Contract Issues - Structural Steel Connection Design in the Midwest 3

[Loui1]

I am located in the midwest (New Madrid Fault Zone).....An area that is relatively new to serious seismic design. On several projects we have fallen into the scope of the AISC seismic provisions. Because our industry here is set up in a way that is not conducive for the EOR to design lateral force resisting system connections, the connection design responsibility is forwarded on to the fabricator. Bottom line, the EOR is not given enough time and/or fee to do connection design. It it becoming more and more common that fabricators are coming back during the project demanding more money (hundreds of thousands of dollars) because the connections geomectrically dont work, they didnt budget large connections, or they just make up excuses to cover their losses.

We have tried several ways to make it clear to the fabricators that the connections will be larger than normal, and that the seismic provisions are to be followed. But it still ends up that the fabricator claims that he/she cannot properly bid the connections because it takes the technical experience of an engineer to do so.

Has anyone else run into this problem and how have you remedied it? Anything to put on the contract documents? Any good published information out there? I assume this problem extends over to the east coast.

 

[nutte]

I perform this exact connection design for fabricators. The ones who have done these types of jobs before know what to expect and bid it accordingly. The ones that don't research the matter, discuss it with us, and determine how much to include in their bid for connection design (usually we submit to them a bid for connection design, with estimated gusset plate thicknesses, brace reinforcement plates, etc).

A good set of drawings will indicate the seismic requirements. They'll also show details with large gusset plates, bend lines, brace reinforcement, etc. An experienced fabricator and connection engineer should be able to bid these projects accurately.

As far as communicating this to the fabricator, the only thing I can think of is to make specific mention of the seismic requirements and show some details that truly represent the larger connections that will be required.

 

[archeng59]

similar issues in my area (oklahoma). however, I've increased fees as needed to cover the connection designs and seismic analsyis. explained to clients that code changes for seismic design increased my scope of work and it takes more effort to do the engineering. told several clients that I can do the design now or they can pay more to have the fabricator pay another engineer to do work I have to review. some grumbling at first until they realized they would be paying two engineers to essentially do the work of one. it might take time, but you should be able to convince clients to give you the time and fee to do connection designs.

 

[Loui1]

Unfortunately we are the little e in an A/e firm, therefore its much more difficult to convey these structural issues to the clients. The architects (ones who usually deal with clients) tend to agree to anything and everything the clients demand...including outrageous deadlines and barebone fees. Telling them that it will take the engineers more time and money will be nearly impossible. I think the only way to change the situation is for me to quit and change companies

It feels like the more typical connection details we put on the drawing, the more problems we get...the fabricators tend to look at them as set in stone. When the fabricator's engineer turns around and tells them the connection cant work exactly as the detail shows it, they proceed with a change order.

 

[nutte]

You have to make the point that the detail is schematic only, just to show intent, and that connection design shall be performed by the fabricator and sealed by a licensed engineer, etc......

The beefed-up seismic connections would be quite a shock to the contractor/fabricator/detailer if they weren't shown on the drawings, so to keep from surprising anyone (and incurring extra costs), I think you have to show some sort of typical detail.

Doing connection design, we see drawings from engineers all over the country. This is pretty standard practice, and an experienced fabricator shouldn't be surprised by this.

 

[Ron]

Loui1....one of the most significant problems we face today as engineers is the erosion of respect for what we do. Part of this comes from others not understanding what we do and its importance in the whole scheme of things, but most of it comes from within. We are our own worst enemy.

I would suggest that you take a stand with your firm and get what's necessary to do a proper design. You've already laid out the case (change orders/claims). If they are unwilling to listen, then yes, it's time for you to move on. You're in a situation that could easily attempt to compromise your integrity and that of the engineering profession.

 

[Loui1]

I dont think that we are our own worst enemy. I feel the problem lies in the fact we have no unified entity representing the profession that takes a strong enough stand on certain issues. It feels like we're a broken flock of sheep meandering around the contryside. The profession is asking us to act as individuals to change the industry, which is just ludicrus when you look at our salaries. I know I dont have the monetary luxury of risking my job. What ever happened to "strength in numbers"? Each state has its own structural engineers group that kindof tries to organize us locally but doesnt always have the necessary clout, and ASCE seems to be too broad to represent us well....the reason why I toss the weekly ASCE "come join us for $200" mailing.

That's beside the point. I was just looking for ideas of what to put on the contract documents to avoid these change orders. It appears we're doing all that we can do at the moment in the current situation.

