Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / General Welding Discussion / undercut fix
- - By cbutts (*) Date 07-16-2019 08:38
where I work we are trying to follow aws D1.1, we run 2 shifts, we seem to get a lot of undercut on welds in the 2g position when they run the cover passes... I think it is from excessive heat and not letting it cool to the minimum interpass temp.... our day quality guy thinks its okay to just grind the undercut that in most cases is over 1/32" deep on 1/4" grade 50 plate, I disagree.... I can't seem to find anything online on how to properly fix undercut after its been done... I tell him we need to grind the last pass out and re weld it but I continue to walk into it just being "blended" we are welding a bevel male connector into a 3/8" plate wall that is capped with 1/4 plate... can anyone refer me to some literature or facts on where to look up repairing undercut?
Parent - - By jwright650 (*****) Date 07-16-2019 11:01
Undercut is a LOSS of material, so grinding isn't helping other than blending the notch out of the material. If you want to grind the notch, then place a quick pass back over where the missing material is.

Suggestion: Rather than looking up literature on how to repair undercut try looking up preventing undercut so that you don't have to repair it in the first place.

Back many moons ago when I was handed the QC position at a local fab shop, I inherited an undercut problem. We worked with the welders until they could produce welds that didn't have any undercut that you could hang your fingernail on. We used a large diameter FCAW wire and our problem was primarily gun angle and travel speed.

https://www.weldpedia.com/2015/05/7-effective-ways-to-prevent-weld-undercutting.html
Parent - - By cbutts (*) Date 07-18-2019 04:46
we have been working with the individuals on preventing undercut, but when it does happen from opposing shifts I can not be there to train, so with that said I need to educate the day shift quality in what is correct and what is not correct.
Parent - By Lawrence (*****) Date 07-18-2019 12:03
Just going over the original post, several thoughts come to mind that may not have been addressed by the excellent responders so far.

cbutts   mentions 2G cover passes are the primary undercut conditions...  

Ok, some pictures of some actual weld profiles would go a long way to determining root cause.....  Because as mentioned already, how to repair the undercut is of much less importance when compared to how the undercut is getting there in the first place.

Rework, even when the part is still on the factory floor is typically a 5X cost increase...  !!!!!

Do you want Johnny welding on base plates and stiffeners or do you want him grinding undercut or adding extra weld passes to cover undercut ?  

How many manhours annually do you spend repairing undercut ?     How many field service claims related to bad welds are you paying backcharges for ?       Put those costs against the cost of a little training and increased supervision on the shop floor ….  Training of MANAGERS  not welders...   The welders will do what they are told..  

This is a management problem not a welding problem....  Get that straight  8: O'clock Day 1.

Back to the 2G Horizontal cover pass...…..   No matter if GMAW, MCAW, FCAW or SMAW,,,, Gravity plays the biggest role in undercut, technique comes a close second.   

Are the welders trying to weave a second pass horizontally over a first pass?   Are they placing stringers?    A look at the welds themselves and the undercut is going to say a lot about root cause...… Prolly everything needed to provide a solution.

I won't rule out overheated base metal, but I give that phenomena absolute last place on root causes for undercut.   A 300 Amp horizontal stringer bead with GMAW or FCAW can be place on a 500F weldment without undercut if the technique is correct...    Unless it's glowing red, this is not likely the primary cause...  With all due respect to my betters who have already spoken to this.
Parent - - By jwright650 (*****) Date 07-18-2019 12:30
cbutts
I feel for ya with the two opposing shifts.....BTDT. I used to get the ole "first shift left this for me"...."this was waiting here for me to fix when I came in this morning"....and various versions of the same blame game. I felt the best thing that happened was when we eliminated the production welding on 2nd shift, man did that free up a lot of repair work. I understand that you have a second shift and now have to figure out a course of action to get the undercut under control.

