Alignment of Shear Tab Connection Welds on Opposite Sides of Column Web

[FoxSE14]

Hi All,

Please see attached sketch. I think we have successfully avoided situation #2 (bottom) where the shear plates and welds are exactly aligned on both sides of a column web, however, we ran into a close call earlier today that had a few of us sweating.

How often do you direct the detailer/fabricator to offset shear plates which connect to opposing faces of the same column web? If a detailer shows plates/welds aligned on a set of shops, do you direct them to change it such that the plates/welds are offset?

I'd like to think we would have been ok "in the real world" since we have 1/2" stabilizer plates above and below that would have assisted in force distribution, however, when we tried to quickly find references/guides directing to keep shear plates misaligned, none of us were able to quickly locate anything. Anyone know of the reference/code section where aligning welds on opposite sides of a column web is 'frowned upon'?

 

[DaveAtkins]

I have never heard that this is a problem, or that it is prohibited by code.

Is this a lamellar tearing concern?

DaveAtkins

 

[FoxSE14]

Dave - Yes, I believe that is the concern.

 

[KootK]

I also didn't know that this was considered a naughty condition.

How would lamellar tearing arise? I don't see the source of restraint, particularly if the stabilizer plates are added after the primary plate.

I imagine you get one heck of a heat affected zone between the two shear plates but, then, I'm not sure that the slight offset obviates that concern.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[sandman21]

Unless a special condition occurs (rare) I have never had them offset. The offset would not significantly reduce any lamellar tearing, if you were to have any which typically you wont. The stiffener plates wont aid with lamellar tearing.

 

[JAE]

Never worried about it. Never heard it was an issue even.

Isn't lamellar tearing only an issue in plates > 1 1/2" thick?

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[271828]

I have also never heard of this being a concern. I think these conditions are common.

 

[271828]

The web is a little worse off with the plates offset as shown in your plan detail. The horizontal stresses in the two shear tabs are offset a little, putting a small section of the web in out-of-plane shear.

 

[Agent666]

From a loadpath perspective (resolving tension in the beams potentially) I'd always align them. Provided you are checking the web yielding locally for the load form both sides I don't see any issue with them aligning or being offset. Often the erection sequence will dictate which side is preferable.

 

[TehMightyEngineer]

My understanding is this is not a lamellar tearing issue, it's a base metal strength issue (yes, base metal can control in a fillet weld). The issue here is the welds are sharing base metal and thus require double the normal minimum thickness of the web.

I'm sure you also get some interesting effects with shrinkage causing potential lamellar tearing.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[JoshPlumSE]

I'm pretty sure the OP was concerned about Lamellar tearing. Anyone want to pipe up about why this detail isn't susceptible to lamellar tearing? Conceptually, the way it's configured strikes me as exactly what could lead to lamellar tearing. The steel is highly restrained on either side by the welding.

Maybe, the fact that it's a web and is not very thick helps us out?

But, gosh, this question of lamellar tearing makes me wonder how any of our SMF welded frames work when we have continuity plates. Perhaps, this concept isn't as much of a concern anymore because we're dealing with newer more ductile steel. Better metallurgy certainly means less lamellar tearing.

 

[WARose]

Anyone want to pipe up about why this detail isn't susceptible to lamellar tearing?

.........................

Maybe, the fact that it's a web and is not very thickness helps us out?

You answered your own question. I normally don't even think about lamellar tearing until about 2" of thickness. (Unless we are talking a very small weld at a very stiff point.)

 

[JAE]

That's what I mentioned above, WARose, that this is a concern in thick plates, not typical wide flange webs.

However, some heavy W14's, etc. have pretty thick webs where it might be a concern.


Check out Eng-Tips Forum's Policies here:
faq731-376

 

[DaveAtkins]

OK, that page from the steel manual does make sense--it is sort of like having bolted connections on both sides of a column web--bolt bearing on the beam web becomes an issue.

But I don't think the type of connection mentioned in the OP should be avoided altogether. With 50 ksi steel, my guess is a fillet weld on both sides of a column web will still be governed by weld strength.

However--I will start checking this in the future.

DaveAtkins

 

[JAE]

Dave,
I think after discussing it here in our office - we were mostly concerned with the use of single shear connector plates on each side of a web.
With the typical plates, and welds (3/16" and 1/4" fillets) used on these types of connections, the supporting member web doesn't usually get close to controlling in 99% of the cases.

For high shear loads, with deep, thin webbed supporting girders, then maybe look at it but even then it doesn't usually matter as the weld capacity of the connector plates isn't the controlling item anyway as the bolt shear or plate shear/rupture is lower.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[WARose]

I think (the one time I looked at this), the stress in the web was limited to the fact of the web rotating to match the (deflected) slope of the beam and connector.

With a flexible web and little deflection.....i wouldn't think it would be that much of a issue.

May take another crack at this today to confirm (it's been years since I've thought about it).