 

[dik]

As far as documents go, you would have to put on all forces that the fabricator is to design to and combinations, thereof. In addition, you should clearly note that the fabricator is responsible for the design and procurement of a registered engineer to undertake the connection design and that no additional fees will be entertained in the design of same. You should also show typical details of the various types of connections required for the various conditions.

As far as being our own worst enemies, I think the thought comes from the idea that I can design something for a cheaper price than Ron can and I'm prepared to do this (not serious, Ron). This, in my opinion, has caused the decline in the quality of trades available, too; contractors cannot afford to hire quality people. In our environs, a real estate agent can make more money selling a building than the engineer that designed the building.

Dik

 

[RARWOOD]

I think it is irresponsible for any EOR to pass on the lateral design of a building to the fabricator. The lateral bracing system of a building involves multipe elements, metal deck, elevator shafts, braced frames, rigid frames and foundations. Also the building geometry has a huge effect on the bracing systems.

When a job gets to a fabricator the above items are already established. This may result in a geometry which may makes it difficult or almost impossible to design the connections. When a fabricator encounters this problem he doesn't have the option of adding another frame, increasing the column or beam size, ect.

Unless the EOR has atleast done some preliminary designs on the critical connections to verify that they workable and reasonable, he may be creating a problem that can't be solved by the fabricator. A problem which could be avoided if anticipated by the engineer, who can adjust elements before the job is bid.

A simple example is an engineer who decides to pin the base of all columns. I have dealt with wood jobs where the engineer designed all column bases as pins and assumed that the wood beams and columns where acting as rigid frames to resist the lateral load. On one job I indicated to the engineer that I could develop fixity at the top of the columns but that the architect and owner would not be happy with the amount of steel required to solve the problem.

 

[nutte]

RARSWC, we're not passing on the "lateral design of a building" to the fabricator. The lateral force resisiting system should always be called out on the plans. We're talking about the design of connections of these elements. If you've ever gone through the design of a seismic brace connection, you know it's quite a lengthy process, and the cost to do that is rarely included in the engineer of record's fee.

 

[JAE]

The lateral design is NOT JUST THE MEMBERS. The lateral design is also the connections of those members.

Indeed, I recall Edward Phrang once spoke at a seminar on the Hyatt Regency disaster in Kansas City in the 70's and his concluding remarks were something like: [green]"We engineeers need to get past this notion that we are designers of frame members that just happen to be connected. We should rather think of our structures as a multitude of connections linked together by members."[/green]

Passing on connection design (the most critical part of any structure) is indeed passing the lateral design of the building off to someone else. Its really what "caused" the KC disaster.

 

[nutte]

You make good points, but this is the industry practice. And unfortunately, most of the engineers that I've come across don't have the slightest clue about connection design. Much more technical knowledge and expertise goes into the design of brace and truss connections than goes into the member design of a braced frame or beam framing. I would say that a good percentage of engineers that I've worked with aren't qualified to do the type of connection design we're talking about.

I believe the more appropriate "cause" of the Hyatt collapse (in 1981, by the way) was a combination of poor design in the office and poor communication during construction as the contractor tried to provide a "better" (at least more erectable) connection.

And the Hyatt collapse was in 1981.

 

[nutte]

Sorry, I forgot to delete that last sentence. I'm not trying to harp on the date!

 

[Loui1]

JAE, Yes you have good points of what should be done, but I dont believe you understand the difference between industry standard and "the right thing to".

I understand and believe that connection design should be completed in the bid documents. I'm not even arguing this point...AND THAT IS NOT THE PURPOSE OF THIS THREAD. In the midwest for typical five story or less structures, the industry has pressured the engineer into passing connection design off to post bid work. I cannot change the industry....Only large groups of engineers can. Therefore I'm F*'d into doing it this way. If you and all the respected senior structural engineers cant gather enough clout to change codes and remedy this solution then please help me deal with what I've got.

What Edward Phrang said makes no difference because in the 25 years since his quote, nothing in my area has changed because of it. Edward Phrang should have worked with ASCE or some other organization to change the situation instead of placing the responsibility to fight the battle on the individual engineer. Also, I believe the KS failure was due to a fabricator changing a detail for constructability, not the engineer failing to design the connection.

You can call connection design part of the lateral design. But in no way can you say the lateral design of the building has been completely passed off to the fabricators engineer.

 

[archeng59]

Loui1, are you a member of a state chapter of NCSEA? If not, I recommend joining and taking up your discussions within that national engineering organization.

This is my opinion for what it's worth: If you think the connection design is your responsibility, then design the connections. If you are sealing and signing drawings, the entire structural design is your responsibility even if the contractor/detailer provides connection design. you are not off of the hook for liability and protecting public safety just because the owner has a tight schedule and doesn't pay you enough to design connections.

