Seismic Retrofit Argument 9

[bookowski]

Probably no b&w answer here, just looking for opinions for and against the following argument.

I've got a client looking at a 3 story conc building that is approximately 100yrs old and was originally a manufacturing plant. Based on the column and beam sizes it was intended for very heavy use. Given the age there would not have been any explicit LFRS that would comply with current codes. This is in a relatively high seismic area.

Here's the catch. There are photos and documents that clearly show that this was originally a 4 story building and for whatever reason the top floor was removed at some point. Based on the photos the top floor appeared to be less heavy duty, still the same beam and slab construction but smaller members and a shorter story height - maybe offices for the factory. The project only makes sense for the client to purchase if they can add back the 4th floor. My first opinion was that they'd have to do a seismic upgrade but after some cajoling from them I'm considering if there's a valid argument to be made to the local bldg dept that we are not increasing the seismic demands. The EBC states that when evaluating increased demand in lateral loads "For purposes of this exception, comparisons of demand-capacity ratios and calculation of design lateral loads, forces and capacities shall account for the cumulative effects of additions and alterations since original construction." Usually this would happen in the reverse direction but in this case the cumulative alterations have reduced the lateral demands and we'd only be restoring them. Either way I've told them that they need to get the ahj to opine, it's too atypical for me to guess how they'd treat it. But looking for any opinions here on how reasonable this argument sounds to an engineers ears.

 

[structSU10]

I'm guessing you do not have original drawings available for this building?

Are you comfortable with the building if you add a story on? I would imagine the new floor would still be built to current codes. I think what is described seems somewhat reasonable, but you are taking full liability for this building.

 

[bookowski]

No original drawings. Yes, taking the same liability as any similar renovation project. I'm only interested in how reasonable the seismic upgrade vs no upgrade seems. Sounds like 1 vote for somewhat reasonable.

 

[KootK]

1) Should you expect your strategy to result in a building capable of surviving a significant earthquake? I expect so as long is the end result is the kind of forgiving building that winds up using every single column as part of a distributed moment frame. Or, conversely, if there's so much solid concrete wall around the perimeter that it couldn't possibly fail.

2) Is this strategy a good way to serve your client's interest? Absolutely. And thus you should pursue it. You're odds of being successful with the AHJ will be highly dependent on which AHJ but I would be inclined to give you somewhere between 30% (San Andreas) to 70% (New Madrid) odds.

3) Obviously, a 100 yr old building with no designated LFRS is unlikely to have the detailing required to check out using modern, high seismic, prescriptive design methods. There would be load path holes. If forced to evaluate the building numerically, something like performance based seismic design might suit. Obviously, that would entail a lot of data collection and a healthy fee.

In my heart of hearts, I feel that you'd almost have to try to screw things up in order to construct a three story, CIP beam and column building that wouldn't survive a big earthquake from a life safety perspective. If the occupancy would be sensitive to drift induced non-structural damage issues, that might be something for the owner to consider.

 

[Greenalleycat]

Interesting responses - clearly some big differences between countries
The answer here is 110% a Big, Fat, NO WAY

I would be astounded if someone accepted the argument that you can arbitrarily add another story to a building just because '100 years ago it had one'
The building is probably already a hazard and past its use by date, now you want to add significant additional load to it without strengthening?
If that big earthquake happens and your building collapses and kills someone do you think you could sleep at night? Could you justify your decision in court? Could you look a family member in the eye?
"It had a fourth story 100 years ago and the columns were pretty chunky so I figured it was OK".....
This is not engineering, this is wishy washy client shit. You don't even know what the building is made from, how can you justify this?

