Industrial buildings in the US 6

[Tom Livingston]

Hi All,

My company has just become involved in a US- based project which requires design of a number of low-rise industrial type buildings. My query is regarding the type of steel frame construction commonly used in the US for this type of building. In the UK we normally use straight hot-rolled beam and column sections for the portal frames, and add sloping tee sections to create haunches to the undersides of the beams at the connections to the columns to get deeper bolt groups etc. I understand that in the US it is quite common to have the whole of the members (both beams and columns) sloping rather than just using locally deepened haunches at the beam supports. Can anyone advise whether both types of construction are commonly used in the US, or is there a preference for one or the other? Thanks, Tom

 

[Kuhladula]

Hi Tom,

You're correct on a lot of the USA using fully-tapered columns & rafters for industrial buildings, mostly being PEMBs. They're done with a proprietary software that analyzes the tapered members at intervals as small as 12", so it's a lot easier to do the design. You can also follow AISC Design Guide 25, but that can get a little convoluted. There's nothing wrong with using straight hot-rolled members or built-up members, it just depends on what's best for the situation or the most economical. I will note that a lot of the industrial buildings around here use cable/rod tension bracing over portal frames whenever possible for economic reasons and ease of erection, but portal frames are not unheard of when larger openings are required. If you're going to get more into the industrial design aspect, I would recommend looking up the "Metal Building Systems Manual" 2012 or 2018 version by the MBMA. You should be able to find some copies online.

 

[SandwichEngine]

Tom,

I work for one of the companies that does fully tapered members. Like the previous poster stated, they're referred to as PEMBs, which stands for pre-engineered metal buildings. I use a software for some applications that has the ability to design using straight hot rolled sections with just the connections widened but I'm not aware of this ever being done in the US. The software I have is European based.

 

[le99]

These days, in the US, hunches or corbels are not popular for a simple reason - higher cost in fabrication and assembly. Also, most industrial buildings use simply supported roof trusses supported by columns for ease of design and construction. If moment capacity is necessary, it can be achieved by using the tapered beam-column connection for high pitch roof, or simply increasing the depth of the beam or column, or both, for flat to the low pitch roof.

 

[dik]

Unless you are using conventional rolled sections, it is likely less costly to use a PEMB (I don't like them, but they are 'marvels of engineering'. You can stipulate additional loading to improve their utility, but I consider them 'throw away buildings'.). Labour costs are higher and fabricating a building using tapered sections may not be economical, at least in Canada.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Do you feel any better?

-Dik

 

[Tom Livingston]

Thank you all for your responses. It seems that the sloped-beam type of construction is a highly engineered and specialized type of manufacture specific to industrial buildings which is not so much like ordinary hot rolled steel construction. In our case it is required that the building structures are designed as blast-resistant which normally requires the section classifications to be Type 1/plastic as plastic deformation is usually allowed for this. If the sloped beam construction is a type which is intended to be engineered virtually "down to the bone" to get the most efficient steel weights for normal type loading conditions, I am thinking that this type of structure might have relatively slender section elements and probably would not be suitable for a situation where the blast design scenario (requiring the ability of the members to deform plastically) is what governs the design of the sections. Can anyone say if the above is correct, or has anyone come across this type of structure in blast resistant buildings? Thanks again for your assistance.

 

[SandwichEngine]

You can design PEMB structures for blast, at least for the methods we use in the US. I personally designed one and saw several done at my previous company.

PEMB frames are extremely slender but that's not a limitation per the ASCE method of blast design.

In many ways, the tapered sections with changing flange thickness reduce the cost but they also spread loading to the entire structure, not just to fuse locations. This creates much more damping than with a conventional structure. The video below is for seismic loading but the impact of a blast would spread in the same way.

https://youtu.be/tq9_UAcKq3M

 

[Tom Livingston]

Thanks for that SandwichEngine. Can you tell me, is the single storey sloping member type of construction normally only something which is provided as part of a "system built" or proprietary type of building, or is it also a structural form which Consulting engineers working with architects would commonly adopt for a bespoke building design in the US?

