Steel fabrication costs 3

[TLHS]

I guess this is a question for the folks that have worked on the fabrication side of things. I'm hoping to start a discussion, more than answer a specific problem.

On large projects, fabricators will sit down and sharpen their pencils a bit. Bills of material will likely be created, and details examined to some degree.

On small scale projects, it seems to me like it's generally a straight up per weight cost that maybe got eyeballed with a modifier for overall complexity.

So, on small scale projects am I costing my clients money by doing heavier things that may simplify fabrication but don't necessarily jump off the page. I'm thinking about things like using a slightly heavier column instead of a couple of stiffeners, or sizing up a beam on an access platform to eliminate LTB bracing with lots of connection points. More subtly, there are things I can do that will simplify connections that aren't even on the bid drawings, because they'll be designed by the fabricator's engineer later on. The fabricator's engineer is often not even involved until the fabrication drawings are partially complete, so they're likely not having any input on a bid. Because of the way things seem to be estimated, I'm a little worried that despite having good intentions, on some of these projects I could be costing my client extra money while saving the fabricator costs.

I know it's good practice to do things like this, but does good practice necessarily follow through to real life cost savings to the party I'm actually doing the work for?

 

[TomDOT]

While the subject has always been larger projects, the key items that fabricators have emphasize to me to save cost and time are:

1) Use standard thicknesses.

2) Use consistent thicknesses, widths, etc - both within the same article, and between articles. The added welds cost more than the savings in weight.

Additionally, simplicity (such as your example of going slightly heavier instead of adding stiffeners) reduces long-term maintenance to some extent, and may improve constructability in the field, as there will be more "elbow room" for assembly.

 

[KootK]

I've been meaning to post this same question myself. I would have reponded back in August were I not out of the country at the time. Anyhow, I'll respond now in order to add my two cents and to give the thread a new year's bump. Here are my thoughts, organized by the parties with vested interests:

1) The environment. I'm not sure what is best for the environment: reduced steel weight or reduced fabrication effort. I suspect the latter.

2) The owner. Based on the feedback that I've gotten from the handful of fabricators that I've queried, pricing is largely done by weight and, where fabrication savings are generated through intelligent design, they are generally not passed along to the owner. At best, I've been told that certain design firms may befefit from having a general reputation for efficient design which may lead to improved pricing in a non-project specific kind of way.

3) The steel supplier. Every year or so a new article emerges touting the 475 latest tips for efficient steel design. It's a bit like Cosmo's never ending stream of tips on how to better please your man. Of course, the articles are published by AISC whose membership is steel suppliers. The tips are clever and I feel like a better engineer for knowing about them. However, I suspect that it is primarily the steel fabricator who benefits the most. If they bid a project assuming a 15% premium for stiffeners, and there are none, it's a windfall for them.

4) Structural engineers of record. We lose out on this arrangment more often than not. Making clever, efficient design desicions often involves a greater investment of designer time. This is particularly the case now that much of our design is automated. Often, the intelligent steel design choices are ones that require the attention of an actual human being which slows things down. The extra effort might be justified, from a business perspective, if we could legitimately claim that the extra effor was generating savings for our clients. As I've described above, however, it's hard to make a convincing case for that. This is why, no matter how many times AISC tells us not to, many of us still specify algorithmic shear reactions for our connections rather than actual design loads.

I realize that I sound pretty cynical here. That's because I am. Hopefully some of our members who are a little closer to the steel industry will surface and refute some of my pessimism.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[JedClampett]

As background, I'm in an industry where we have multi discipline process facilities and the structures are mostly boxes to hold them.
Every time I've tried to out-think fabricators and contractors, it's bitten me in the behind. I've come to the brutal realization that:
a) The costs I can control aren't that significant.
b) Most of the times I do something for ease of construction or lower costs, it's not appreciated or worse, ends up with a complaint (why did you do it that way?) or problem.
I try not to throw away material. But going from a W8 x 21 to a W8 x 18 by adding stiffeners just isn't worth it. They might get forgotten, they might be put in wrong, the welder might undercut the welds, any number of screw ups might happen.
For instance, I never use W8 x 10's as load carrying members. The flanges are so narrow that it's hard to weld deck to them. And the webs are thin, so if they're coped, there's nothing left. Up size to a W8 x 18, maybe change the spacing, but don't worry about the few extra dollars and think a little long term.