There was a determination on a similar case here by the national building department recently
A developer wanted to seismically strengthen their existing building to 80% code [allowed by specific legislation] but to make it economic they wanted to add 3 more stories at the same time
They wanted to design the new stuff so the end result was 80% rather than 100% code for new work + strengthening - bearing in mind that the loads on the strengthening would be higher due to this additional 3 storeys
Council refused it and the ultimate conclusion supported council's position - the new work MUST comply with the building code, and the building code doesn't allow you to design to 80% for new work
If they wanted to add 3 storeys then everything, including the strengthening, had to be done to 100%

https://www.building.govt.nz/assets/Uploads/resolving-problems/determinations/2024/2024-034.pdf

 

[TLHS]

It feels wonky to me unless you've got some better records than photos. I've had this same discussion with people regarding buildings that were built with future expansion capacity. It gets tricky.

If you're relying on the 10% clause, how are you going to show that you're not exceeding it when you don't know what the structure was, what load path it used, what it was used for, or how heavy it was?

My general thoughts are that you shouldn't go down the grey area road because it sounds like the type of building that you could qualify fairly reasonably anyway if it's robust and has pretty regular load paths. You're going to have to do engineering and construction to reinstall that top floor, so are some moderate upgrades to load path really a big deal?

 

[JAE]

I'm kinda leaning towards what Greenalleycat says.

1. You don't have original design drawings so you have no clue what level of seismic event your building can resist without collapsing (life safety).
2. Do you know in fact that the fourth floor was ever actually built?
3. If the client has the money to pay for a thorough structural takeoff on all beams and column reinforcement, you could perhaps develop some idea of your seismic capacity - otherwise you're just guessing as to item 1 above.
4. My opinion of the code allowing some level of exceedance with older buildings is based on you knowing the demand/capacity ratio in the first place - which you don't.
5. Sounds like your building is very heavy - thus resulting in a very high seismic demand.

...just a few gut feel thoughts.

 

[Tomfh]

Given the original top floor is gone, I would think "original" consists of what you have right now, and your new top floor is an addition.

 

[bookowski]

Nice spread on responses. If I take 0 as neutral, + as do it, - as don't do it: koot (+2), struct (+1), tlhs & jae (2 x -1), green (-2).

@Kootk: I'm roughly where you're at right now. One thing that concerns me wrt to the likely detailing is that in my experience with old concrete buildings there is often very little beam shear reinforcement. I can have whatever probes, scans, tests that I need so I'll be able to see if this is in fact correct.

@Tomfh: The way the code reads is that it's wrt all cumulative alterations, i.e. measured relative to the original.

@JAE: Yes, I know that it was built based on many photographs. This was a car manufacturing facility (has original internal car elevators) and there are decent historical photos and documents that the 4th floor was in place for a long time.

@GreenAlleyCat: I get the argument but I'm nowhere near as convinced in the certainty of it. I'm a bit jealous of how black and white you're able to see things, my problem is always the opposite. Could I sleep at night? Yes, I think so and if I didn't think so then I wouldn't be considering it. I disagree that this isn't engineering, I'd say the opposite - it's going beyond applying a set of prescribed calculations to evaluate the particular risk vs benefits of a project. I do not mean my risks vs benefit, I am referring to the project and public. As to "you don't even know what it's made of"... I assume that this comes from not much work on alterations. We can measure all geometry, we can test the concrete, it's a very regular layout (repeated grid, manuf type layout) so exposing the rebar on several beams, columns, and slab and extrapolating that is not unreasonable. It's true that we will never know 100% of every detail, but I can definitely sleep at night taking a reasonable amount of info and extrapolating it. Definitely a difference based on country. I've got a good friend that works at holmes and you'd think that the seismic gods were out to get him in particular, he'd retrofit the shocks on your car if you let him.

 

[Brad805]

We all have read your posts and I am confident you will navigate around the drawing issue fine. The question of the level of upgrade comes down to a risk discussion with the owner. Given it is a realitvely high seismic zone I would expect a large amount of damage or total buidling loss if that were to happen. A seismic retrofit could reduce the amount of damage signficantly. If this building is critical to the owners operation and is filled with expensive equipment, a significant upgrade could be of great value if the risk is high. They might want to look at the fine print in their insurance policy. We are working on a project now where the potential for liquefaction is high. The geotech suggests up to 500mm of deformation in the event of a worst case event. We outlined three foundation strategies and advised that the most cost effective foundation would most likely result in a complete building loss if that event were to happen. They can still get out, but the building will be history.