Lawrence we used a 3/32" FCAW and it was pretty brutal trying to weld the second side of 1/4" and 5/16" thick material without a few minutes in between passes and keep it undercut free.
Parent - - By Lawrence (*****) Date 07-18-2019 14:40
That's some sizable wire there John

Big lure.... Big Fish  :)
Parent - By jwright650 (*****) Date 07-18-2019 15:20
Fish on :razz:
Parent - By praga312 Date 07-19-2019 02:38
I thought undercut is an unavoidable problem. Can this be avoided during plasma welding of sheet metals ?
Parent - - By WeldinFool (**) Date 07-16-2019 19:14
To repair undercut, D1.1 clause 5.25.1.2 states "the surfaces shall be prepared (cleaned, see 5.29) and additional weld metal shall be deposited". Unless you have an Engineer's approval to do otherwise, this is the only code approved way to repair this discontinuity.
Parent - - By SWN1158 (***) Date 07-16-2019 19:55
In 5.25 Repairs, the current 2015 edition of D1.1 states “The removal of weld metal or portions of the base metal may be done by machining, grinding, chipping, or gouging. It shall be done in such a manner that the adjacent weld metal or base metal is not nicked or gouged. Oxygen gouging shall only be permitted for use on as-rolled steels. Unacceptable portions of the weld shall be removed without substantial removal of the base metal. The surfaces shall be cleaned thoroughly before welding. Weld metal shall be deposited to compensate for any deficiency in size.”

I typically see undercut on the edges of shop welded framing L’s. I’ve always removed this undercut, when it’s not really deep, by slightly grinding the edge of the angle profile, which does not affect the angle thickness. This, in essence, does not, in my humble opinion, remove a substantial amount of base metal of the angle.
Parent - - By jwright650 (*****) Date 07-16-2019 22:22
Scott, I have seen that too. Shear tabs have given  our shop fits until I suggested that we weld the fillet weld on one side and move on to the next shear tab and then the next and so on, then come back and place the fillet weld on the otherside. This gave the shear tab time to cool off a bit and made placing the fillet weld on the other side manageable.
Parent - - By SWN1158 (***) Date 07-17-2019 00:46 Edited 07-17-2019 12:35
The AISC states that "an accepted practice to avoid notches in base metal is to stop fillet welds short of the edge of the base metal by a length approximately equal to the size of the weld", which I encourage in the shop. The weld is started at the top of the shear tab and stops one weld size from the bottom, where welders tend to hold back a little on the weld progression to allow the weld to build up so as to eliminate the crater. The problem has been they hold back long enough to create undercut. But stopping one fillet weld distance from the bottom of the shear tab eliminates this issue.
Parent - - By jwright650 (*****) Date 07-17-2019 11:50
That is a good practice(on thin shear tabs and stiffener plates).

Problem we had was undercut for the entire length on the second (back) side due to the shear tab having the heat still in it from making the pass on the front side and the welder not paying attention to the fact that he was leaving undercut in the shear tab along the top toe of the fillet weld. Having multiple shear tabs on one side of the beam web, they were able to weld one side only on each shear tab and then come back after a few minutes and weld the other side which seemed to help eliminate the majority of the undercutting.
Parent - By SWN1158 (***) Date 07-17-2019 12:35 Edited 07-17-2019 12:41
We rarely have undercut in that situation, but moving forward, when we do, that is definitely something I need to remember. Thank you.

John, I just realized a typo on my last post, where I referenced "by a length approximately the SIDE of the weld." It now says "the SIZE of the weld."

Another affirmation that size does matter
Parent - - By LToca85 (**) Date 07-23-2019 19:32
Scott could you tell me where i can find this on AISC, I want to show this to my boss, we have been getting some undercut on the edge of shear plates too and i really need to change this.
Parent - - By LToca85 (**) Date 07-23-2019 19:37
Never mind found it, thanks
Parent - - By SWN1158 (***) Date 07-23-2019 20:11
For anyone else interested, this is in the AISC Steel Construction Manual, at the bottom of page 16.1-120 in Section J2........

User Note: Fillet weld terminations should be detailed in a manner that does not result in a notch in the base metal transverse to applied tension loads that can occur as a result of normal fabrication. An accepted practice to avoid notches in base metal is to stop fillet welds short of the edge of the base metal by a length approximately equal to the size of the weld. In most welds, the effect of stopping short can be neglected in strength calculations.

The user note also goes on to indicate two common details where welds are terminated short of the end of the joint.
Parent - By Shane Feder (****) Date 07-24-2019 11:51
To summarise everything that has been noted above.
Undercut is a discontinuity / defect that can be eradicated easily by three basic criteria - correct wire / electrode size, correct interpass temperatures and correct technique.
Cheers,
Shane
Up Topic Welding Industry / General Welding Discussion / undercut fix

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill