Fillet Welds in OMF

[TheLinker]

I am looking at designing an OMF per AISC 341-05 Section 11. This will be the first time I do a full blown design to 341 and I have a few questions I would appreciate any input on.

I have an exterior pipe rack in SDC D (non-building structure similar to buildings) designed to R=3.5 (using ASCE 7-05 table 15.4-1 note “d” for acceptance). I cannot use a lower R value otherwise the foundations get huge.

I plan on designing the connections per sections AISC 341-05 11.2a, 11.2c and 11.5 as opposed to using a prequalified SMF or IMF connection per AISC 358.

Q1: Can a double sided fillet weld be used as the FM moment connection between the beam and column flange?
Referencing 341 section 11.2a requirement number (3) indicates required strengths for double sided PJP and double sided fillet welds, which leads me to believe fillet welds are acceptable for OMF’s.

Q2: If a fillet weld can be used, does it have to be demand critical?
Section 11.2c indicates only CJP’s for OMF’s to be demand critical so I would assume fillet welds do not need to be.

Q3 (Hypothetical maybe): Say a PJP weld was being used that was demand critical, since this is an exterior application, say 0deg F is the lowest anticipated service temp (LAST) per section 7.3b a qualified weld filler material to be used would have to be in accordance with 341 Appendix X for 0deg +20deg = 20 deg F for the notch toughness test. How is this accomplished?
Can the design document specify the required temperature and a filler material be purchased by the fab shop that already has been tested per Appendix X requirements? Does the fab shop performing the welding need to perform the verification test?

Q4: 341 section 5.1 Structural design drawings, How does one identify the SLRS members on a drawing?
I do not believe I have ever seen this done before and do not know if this is done in a typical or standard way. Can this be as simple as putting an “*” by the member sizes in the SLRS and adding a note explaining the “*” is part of the SLRS. Similarly for the connections and demand critical welds, can this be accomplished with an “*” or “SLRS” note pointed to each of the connections and welds.

Appreciate any / all input and comments. Thanks

 

[jdgengineer]

Let me take a stab at your questions.

1) Yes, I do believe a double-sided fillet weld can be used. I've never done it but it's allowed.

2) Yes, Fillet welds are not required to be demand critical -- I am not sure the reasoning for this as it was my understanding that whenever the failure of a connection results in the failure of LFRS that it should be demand critical weld. I'm assuming that it is only because Fillet Welds are only allowed for OMF and therefore are not expected to have the same level of ductility as a IMF or SMF.

3) I'll let someone else answer this as I'm not entirely sure. However I don't believe that the PJP weld needs to be demand critical in an OMF.

4) I think that most of these are handled with general notes / details of your lateral elements. Typically, for a moment frame I'll show the moment symbol (and have that shown in the legend) at the end of the beam and then draw an elevation of the frame as well. The elevation will then reference the moment connections in detail. The demand critical welds would typically be called out in general notes or on the moment frame detail. I don't think you need to add a "*" next to your beam to say its part of the moment frame, but it also wouldn't hurt if you wanted to.

 

[connectegr]

With an R = 3.5, are you sure OMF is acceptable. Typically we see anything with R > 3 as IMF or SMF.

http://www.FerrellEngineering.com

 

[jdgengineer]

Yes an OMF qualifies as an R of 3.5 -- See ASCE 7. OMF is specifically addressed in the AISC Seismic Provisions, and using an R > 3, the requirements in the Seismic Provisions must be followed. However, for an OMF they are not all that onerous.

 

[connectegr]

I don't think you can use fillet welds. See Fig 4-4, AISC Seismic Provisions. Although the prequalified connection details in AISC 358 are not required. The OMF moment connection detail is provided in the manual. The CJP of the flanges and web are required and are demand critical. Along with removal of the backing bars and addition 5/16 reinforcing fillets at the flanges.

http://www.FerrellEngineering.com

 

[jdgengineer]

I believe the detail shown in Figure 4-4 is merely one method of completing the moment frame welding for an OMF (although it is the most common). Figure 4-4 is not part of the AISC Seismic Provisions and is referenced as a Design Example. I think a lot of the time engineers specify this welding procedure for OMF (I know our office does), but I don't think it's the only option. See Section 11.2a of AISC 341. If you are using complete joint penetration welds, then use you have to follow this procedure. But per item 11.2a(3) double-sided fillet welds are allowed.

In the end it may just be semantics. When I have calced a double-sided fillet weld the weld size was large enough that it seemed to make sense to just use the complete joint penetration weld.