 

[Rabbit12]

I don't think I like this. The main thing that gives me pause is the fact that the rebar detailing is not likely to provide sufficient ductility to prevent significant damage or collapse during a design seismic event. I think you are taking on a large amount of risk here. I'm generally a little less risk averse than some of my colleagues but this one is too much for me.

If you do this and a couple years down the road you get a design seismic event and the building collapses and kills some people.....I'd certainly not want to be in your shoes.

If you knew this was detailed properly to get the proper ductility, I'd say your argument holds water.

 

[lexpatrie]

Having "done" this once (it was a building in Ohio designed for vertical expansion before meaningful seismic codes, in concrete),

This seems unlikely to pan out. The bars will not be where you need them and will not extend as far as you need to develop them. (This wes before the IEBC really, as it wasn't adopted yet in the area).

While restoring the fourth floor might be okay for gravity (presuming it was correctly designed in the first place), the lateral situation is changing because you are increasing the seismic load on the building. I doubt "it was there before" holds water regarding the IEBC because it's an alteration (at least) now.

The AHJ may not be astute enough to catch on, which really isn't relevant as the onus is on you (or the relevant P.E.) to protect life safety, health, and welfare.

 

[TLHS]

Brad, I'd disagree. Seismic risk specifically is a tricky thing and owners aren't really in a great place to make informed decisions on it. It's the only area of code where we basically say that a design case is allowed to be broken, it's just not allowed to fall down and kill people. Especially with a retrofit that will have detailing you can't fix, explaining it to them in terms of property damage really doesn't feel like a good way to go. Those areas without great detailing are more likely to break in a way that's a hard evaluation/fix. It's writing cheques that you can't cash in terms of promises.

Like, it's not unusual for a retrofit to use different materials. If you're putting in a bunch of steel stuff that will only engage after the concrete cracks to hell, your design level earthquake with the reinforcement may not fall down but it might be in a condition where it's a complete teardown anyway. That's still likely code compliant to design though.

Basically, the ASCE provisions with a normal importance factor are intended for life safety with a design quake and collapse prevention with the MCE. You can obviously explain to an owner that various design choices have different benefits, but unless you're actively designing for a higher performance I really don't like implying that earthquake design is giving more than life safety. For the most part, improved seismic design is going to give better performance in earthquakes less than the design quake, but even that isn't necessarily true for all systems.

 

[KootK]

With regard to the caliber of the existing rebar detailing, I feel that the only rational approach is for the engineer to just assume that it would be deficient for high ductility. If the building is to calc out without retrofit it really has to be the case that expected strains will be so low that high ductility is simply not required.

 

[TLHS]

You can try assuming a fully non-ductile system as a worst case. If that explodes, then without retrofit you can also do the drift based evaluation stuff in ASCE 47 with the values for the system you get from the screening. Haven't used the latest one, but it's not crazily hard for fairly regular systems if I remember correctly.

 

[Rabbit12]

KootK, I'm not sure adequate information exists to even do an analysis. I think this is a case of wanting to only say the 4th story existed before it was fine we are going to put it back and still be fine.

If there are sufficient drawings to analyze the building assuming ordinary concrete moment frames, I think your approach would work but there needs to be some analysis so confirm this is adequate under seismic loads.