 

[JAE]

nutte and Loui1 - good conversation here. Here's an excerpt from the AISC Code of Standard Practice (underlines and bold parts by me)

[green]When the Owner’s Designated Representative for Design shows the complete design of the Connections in the structural Design Drawings, the following information is included:
(a) All weld sizes and lengths;
(b) All bolt sizes, locations, quantities and grades;
(c) All plate and angle sizes, thicknesses and dimensions; and,
(d) All work point locations and related information.

The intent of this approach is that complete information necessary for Connection detailing, fabrication and erection is shown in the structural Design Drawings. The Steel Detailer will then be able to transfer this information to the Shop and Erection Drawings, applying it to the individual pieces being detailed.

When the Owner’s Designated Representative for Design allows the Fabricator to select or complete the Connections, this is commonly done by referring to tables in the Contract Documents or in the AISC Manual of Steel Construction, or by schematically showing the types of Connections required in the structural Design Drawings. The Steel Detailer will then configure the Connections based upon the design loads and other information given in the structural Design Drawings. If the desired Connection is not covered in those tables, a detail of the “special” Connection should be contained in the structural Design Drawings. This detail should provide such information as weld sizes, plate thicknesses and quantities of bolts. However, there may be some geometry and dimensional information that the Steel Detailer must develop. The intent of this method is that the Steel Detailer will select the Connection materials and configuration from the referenced tables or complete the specific Connection configuration (i.e. dimensions, edge distances and bolt spacing) based upon the Connection details that are shown in the structural Design Drawings. This method will require the skill of an experienced Steel Detailer, who is familiar with the AISC requirements for Connection configurations, capable and experienced in the use of the Connection tables in the AISC Manual of Steel Construction and capable of calculating dimensions and adapting a typical Connection detail to similar situations. Notations of loadings in the structural Design Drawings are only to facilitate selection of the Connections from the referenced tables. It is not the intent of this method that the Steel Detailer practice engineering.[/green]


Yes, there are two parts to this topic - what is the "standard practice" which I've posted above AND what is the "prevalent practice" and how can we effect the politics of it all to allow better control over our designs. I would recommend and refer you to archeng59's post above - we aren't that helpless.

I also work in the Midwest US and I guess I haven't found that much resistance to setting a fee appropriate to do the connection designs. I have to admit we don't do a lot of work in southern Illinois, but we have done work in San Francisco and LA and detailed the connections.

nutte - the Hyatt Regency collapse, according to Jack Gilliam, the EOR, and the multitude of reports that came out of it, was that Gilliam didn't design the connections, he only showed a concept with a design load and expected the fabricator to design it.

The fabricator started the shops, and a bit of the design, but half-way through the process, farmed the work to another fabricator who ASSUMED that the design had been completed by the first fabricator and simply passed the un-checked un-designed connection through their process.

Gilliam's firm quickly reviewed the shop drawings and assumed the connection was also designed and didn't check it closely.

This original design (the continuous rod) wasn't ever built but was later determined to have been very poor and lacked sufficient capacity had it been built. The Contractor's alternative (splitting the rod) only made a bad detail worse.

The whole point here was that farming off a connection design to a fabricator carries additional risks for the EOR and I agree with Loui1 that there's frustration out there in our engineering industry in that we are being squeezed to design less due to tigher fees. But is it a problem or a challenge for us?

 

[Loui1]

I am very familiar with the membership benefits of my state NCSEA. Several coworkers are members and I have sat in on their meetings if one cannot attend. I just attended the annual convention and posed this same question. The reply I got was that, yes it was an industry problem, and "they're looking into it." The organization is very fragmented and I cant figure out what they really do other than collect membership dues.

Yes, engineers in my state are pretty much on their own. If I go ahead and start charging adequete fees and time to design all the lateral system connections I will be out of a job because there are plenty of engineers in my state who will gladly outbid me and farm out the connections and risk the lawsuit later. I have no options but to go with the industry or move out of this state.

The jobs done in california, yes, of course you designed the connections....because THEY'RE REQUIRED BY THE STATE CODE and the industry is already set up correctly there!!! I dont think CA even allows the EOR to farm out the connections. I'm talking about the midwest, not the west.

I totally understand your reasoning, quotes from the Code of Practice, and ethical ideas, but you still have not given me any input to my initial question other than trying to unknowingly convince me to place my balls in the guillotine I will try to change my local NCSEA when the time comes and that's about all I can do. This is a industry problem, not a personal challenge. Maybe you have to come visit and see for yourself how messed up things are here.

