Deficient existing building

[Once20036]

I was recently hired by a client to retrofit an existing building to support their equipment. They're using roughly 1/2 the building and the other half will be vacant for the time being. It's an existing structural built in the early 80s, joist & joist girder construction, braced frames, cmu exterior, etc. There are no existing drawings.

The issue is that most of the perimeter girders have significant existing overstresses under code required snow loading.
They're 36' long with a wind column in the middle and support 40' long joists. The "girder" is a W12x14.

Checking it as a simply span, i get 500% overstresses.
The slotted connections on the wind column are all bottomed out, so there's gravity load being transferred into the columns which may or may not have existing foundations. Analyzing it over this center column, its roughly 30% overstressed if you count inflection points as braces, or 200% if you ignore inflection points.

There are also uplift issues for the joists and joist girders along the entire perimeter of the building.

I`m planning on writing a letter to my client and the owner pointing out the issues that I've found, but I`m conflicted as to what that letter should recommend. Obviously, this has the potential to be an issue, however, the building has stood for the past 30 years without failure including last year's record snows. It's tough to say that they need to do anything, but on paper this could end poorly.
I`m tempted to indicate what I've found and strongly recommend they hire someone to evaluate this, as it's outside of my scope.

Has anybody run into something like this before? I`d appreciate any advice, insights, or shared experience.

 

[JAE]

In the early 80's I think many engineers considered the inflection points as brace points for negative moment checks. So the 30% overstress is reasonable to see.
With the full length are you calculating out the actual Cb ratio for the span? Going from 30% to 200% overstress doesn't sound right.

In terms of what to say or recommend in your report, you need to state the facts as you find them and offer possible remedial actions.
The fact that the building has stood for 30 years doesn't make it safe. You can't sign off on 30 years of "not falling down" as that is quite unprofessional.

For the perimeter W12, a 36 ft. span is a bit much, even if it is parallel to the interior roof joists.
You could research what sort of footing there might be under the wind columns, and then possibly add footing to the existing gradewall footing, stiffen the columns if necessary, and brace off the top connection to ensure that the wind column can serve as a gravity column.

For the joist uplift - can additional bridging help?



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[msquared48]

500% sounds like an awful lot, even considering the snow drift.

If that is the case though, perhaps you could field weld a tube section to the bottom flange to stiffen it?

Mike McCann, PE, SE (WA)

 

[JedClampett]

Here's a similar issue, discussed on a different board. thread765-368271
My advice is the same as I said on that thread. Once you're sure, you must notify the owner. You must document your notification. If it's overstressed (and a 12 inch beam 40 feet long is overstressed), it's an ethical obligation to tell the owner.
You might make an enemy (hopefully not; you're not doing this out of spite), but if the beam failed and you were brought into court, the "...it's been standing for 30 years, so it must be OK...." defense is going to cost you a lot of money.
And your letter doesn't need to recommend anything, just put the owner on notice. Unless you've done a design you could seal, you probably shouldn't include a fix. If you did, the owner might just do it, without talking to you. If you conclude with "I'd be happy to help resolve this," that's not out of bounds.

 

[Once20036]

Thanks for the replies,
I`m not concerned at this time with the repairs as its an existing condition that we're not changing. Every location we're adding load is being strengthened if there's a new or existing overstress.

JAE:
I agree using the inflection point was common which is why I checked and assumed I'd find zero overstress. Using the inflection point and still being 30% over was surprising to me. Regardless, this is using the wind column to support gravity loads which , a) doesn't eliminate the overstress, and b) puts gravity loads where there may not be a significant foundation. I ran the calculations a couple weeks ago and was holding off on the report pending additional field investigation, so the 200% was from memory, however, without the inflection points there is no bottom chord bracing. I`ll double check the percentage and the Cb ratio and post them. I haven't been asked to sign off on the building and obviously wouldn't, but I`m grappling with the tone/sense of urgency that the letter needs to convey. Essentially the two ends of the spectrum are: here is a list of the issues we've found, or, these are the issues that we believe need to be fixed immediately and we're recommending our client not occupy the building until the work is complete.

ms:
500% *is* an awful lot. That's why I would never use a w12x14 girder to span 36'.

Jed:
I read the other thread before posting this - the question over there seems to me, "Should I tell the owner". I`m already sold on that one. My question is more, does this need to be fixed? I have no idea who the original engineer was and (somehow) the building may have been code compliant at the time it was erected. I have no way to determine how it was designed and intend to stress language in my letter that the building does not meet current standards. It's possible that the wind columns are adequately tied into the foundation wall which distributes the load over a sufficient area. I don't have drawings, time, or a budget to perform that field investigation. If the building was code compliant at the time of construction - I feel I would be out of line using strong language or pushing for repairs. The building has stood for 30 years... Again, I know that's no reason to sign off on it and it's a terrible argument, but it's one that the owner would surely raise if I began demanding they spend big money on repairs.

Of course, I`m planning on checking, rechecking, and getting a coworker to perform an indepedent check prior to taking any actions...

 

[JedClampett]

The answer to does this need to be fixed is, "yes, unless you can avoid snowfalls and other code design cases" But that's up to the owner. The numbers tell the answer.
Favorite Overstress Story:
Guy I knew worked for a company that designed storage racks. He said these things are designed to the gnat's buttocks. F.S of 1.0 and no more. So he designed a rack for Frito Lay's Bean Dip Line. And he made a mistake. No amount of pencil sharpening was going to make it go away. So it was time to tell the boss. He was worried he'd be fired. So he explained to the boss what happened and when the boss figured out that he had, indeed, screwed the pooch, he called his Frito Lay contact. He said hello, then said, "You know that rack we designed for bean dip? Use it for chips!" and hung up the phone.