 

[Greenalleycat]

Book, my perspective comes as someone with extensive experience in seismic assessment, retrofit, and alteration in a high seismicity country
I got into structural engineering specifically because of the time that I spent working in and around damaged and destroyed buildings following my city's earthquake in 2010
So your assumption of my experience is wrong
Perhaps you are simply more blaise than me about this because your city has not been flattened yet

It is not engineering to say "100 years ago they stuck a 4th floor on it so she'll be right to add it back"
Engineering is doing the numbers to prove that it works

We can measure all geometry, we can test the concrete, it's a very regular layout (repeated grid, manuf type layout) so exposing the rebar on several beams, columns, and slab and extrapolating that is not unreasonable. It's true that we will never know 100% of every detail, but I can definitely sleep at night taking a reasonable amount of info and extrapolating it. Definitely a difference based on country.

I do agree with this, and if you do those investigations, get the details, do your numbers, and prove that it works - more power to you
That is engineering

The difficult part from there is as discussed in that Determination - your new story must be to Code, and how can it comply with the seismic provisions when the stuff below it doesn't?
There's no doubt that adding an extra floor will also significantly punish the levels below, so there is no way you can do this without a significant increase in the risk level of the building - without strengthening it too
This may be where your regional differences come into play - we would never get it through Council, but perhaps your building authorities view things differently

I have several good friends at Holmes too - one of my good friends is a TD there specialising in Risk Engineering, funnily enough
I'm not going to pretend I agree with everything Holmes does, but they wouldn't agree with everything I do either
They are a good firm though and, as with me, their experience is shaped by what we've all learned from the earthquake here

 

[KootK]

I think your approach would work but there needs to be some analysis so confirm this is adequate under seismic loads.

For sure it would require some analysis. My point is really that I don't feel that it is productive for us to be overly focused on whether or not the 100 year old building has modern rebar detailing appropriate for high ductile seismic design. I guarantee that it does not. Whatever path forward Bookowski takes, it shouldn't be that one.

 

[TLHS]

Greenalleycat,

I generally agree with you, but to defend bookowski a little bit, the argument isn't that the building is okay for seismic loading because it used to have a top floor. It's that the building is potentially grandfathered into not necessarily needing to be 'okay' because it's effectively in its original loading condition. That the lateral demand is effectively the same as the original condition was and that it therefore doesn't need to be seismically upgraded.

Basically is the original condition we're comparing against 4 floors, in which case the argument is that 4 floors - 1 floor + 1 floor = 4 floors, or does something (time elapsed, differences in construction, etc) mean that the base condition is 3 floor, in which case we're talking about 3 floors + 1 floor.

In both conditions the building is not going to be compliant with current code. With zero time gap and an exact duplicate replacement top floor you have either a renovation or an addition where the new load is fully offset by a removed load, so you can pretty reasonably argue that the demand has increased by less than 10% in the lateral system and you don't need to do a seismic upgrade. This still may not fly everywhere, but this is a pretty reasonable position to take. Now the question becomes, how much time lag is acceptable? Does it matter? How similar does the replacement need to be?

I, personally, think that the time gap means that this work isn't really falling under the intent of why those exceptions are there. The goal is to not punish people trying to do minor and moderate work to a building. This is significant and heavy work. I also don't think you're going to be able to properly establish what the base condition of the building was, so you can't determine whether your new configuration is the same and what the comparative effects on the lateral system are. I might change my mind on that, though, with a solid diaphragm on the new work and a well considered way of making sure there are redundant load paths for new lateral and vertical loads.

I admit that this is partially gut feelings. If I had original drawings that said that an area was designed for a storage load of some large magnitude, I'd be okay derating that storage load downward and using it for something else without looking at the lateral. Doing this with an entire poorly documented floor just feels a bit cheeky. I might feel differently if I had a drawing or calc set showing that floor, though.

But yeah, the discussion here isn't whether the building 'works'. It's whether the owner is required to make it 'work' or not.

Another question is, why did they take off the top floor? That's a bunch of work. Are you sure they didn't remove it as part of other changes that might have changed the loading?

 

[Brad805]

TLHS, it is their money and they have the most at risk. All we can do is present the options and it is up to them to decide. I am not suggesting anything less than the minimum for basic life safety. Few understand we mean life safety when we design for seismic.