Best practices for reporting steel connection loads to a connection engineer 3

[jochav5280]

Good Day!

I would greatly appreciate if you could provide me with some help in identifying the best practices for reporting steel connections loads to a connection engineer. On our most recent project, simplifying/shortcut methods such as 50% UDL was used for determining the required shear force to use at Beam-to-Beam connections, but this led to a lot of ridiculously expensive/beefy connections that were far too conservative. I'd like to eliminate this.

Is it possible to just give the connection engineer your structural model in order for them to withdrawal the forces themselves as well as calculate the transfer forces? I'm not a fan of putting loads on our drawings as it complicates them and makes reading them difficult.

I'd appreciate any ideas and references that I could follow. Thank you very much!

jochav5280

 

[DETstru]

I've never given them the model for the purpose of getting reactions and I wouldn't really recommend it. The documents are the contract.

Typically what I do is classify them by beam size. For example, I'll make a schedule with various beam sizes and the appropriate LRFD reaction.

W16______22K
W18-W21__45K
W24X76___65K
etc...

Then I'll add a note saying that these are the connection forces unless a connection is specifically detailed or there is a reaction called out on the framing plan. For those oddball beams that don't fit the schedule I'll add the load next to the end of the beam on the framing plan.

For drag or axial loads I indicate them on the plan, detail, or framing elevation.

Moments (if you're doing a moment frame building) get annoying because it's hard to schedule it so they usually end up all on the framing plan (or elevation if you have one).

 

[sbisteel]

The best jobs I've worked on have a table that list a load for each beam depth. It's fast, conservative, and eliminates all the issues that the UL/2 method causes.

Transfer forces (I assume you mean axial loads from drag struts and vertical bracing?) should just be marked on the drawings for each beam where applicable. I've seen them called out on elevation section cuts, which helps keep the plan views cleaner.

 

[jayrod12]

No.

If you want them to design for specific connection loads you must show them on your drawing. Otherwise they're designing for 50% of the UDL.

You cannot expect them to pour through your model and essentially design the project for you. Nor would I trust some other person of unknown experience (and likely even an experienced one wouldn't be proficient at design) to determine the loading.

Remember, these steel guys are detailers, not engineers. Their experience and knowledge is invaluable when it comes to figuring out the best way to construct something, but that doesn't mean they capable of determining the sizes of the members going into the project.

 

[jochav5280]

Thank you Gents,

I appreciate your responses!

In regards to calculating Transfer Forces, could any of you please describe how you accomplish this? Do you use a spreadsheet to calculate the transfer force for each load combination in order to find the worst force to report on your drawings? Any other methods? We use a spreadsheet combined with STAAD.Pro, which leads to very large files since there is so much data to consider, (in some cases, we have (2) vertical braces and (2) horizontal braces framing into our beam-to-column intersections, which requires that we bring the forces in for all load combinations associated with these (4) members.)

Hello jayrod12,

Understood; the idea to just transfer our model was the idea of our most senior engineer, so I wanted to see if this has been done before. Additionally, our connection engineering will not be done by the fabricator's detailer, it will be done by a licensed P.E. that will stamp his/her work. Nonetheless, I understand your point.

Thanks again!

jochav5280

 

[jayrod12]

Additionally, our connection engineering will not be done by the fabricator's detailer, it will be done by a licensed P.E. that will stamp his/her work

This is typical, but if you think that the connection engineer is doing all of the figuring out then you don't understand the process. The detailer comes up with something that should work, the connection engineer just confirms it does, or in the case where the detailer has no idea what to do then the engineer comes up with something from scratch.

Most detailers just put 3 bolts for a W300 or W360, 4 bolts for a W410 or W460, 5 Bolts for a W530....etc. Then they check whether they need a double angle or single angle connection based on the 1/2 UDL of the beam in question. The engineer then looks for anomalies. At least that's how it goes in the great white north.

 

[jochav5280]

Thank you jayrod12,

My experience in our last few projects is that our detailer's connection engineer took our information, engineered the connections and passed their summary onto the detailer for detailing, so it seems things are different in your area.

Thanks again,

jochav5280

 

[jayrod12]

What I wouldn't give for the connection engineer to see the drawings before the detailer. It's so bad around here that often we are reviewing shop drawings that haven't yet been looked at by the connection engineer. They're always revise and resubmit due to the missing engineer's seal.

 

[jochav5280]

Hello jayrod12,

Ah, yes, that sounds very painful!

We had difficulty with our situation too, even thought the engineering came first. The problem was that our connection engineer considered all of the connections independently of one another, so when it came time for our detailer to implement the connection summary, there was a ridiculous/inefficient back and forth between them and the connection engineer to address all of the interface issues. For example, our connection engineer would call out (4) rows of bolts on one side of a beam-to-girder clip-angle connection and (3) on the other, and the fourth bolt row on the deeper beam would interfere with the flange of the shallower beam; this lead to the connection engineer having to eliminate the interfering bolt and revise their calculation to confirm the new situation could support the load. In short, in order for connection design to proceed efficiently, I believe it's best for the engineer and detailer to be in the same office to address issues as they come.

Our office is moving towards taking on engineering of the connections and applying them to our Tekla models in-house in order to gain back control. Obviously there is no free-lunch here, so our hours will increase, but we believe overall our efficiency and schedule reliability will improve. We're working on developing custom Tekla components via API programming to enable our designers to just "click-click" to pop the connection in. Tekla has the ability to jump out to Excel spreadsheets as well to perform engineering calculations, so we are also working on developing these spreadsheets for the components; it's our hope to provide our fabricator's with fully connected models, and eventually the shop drawings themselves.

Thanks,

jochav5280

 

[hokie66]

jochav5280,

That sounds like "back to the future", the way we used to do it before there were "connection engineers", but also before there was Tekla, and spreadsheets for that matter. I never did understand the philosophy of hiving off the control of connections.

 

[cliff234]

Follow the AISC Code of Standard Practice. (See the COSP for a list of things that the EOR is responsible for showing - stiffener plates, etc.)
Show the reactions, moments (for moment connections) and axial loads (for trusses, braced frames, etc.)and make it clear as to whether they are service level or factored.
Provide details showing how you want unique or unusual connections designed. (I call these connection concept details.)
List restrictions (where applicable).

 

[JedClampett]

When I was in the power plant business, a long time ago, we would post all the reactions and transfer forces when they exceeded the capacity of a standard AISC Type II (bolted bolted connection) with 7/8 inch A325 bolts.
When I look at AISC now, there's no more Type II connections (I don't know why), but couldn't you make a typical detail for a bolted bolted or bolted welded connection and post loads when it exceeds its capacity?
It seems like deferring the connection design exclusively to the detailer just defers the pain.