Delegated Design of Steel Joist to Joist Girder Connection 3

[Pjstol]

A joist manufacturer on the west coast is refusing to design these connections.

They claim they can't do it because of code (even though a joist manufacturer will design these connections on the east coast).

so is it mandatory for the EOR to design these? if the joist contractor won't do it, can the EOR delegate these as typical steel connections to a third party?

 

[SteelPE]

I'm a bit confused.... what exactly is there to design for this connection. Doesn't the joist sit on top of the joist girder? Wouldn't you just be welding the joist to the top chord of the girder?

 

[Pjstol]

Yes. That weld.

 

[TOLstryk]

If the joist isn't acting as a chord or a collector, then isn't this just the standard weld detail in the joist catalog (1/8" x 2" both sides) checked against your uplift? I've rarely had to deviate from that.

 

[271828]

Did they give a code reference?

 

[271828]

Is this a govt job? I have seen a navfac requirement that the LFRS connections not be delegated.

 

[Standard American]

Some joist manufacturer's are quite the prima donnas. I had one call me on several different jobs asking for the net uplift. I told him the gravity and wind load schedule is on the drawings, just take 0.6D - uplift. He told me he can't do the simple math to calculate and I need to provide it.

Long story short, if they are pushing back just provide a typical detail for the connection and move on. Somebody probably told them their insurance won't cover it or something and will not back off it. You will spend more time fighting with them then it will take to check a small weld.

 

[Aesur]

I told him the gravity and wind load schedule is on the drawings, just take 0.6D - uplift. He told me he can't do the simple math to calculate and I need to provide it.

Too bad we can't say that to the PEMB manufacturers where their loads don't make sense..

 

[Once20036]

The SJI codes of standard practice says that the EOR is to design the connection of the joists & joist girders.
In cases of joist seat rollover, or any additional axial loads (to stabilize the tops of columns under wind loads, for example), the SJI minimum welds might not be enough and they need to be checked.

I customarily push this back onto the steel contractor who is responsible for connection designs.
I`ve seen them use the connection designer, and I`ve seen the joist manufacturer provide the design.

In my experience, as an EOR, I don't have enough information about the joist seat width, the joist seat holes, or the joist seat thickness to be comfortable handling this connection during the design phase. I`m not sure how SJI expects that work to be performed in the absense of detailed joist information.

 

[haynewp]

I've never had an issue with a joist designer and designing something that basic. Even when there were drag forces or rollover forces. If SJI expects the EOR to design the connection, then by default they have to design the joist seat and therefore are doing part of their joist design. Seems asinine and a good way to hold up a project.

 

[GC_Hopi]

SJI is heavily influenced (supported) by the manufacturers so they do not mind contradictions that put the responsibility on the EOR and give them an out.

 

[Aesur]

In my experience we almost always design these connections, if there are axial loads, assuming being used as a drag, are you not just splicing across the joist girder (bypassing it) connecting the top chords of the joists together with a plate welded to the chords? I don't rely on the connection to the joist girder to transfer these loads, because it won't. I am struggling to think of a condition that I have used thus far where this wasn't the case. I believe you would almost always have all the information you need, including minimum seat sizes from the joist manuals to design for uplift loading.

 

[Aesur]

One thing that I have always wondered about however is, if they bid a project, they are bidding to fulfill their part of the project, if you defer or delegate these connections then I would think that should be part of their scope if they bid per the contract drawings. I could see an argument being made that they or the contractor then have to hire another engineer to design these connections if the sub refuses to do so. Curious what others think about this.

 

[JedClampett]

One thing that I have always wondered about however is, if they bid a project, they are bidding to fulfill their part of the project, if you defer or delegate these connections then I would think that should be part of their scope if they bid per the contract drawings. I could see an argument being made that they or the contractor then have to hire another engineer to design these connections if the sub refuses to do so. Curious what others think about this.

Got to love your sense of humor! If it's anything like my experience with PEMB anchorage, it will be a cold day in h___ when any of these happen.

 

[Once20036]

Aesur
- I agree an axial transfer force would be best transferred by top chord plates and by pass the seat connection.
- I`m not familiar with minimum seat sizes from SJI. That includes the width of the outstanding leg, thickness of the outstanding leg, and some sort of maximum slotted hole width/length? Can you advise where that information is located?4
- I agree that we typically delegate the connection design to the steel fabricator's engineer. They can certainly design this on behalf of the joist manufacturer. I also specifically point out this deviation from SJI standards prior to that contractor being awarded. It all depends on how the bid package was worded. On the west coast, I think there may not be a connections engineer, so it's either the EOR or the joist contractor. Again, it all depends on how that deviation from SJI standards was incorporated into the project. I think they`re right to stand behind SJI unless the conenction design was specifically required.

 

[haynewp]

In my experience we almost always design these connections, if there are axial loads, assuming being used as a drag, are you not just splicing across the joist girder (bypassing it) connecting the top chords of the joists together with a plate welded to the chords? I don't rely on the connection to the joist girder to transfer these loads, because it won't. I am struggling to think of a condition that I have used thus far where this wasn't the case. I believe you would almost always have all the information you need, including minimum seat sizes from the joist manuals to design for uplift loading.

If it's a few kips of axial they can take it through the seats if the joists are beam supported. WF beams for girders are typical for almost all of the cases on the building types I do and I usually switch to a collector WF beam if the axial is higher. That's probably why I haven't run into a problem with the axial transfer.

I can see if it's a warehouse with joist girders that you will always need a tie plate or tie angles. The tie plates should be less than the joist top chord width whatever it will be so that there aren't underhand welds. And the welds sizes should be somewhat compatible with their thickness of joist chords so you don't have like 1/4" fillet weld on a 1/8" thick joist top chord.

 

[Aesur]

Got to love your sense of humor! If it's anything like my experience with PEMB anchorage, it will be a cold day in h___ when any of these happen.

Humor is how you stay sane in this business!

Can you advise where that information is located?

See page 4 of Link Technically this is the max, but gives some rough guides of what you are allowed to go up to and they could accommodate for your drags.

On the west coast, I think there may not be a connections engineer, so it's either the EOR or the joist contractor.

That still blows my mind that it's common on the East cost to defer connections for the majority of structures. I can see it on huge structures where connections can be optimized by a specialty engineer, but for the other 99% of structures I suspect the EOR could easily design them.

 

[jerseyshore]

It's been much more popular in the last 5+ years out here. I remember a webinar we had actually encouraged it. If you have simple connections with 1 or 2 moment connections it's easier to just have EOR design it.

But those bigger jobs with maybe 6+ different MC designs and various end reactions, put the loads on the plans, show a typical detail, and review on shop drawings.