Max Weld size Question 4

[Bagman2524]

I'm connecting a 1/8" thick tube to the bottom flange of I-beams varying in thickness from 1/4" to 7/16". Sec J2.2b of ASD states that the min weld thickness depends on the thicker of the 2 parts joined and is per table J2.4.
For the max size fillet weld it states that for material less than 1/4", the weld is not greater than the thickness of the material. Does this mean the thicker part also? or the thinner part? If it goes by the thinner part than my max weld would be 1/8", but for I-beams flanges above 1/4" thick, my min would be 3/16" (greater than the max).

 

[recentgrad]

What you wrote is correct. For 7/16 thick flanges minimum weld is 3/16". However for 1/8" thick material, maximum weld size is 1/8". But it doesn’t mean that code is wrong. I have a feeling that you cannot furnish this connection as per code without reinforcing HSS web by adding plate. Check with AISC Design guide for hollow sections (D/t limitations etc).You will see that all will add up. Also look into the capacity of welds little bit. Like a fillet weld with one leg 1/4 and other 1/8 will have capacity of 1/8.If you can provide me connection lay out, I can be more specific. Excellent source is Descon software, it will take you step by step to options you can have.

 

[bjb]

I am assuming in the connection you describe that the tube walls are perpendicular to the beam flange, so that you have a Tee joint. The max. size of fillet welds in J2.2b is intended for lap joints, not for plates that are lapped. The reason for the max limitation is that the exposed upper corner of the plate edge tends to melt into the weld, and this can reduce the effective throat of the weld. Volume II Connections explains this on page 2-44.

If you are lapping thin plate to a relatively thicker beam flange, the weld size doesnt exceed the thickness of the thinner part, but there has to be adequate preheat to avoid cracking. Hope this helps.

 

[HgTX]

Someone more qualified than I am should write a max weld size FAQ.

Hg

Eng-Tips policies: faq731-376

 

[bjb]

I just reread my post and noticed a huge typo.

In the first paragraph, the second sentence should read "The max size of fillet welds in J2.2b is intended for lap joints, not for a Tee connection where the plates are perpendicular to eachother".

With Tee joints, you don't have melting of the corner as when the plates are lapped.

Sorry for the confusion.

 

[Bagman2524]

Thanks bjb. The joint is a "T" joint.
This clarifies the "that is permitted along edges of connected part" in J2.2b for the maximum size of fillet welds permitted.

 

[henri2]

Just a few points to add:

1. In the 9th edition of the AISC-ASD Spec, min size(leg) of fillet weld per Table J2.4 is based on material thickness of thicker part joined.

2. The AWS reference the 9th edition of the AISC-ASD Spec for bldgs was based on is one from the 80s.

3. In the latest AISC Spec for Structural Steel Bldgs (2005), the min size of fillet weld in Table J2.4 is based on material thickness of thinner part joined

4. The corresponding Table for min fillet weld size in AWS D1.1 (96 edition onwards) is Table 5.8..variables like low hyrdogen/non-low hydrogen, preheat etc can determine whether the thicker or thinner member joined govern the min size of weld. AISC modifies this table.

Question: Even though AWS D1.1 does not stipulate a maximum requirement for fillet welds used to attach tubular members to planar surfaces in T-joint configuration, designers are expected to use their engineering judgement to limit undesirable effects of shrinkage, distortion etc.

What is an acceptable plus tolerance for specified fillet weld size in the scenario described above...or is one necessary?

 

[samdamon]

Bagman- Yours truly recently specified an 1/8" plate to be welded horizontally to the bottom flange of a W16x26 (for supporting some light gauge), with a one inch long 1/8" fillet weld spaced six inches apart. The welds ended up looking pretty ratty (I think because the 1/8" pl thickness was a challenge for this particualr welder), and the 1/8" plate ended up badly wrinkled and distorted by heat (to the point it had to be cut off and redone) after welding. If the fabricator is on board suggest discussing that with him. Consider using a thicker plate just to minimize these issues, even if it isn't the most efficient use of material. Regards

 

[J1D]

In my impression, the maximum all-around fillet weld size of HSS connections is not limited the the thickness of the thinner part. The weld size can be larger than the HSS wall thickness. I saw some tables like that. Correst me if it not right.

 

[Bagman2524]

Figure 6 on pg 7.4-3 of Blodgett's Design of Welded structures seems to confirm what bjb is saying about the max weld size being for lap joints.

 

[AggieYank]

J1D, the maximum fillet weld size is controlled by the thinner part.

For thicknesses up to and including 3/16" the maximum weld size is the thickness of the material.

For thicknesses greater than and including 1/4" the maximum weld size is the thickness of the material - 1/16".

So the maximum thickness of fillet weld for a 5/16" angle welded to a 3/16" tube is 3/16". And the maximum weld for a 3/8" tube fillet welded to a 5/16" tube is 1/4".

 

[Bagman2524]

AggieYank, Where are you getting this information from? This seems to contradict most of the previous posts.

 

[HgTX]

The max thickness limitations only apply to lap joints. AISC language reads "along the edges".

Maximum useful fillet weld size in a T joint is 1.414 times the thickness of the thinner part, since that gives an effective throat the same as the thinner part.

Hg

Eng-Tips policies: faq731-376

 

[henri2]

HgTX, so if we were to join a 3/8 in (0.375 in) plate to a 1/2 in plate in a lap joint using fillet welds, the maximum useful fillet weld size would be 1.414[0.375]= 0.53 in..No? This is larger than the thickness of the thicker plate...perhaps there is a further provision.

From what I recall, 1.414 x size(leg of convex fillet weld) = face width of inscribed triangle in fillet weld cross-section..and 1/2 of that (0.707 x size) = theoretical throat..assuming isoceles inscribed triangle.

 

[henri2]

HgTX, please ignore my previous post...thought you were referring to lap joints.

 

[shepherd]

I agree that the maximum size specified by AISC is only for edge of lapped joints. In the case of welding to a tube wall, I don't know of any provisions that prevent you from providing a weld larger than the tube wall thickness. For larger tubes with thin walls, sometimes larger welds are required because the maximum weld stress is significantly higher than the average stress due to tube wall flexiblity (See AWS D1.1).

 

[HgTX]

No. As I said, the maximum provisions DO apply to lap joints.

In a lap joint, the maximum fillet weld is 1/16" less than the thickness of the plate along whose edge the plate is made, unless the plate is less than 1/4" thick, in which case the weld can be up to the thickness of the plate. The effective throat will be less than this, but that's all the weld you're allowed to put in there (and in the case of <1/4", all the weld you can get in there).

Actually I misspoke for T joints--although you're not going to get additional capacity by making your effective throat any larger than your material thickness, you may want to make that a minimum rather than a maximum, depending on your design & application, so that failure would occur in the base metal rather than the weld. Either way, the fillet weld leg size that has an effective throat equal to the base metal thickness t would be 1.414t (so that when you multiply size by 0.707 to get effective throat you're back at t).

HG

Eng-Tips policies: faq731-376

 

[bjb]

I agree with you HgTX. A limiting condition for a Tee joint is the shear capacity of the base material. If this is less than the shear capacity of the fillet weld, the oversized weld isn't helping you.

Using ASD '89 specs, the shear capacity of a 5/16" fillet weld is 4.64 kips per inch. If you were welding to 1/4" A36 plate, the shear capacity of the plate is 14.4(.25)=3.6 kips per inch.