 

[SandwichEngine]

Tom,

PEMB structures are always provided as "system built" and aren't something that can be designed by regular consulting engineers. Each company either uses proprietary software or the industry standard Metal Building Software, which is highly customizable and will be different for each company. The reason they can be so different is that the codes aren't written for this type of structure so each company has to do their best to interpret the codes in the way they think makes the most sense and different engineers will come to different conclusions on some matters. It also has to do with each company's manufacturing capabilities. At my previous company, the maximum depth of a member was 60 inches with a 12"x1" flange. At my current company, our maximum depth is 78 inches with a 20"x1.5" flange.

That being said, most of the buildings we do go through the normal consulting engineering process. Consulting engineers with experience working with PEMB structures can do a good job guessing at member depths and foundation reactions to get the foundation and building package to bid. Sometimes they adjust their foundations to the actual reactions, sometimes they do a good enough job that they don't have to, sometimes they probably go conservative on the foundation to avoid any chance of a redesign.

 

[CrabbyT]

I've seen it done many ways (tapered beams/columns, straight beams/columns, tapered columns w/ straight beams, etc.).

It's mostly a matter of means and methods for the PEMB manufacturer. From their perspective, they're going to use whatever combination they can so as to obtain the highest profit margin.

 

[SandwichEngine]

Previous posters are correct that you can do PEMB beams or columns straight but typically that's only done for aesthetics which wouldn't be a concern for an industrial building. The more important distinction is that they will always be 3 plate built-up, not hot rolled I beams.

I'd also agree with previous posters that some PEMB structures are built to be the absolutely cheapest thing out there and aren't intended to last more than about 25 years or so. However, that comes down to what's spec'd. If someone convinces you to to a metal panel roof that's through-fastened, expect the holes to wallow out over time due to thermal expansion and contraction of the roof and to develop leaks. Going to a standing seam or built up roof will significantly lengthen the expected lifespan.

In fact, many PEMB structures are used for schools, churches, government buildings, etc. You just might not recognize them as such.

To address some of the concern I've seen on here that implies that PEMB companies possibly skimp on designing things robustly to obtain a high profit margin, let me discuss. PEMB companies bid to furnish and erect buildings based on architectural drawings, structural drawings and specifications. To do that, we have estimation departments that have to do a quick and dirty design of the building to figure out how much that material will cost. So if the specification and the contract say the floor needs to hold L/360 deflection, an individual engineer doesn't have the ability to decide that their own personal limit is L/600 because the extra material required isn't in the budget.

Because we bid against other manufacturers for work and price is mostly tied to the weight of the buildings, we do try to design with as little material as possible. Throwing in additional extra material as a result of more conservative engineering judgement would result in not winning jobs. This bidding process and the use of as little steel as possible results in small profit margins and with savings passed on to the customer. It's just like with any other sub. If you were doing your own building, you wouldn't want to just get a bid from one HVAC company only to realize that they could have designed a much less expensive system.

So the quality of your PEMB building comes down to the choices of the architect/design team/owner. Specify a building with stringent deflection limits and roof and wall systems you feel good about, get a building that stands up just as well as a conventionally designed steel structure. Allow the drift to be H/60 with a through fastened roof, get a much less expensive building but one that won't be expected to last as long or perform as well.

 

[jimstructures]

SandwichEngine,

I must strongly disagree with your statement about Hot Rolled Steel shapes vs 3 plate steel sections. I can only speak from my 30 years of experience with Butler buildings ( 10 years with Butler Manufacturing Company and 20 years with a major Butler Builder). If you want Hot Rolled sections or HSS tubes or columns, they are available at an additional cost. Tapered member are more efficient. Butler and other MBMA manufacturers were specifically AISC certified for many years (now another certifying agency).

Jim

 

[OldDawgNewTricks]

To address some of the concern I've seen on here that implies that PEMB companies possibly skimp on designing things robustly to obtain a high profit margin, let me discuss. PEMB companies bid to furnish and erect buildings based on architectural drawings, structural drawings and specifications.

