriously, this stuff is wrapped in burlap, and the label just says "HiTest".  But the unknown 1/8" seems to be fine.  I'm getting some big silicon islands on the hot pass, but no bubbles!!  Thanks yojimbo for the advice.  That big gap is scary at first, but once I figured out the pressure on the rod, it wasn't so bad.
  Speaking of how to feed the rod...lets say you are doing 5G, welding at 12 o'clock.  Where do you let the filler enter the pipe?  I've been doing it around 10 or 11.  Would coming down to 8 or 7 work better?  Of course, once you are finishing up the root, you have to gradually get the filler out of the pipe.  When does that transition begin?
Parent - By yojimbo (***) Date 12-04-2012 18:36
devo-
Congrats on overcoming your fear of the gap.  Some welders never do but the gap is your freind- it lets you get enough filler into the root by not closing down so much you cant feed it in and is wide enough to give you a good view of the root penetration you've alread welded.  I've kown welders that could feed the filler wire straight through the pipe as in feeding from the 11:30 position to weld the bottom at 6 oclock and feed the root from across the pipe all the way up.  Their control of the tig torch ie. angle of tungsten, distance from weld puddle and control of arc was something to watch and I never mastered it- the best I saw was a guy who did this on Inconel as easy as slippng his boot on, his competence was a performance to behold.  To answer your question of filler wire position:  Either of the two you are using will work, you can feed down into the puddefrom the 10/11 oclock, or up to it from the 7/8 oclock, the primary idea is to be feeding the puddle from this inside or backside leading edge.  A basic trick we were taught is to let your filler wire rest on a tack for balance to steady your feed rate.  Another good self monitoring is to be aware of how tightly you are holding the wire, it should be a light, loose grip on the filler.  Too much hand/finger pressure will translate to a shakier feed rate because of muscle tension.  The tight grip is usually caused by nervousness or concern about screwing it up.  Relax, breathe, hold the wire firmly but lightly and your feed rate of the filler will be smoother and provide better heat contol of the puddle with increased amount of filler acting like a heat sink, which also means as you get this down you will be able to increase your amperage [and therefore your heat which you will then control with increased feed rate] which will improve fusion.  Some guys can feed using an entire length of filler wire- they have enough balance.  I always had to cut mine in half finding a full length too much to manipulate.  Either way, another good trick is to hold the wire as close to the end away from the pddle as you can and heres why:  If ou hold the wire halfway down to the puddle you will only be able to feed that much of it into your puddle before moving your hand/fingers away and you will tend to starve your puddle.  I would reccomend learning and practicing holdng the wire at the very end, starting with half cut lengths of filler wire, and developing the coodination and balance needed tofeed smoothy.  Remember it's all in the feed hand, you probabl already have the torch hand down.  Feather your tacks just like you would on a 6010 root for two reasons- it allows the same smooth transition of fusion into the tack and gives a smoother surface to rest your filler on for balance and feed.  Also as you come up to a tack angle your filler at a steeper angle and learn to feed it into the root as you close it up at the tack.  A little tricky to learn how much pressure you need to apply and how far in to push the wire when you cant see the opening on that last 1/4", just needs practice.  Lastly, you'll make yourself crazy trying to weld with an unknown filler wire- that sounds like 80% of the problemes you mentioned.  Would it be worth the $50 or so investment to buy 10# of ER 70S2 1/8" so you know what your dealing with, or maybe get your employer to purchase it to help you develop your skills so he can benefit from them as well?

Finally- don't be afraid of too much penetration on your root.  A 1/16" root reinforcment is within guidlines for ASME process pipe.  It akes a lot of feed to get that much penentration but the benefit is that root will be solid and wont suck back on you.  Hope any of this can help.
Parent - - By yojimbo (***) Date 12-04-2012 18:59
devo-

I replied to your other post first, to respond to your questions.
Can you discconect the foot control altogether and just hook your leads straight to the power block on your torch?  You will likely never see a foot control on a construction site and besides for this kind of work they are a hindrance and offer no advantage.

1/8" tungsten will stay cleaner.  I would suggest 80-85 amps on the root with 1/8" filler wire on the larger gap.

Tacks- Yes you backfeed those as well.  Start your arc on the bevel and wash it down over the filler which is held just a hair inside the opening or even touching the knife edge of your bevel and bridge the gap in one motion.  It will require sufficient feed rate to bridge the gap.  Once gap is bridged your on your way.  Trick- if the puddle starts to get to hot/liquid- likely due to starving the puddle, bring the arc back up onto the bevel and out of the puddle for a second or two alowing the puddle to cool and you to reset.  Welding takes a lot of anticipation- you are learning to control a liquid mass of molten metal, once you see how they are going to behave you can anticpate their behavior and adjust your action to control the puddle behavior.

3/32 wire on 1/4" carbon steel will starve the puddle and you will lose control of it.

Backfeeding 24" pipe is done every day of the week.  Read the tip I sent in the other post regarding resting your filler wire on the tacks for balance.  On big bore I would normally put in more tacks to prevent root gap shrinkage anyway and would also have more places to rest/balance the filler rod.

I don't know if it is still in operation, but Fluor Daniels used to have a training center in Tenn and offered the same 12 week free course of instruction for advanced pipe welding.  They had some very good instructors.
Parent - - By jsdwelder (***) Date 12-04-2012 20:54
I have seen people try to TIG weld with RG45 rod with similar results. That rod is not intended to be used as TIG filler. Could be the problem you were having.
Parent - By Blaster (***) Date 12-05-2012 05:40
I have seen that RG45 or RG60 mistake too.
Parent - By devo (***) Date 12-05-2012 02:44
Thank you for taking the time to spread some knowledge and help a guy out.  All of it helped in one way or another, now it's just a matter of keeping the hood down and practicing.
Up Topic Welding Industry / General Welding Discussion / Tig root porosity...please help me kill the gremlins

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill