Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Do you guys defer steel connections? Should it be on the contact? 1

reverbz

Structural
Joined
Aug 20, 2024
Messages
94
Location
US
Hey Guys,

So I was just wondering do you guys always defer steel beam connections? Even on more simple projects? Also, do you think this should have been specified on the proposal or is it standard practice?

Thank you!
 
I'm on the other end. I receive the contracts and do the steel connection designs for a fabricator. Delegated steel conx design seems to be mostly an east-coast USA phenomenon.

"Should it be on the contract?

Yes. Absolutely include the responsibilities and scope within the Original Contract Specifications. Usually, the delegated/deferred connection designs are explicitly stated to be in accordance with AISC 303 "Code of Standard Practice," Section 3.1.1.(3) (page 16.3-11 in the blue 15th manual) and the min fabrication inspection requirements per AISC 360 Section N.'

There are a lot of webinars dedicated to the subject of delegated structural steel connection design out there. They all go over the basic requirements for contracting and what to expect in the workflow.
 
It is very common practice on the east coast of the US. I, however, despise it. It has it's uses - especially on very large projects or projects with unique connections. Fabricator input early can be critical for coming up with a really efficient design. But generally speaking, I find it's more work to specify all the loading and then review the calculations than it is to just do it myself. There's no real cost savings for the engineer, though shifting effort from design to CA can have benefits for scheduling if your firm is good at managing them. IF.

I have a clause in my proposals and service agreements that states that I reserve the right to delegate certain specialty aspects of a structure in accordance with standard industry practice. This include, but are not limited to, steel connections, open web steel joists, metal plate connected wood trusses, etc.
 
Most of my jobs are small enough that I do my own connection design. I agree with everything Pham stated..
 
You should definitely specify in your Agreement if you won't be designing the connections.

I like to design the connections, or at a minimum make sure a reasonable connection should work for the member sizes shown on the plans. For example, with really shallow I-beams (like W8s and W10s) it's not uncommon that a basic shear tab connection with (2) bolts won't work. I'd rather know that up-front so I can either choose a deeper beam section (if possible) or a different connection type, rather than wait for the steel fabricator to (hopefully) catch it.

I'd also add that I've seen a very wide range recently in terms of the quality of work done by steel detailers and fabricators. It makes me more cautious when it comes to handing off this very important part of the design.
 
It's common where I practice to defer steel connections for any steel building.
 
You should definitely specify in your Agreement if you won't be designing the connections.

I like to design the connections, or at a minimum make sure a reasonable connection should work for the member sizes shown on the plans. For example, with really shallow I-beams (like W8s and W10s) it's not uncommon that a basic shear tab connection with (2) bolts won't work. I'd rather know that up-front so I can either choose a deeper beam section (if possible) or a different connection type, rather than wait for the steel fabricator to (hopefully) catch it.

I'd also add that I've seen a very wide range recently in terms of the quality of work done by steel detailers and fabricators. It makes me more cautious when it comes to handing off this very important part of the design.
W8x10 and W10x12 are almost always very short, so little of the cost is from tonnage. The weight difference is small between those and W12x14, which almost never causes any connection difficulty.

I used to be an EOR and have worked as a connection designer in recent years. Knowing what I know now, I would almost never use a W8 and might even use W12x14 as the typical minimum.

W10x12 is often OK. In almost every case I've seen with a W10x12 causing problems, the EOR said to design for 1/2 the UDL. The solution is for the EOR to say design all W8 or W10 connections for 10 kips (or whatever) unless noted otherwise.

I guess it's obvious that a double-cope or deep single-cope causes a W10 to be problematic. Going to W12x14 as the min is a lot more forgiving for those.
 
Last edited:
Knowing what I know now, I would almost never use a W8 and might even use W12x14 as the typical minimum.
The W8s and W10s I mention are common in residential construction where a shallow beam is needed to fit in a limited floor depth. Without such a constraint, I normally use the deepest section possible.
 