You seem to be an ethical engineer, but I believe that not everyone is on the same page as you. On many PEMB project where I have specified design criteria that exceed the bare minimum (e.g. tighter story drift limits in order to be compatible with architectural finishes), I have received construction submittals that do not meet the specifications and instead only provide the bare minimum criteria that is the manufacturer's "standard". I believe that they are not doing this out of an honest mistake in misunderstanding the requirements, but they are intentionally skimping on design for financial or competition reasons. When called on it, they may complain at first that it's not in their budget but then ultimately comply with the specifications that they bid on in the first place. Like I said, you are probably with a more ethical PEMB company but there seems to be a few bottom feeders who give the rest a bad reputation.

 

[CrabbyT]

PEMB companies bid to furnish and erect buildings based on architectural drawings, structural drawings and specifications... the quality of your PEMB building comes down to the choices of the architect/design team/owner.

Your post is refreshingly honest. The way I see it is, if you're buying a metal building then you better know what you want. And if you don't know what you want, you better make sure you ask for extra capacity.

I wish that PEMB manufacturer's would include a checklist of options for the customer. Things like:

[ul]
[li]Would you ever like the option to add a hoist beam at some point in the future?[/li]
[li]Are you ever going to cut a hole in the roof to remove a piece of equipment, and do you want a roof hatch?[/li]
[li]Do you intend to hang a ridiculous amount of piping from the roof?[/li]
[li]Are you ever going to remove bracing because it's in your way?[/li]
[/ul]

My only real gripe with PEMB manufacturer's is the anchor bolts and the base plates. I'd love to see the math behind em to see whether or not they use anything from the AISC or ACI to calculate the capacity.

 

[Kuhladula]

Crabby, most building manufacturers use MBS, Metal Building Software, which follows the AISC Design Guide 1 (first or second edition, depending on set parameters).

 

[SandwichEngine]

Thanks for all the positive replies.

To illuminate further, the design criteria isn't up to the PEMB engineer. It's negotiated during the sale. A common example is collateral loading. Collateral loading is removable dead load. Lights, HVAC, sprinklers, etc. It's broken out from regular dead load because much of the roof is controlled by uplift. For a worst case downward combination, I want to assume my roof member is loaded by those items so collateral load is part of the combination. But for a worst case uplift combination, I want to assume it's not present on this particular member so that combination will be wind uplift plus DL but COL dead load isn't present.

Sometimes a specifying engineer will call out an extraordinarily high dead or collateral load, presumably in the interest of providing a more robust structure. Consider a finished out school building that's 25,000 square feet. Accounting for the true collateral loading will usually take 7-10 psf. However let's say a specifying engineer calls out 20 psf. That would definitely produce a more robust structure but it's also forcing the end user to pay to support 250,000 lbs of hanging material that isn't present. PEMB sales teams will often provide quotes that deviate from the 20 psf number to something more realistic. Typically, they'll provide these numbers as an alternate so the end user/GC/design team understands the amount of cost associated with beefing up the structure beyond what is intended to be there.

From there, it's the choice of the end user/GC/design team to decide what they want to go with--at least at the companies I've worked for. From some of the posts above, it sounds like either these deviations aren't being communicated back to the specifying engineer at times or maybe there are some companies intentionally hiding their deviated numbers to win the job. If the latter is true, I'm thankful to not have worked for one of those companies.

To address the baseplate question, If you're seeing thin baseplates, that's likely because they are designing their baseplates to AISC design guide 1 first edition, instead of the second edition. The way it was related to me, the writer of design guide 1 realized at some point that it significantly under designed the baseplate thickness so they published the 2nd edition which basically doubles the required baseplate thickness. By then, the 1st edition was already programmed into the proprietary software of many of the companies and had been working for years so there wasn't urgency to change it. That's not how my company works but I'm aware of at least 2 where that's the case.

Another reason you could be thinking that the anchor bolts don't work is if grout is present and you check the anchor bolts with Hilti Profis. Both design guide 1 and the ACI code say to take a 20% reduction in the shear strength of your anchors when grout is present. Hilti Profis ignores the grout and designs the anchor bolts in bending which will always show them failing. That's one of many reasons why I prefer the Simpson Strong tie program. It checks the anchors per the codes.

