East Coast Connection Design 2

[thoughtofthis]

Without getting into the debate a few posts below, I am interested in how your companies delegate connection design. Currently, we design shear type connections and provide loads for moment connections. For a variety of reasons, I would like to see us delegate more of this to the fabricator.

Internally, there is some resistance and I can only speculate the reasons; however, a well thought out work flow might help things along. I am interested in what works well and what does not. At what stage of the design in a typical Design-Bid-Build do you involve the fabricator? How does it affect your bottom line and does it contribute to everyone on the project being profitable? How does it effect your risk profile? How do you guard against liability issues with your loads being published on a drawing or spec? How do you handle a building that gets re-purposed later (potentially impacting loads)? I'm interested in the good and bad, but not interested in a debate on whether it should be an accepted practice. Thanks!

 

[thoughtofthis]

Thank you connectegr, I would be interested in that. Does anyone have experience integrating this process with BIM? We are still using CAD for the most part, but are in the process of transitioning to Revit. It would be a convenient time to address these issues.

For those who have made this transition, how did you suggest the process to architects, owners, and/or contractors (for design build) to get the team on board?

 

[ukengineer58]

i cannot see one advantage of designing all the connections myself. most faby in uk have in house designers as part of the overhead so included in the steel cost. all i see is hours and fee eaten up by doing simple connection design. producing details that the fabricator will redraw on shop drawings so double work there. uneconomical designs as the fabricator may have to order in materials not in his stores. rfi's and queries to change the design to suit the fabricators prefered. just show the forces and review the calcs and fab drawings when they come in. does anyone really believe that takes longer than producing all the details in house.

 

[mijowe]

some/most of the fabrication software has connection design with back up calcs included in the software (clear comprehensive checks). getting the guy that inputs the information to do it correctly is a different story.

as far as revit goes, with limited experience, the steel fabricators have not used my revit model. they start from scratch using their own BIM software.

 

[connectegr]

We can also discuss your BIM concerns.

Most detailing software does NOT offer connection design. As with any engineering process there is packaged software for simple conditions. And there are very expensive software packages ($20k +) (for example SDS) that offer very good connection design information integrated into the model and shop drawings. For the best software the limitation is the experience of the user, which is rarely an engineer.

I believe the front runner in the BIM software is TEKLA. We are involved in an increasing number of BIM projects. And honestly I have mixed feelings on the value. But, I believe this will be a big part of the engineer's responsibility in the future. Ideally a BIM would incorporate the models of all the trades and provide an immediate representation of interferences and practical concerns. However, many of the modeling softwares do not communicate well, so an additional party must combine the information. Also structural models lack the geometric precision required for final shop drawings. And today a fabricator may have invested in SDS and the BIM may be in TEKLA. I can see a future where the fabricator is awarded a project with a BIM package. Having the appropriate software, the fabricator can push a button and generate the shop drawings. This will create some new contract clauses over erection and fabrication fit-up and tolerances.

It is coming, but I don't think it is mainstream yet.

http://www.FerrellEngineering.com

 

[thoughtofthis]

A few follow up questions:

1) How do you handle shop drawing review regarding connections, RFIs, and SI? I am thinking specifically about the workflow and who does what (in the case of the fabricator providing the engineer). In some cases we are also the SI.

2) How do you show connections on your structural drawings? For certain projects we show sections at the same locations as the arch and additional sections and enlarged plans at complicated areas. If you show a representational connections do you just not provide notes? or do you show it as hidden or not at all?

3) In my state, state projects are required to undergo review by the state construction office. Is this an issue on these projects because technically none of the connections will have been designed at this point?

 

[TLHS]

As far as detailing on the design drawing, if there's a reason you need a connection to go together a certain way you either provide a conceptual drawing showing the requirements with only the items you require noted, or you detail it all the way and take responsibility for it.

Otherwise you might artistically draw something in so that a zoomed in detail doesn't look stupid, but you wouldn't provide any sizes, dimensions or other information about the connection other than the required loads.

A problem I've often seen is that the detailers in the design engineer's office get overexcited and put dimensions and details all over connections that are supposed to be designed by others. The engineer stamps them without doing much of a check because he figures it's the fabricator's responsibility to check. The fabricator then looks and sees a stamped detail with all this information on it and just builds it because it's already designed. In this case this is an oversight of the design engineer, but it's a dangerous situation no matter who is at fault.