 

[darthsoilsguy2]

i haven't done much steel design but have inspected steel for a decade. speaking with the erector foreman and steelfabs, my impression has been tonnage and crane days being the variables looked at by estimators and the rest being noise. They build so much from so many designs and have a few baseline $/ton depending on the project size. seems like this kind of question is hard to get a straight answer on... steelfab/erectors definitely tell egrs that constructibility is factored in the cost at those AISC sponsored pdh sessions...

 

[SlideRuleEra]

On industrial projects where modifications and additions are common, having slightly oversized structural members is an advantage for future changes. It also has value as a small corrosion allowance. To encourage this design approach on electric generating stations, we (Owner) specified a minimum thickness for any steel member. If memory serves, it was 3/8".

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[KootK]

A few years back, I worked on a Gehry-esque steel building project. During the design phase we got to have one of North America's premier steel fabricators on board in a design assist role. Great, I thought, this will be like going to "steel university"! Real time pricing on design changes is tough to beat. Here's how it went:

1) The fabricator was helpful but there were few big revelations that I hadn't heard of before elsewhere.

2) Following the rules for efficient connection design and fabrication design barely seemed to move the needle at all.

3) Contingency allotments for drawing incompleteness seemed to outweigh everything else by a huge margin right up until IFC.

4) Many of the legitimate money saving ideas centred on erection and worker safety during erection which don't seem to get much press and tend to vary from supplier to supplier and project to project.

5) I turned the floor and roof deck something like 50 times which is a big no-no supposedly. The cost impact was 15% on the install portion of the deck supply and install contract. Basically, it didn't affect the steel contract at all.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[manstrom]

Interesting info. I am getting more into steel design and this interests me as well.

In my designs, I try to minimize the variety of steel members and keep field welding to a minimum. From a pure pricing point of view, I am sure that least tonnage is the cheapest answer (but this may add complexity). From a construction point of view, simplicity is best (but this may add cost).

I am not sure there is a magic bullet. A lot of times, my mentality is: easy to engineer means easy to build.

When I am working on a problem, I never think about beauty but when I have finished, if the solution is not beautiful, I know it is wrong.

-R. Buckminster Fuller

 

[JedClampett]

manstrom, your cost saving suggestion of minimizing field welding is a good example of no good deed going unpunished. I've tried that. So the steel gets to the site and holes are not aligning (heck, I don't know how they ever align them), so the RFI requests to field weld the connections. So all my fussy little bolted steel details, checks of bearing, torque and pretensioning requirements are worthless.
I've come to the realization that the I'm worried a lot more about the contractor's ease of construction than they're worried about putting it in right.

 

[TomDOT]

Interesting. Bridges in Texas almost never have field welding. All bolted. They are supposed to do fitup in the fab before the girders are shipped.

The Margaret Hunt Hill bridge was field welded, but that was architect-driven.

 

[drawoh]

manstrom, your cost saving suggestion of minimizing field welding is a good example of no good deed going unpunished. I've tried that. So the steel gets to the site and holes are not aligning...

This sounds either like a dimensioning and tolerancing issue or a quality control issue. If your manufacturing is not accurate enough to centre regular sized holes on each other, you are going to have to enlarge the holes. Is this structurally acceptable? Are slots acceptable?

--
JHG

 

[darthsoilsguy2]

jed's comment sounds like reality. at least 95% of the welds i have called out as failing are missing, and half because the fit-up didn't work. at least 80% of the bolted connections i've called out as failing with mostly had problems from fit-up. the problem is usually not a fabrication control, but the plumbing tolerances built in to the erection process for buildings and the fact that anchor bolts and embed plates are a dice game. Most anchor bolt problems are 'worked out' in the plumbing. bridges girders definitely have better contractor QC in both the shop and field than building structural.... Now, if you want to feel better about your structural steel QC, hang out with the cowboys they sub stairs and misc steel to. in my experience many egrs in the region i've worked have gone to oversized AB holes with welded on plate washer details. i haven't crossed any that went slotted or beam oversize holes except for targeted locations. I'm speaking strictly structural... i've seen utility framing mix it up more.

 

[JedClampett]

As far as the comment about it being a tolerance issue, sure. But what do I do about it? We call out standard AISC tolerances. If the steel fabricator can't line the holes up, I don't need someone to blame, I need the pieces attached together. And I can't just say, "they're out of tolerance..." and walk away.
If you're looking for me to say the fabricator is incompetent, I'll heartily agree. But we can't get the AISC Certified shops interested in our squirrely little projects and a bid project encourages the General Contractor to use the sloppiest, cheapest, dumbest steel fabricator around. Money talks. And the GC doesn't care. They get a lump sum price from the steel guy, and if it doesn't fit, the steel guy has to fix it on his nickel.