 

[connectegr]

Using connection design forces greater than the capacity of the beam, will result in extremely large fillet welds. I believe the detail provided is more than a simple suggestion. After the Northridge earthquake these connections were tested extensively. Many of the welds in this seismic event experienced cracks. This resulted in the FEMA documents and connection recommendations, which eventually became AISC 341. This also resulted in the development of AWS D1.8, and specifications of demand critical welds and filler metals. This said these design requirements are a "moving target", with more changes coming in the near future. Our policy has been to select the seismic detail with the most technical support. In the case of your OMF connection that is Figure 4-4. Modifications or deviation from this detail may create a difficult position to defend if a failure occurs.

Using R = 3, and avoiding AISC 341, is always an option. Depending on the number of seismic connections, the cost of larger foundations may be reasonable. These details, even OMF, have significant fabrication and erection costs.

http://www.FerrellEngineering.com

 

[TheLinker]

connect: I do not thnk I can use R = 3, per ASCE 7-05 table 15.4-1 my only option to stay in AISC 360 is R=1 which would result in huge foundations. The other options are R=3.25 (per 341), R=2.5 (per 341) and R=1 (360). I am not sure why R=2.5 would require 341 provisions, thought they were only for R > 3.0.

jd:you have a good point, the prlim beam size is W14x30 (it a small pipe rack) and per section 11.2a the required flange tension force is 105k +/- which would result in a 5/8" fillet weld which is greater than the beam flange thickness.

I may look at a few more beam sizes before I completely ditch the fillet weld option. A large fillet weld in the shop may be worth it if it results in not having to use demand critical welds and the extra fab costs of the weld access holes per 350, etc.. .

 

[jdgengineer]

Just to close the loop on the discussion. connectegr, I agree with you that the complete joint penetration weld procedure and access holes that are shown in AISC 341 are extensively testing as part of the SAC project. FEMA 350 provides a lot of good background information. I agree that in IMF or SMF that we are much more concerned with ductility and that ductility at the beam/column interface is more important (as was found in the 1994 Northridge Earthquake). However, with an OMF I believe there are other options since the ductility requirements and performance expectations are lower.

To confirm my thoughts, I e-mailed the AISC Steel Solutions center to ask for their interpretation of the code section. My question/response is as follows:

Question: I have a couple of questions regarding the welding requirements of an OMF. Section 11.2a (3) states that double-sided fillet welds need to be sized for 1.1RyFyAg of the connected part. Is this statement intended to imply that a double-sided fillet weld may be used in the connection between a moment frame beam and a moment frame column in lieu of the complete joint penetration connection that is more typical? Or is this statement only intended for fillet welds in continuity plates, cover plates, etc.

Answer: Fillet welds are allowed at the beam-to-column flange connections.

 

[TheLinker]

Thanks for the follow up jdg.

I have looked at this alittle more also and I am unable to see how a fillet weld can be sized for the 1.1RyFyAg however. I think I am missing something.

I am assuming the Ag = the flange area of the beam (delivering the tension to the column). Is this correct?

Using ASD with Ag = Area of the beam flange,
T =(1.1/1.5)RyFyAg = with Ry = 1.1 and 50 ksi = 40.3ksi*Ag = 40.3ksi*(bf*tf)

For sizing a weld (360 page 8-8) Rn/FS = 0.928D*L(length weld)

L = say 2xbf (will always be less due to k1)

setting T = Rn/FS
40.3*(bf*tf) = 0.928*D*2bf
21.7*tf = D (in 1/16 inch)
1.35*t.f = D (in 1 inch)

Which means the required double sided fillet weld size is always 133% +/- of the flange thickness. I do not see how this could work.

 

[connectegr]

Not that I recommend using fillet welds. But, have you considered using the k = 1.5 strength increase for fillet welds loaded transverse to their length. This may be a problem with welding the web for shear. The flange fillet welds may require yielding to transfer shear to the beam web.

The Steel Solutions response may be an impractical suggestion.

http://www.FerrellEngineering.com

 

[jdgengineer]

Yes, your logic appears to be mainly correct. One thing to note -- the weld is not longitudinally loaded so therefore you are allowed to increase the strength of the weld per equation J-5 of AISC 360 (50% increase in strength for transverse loading).

While others may disagree with me, I think that even if the weld was thicker than the flange thickness that it would be "ok". Per AISC 360 you are supposed to check the weld strength based on the lower of the base metal strength or the weld strength. The code is essentially making you design on a capacity based approach to develop the full-strength of the base metal. Therefore, your weld strength can be based solely on the weld thickness. I know a lot of engineers don't like to size fillet welds larger than the base metal, but I don't think its technically not allowed.

All of this essentially points to the fact that maybe a double-sided fillet weld is not the best option, although it is technically allowed.

 

[StructGuy123]

You can also just design the connections using R=1 for OMF instead of developing the entire strength of the beam. See the commentary in AISC 341