Fixing Undercut in 3G Cover Pass

admin

I'm working on my LA city structural welding certification. One of the requirements is 3G test plate welded with stringer beads, at least on the cover pass. I've gotten my beads pretty consistent but I keep having little cuts, lines or undercut as I'm stacking the beads over each other. I've tried everything to fix the problem. My instructor said that I might be running too hot, so I've reduced amps. I've set the machine for a soft arc, I've changed the angle of my rod. I've tried welding faster and slower. I've tried more and less bead overlap. Take a look at the photos below, and I'd appreciate any help.All welds were made vertically, using 1/8" 7018 ESAB Atom Arc. I dropped amps down to around 77-80 on cover pass in order to try and correct undercutting. Base metal is 1" thick a36. The Miller welder was set to "soft" arc setting. By the time I had gotten to the cover pass, my plate was really hot. Could that be causing the problem?
Reply:It looks and sounds like the plate is getting too hot. When you're welding it out, try letting the plate cool down to about 250 degrees Fahrenheit. If you don't have a temp gauge, you can pick one up at Home Depot that goes up to 600 degrees for about $30. Also, you shouldn't have to drop your amps at all for the cover.Amanda
Reply:For years in the Seattle area the structural welding test was 2-G, 3-G, and 4-G. 1-inch thick. This somewhat gave the plates time to cool, because you made 3 or 4 passes in one position, then moved to the next position. But before the cap passes you had to take a break and let things cool down. Then for some reason they went to just 3-G, and 4-G. So the wait time for things to cool off was longer. The vast majority of testing facilities I’m been in, there just isn’t a lot to look at, so waiting for the plates to cool down, is very boring!Don’t pay any attention to meI’m just a hobbyist!CarlDynasty 300V350-Pro w/pulseSG Spool gun1937 IdealArc-300PowerArc 200ST3 SA-200sVantage 400
Reply:Originally Posted by amandalarson34It looks and sounds like the plate is getting too hot. When you're welding it out, try letting the plate cool down to about 250 degrees Fahrenheit. If you don't have a temp gauge, you can pick one up at Home Depot that goes up to 600 degrees for about $30. Also, you shouldn't have to drop your amps at all for the cover.Amanda
Reply:use 3/32
Reply:The test needs to be done using 1/8" or 5/32" low hydrogen electrodes.
Reply:What size root gap are you starting with on your plates?Your cap looks real wide. Looks like I'm seeing 8 stringer beads on your cap? The structural tests I've done had cap limits of 1/8" wider than the top edge of the plate on each side and a max of 1/8" crown reinforcement. You need to slightly and consistently oscillate your rod and hold the sides, this fills the undercut your getting.JasonLincoln Idealarc 250 stick/tigThermal Dynamics Cutmaster 52Miller Bobcat 250Torchmate CNC tableThermal Arc Hefty 2Ironworkers Local 720
Reply:1/4" root opening and 45 degree included angle. As far as the amount of beads, my instructor asked me to put in more beads, with more overlap. He asked me to try and cover 5/8th the previous bead. I could probably do a cover with 4-5 passes, but he prefers to see more.Yeah, I have more reinforcement than would be allowed on an actual test. I welded over the previous cover pass to save time, in not having to put together a new test plate.I'll try using more of an oscillating movement next time I do a cover. But it can't be so big as to become a weave, the beads need to be pretty small in width.Last edited by SteelMind; 12-10-2015 at 05:08 PM.
Reply:Make sure your arc length isn't getting too long.  Long arcing will increase the odds of having undercut.  I'll second the recommendation for a little bit of wiggle or side-to-side weave as well.  Not alot; maybe 1/32- 1/16".  A tremble to your hand holding the stinger.  This helps me flatten the cap out and cure that undercut you're seeing.Benson's Mobile Welding - Dayton, OH metro area - AWS Certified Welding Inspector
Reply:Originally Posted by SteelMindThe test needs to be done using 1/8" or 5/32" low hydrogen electrodes.
Reply:Wanted to post an update on my results. I started with a cold plate, and used a slight side-to-side movement. I was still getting undercut. I asked my instructor what to do about it, and he said that I should pause slightly on the left side, where the undercut happens, as I'm moving the rod from left to right. Take a look at the pictures. The first one is with no pausing and there is still undercut, even with a cold plate to start with. Second picture is with same technique but with a slight pause on the left side. Aside from a problem at the top right, I think it looks pretty good. I welded over top of existing cover passes in order to save time and metal.