From the economies of scale standpoint, specialization is the way to go. I personally would like to defer connection design as much as possible.
 
If you are using Revit for your structural design, you can export to Advance Steel (also autodesk) and model your connections. Advance steel also can design the connections. If you take it a step further, you can learn to generate the steel shop drawings which turned out to be a giant cash cow. I will never understand why engineers on the east coast don't design the connections.
 
I design the connections for everything and add a table to one of the connection details showing the number of bolts for each beam size. As part of the design, I make sure that the number of bolts shown in the table will work for the beam sizes listed. It's easier for me to do that than review all of the connection details in the steel submittal weeks after the design is finished and I have to remind myself about the loadings & reactions for each condition. I detail bracing and moment frame connections, too.
 
I design the connections for everything and add a table to one of the connection details showing the number of bolts for each beam size. As part of the design, I make sure that the number of bolts shown in the table will work for the beam sizes listed. It's easier for me to do that than review all of the connection details in the steel submittal weeks after the design is finished and I have to remind myself about the loadings & reactions for each condition. I detail bracing and moment frame connections, too.
As Engineer Of Record, don't you still have to review the steel shop drawings and make sure what is shown in the steel shop drawings (like number of bolts at each connection) matches your table in the detail you provide in your structural drawings?

Since we were reviewing the steel shop drawings anyway, we started taking our Revit model, and exporting the structural elements to advance steel. In the time it would take to review the shops, we could generate them. It was a total cash cow, especially if you used the built in dynamo tools to automate the process. I've never had a project that didn't assemble perfectly together. The architect still had to review the overall dimensions. Contractor was responsible for providing the bottom of base plate elevations. It's a no brainer. This whole process only works if you are modeling your structure in Revit accurately. Garbage in = garbage out.
 
As Engineer Of Record, don't you still have to review the steel shop drawings and make sure what is shown in the steel shop drawings (like number of bolts at each connection) matches your table in the detail you provide in your structural drawings?

Since we were reviewing the steel shop drawings anyway, we started taking our Revit model, and exporting the structural elements to advance steel. In the time it would take to review the shops, we could generate them. It was a total cash cow, especially if you used the built in dynamo tools to automate the process. I've never had a project that didn't assemble perfectly together. The architect still had to review the overall dimensions. Contractor was responsible for providing the bottom of base plate elevations. It's a no brainer. This whole process only works if you are modeling your structure in Revit accurately. Garbage in = garbage out.
This is an interesting approach as an EOR. I'm curious how this arrangement is structured contractually. Do you provide the shop drawings under a separate contract with the GC or architect, or is it part of your basic design contract with the architect? Do you deliver the shop drawings as a separate drawing package later or do you include them in the CD set?
 
I've done this a few different ways. The simplest way was to include the steel shop drawings in our structural engineering contract. Then the work was guaranteed and we made sure to model any steel with the shop drawing process in mind from the beginning. The drawings were provided as a separate drawing package, but the slick part was we could include the steel shop drawing 3d model in our CD's, which made the structural drawings look so much better and coordinated. We started doing this on every job and was got so busy, we spun off a separate steel detailing company with the vast majority of our work being our own structural projects. I think looking back on it, that this was unnecessary overhead and burden. There was internal rational about limiting liability but I think this was actually asking for trouble, as now you have two companies someone can sue. It was never an issue so I don't know how that would ever play out in real life.

I'm retired now, but I think if I were to do it again, I'd keep things much more simple. Just provide the steel shop drawings on your own projects. If you know the fabricator early, you can customize the look, numbering scheme, and other graphics to that shop. The value added to the project was off the charts. I had one huge industrial project where I made more on the steel shop drawings than I did on the engineering, which was a great fee to start with. BIM changed the game and also the rules. If you are still using autocad to draw anything, I'm sorry to inform you your companies days are numbered. Just wait till AI can generate revit models...

We tried to avoid contracting with the architect as much as possible. Most just picked the cheapest firm and marked it up on their end (different rant for a different day). If possible, we contracted directly with the owner across the board which skipped quite a bit of paperwork / drama.