Personally, I never provide any extra information at the connections unless it's conceptually necessary to make my member arrangement work. I also have a note on every drawing saying that, unless specifically noted, all connections shown on the design drawings are conceptual in nature only and detailed design is the responsibility of the fabricator's engineer.

Specs from large clients drive me crazy, because they tend to be full of really poorly worded connection design specifications where it tells the fabricator to design to the greater of, or least of, or some other series of really complicated tests that you'll never be sure are being used properly and have no relationship to the actual forces in the structure.

 

[TLHS]

Oh, and as for shops... For relatively standard connections, the connection review is part of the shop drawing process. All the fancy detailing software is generally pretty capable of doing things like double angle connections. So you get a whole package of shops including all the connections and hopefully some supporting information or specs about the connections.

Other times it'll come in two stages and you'll get a copy of some connection calcs first.

Sometimes you never really see much in the way of calcs and you just get a pile of shops stamped by the connection design engineer.

 

[paddingtongreen]

Before the lawyers and bean counters started to restrict what we could do and before we had computer programs, I worked on big stuff, two unit power plants anywhere between 300 and 800MW apiece. As soon as we had a fabricator on board, I contacted his leas connection engineer/designer. After a chat where we set some procedures in place, the first being that I sent him my wind and earthquake sheets. These showed each bracing bay, separate sheets for each case. On the drawings we drew up a connection sample set that covered most types on the job. Simple beam conns and column splices were by the book. Then sketches and phone calls nailed the whole thing down.
Most jobs don't have enough steel to make that work in full, but all of my jobs were based on maximum information and communication; by the end, these engineers and I could finish each other's sentences, but we were, of course, too respectful to do that.

When the lawyers and bean counters moved in, we were to "not interfere in means and methods" because they were the Constructor's job. It is surprising what they can say does, or may, interfere with means and methods.

As I approached retirement, I tried sending complete copies of the program results, but "that belongs to us" and we won't give it away.

I should perhaps have mentioned that I detailed steel for a while.



Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[ToadJones]

Pad-
I think we are a generation apart on big coal plants.
I started off working on plants in the range from 600 to 1250 MW.

Connect-
I last used TEKLA in 2007.
At that point the shop drawing you could "scrub" out of a model were absolutely horrible. I would be embarrassed to give them to a shop for fab. They were cartoonish and had no where near the info you'd find on the good ole hand drawn or CAD shop drawings (I was always partial to American Bridge drawings).
QC on the drawings was almost impossible....The drawings had to be checked against a model....oops, hope no one changed my connections from the last night!
Often times I would open the "official" model to go back and check the connection and OH SH&^%!!!! there is no connection at all. The documentation process was non-existent and once we finally had some in place it TOOK LONGER THAN JUST DETAILING THE CONNECTIONS IN CAD!!!! AHH I hated it at times.

The BIM thing is also hard for me to believe in. We'd have entire power plants modeled in TEKLA/BIM and when you ran an interference check you would literally get millions of "interferences". Sometimes an interference between a piece of steel and a flue would show up as 500 separate interferences. Again, tracking, finding, and documenting these things was next to impossible.
I really think a lot of this stuff comes from owners wanting to do things quicker than the technology will allow.

 

[connectegr]

Toadjones
Many of our customers now use TEKLA or SDS for their shop drawings. We do not review connections within the model, so shop drawings must show sufficient information for review. I have received useless drawings, but simply mark the revise/resubmit. But, the traditional shop drawing information is becoming less important and less common on the drawings. But the internal detailing within the model shop be done be an real draftsman. These skilled people are disappearing. The fabricator wants the CNC data files for their equipment. A beam line with do all the drilling/punching and cutting from the magic uploaded file. Similar for the cutting tables. The shop drawings are primarily for the miscellaneous fitted material. Shear tabs, stiffener plates, gusset plates, etc.

I recently bled red marker over a set of shop drawings. The obvious fit up and detailing errors were to obvious to not mark. I charge to much per hour to be a detailing checker. But, when I asked if the drawings had been checked before submitted, the response was "we have never used a checker". They expect the software to catch the geometry issues. The fabricators are increasingly relying on this mentality and accepting a language barrier with the lowest bid detailers. But, in fairness they are not all this bad.

http://www.FerrellEngineering.com