 

[Hokie93]

If the "wind columns" were indeed intended to support gravity (or net uplift) loads, then I second JAE's comment about the C_b factor. The Commentary to AISC 360-10 (in the 14th edition AISC Steel Construction Manual) includes an expanded treatment of the C_b factor for the situation you describe. Using Equation (C-F1-5) may provide some additional capacity. Or it may not if you are already at ΦM_p or M_p/Ω. The Commentary also includes a C_b factor for a net uplift load when the top flange is laterally braced (C_b = 2.0). Yes, these are Commentary equations, not Code, but I would not hesitate to use them. The equations were developed by Joseph Yura and fall well within the "alternative methods of analysis and design" permitted by the AISC Specification. Good luck.

 

[CELinOttawa]

My first though was the inflection point in a girder being a brace point (aka Gerber System? or is that a memory issue?) system issue, as was JAE's reaction. Second is to run through the following:

- Some engineer's will consider the span as being d from the face of support. Others will argue the span is to the face of the column. Still others will be conservative and consider the span to be centre of supporting column to centre of supporting column (typically only when fully fixed ends).

- Was fixity considered in the beam design? Can you do so with the detailing as built?

- Has any composite action of the deck been considered? Can you?

- You say CMU exterior, but is the slab continuous across the wind column? Could tension bars and shear transfer be present? I have seen a french system used in New Caledonia which did exactly this and resulted in very wrong looking beam sizes.

Effectively I am asking if you know your facts. I am not meaning to belittle your work, but there are a lot of tricks I've seen by traveling that I don't know I'd have thought of/been aware of if I had stayed in Canada. Be careful when you criticize, and please let us know where this goes. Very interesting case!

 

[Ron]

I`m not concerned at this time with the repairs as its an existing condition that we're not changing. Every location we're adding load is being strengthened if there's a new or existing overstress

You have to be concerned about it. You know about it, therefore you have an obligation to point it out and recommend further evaluation at the least. Let the owner know that repair is necessary, but because you have not done a complete analysis, you cannot determine the extent of the necessary repairs. I don't see the need to get another engineering firm involved to evaluate this just because it isn't in your scope. If you have the capability to do the evaluation, ask the client for a change in scope to complete the evaluation and provide repair recommendations.

 

[Once20036]

Some of the details:
(2) 18'-8" spans, joists @ 5'-7". Joists have 42' span, 20psf dl + 20psf ll
W12x14 Bm (36 ksi? 50 ksi?)

As a simple span:
M=149ftkip
Mn/omega(36ksi, cb=1.14, Lb=5.58 = 35.17, 423%)
Mn/omega(50ksi, cb=1.15, Lb-5.58 = 44.77 333%)
Previous calcs used Cb=1.0, hence the 500%

With gravity loads into the wind column:
Mpos = 20.9ftkip, no issues @ 36ksi or 50ksi
Mneg = 37.7ftkip
Mn/omega (36ksi, cb=1.0, Lb=5.0 (infl point) = 32.14, 117%)
Mn/omega (36ksi, cb=1.95, Lb=5.0 (infl point) = 36.11, 104%)
Mn/omega (36ksi, cb=2.15, Lb=18.67 (no infl) = 17.74, 213%)

SO - it looks like my original calcs didn't include Cb for the case with inflection point bracing. Including this factor it looks more favorable for the beam.
The issue still remains that in order for this system to work, the wind column (as evidenced by the slotted connections at the top) needs to take gravity load, and transfer it to some unknown foundation. There is the issue with inflection point bracing, there's the issue with joist and joist girder uplift, and possibly an issue with uplift at the perimeter beams...

At this point, I`m *guessing* the construction drawings were just a couple sheets with a couple details and a couple beam sizes. The contractor was left to guess which details to apply where and made some... interesting guesses. It doesn't make sense for these to be wind columns, but I've attached a shot of the top connection in case anyone else wants to weigh in. You can clearly see that there are slotted connections and they're bottomed out.

CEL:
These calcs are based on the center of column dimensions. Beams frame into the column webs so face of support would be the same. Personally, I've never used d from the face.
Connections are double angles with a-325 TC bolts. There are no welds to indicate fixity.
The roof is 1-1/2" metal deck with no concrete on top. I`m not sure you can consider anything as composite in this situation...
The only facts that I have are the ones that I've field measured. There are no design drawings and the originally designer hasn't been/can't be determined. I`m trying to work through these issues before saying anything that could be interpreted as a critisism.

Ron:
Our client is leasing the building from the owner. They aren't responsible for understanding or repairing deficiencies in the owners building. In the letter to the owner, I can certainly indicate that my firm is capable of doing a complete analysis and offering to assist, but they could certainly seek advice elsewhere.

 

[CELinOttawa]

No concrete, no composite.. You're right on from what I can see.

FYI: I would never consider d from the face, although this can be justified in deep beam concrete. At that point, however, I prefer to use strut and tie to cut out the middle man.

The application of d from the face is only really legitimate in timber, and at that I only really like it as a shear span. I have seen old calculations (family archive) which used this, as well as met older engineers who's attitude could be summarized as "the shear capacity of steel is such that it behaves like a support within d of the face". I suppose if the section was also squat and detailed for this, ie: there is no deflection in this range, I could buy into this, but I don't like it.

Jeez, I know I am one to digress, but this one was crazy! *sigh* My apologies, OP.

 

[JAE]

The only problem with your new calcs you provided is that the inflection point isn't a brace point and shouldn't be considered as one.
However, it is usually a simple procedure to add lateral braces along any beam.

Check out Eng-Tips Forum's Policies here:
faq731-376