General work flow:
1. Build structural 3d revit model including accurate steel sizes, locations...etc. At the time, we didn't model connection in Revit, although this process is much better now.
2. Export the steel elements to advance steel and add connections.
3. Save this file as a plain 3d cad (acis solids) file and insert into a blank revit generic family, which is then inserted into your structural model. This was important because of the way revit cuts 3d cad elements in views.
4. Drawings are now permit ready. Get paid for structural engineering portion of project.
5. Have advance steel generate your shop drawings off of the advance steel model and send off for approval to architect and contractor. I got this process down to a few hours even on a large project. Mainly checking each sheet to make sure the views that were auto arranged on the sheets were in the right place.
6. Once contractor provides final base plate elevations, you can generate the "For Fabrication" drawings. Get paid for shop drawings.

Each release of revit gets substantially better. The writing is on the wall that steps 2 and 3 will be done in revit in the near future.
 
Advance steel may work for some fabricators, but not all. Keep in mind this software is used to control all that fancy equipment. If someone is not an Advance user they may need to re-do the part drawings. At that point, that model is not a great time saver. Most fabricators have their own standard practices as well. It can be a problem if those things are not followed. The fee for the shop drawings is another issue. In all my years I have been provided some very terrible models from EOR's and architects. I would be very hesittant about a new teams model. That does not seem the case with Erik. We are working with an architect now and their CAD file has 1/4" errors. That stuff matters. The problem with Revit is the amount of 2D CAD work that is used. Few invest the time creating the families and the 2D CAD details are not coordinated well.
 
Advance steel may work for some fabricators, but not all. Keep in mind this software is used to control all that fancy equipment...
Fully agree, but, at least in our area, it wasn't an issue. Advance steel generated the NC / NC1 files needed for the machines local fabricators had. DXF files were used for the burn table machines that the plate fabricators used. There was one guy that had a 6 axis Python machine that used the NC1 files. What used to take 1 hour to fab was reduced to 8 min. FYI, nothing against Tekla, but advance steel was included in our subscription and it seemed to get the job done.
In all my years I have been provided some very terrible models from EOR's and architects. I would be very hesitant about a new teams model.
Also fully agree. I quite working with architects using autocad, but still had to work with garbage architectural revit models all the time. I still modeled everything accurately in Revit based on the dimensions called out in the architectural CD's and tried to avoid showing elements from the garbage architectural model where possible. Please note that as the EOR, I wasn't giving my revit model to a fabricator or detailer. I was the detailer. Nothing was lost in translation this way. The only model that was shared with the rest of the team (architect, mech, plumbing), was the structural revit model, which had my 3d cad solid steel model linked in, which was basically un-editable by anyone else. It was only provided for coordination with other disciplines like mechanical / plumbing.

Overall I really recommend engineers zoom out a bit and see what opportunities are out there given all the new fancy tools we have at our disposal now.
 
I understand what attracts people to Advance steel. It is included with our license as well, but I never use it. I work with a few that do. If you are acting as the detailer, it makes sense. The details is what takes the time as you know, so getting the typical Revit models I see from an EOR with the beams and columns does not save much time. Most still have 2D CAD departments, and I do not see that changing anytime soon.

The fact you mentioned dynamo gave it away you are not following the typical methods. We use grasshopper to drive models these days. That way we can change a few variables and all the details update when our flaky archs change something. There are a lot of things one can do with software.
 
You guys have problems with the models from the EOR? I've had no issues and seem to get better results handing our steel models in .ifc to detailers. Then again, this is industrial, arch is not making the models.
 
You guys have problems with the models from the EOR? I've had no issues and seem to get better results handing our steel models in .ifc to detailers. Then again, this is industrial, arch is not making the models.
We don't generate models as EOR, aside from 2D CAD files. The only models I get are from our steel fabricator clients when they retain us to do calculations; they model it in Tekla and give me the .ifc.
 

Part and Inventory Search

Sponsor

Back
Top