Welding in the 'k-area' 1

[GarethJohn]

I am designing a welded moment connection using Australian Standard AS4100.

Common practise in Australia is to fully weld continuity stiffeners at moment connections to the column flanges and web. However, the US company we are working with on our project have expressed concern about welding on the 'k-area' of the column (apparently AISC recommends against it).

Is this a recommendation only or is it mandatory when designing to US design codes? What do the US welding codes say on the matter?

I have not seen anything like this in the Australian Standards and am reluctant to clip or scallop stiffeners for two reasons:
1. it is harder to maintain continuity of welds through the scallops
2. surface preparation and painting of the steelwork around the scallops may be compromised - i.e. harder to paint in and around scallops.

Has anyone come across a requirement to avoid welds on the 'k-area' in the Australian codes. How do US fabricators get around the two issues above?

 

[SteelPE]

I'm not really an expert on they type of welding you mentioned but I will say that every seminar I have gone to (given by the AISC) recommends to not weld in the k area. From what I recall, residual stresses left over in the beam due to rolling can cause cracks if you weld in this area. Now, the AISC does allow some items to be welded if required. One item is web doublers. In this instance, the fabricator has to be diligent and check for cracking during the fabrication process. If cracks occur they need to be repaired.

Usually continuity plates are designed to transfer beam flange forces through the column to another beam or to the web of the column. In these instances the welds along the column flange and web are sized appropriately to transfer these forces taking reductions for the "clipped" continuity plates.

I don't think may would worry about painting as the connections are usually buried within the building are are not exposed to the elements.

 

[dik]

I try to stay away from the fillet area... triaxial stress states and can initiate a brittle fracture in cold climates... only seen this once in 45 years... so may just be a personal 'quirk'.

Dik

 

[Lion06]

I've always heard and read to stay away from the k area. We show clipped corners for all our stiffener plates.

 

[JoshPlumSE]

In addition to brittle-ness, isn't there a cost issue? It's more expensive to fabricate exact dimension plates like this. And, what extra capacity are they really providing? If you just clip them in this area, then everything is cheaper and easier, with no real loss in capacity.

 

[EngineeringEric]

I have far less experience than others here, but it is my understanding that this is done to assist in a controllable plastification of the beam. If the section is welded the beam will transfer unintended forces in unaccounted fashion.

I want to say that some of these details were developed for AISC post Northridge/Kobe EQ.

Again, this is based on memory and only my 2-cents as someone who follows AISC and recommends one to follow them...

 

[dcarr82775]

I believe Northridge showed cracks originating in the K zone. If you do not have to worry about fatigue of cyclic loading welding the K zone probably isn't a big deal. If seismic is driving the moment frame I would stay away from it. You are not getting much out of that bit anyway.

 

[TLHS]

I clip them because it's generally just a pain in the butt to work in that region. It's not a huge percentage of the plate width or length, so it doesn't make a big difference from a weld standpoint, and that part of the section likely doesn't control your plate sizing.

So basically even if welding in that region were fine why would you make them fabricate and weld for that geometry?

 

[GarethJohn]

There are a few reasons why I am inclined to weld in the k area in this case:

1. No mention of not welding the k area in the Australian Standards.
2. The fabricator has advised no additional cost or time impact. The stiffener is clipped and then the small void filled with weld metal during welding.
3. The Client specifications state that all ratholes shall be filled.
4. The project location is hot/tropical and non-seismic. The structure is not exposed to fatigue loading.
5. The structure is outdoors and close to the ocean (higher corrosion risk). There is a risk that painting and surface preparation in and around any void left will not be possible or will be compromised. There is also a requirement for all welding to be continous and it could be tricky continuing welds through ratholes left.

It is correct that the stiffeners are still structurally adequate with the rat holes. However, like most things engineering, this particular issue stopped being about strength a long time ago! Thanks for all the responses!

 

[hokie66]

GarethJohn,
In your situation, if galvanized, they are probably best left open. If other protective coatings, then close them. But if there is a specification requiring continuous welding, you may not have a choice.

 

[dik]

This just about sums it up, "the US company we are working with on our project have expressed concern about welding on the 'k-area' of the column (apparently AISC recommends against it)." If problems develop because you have welded in the 'fillet' area... you have bought yourself a bucket full of liability...

Dik

 

[wannabeSE]

GarethJohn,
Here are some links with AISC information on welding in the k-area

K-Area Welding in 2/14 Modern Steel:

http://www.modernsteel.com/Uploads/Issues/January_2014/012014_si.pdf

. Synopsis: welding in the k-area is discouraged but not prohibited.

AISC Advisory Statement on Mechanical Properties Near the Fillet of Wide Flange Shapes and Interim Recommendations, January 10, 1997:

http://www.modernsteel.com/Uploads/...7_Modern Steel Construction February 1997.pdf

The entire issue of Modern Steel is at this link (slow download). The information on k-Area welding is on page 18 (page 20 of the pdf).

The Commentary to AISC's specification has some information in Comm.J10.8 on page 362. The entire spec can be downloaded at:

http://www.aisc.org/content.aspx?id=2884#spec

You might be able to appease your US partners by inspecting for cracks with MT after welding and/or using fillet welds if you aren't already.

 

[SAIL3]

this is the first time I have become aware of this issue....if it is due to residual stress, then one would expect the same concern/issue with welding on the outer surface of the fla in the same area....

 

[CELinOttawa]

SAIL3: I don't think so... Opposite issue, ie reversed stress. It is a problem in the k-zone because you are heating and then cooling once again in an area under tensile forces and which is constantly in a state of wanting to draw the rest of the section in. That's not the case at all in the outter flanges which, while under compressive strain due to the material in the k-area, are more "set" in their state of stress and have no 'desire' to pull away/apart from new weldment material.

 

[SAIL3]

Cel.....I meant the outer surface area of the fla..say on the centerline of the member....anyway, I suspect it may also have something to do with welding a member/or portion of member that is under tension(from residual stress) and subject to cyclic/fatique loading, although, one would not expect the no of cycles to be that high in a seismic event....one would think that stopping and starting a weld in that area would lead to it's own problems with the probabilty of having a weld crater in that zone...

 

[dhengr]

GarethJohn:
It seems to me that that is going to be a very difficult detail to weld full, and get at during welding to get good penetration, particularly into the stiff. pl. And then, there is some chance that you will not get a good weld to the col. fillet from both sides, at the root, even if this is done in the shop. If they’ve clipped the stiff. pl. to miss the radius, they probably have to bevel the clip so they can get weld pen. to the pl. If they had actually fit the plate to the radius, a tedious process too, they could at least apply a continuous fillet weld. So, they haven’t saved any time or money either way, and end up with a crappy weld detail.

The fabricators like the clipped corner because that stiff. pl. is easier to fit-up, and they don’t have to weld into that corner which always leads to tri-axial stresses, compound residual stresses and the potential for an inferior weld detail/condition at that corner. You really must be very careful welding around the edges of these pls. too. They almost always leave notches at the vert. edge corners of the stiff. pls. where the weld face meets the edge corners.

High likelihood of weld defects, tough to inspect properly, tri-axial stress in a corner, combined residual stresses (rolling and welding), are just not a good condition. But, I have done it. The primary Northridge failure detail was different than yours. But, many of the failures were due to crappy weld details in the fillet area, a very stiff area of the member which makes it more susceptible to any defects and high combined stresses, and your detail does have that in common.