Weld 1

[StrP88]

I appreciate your input
I have the option to locate the weld parallel to the load "P" or perpendicular to the weld, any of the choices is better or recommended from the other?
Thank you

 

[Teguci]

AISC - Table J2.5
- In short, for fillet welds, parallel is superior.

 

[StrP88]

Teguci (Structural)
Thank you for your response. I have the AISC 9 Edition old book and didnt see table J-2.5
I guess is a different version. can you possibly attach that table, I would greatly appreciate it
Is it directly stating the weld parallel to load is better?

Thanks

 

[Teguci]

Is that the Green book? Goto specifications chapter on connections - subpart welds - look for allowable stress table for different welds.

Paraphrasing for fillet welds - Tension or compression parallel to weld axis - this loading need not be consider in design of welds joining the parts.

 

[MotorCity]

Actually, you can get up to a 50% increase in weld strength for a perpendicular loaded fillet weld compared to the same longitudinal loaded fillet weld. The caveat is that you need to consider a decrease in ductility, but it is usually not significant for a single line weld.

 

[Hokie93]

If I understand the load transfer in question, MotorCity has it right. That is, I am assuming you are transferring loads such that the weld will be in shear, either parallel or perpendicular to its longitudinal axis.

A great explanation of the condition Teguci mentions can be found in an article (Welding Wisdom Part One) by Duane K. Miller in the August 2015 issue of Modern Steel Construction magazine. The article is available here:

http://msc.aisc.org/globalassets/modern-steel/archives/2015/08/welding.pdf.

 

[Teguci]

That's scary.

I will see your Duane K. Miller and raise you Omar Blodgett - Please note his interpretation regarding parallel loading. Please also note that the conclusion is that fillet welds loaded parallel are weaker. Muy mal. In this case either AISC is wrong or Omar is (and I definitely am). I do not accept Miller's interpretation because we don't design welds for non-existent loads (in his figure 1 and 2 because there is no load transfer therefore the welds are good doesn't pan out).

http://weldingdesign.com/blodgett/wdf_18043

 

[nutte]

Teguci, remember that fillet welds are always (almost always?) loaded in shear through their effective throat. Every one of those diagrams in your link load the fillet weld in shear on its effective throat.

 

[StrP88]

Teguci (Structural)
May I ask where (what page and what paragraph) in the Duane K. Miller paper states that weld perpendicular to loading is better than parallel?

Guy, please help me to get the bottom line, I am confused now
Dont need arguees, I was simply asking for some direction
I am an EIT

Thank

 

[Teguci]

@StrP88 - At this point, don't worry about the direction and use the same allowable stress on the weld for both directions. This is conservative.

Regarding the articles, figure 1 and 2 of Millers article show a welded box shape with no shear transfer between elements (the welds at the corners transfer no significant load = therefore OK). Blodgett's article says transverse loading will create a lower stress on the weld than a parallel weld. AISC, states for fillet welds with tension or compression parallel to weld axis - this loading need not be consider in design of welds joining the parts.

I consider AISC as the legal authority on this while I'd consider Blodgett the preeminent expert on all things welded. I am unfamiliar with Miller, but Modern-Steel magazine is reputable. Therefore, even the experts are at an impasse with no discernible bottom line.

 

[Lomarandil]

I also understand it how MotorCity explained, and I don't see in the Blodgett article where parallel welds are stated to be superior to perpendicular welds.

(assuming as before, we're talking about shear through the weld throat)

 

[StrP88]

Please tell me which one is preferred?

Thanks

 

[StrP88]

Sorry forgot the attachment

 

[nutte]

StrP88, if you're an EIT, you need a more current book. A fillet weld loaded perpendicularly to its axis is 50% stronger (but less ductile) than a fillet weld loaded parallel to its axis. This phenomenon has been known to engineers for decades, but it was not in the old green book (9th edition Manual of Steel Construction). It is in the AISC Specification contained in the current Steel Construction Manual (and has been since 1999, I believe).

 

[Hokie93]

Properly sized and detailed with all applicable limit states checked, either weld configuration can work. Per AISC 360-10 Section J2.4, you can get a 50% increase in strength for a fillet weld loaded at 90-degrees to its longitudinal axis. If you don't need this additional strength, then orienting the welds parallel to the load should be fine (subject to the weld being properly sized, applicable limit states checked, etc.). As nutte indicated, the 50% increase is not in the 1989 AISC Specification.

 

[StrP88]

nutte (Structural)
Thanks for response, please reference which AISC edition and what page or what table or paragraph
Thank you so much

Any other references stating this please attach only that page or paragraph

 

[KootK]

Another advantage of the perpendicular welds is reduced shear lag in the connection. That said, I feel that the parallel welds would be the better choice in most situations. You might have more weld but, I suspect, a cheaper connection. With the perpendicular welds you'd have to flip the part on the table (shop weld) or relocate to the other side of the part (field weld). Either way, I believe that the welding would be more economical were it all to be done from one side of the part as in OP's scenario #1. And who doesn't dig a little extra ductility on the cheap, am I right?

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.

 

[StrP88]

Thanks Hookie for clearing the mud and thanks everybody (including Kootk,nutte and Teguci)for bringing valuable points to this forum.

Now the question here is when you have a beam sitting over another beam crossed (assume both are w shapes and same size) and to provide a lateral support for the below beam (which supports above beams sitting on it) by the above beams where do you locate the welds parallel to above beams or parallel to the below beam?
Please draw it so your answer is clear to be understood.

Very much appreciate it for it
I look forward for your comment

 

[KootK]

Parallel to above. Parallel to below means welding in the overhead position which is to be avoided whenever possible because it's very difficult to do. I've disregarded your mandatory sketch requirement because time is limited and my advice is simple. If that's a problem, just ignore my advice.

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.

 

[Teguci]

Enter "Welding Across Flange" discussion here -

http://www.eng-tips.com/viewthread.cfm?qid=45466

As for me bringing valuable points, I feel that all I did was muddy the waters. Thanks Hokie for cleaning up.