Beam to beam welded joint 2

[michelemarchiol]

Dears,

I have a question regarding this kind of connection.

It's a welded connection between beams, as shown in the image uploaded.

Secondary beams have notches in both flanges. Secondary beams are welded to the principal beam in 3 points: (1) in bottom flange; (2) in the web; (3) in top flange

A designer proposed me to use this kind of joint, but to be honest I have never seen this kind of solution before and I don't really like it (due to the "holes" between secondary beam web and principal beam flanges).

Is this a solution that you have just seen before? Have you some reason to avoid this kind of solution?

Please tell me what you think about this joint.

¯

Thank you in advance.

 

[GeorgeTheCivilEngineer]

Is it a nominal pinned or fixed connection?

Is this to be welded in a fabrication shop or on site?

Notches (or 'mouseholes') are normal and assist with fitting, welding, fatigue or drainage... but this is an unusual arrangement.

 

[michelemarchiol]

Thank you for your answer, George.

It's a fixed connection.
It has to be welded in workshop, not on site.

I agree it is unusual and I don't really like it, I have always seen bolted solution (considering pinned connection) if the beam has notches.

 

[SWComposites]

The flange weld config's look bad, particularly on the side with tensile loads. Need some sort of welded or bolted straps spanning over the butt joints at the flanges.

 

[JAE]

The openings in the web are typical but aren't drawn/sized correctly - see the image below for a typical connection using bolts for the shear connection.
It's murder trying to hold the secondary beams in position and start welding on a web in the air.

 

[Enable]

Also notice that the shop prepared flange bevels in JAE's detail are oriented the same way (not mirrored). That way the welder can weld both the top/bot flange from the same position above the connection. It's much more efficient that way.

 

[dik]

Best to have bolted connections for shear... holds the beam in place while it can be welded.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[JAE]

Ditto on what Enable says - top welding is always preferred - overhead welding is more difficult and has less general quality.

 

[JStephen]

If the depths or the flange thicknesses of the beams vary, it would complicate the detail.
You're cutting the secondary beam in 3 different places and assuming all of those cuts are perfect to fit up like that.

 

[TLHS]

My opinion depends on who the designer is and who you are. Is this someone in your office proposing that you use it? Are you a design engineer or are you working for the fabricator?

This is all assuming that we aren't talking about a high criticality connection here, or anything loaded significantly dynamically. I see this type of detail in things like skids a lot and they are normally just seeing some small fraction of the moment capacity of the member.

-If this were from a client or fabricator asking me to approve it, I likely wouldn't blink other than pointing out the weld position issue. It'll transfer load and if they want to do it in a not perfectly ideal way then I'm not going to worry too much.
-If I were working for the fabricator and had a relationship with them, I'd likely try to have a quick conversation about constructability and maybe suggest something
-if this were my detail as a design engineer I might add a shear tab, I'd likely redraw the way the cuts look, and I'd change out the weld callouts to give the fabricator more leeway to pick their methodology. I'd just tell them that I need a complete joint penetration weld (or a PJP with a required throat) on the flanges and then specify the fillets or whatever is required on the shear connection. Unless I need a specific procedure for a good reason, or have a strong opinion on the easiest direction to weld from, there's no reason to force the fabricator into my method.

 

[Flotsam7018]

In a past life I was a structural certified welder. In order to receive the AWS D1.1 unlimited certification, you have to be able to achieve a CJP weld on a 1" thick plate in the vertical and overhead positions which then undergoes destructive testing.

A certified welder should be very comfortable welding overhead and in my experience there were many moment connection CJP flange welds where I would've much preferred to weld the bottom flange overhead as there are no obstacles, awkward rod angles, or additional stops/starts required. You've just got to remember to wear your leathers.

 

[michelemarchiol]

Thank you for all your replies.

Anyway, the welds have to be done in the workshop, not in site. So it's not necessary to have bolts for shear. Also the discussion regarding overhead weld, that's should not be relevant in my situation.
I was focused on the load transmission here. I don't really like the design, since it increase the stresses from M/N in the flanges (you have to avoid the contribute of the web, which we know it's small but not null).

Anyway, principal and secondary beams have same section, so there are no problem regarding different height and thickness of flanges (except tolerances).

Here I'm the project engineer, I have to check and approve the design of an external technical office.

 

[DrZoidberWoop]

There is nothing wrong with the shop-welded node connection that has been proposed.

The web-to-web connections should be fillet welds, unless 1"+ fillets are needed to develop the web (convert for yourself if you're a dirty metric user).

If you can, try to design the flange-to-flange welds as PJPs for the design moment. If CJPs are req'd, spec proper weld access and a single bevel weld w/ shop back-gouge (no backing).

Probably most importantly, make sure your steel piece's transportation width is deliverable from shop to worksite.

 

[JAE]

I still maintain that overhead welding is more difficult since you are fighting gravity and it takes a more talented/skilled welder to achieve a proper weld without voids or flaws.

Also, for building or other vertical construction with the welding in the field and the beam in place - changing from an downward weld on the top flange to an overhead weld on the bottom flange requires the welder to change position, which is time consuming and more costly - so why do it?

For the OP's situation, apparently the shop welding can allow for the assembly to be turned over so everything is downward.

 

[dik]

JAE... I work with metal fabricators daily, and you are correct.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Flotsam7018]

JAE/dik - agreed that overhead welding is more challenging and if we're talking about shop welds then rarely would there be a case in which I would prefer an overhead weld.

I have personally experienced field moment connection welds in which I would prefer to weld the bottom flange overhead and welders with the correct qualifications should have the skills to perform overhead/vertical welding without much difficulty (vertical is often more difficult than overhead for many folks). Does it require a higher level of skill, yes - should we be able to trust that the certifications required in order to be able to perform such a task makes them eligible to perform that task - yes. Therefore, I will specify a field weld out-of-position if it makes a simpler detail, while I have known some engineers to go out of their way to avoid such a condition. Perhaps if we have more faith in our builders, they may have more faith in us.

Stepping off the soapbox as I realized this has nothing to do with the OP's detail.

 

[michelemarchiol]

Thank you again.

I still don't really like this kind of joint, but looks like it's admissible (as DrZoidberWoop said)

What if the beams have not H section but have C sezione? Since generally the internal edge of the flanges are inclined, it looks pretty hard to perform the chamfer and the CJP/PJP there.

Do you agree?