 

[nutte]

Not to get too far off topic:

The Hyatt connection originally called for one continuous rod, threaded along the entire length, supporting multiple levels. The contractor realized the difficulty in threading nuts the entire length of this threaded rod (bad original design) and proposed a new connection with two separate rods (really bad design). The revised design fell through the cracks in communication, got approved by the EOR, and was built, resulting in the disaster. More here:

http://en.wikipedia.org/wiki/Kansas_City_Hyatt_collapse

The original design was structurally adequate but not erectable. The connection design was not farmed out to the fabricator. In an effort to make the connection erectable, the contractor proposed another detail, which was wholly inadequate, that the EOR carelessly approved. The problem here is the poor communication, not the delegation of connection design.

Back on topic, I think there are benefits to the way we do things, that is letting the fabricator perform connection design. For simple framing especially, this allows the fabricator to choose the best conneciton for them (bolted clip angles, welded clip angles, shear tabs, end plates, etc). Similarly with bracing connections, the fabricator can choose one of several methods to connect the pieces to expedite the entire construction process.

I also feel (as I mentioned above) that a lot, maybe most, of the structural engineers out there who don't regularly performc connection design know very little about the process. If a small-time engineer doing convenience stores checks a shear plate connection, does he check bolt shear and plate shear and stop? Does he even know about the other limit states (plate bending, both gross and net section, bolt bearing, eccentric bolt shear, eccentric welds, bending/buckling on the beam cope, and so on)? Maybe he has a good grasp of these processes. What about the Uniform Force Method? Designing truss connections? I think we're better off leaving connection design to engineers that specialize in it.

The challenge is to provide good information on the structural drawings so the fabricator can design the connections economically and efficiently. Get rid of the 50% allowable load capacity crap. Give reactions on beams, or at least a table that's not so conservative it's useless. We go round and round with EOR's that insist we develop 25 kips shear in the W8x10 filler beam that spans 3', just because their standard notes, which haven't been updated since 1980, say to do it that way. Show the types of connections you're looking for, with the note that they're representative of the type of connection you want. Make the fabricator provide connection design calculations that you can review. You'll learn real quick if the connection engineer knows what he's doing or not.

And to get back to the first post, show a sketch of one of these super-strong seismic connections so everybody's bidding the same thing and you don't get hit with extras at the end. That way we, as engineers, can exert some influence and weed out inferior fabricators in the beginning. Otherwise, the good fabricators are priced out of the job, and you get an inferior product that winds up costing the same or more than the good fabricator with considerable more headache.

 

[JAE]

Loui1 - I guess I honestly don't have an answer for your "original question" other than to agree with earlier posts that you simply must put enough information on your plans to convey the magnitude of the required connections (assuming you are having your fabricators design them). A giant note on the plans in a huge, bold box. But that probably doesn't deal with the construction community you are working with.

In the 25 years + I've been a structural engineer, the common consensus is that fabricators on large projects generally do standard beam connections but for anything unique, the EOR should design it. The extended quote from the Code of Standard Practice that I posted above implies that this is the desired case.

nutte:

Your quote

The original design was structurally adequate but not erectable

about the Hyatt is not correct. The original design (single long rod) was checked and tested by the NBS and found wanting. The split rod concept was simply a worse condition from that.

You also say

The connection design was not farmed out to the fabricator

This is not so. Jack Gilliam stated that his original detail was not designed (this was his primary defence) but that it was a schematic detail with a given tension load that he fully expected the fabricator to design.

Finally, you say

The problem here is the poor communication, not the delegation of connection design.

The entire ramification of this disaster was all about the issue of delegation. The concluding comments that Gilliam makes when speaking around the country on this was "you can delegate engineering tasks to others, but as EOR you cannot delegate responsibility".

 

[Redfish205]

Nutte:

Here, Here!!! Sounds like you run into the same stuff we deal with. I could not have said it better myself. I too am a connection design engineer, thats all we do. I believe the best thing an EOR can do to help with connection design, is do as Nutte said and avoid the 50% UDL stuff, furnish the real reactions, transfer forces, etc. on the drawings, like AISC recommends. And, pay close attention to the R of the building your designing. This too can have an adverse effect on the size of the connections. And lastly, if you are going to specify that seismic provisions apply, at least be familiar with what that means to the connection design. I just had a conference call with a detailer, fabricator, EOR and Myself where the EOR had done a great job of furnishing beam reactions and actually gave moment values for the moment connections, but he specified that seismic conditions apply. We spent 45 minutes walking him through the seismic provisions of AISC and afterwards he still insisted we design the moment connections for the small moment loads given yet provide full column stiffeners as if designing for the full moment capacity of the beam.