To address PEMB manufacturers providing options for future loading, that's done all the time. Most commonly it's designing buildings for future expansions or future cranes but I've seen all kinds of future loads. A recent building has future overhead doors. However, we have no way to guess what future loads someone might want to impose. That's got to get figured out by the design team and clearly communicated to PEMB companies bidding the job.

Jimstructures,

Not arguing, just clarifying. Yes you can get rigid frames with hot rolled sections if they're small enough and tube columns are common for covered walkway, lean-to, interior and other applications. My real point is to address the original poster who was asking if it's common to do rigid frames with regular hot rolled I shapes that have beefed up knee and ridge connection like below.

You may be aware of that type of building being done in the U.S. but I haven't seen it. Then again, what do I know? I just found out this type of building is a thing.

https://www.legacybuildingsolutions.com/

You're also correct about the accreditation process. It used to be AISC but they had a lot of strange requirements that the industry didn't like. For example, you had to have an internal company design manual that just regurgitated the code. The new accreditation agency is IAS which is a subsidiary of ICC.

I would definitely suggest making MBMA membership and IAS accreditation a non-negotiable part of your specification. I can't say that will save you from all of the issues that have been mentioned but it would definitely rule out fly-by-night companies and anyone else who would be much more likely to have questionable business/fabrication/other practices that would preclude you from IAS accreditation.

 

[Enable]

Nothing productive to add. Just wanted to stop by and thank SandwhichEngine and others for shedding some light on the PEMB side of things.

Also, wanted to say those pictures of PEMBs are damn sexy especially for PEMBs! Hats off to you sir

 

[dhengr]

SandwichEngine:
I agree with Enable, and think you are a darn good spokesman for reputable side of your industry. Keep up the good and honest engineering work and thinking that you are doing. Your comments and explanations are right on the money, actually most of them are things we should all know and be thinking about in our design work. But, we kinda come from two different worlds in that you specifically sell economy of steel structure down to the col. base pls., and that’s it; you are set up to design/engineer and purchase/fabricate that way, and you include the no frills engineering in the cost of the bldg. you supply, so it kinda goes away (hides), as does everything below the col. base pl. Consulting engineers in private practice must cater to an Arch., client, contractors, their RFI’s, AHJ, etc., and their every whim, change, code interpretation; and on top of all that their measly ‘percentage of bldg. cost’ fee is looked at as an up front cost the owner would just as soon avoid/minimize if he could, he’s already over paying the Arch. and other consultants. Then, most designs use std. shapes and sizes and must cater to all fab’ers. who might bid, but are not set up the way the PEMB shops are.

 

[SandwichEngine]

Again, thanks for the kind words.

I see a lot of stuff in structural engineering discussions about PEMB structures that sounds to me like-"I bought a Honda Civic but now it won't tow my boat." They can do whatever you want within the limitations of their cost efficiency and code restrictions.

Here's a few more that you might like.

 

[chris3eb]

Definitely good stuff, Sandwich. I think a lot of consultants' views of PEMBs are influenced by two big things:
[ol 1]
[li]The average PEMB is designed much less robustly than the average traditional building. This is not to say that a PEMB can't be designed as robustly as a traditional building. It's also not to say that it was the wrong decision for the owner to have a PEMB built that was less robust than a traditional building - it might have made perfect economic sense. But when a consulting engineer walks up to a PEMB, they know that on average, it will be less robust than a traditional building.[/li]
[li]When a consulting engineer designs, they also have an internal set of "specifications" they are designing to, but they aren't beholden to them. There are plenty of times when they use engineering judgement or "good practice" to say that they should go above and beyond the "specifications" for a specific member or connection. Possible reasons being that the member is critical, or the smallest size that calcs is too small, or say two bolts calc out but four would be nice for redundancy. The PEMB system strips away a lot of this engineering judgement and "good practice" because it's so difficult to specify when to go beyond the specifications[/li]
[/ol]