Reply:Originally Posted by SteelMindAside from a problem at the top right
Reply:One of my biggest challenges when I took this test for the shipyard was making sure your bead crown was high enough to fill whatever of the bevel was left. Most of my mentors when I was just a helper told me just have the toe of the bead nip the beveled edge. My last fill layer gave me about just 1/16 of bevel left to fill. It took me a very long time to figure out that the very toe of your molten puddle WILL NOT have a 1/16 crown when it solidifies therefore causing undercut. So if you're trying your best to not go too far outside the bevel which alot of beginners like myself when I was a helper was told, don't sweat it. Take that bead out a 3/32 more or so into the base metal and your bead crown will naturally fill the rest of the bevel giving good fusion along the toes of the weld to the base metal with no undercut.Now that I've covered that possibility, if you say you're turned down to 80 amps on a 1/8 7018 and still getting undercut, that is a sign that your plate must be smoking hot. I can throw a good vertical cover pass on at 120 amps-ish when the base metal has cooled to about 150F. What I usually do is wait for the base metal to cool to 150F then throw the first stringer tieing the bead to one test plate and the rest of the center beads until I need only one more bead to tie the other stringer beads to the last plate. At that point I will again wait for the plate to cool to 150F then throw that last bead to connect the second plate. Probably the most important stringers are the ones that tie into the test plates, these are the areas that they look for that 1/32 or undercut or more to disqualify you. So I highly recommend you let your plates cool before throwing those two two stringers on.Good Luck!!!
Reply:Originally Posted by TimmyTIGuse 3/32
Reply:does the weld machine have a 'dig' or inductance setting?  didn't see it mentioned. your in a class, sooo have the instructor lay a bead, how does the instructors bead look?     basically its either your form, (arc length, speed, angle of electrode) or its in the machine settings.        so go thru the process of elimination, is it you or the machine? etc....they say that people that can't do, teach. i hope your instructor doesn't fit that statement.    there are too many possible solutions to the issue, to be sure of what the issue or issues are, have the instructor tell you what the issue is then ask him to show you!
Reply:Originally Posted by tracymobilecncdoes the weld machine have a 'dig' or inductance setting?..
Reply:With my Lincoln V350-Pro I like the arc force set at 0.0 when running vertical up hill with 7018. I like the puddle to flow. I do the side to side motion, holding on each side until the puddle fills to my liking before moving to the other side.Don’t pay any attention to meI’m just a hobbyist!CarlDynasty 300V350-Pro w/pulseSG Spool gun1937 IdealArc-300PowerArc 200ST3 SA-200sVantage 400
Reply:yep arc force, arc control, dig, inductance...  there are allot names for the same function.    as cep said i would set it at 0.   the function has its place but for the most part i believe it should only be used to help people learn to not 'stick' the stick.    stick is a great process because it gives the user so much control of the weld.  very generally speaking as weld control shifts from user to machine production increases. and especially for the purpose of test taking dont give up any useer control, and when learning never use a crutch, or you'll always have a limp.
Reply:I hadn't read all the responses posted here but always check out the 1 subject. 1st off I can't imagine any inspector accepting the fact that you took a 1/8 7018 and ran it down to 70-80 amps. how you ran it is beyond me. 115-125 is the norm. At that rate your not burning off enough rod to fill in the puddle I would think. And yes there should be a a cooling time because you don't wanna over heat the plate.
Reply:Originally Posted by Mark's WeldingI hadn't read all the responses posted here but always check out the 1 subject. 1st off I can't imagine any inspector accepting the fact that you took a 1/8 7018 and ran it down to 70-80 amps. how you ran it is beyond me. 115-125 is the norm. At that rate your not burning off enough rod to fill in the puddle I would think. And yes there should be a a cooling time because you don't wanna over heat the plate.
Reply:I had the same idea about going down that low because even the manufacturer says the rod is designed for 90+. But the WPS for the test did not state a minimum amperage, and my instructor said not to follow any listed or stated amps but instead to pay attention what the puddle is doing. I have no way of knowing the machine is giving me an accurate amperage.  The standard answer for undercut I was getting is that it a sign things are too hot. Response to that is to lower amperage. It wasn't too hard to run that low actually but I don't know if their is porosity below the surface when its that cold.
Reply:i would imagine you could only run that low with the arcforce turned up. again turn it off
Reply:Here is another update: I'm using a temp gun to measure my metal as I put in the beads. I'm welding on a 6 inch long groove this time, all other factors are the same. I measured the metal both directly on the weld area and a couple of inches outside it, and I'm getting easily above 500f after just a few passes. I know a36 can take the heat, but at what temp does it start to affect the ease or quality in which a good weld can be put in? You guys have suggested letting it cool to 150-250f for the cover pass, do I need to worry about the fill passes too?
Reply:Push it in more, or drop your amperage a little (2-5 amps)Looks like you could slow down a hair too.Betting on slightly too long of an arc though.....So either stuff it in a little more at the same heat while slowing down a hair, or drop you amperage (which will make you slow down and stuff in in more to keep from stubbing out and leaving a ropey bead)My .02 anyway.....
Reply:When testing is it allowed to use a chill plate or some sort of a heat sink?Thunderbolt AC/DC330AB/P + Bernard coolerSyncrowave 250Oxy/Ace torchDayton 160 MIG6CF pot with 185 compressorToo much big equipmentEven more small toolsOriginally Posted by OMBWhen testing is it allowed to use a chill plate or some sort of a heat sink?