Need help, root keeps cracking

admin

Alright, I got a project back in the shop this week that I need some help with.  Its a 10' long x 16" dia. x .750" wall thickness, hot oil roller.  Its internally baffled and circulates hot oil ( roughly 600° ) for a lamination process.  A couple months ago, this roller came into the shop for journal repair and inspection of the internal baffles.  A 1.125" dia. hole was drilled in the middle of the roller so we could use a bore scope.  That's where my problem begins.  I have tried to plug this hole twice now, unsuccessfully, and I can't figure out what is really going on.First attempt, a .500" thick plug was made, and drove into the hole.  Weld process was TIG, er70s6 filler, pre-heated the roller, and the root cracked thru the middle.  Boss said weld it out, so I did, and out the door it went.  Crack and allSecond attempt, repair cracked weld in the field  ( Big suprise ).Ground all the prior weld out, pre heat again, but this time 6010 root, 7018 fill, peened welds.  No cracks when I left the site.  Month or 2 later, and its opened back up.Today was my third attempt, back in the shop  All weld and prior plug were removed.  I made a thin 1/8" thick plug, 1/8" gap, 1/8" land,  and ran an open root tig pass.  Instant crack thru the root yet again.  At this point I'm at a loss as to what I'm doing wrong I'm tempted to fill the hole entirely with weld at this point
Reply:I've run into this before. What I did was complete the weld all on one half of the circle first. Then complete the other half of the circle. You have to stager your welds, bottom weld longer than the next, etc, etc. I found if I welded completely around at one time, my weld pulled it's self apart.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:Try SuperMissileWeld filler (Er312), or Er309.And tig it.Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li  ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:The circular weld and thickness of the part are combining to create a highly restrained weld.  The stress created by the weld cooling and shrinking is cracking your weld.  How much pre-heat did you use?  Is there oil residue from the inside of the roller getting into the weld, and adding carbon to the weld deposit?  This would make the weld much more brittle as well.  I'm betting your second repair in the field cracked as it probably cracked and you couldn't see it, or the 6010 wasn't ductile enough to survive the cyclical thermal stress whenever that roller heated up and cooled down.  A small crack grew and failure happened due to thermal fatigue.I'm guessing you need the surface of the repair to be perfectly flush and smooth with the surrounding metal, right? Any rough imperfection in the roller surface will mar the laminate as it goes by.  With Oil at 600°F inside the roller, there's no going with any kind of interference fit like an expandable freeze plug.  A threaded plug would also likely leak as well.I'm with CEP.  I think you need good cleaning around the weld, inside and outside the roller, more pre-heat(300-400F if possible), and staggered weld deposits.  You didn't say what the roller was made of, so we're all assuming that it's a mild steel.  You could try TIG welding and switching to an 80Ni1 filler metal.  The added nickel will make for a tougher, more ductile weld deposit.    You've got to get a sound root deposit in, because any cracking will just propagate through the rest of the weld and you'll be back again in a couple months.  Might be worth welding, say half the joint full and then stopping to check for cracks with dye penetrant and developer.  If that looks good, then warm the area back up again and finish the weld.  I'd also get a machinist to turn a cone shaped plug, that way you have some bevel to work with.  It won't be a perfect fit with the curved wall of the roller, but you'll have better visibility and access to the root for welding.  Make the plug the same size as the hole, 1.125" and taper the top to something like 0.5" to 0.75".  Grind it to fit with a 1/8" root opening all the way around and as little land as you can manage.  TIG the root in with 80Ni1 rod and put in at least one fill pass before you stop to check for cracks(if you're going to).  Then continue filling the weld joint.  Wrap the weld area in insulation and slow cool the weld.That's my best guess on how to attack the problem.  ZAPster might have some better suggestions.  I know he works on repairing rollers all the time.Shovelon's suggestion to use Er309 or Er312 is a good one too.Benson's Mobile Welding - Dayton, OH metro area - AWS Certified Welding Inspector
Reply:Any updates on your situation?
Reply:My bad, I completely forgot to update this.  I wound up having to scrap the idea of using a plug.  No matter what I tried, even the above suggestions, I could not complete the root without cracking.  I assume there was just too much joint restraint.  And there was absolutely no point continuing without a solid root.So, I started all over.  Pre heat was 400°, filler was 309, TIG was the process. Welded out the entire hole with the 309.  Had a slight issue in the beginning with a weld trying to tear as I was closing the hole.  Changed my weld direction to oppose the tear and once I was sure it had stopped, I ground out the tear and continued.Each layer of welds was layed in 90° of the last layer.  After about 2 layers I would have to heat the area back up to temp.  The sheer mass of the roller was soaking up all the heat, even constant welding wasn't keeping her hot.  Took a lot of time to complete, but I'm confident this will be the last time I have to deal with this