Minimum Fillet Weld Size, BTH-1 vs. AISC

[jcorrado]

In determining the required minimum fillet weld size on a Below-the-Hook Lifting Device I see a conflict between BTH-1, Table 3-3 and AISC Table J2.4, 13th edition and 9th edition. Both sources essentially state that the 'minimum size of fillet welds shall be not less than the size required to transmit calculated forces, nor the size as shown in Table X-X’.

BTH-1, Table 3-3 and AISC 9th edition, Table J2.4 both base the minimum fillet weld size on the "Material Thickness of Thicker Part Joined".

AISC 13th edition, Table J2.4 bases the minimum fillet weld size on the "Material Thickness of Thinner Part Joined". The commentary in the 13th edition reasons that it is based on the quench effect of thick material on small welds; very rapid cooling of weld metal may result in a loss of ductility.

AWS D14.1 Table 6 uses the Thicker part but also states that the weld size need not exceed the thickness of the Thinner part. BTH-1 section 1-4.6 for Welding also states: When conflicts exist between AWS D14.1 and BTH-1, BTH-1 governs.

So if I have a 1” thick lift lug being welded to a 1/4” thick flange:
-According to BTH-1 and AISC 9th edition, minimum weld size (based on thicker part) would be 5/16”.
-According to AISC 13th edition, minimum weld size (based on thinner part) would be 1/8”.
-According to AWS D14.1, minimum weld size (based on thicker part) would be 5/16” but per note (b) would need not exceed thickness of thinner member so would allow a 1/4” weld.

1/8” will handle the required forces but was planning on 1/4” welds but based on the BTH-1 Code it appears that minimum 5/16” fillet welds must be used. Anyone else run into this issue? Should BTH-1 correspond with AISC, 9th edition or AWS D14.1?

I appreciate any feedback.

 

[connectegr]

BTH-1 and AISC 9th Edition are both antiquated references. Minimum fillet requirements were changed in the mid 90's. Refer to AWS D1.1:2010, Table 5.8, Note a "for low-hydrogen processes, T equals the thickness of the thinner part joined", Note b (that you referred to) remains.

The table distinguishes between low and nonlow hydrogen processes, and whether loading is cyclical (Note c) to determine the minimum fillet weld size.

http://www.FerrellEngineering.com

Providing fabrication and erection efficient structural design of connections. Consulting services for structural welding and bolting.

 

[jcorrado]

Thanks for the response connectegr.
I agree that AISC 9th Edition is antiquated but ASME BTH-1 is the 2011 edition and is the standard for Below-the-Hook lifting devices which is what is being designed. I had looked at D1.1 as well but BTH-1 requires the use of D14.1. I guess what I'm thinking and what you are saying is that the Minimum fillet weld table in BTH-1 is not up to date with AISC 13th edition nor the AWS standards D1.1 or D14.1 and was wondering if there was a reason why it shouldn't reflect the most recent AISC & AWS specifications?
Thanks again for any input.

 

[connectegr]

Actually I am not familiar with BTH-1, so I don't know their committee process for conforming to AWS updates or other code changes. I am very familiar with the AISC and AWS process, and there is constant communication between the codes. Every effort is made to avoid discrepancies, but the still occur. And our task groups and committees get constant letters informing us of the confusion. The AWS change related to minimum fillet weld size is related to improvements in welding processes and consumables. The intent is to achieve adequate fusion and this has improved with technology. BTH-1 may be intentionally conservative. Good luck.

http://www.FerrellEngineering.com

Providing fabrication and erection efficient structural design of connections. Consulting services for structural welding and bolting.

 

[dhengr]

Jcorrado:
These various codes are changing at such a rapid rate and completely out of sequence w.r.t. each other that it is impossible for them to keep track of each other, and that’s over and above the typical editing errors and corrections which come along with any printing. Many of the changes have much more to do with keeping the publishing cycles happening fast, and the money machine churnnin to their benefit, than they do with really improving our finished products or making our engineering process easier and better. The primary reason for the min. weld size is... The thicker pl. acts as a heat sink, and they want you to use a large enough weld (input enough heat in the welding process) so the cool thicker pl. does not quench the weld as it cools. This quenching (quick cooling) leads to weaker, crack susceptible welds. However, this quenching problem can be overcome by preheating the thicker plate a bit before welding. Then the thicker pl. does not draw the heat away from the weld so quickly.

This doesn’t make the engineering process any less disconcerting, but it is a fact of life these days. And, I think we would all do well to get back to using a little less code minutia and a little more common sense and good engineering experience and judgement in our design process.
Use .375" doughnuts on each side of a .25" lifting lug, or .25" doughnuts and a .5" lifting lug and a .25" fillet weld, and you may have a more balanced design. The only reason you need the 1" pl. is for the shackle and shackle pin, and then it looks crazy down on the .25" flange. The more important thing might be the final quality of your details. Your lifting lug should not be perpendicular to a web below. There should be a nice clean load path from the shackle to the actual load. Welding should be high quality, nice and clean, with out stress raisers, etc.

 

[gtaw]

If and when there is a conflict between governing welding standards, I opt for the most conservative design.

When it comes to welding, there is no such thing as a perfect weld. The weld is a metallurgical continuity. There are always discontinuities present. They can be on the atomic level or the macro level. Some are revealed by visual examination and others are very difficult to detect even with the use of NDE such as RT or UT.

Over welding is not productive, but then again, I tend to be conservative because equipment is always subject to abuse and misuse. Workers often have the mindset that if the devise is rated for 1000 pounds, it will surely pick 1500 pounds this one time. After all isn't that why the engineer includes safety factor? Convincing workers otherwise is a challenge.

Best regards - Al