 

[drawoh]

JedClampett,

I am not structural. I design mostly precision stuff made from machined parts. When I design weldments, and/or stuff out of sheet metal, I open up my holes. If the process is not as accurate, you have to do a tolerance stack-up, and solve the problems. In my world, weldments are fabricated to around [±]1/16".

If you have a bolt located at exact nominal position, your clearance hole must allow a clearance of not less than the positional tolerance your fabricator can achieve. If your fabricator cannot do better than Ø1/4", your clearance cannot be less than 1/4", etc. If that clearance is not acceptable, you are going to have to insist on more accurate fabricators, or you are going to have to make the holes at assembly. The other options are slots, some sort of floating bracket, or just giving up control over the design.



--
JHG

 

[TLHS]

I got incredibly confused when I got emails about this thread out of nowhere. I'm glad to get people's opinions.

On a structural project, there are established tolerances for fitup, but there are all sorts of times where it just doesn't work out. It's not a machining tolerance issue. It's generally an issue of where the shop fabricated stuff hits the real world.

Let's say you've got steel that was fabricated in a temperature controlled building and fabricated to tolerance. You have the anchor bolt locations surveyed and cast, and then go to erect your steel. If your survey location is a little off, your anchor bolts are a little off and you're installing in hot temperature you could have some fitup adventures despite the fact that theoretically everything was within tolerances. Throw in an actual mistake and you really end up in trouble. In one off construction work you're always going to have mistakes of one sort or another, even if you have a really good QA procedure.

An actual analysis of tolerance sensitivity on a global scale is prohibitively difficult for most construction projects. You've got long members that get shimmed, leaned back and forth, installed in different orders, and smashed into place when necessary. Whether something fits could depend on something as stupid as whether an anchor bolt is off plumb a bit and prevents to column from being rotated to meet a connection at the top before all the bolts get tightened. There's just too much going on.

To let us actually get things built, we have industry standard tolerances that will allow reasonable fitup on conventional project a reasonable amount of the time.


Fitup issues are a fact of life in construction. If it's happening a lot with one contractor, it's likely bad work practices and the solution is to stop using them if possible.

I don't think field welding is as big a deal as some make it out to be. Sure, if you can *completely* avoid it, then it saves a lot of cost. The nominal cost for one more weld, if you're using them anyway, isn't a big deal if it's reasonable and fixes an otherwise awkward bolted connection. Obviously you also don't want to be doing welds in spots where it would be difficult to get access to with welding gear. You also need to have appropriately detailed plates to allow dimensional issues to be fixed. Welding members straight to other members is even worse for fitup than a bolted connection that doesn't line up.

 

[moltenmetal]

Manufacturing tolerances and fabrication variability of the sort you've mentioned (temperature etc.) both "stack up". Lots of people THINK you can build things in a shop to precise dimensions and have them fit- precisely- and sometimes indeed you can. But in structural, more often than not, during erection there's a guy with a torch blowing holes larger to make them fit that the designer never sees. Or, in my world, there's a fitter sucking pipe flange faces together with a come-along to get the stud bolts in, or putting a "diamond heat" in with the torch to get something to line up- sort of.

Angular tolerances are the most often forgotten or ignored. They're difficult to measure accurately in a shop fab environment, next to impossible to measure in the field, and hence are very hard to control. The stack-up of angular tolerances is the primary driver for the need for field welds in piping. That we, as a design/build facility, know this and acknowledge it in our design, inserting sufficient degrees of freedom by means of properly selected field welds, is the reason that our piping fits precisely and our flanges don't leak, and why our piping has a hope of performing properly when put through thermal cycles etc.

As to the OP's original question: small structural projects are bid by steel fabricators pretty much purely as a multiple on weight. Yes there's a complexity "weighting factor", but it's usually overwhelmed by their "shop capacity factor"- if the shop is empty they bid a lower cost per pound.

Steel itself is cheap. Many people solve problems, present or perceived future ones, by throwing steel at them. We build our own frame structures when we're not busy, and subcontract them when we are, so we see both ends of this. While solving our own problems by throwing steel at them is often very cost-effective, when we go out for steel fabrication I have to remind folks here that having the extra HSS wall thickness to allow us to drill and tap steel members at will for minor attachments, to add stiffness without adding bracing, or using consistent member thicknesses purely to avoid errors during fabrication, are luxuries that we cannot afford when every pound of steel in the design will be multiplied by the fabricator's per-pound factor.