A question on "suction", ASCE 7, ICC 500, storm shelters. 1

[thewolfofconcrete]

I've been a lurker for years, but the issue at hand has me so turned around I finally decided to make an account and post the question. I'm hoping that the intelligence of the masses can help me out.

Here is the issue: I work for a precast designer and supplier that was asked by an EOR what the "tension" capacity was between our precast plank and a poured and bonded topping, specifically for ICC 500 compliant storm shelters. The engineer (my colleague) who was asked this question didn't have an answer prepared at the time, so the EOR took the position that interface ties (repurposed lifting loops, in our case) are required on a 4' x 4' grid at the precast roof to provide tensile bond between the plank and the topping. Their position is that if a tornado hits the storm shelter, the "uplift forces" will "suck" the topping off the precast plank. There are clear and established standards for plain interface shear capacity, but direct tension on this interface is distinctly different in our eyes.

I wrote a letter attached to a calculation packet that used a first principle physics argument to contend that no interface ties were required, because no such force exists to pull a bonded topping off of precast plank. By nature, gasses and fluids cannot pull on an object. Pressure differentials can push on objects, but they cannot pull on them. In our case, the pressure differential exists between the inside of the building and the outside so it pushes on the entirety of the precast roof assembly. Intuitively, this force could never cause the topping to "fly away". If you have a sub sandwich, and you push on the bottom bun, the top bun will not move, the whole sandwich will.

The EOR rejected this argument and maintained their position that suction does exist and interface ties are required if we cannot provide a calculation or reference for the tensile capacity of concrete topping on precast plank.

Complicating the issue is that ASCE 7 repeatedly uses nomenclature, figures, and sign conventions that suggest or imply "suction" forces on wall and roof assemblies. Reference 7-16 sec. 26.4.1, fig. 27.3-1, fig. 30.3-1, etc...

I welcome any and all comments on this issue. Thanks in advance, guys.

 

[phamENG]

The net effect on the assembly is suction - by way of reducing the pressure above the roof to less than the internal pressure due to the 'speed up' effect of the wind flowing around the building. But yeah, you're right. If there's nothing between the two and no differential pressure across the topping itself, it won't lift off.

You may be right, but what is it worth? The EOR is ultimately responsible for the design of this building. You can fight him/her on it...or you can say okay, send your change order, and be done with it. That way, you get paid for the extra work, know that you're right, and not deal with the headache. Sounds like a winning proposition.

 

[thewolfofconcrete]

phamENG, I appreciate your reply. I disagree that the net effect on the assembly is suction. The net effect is a pressure from the inside of the building, putting the roof assembly into negative bending. I contend these aren't the same thing, especially when it comes to designing the individual parts of the roof assembly. The location and application of the force vector matters.

Regarding "what it's worth", the answer to that is unique for me because I work for a precast supplier and erector. Each one of the interface ties needs to be placed and secured in the product during fabrication. This takes time and energy. I hear from our plant guys and get feedback on the things that increase or decrease efficiency. Additionally, in the field, the rebar in the topping has to be threaded through these interface loop ties. This is a big increase in time and hassle for the CIP installer.

Sure, I agree that we can be compensated for these things. If we're getting paid for the extra time, material, and hassle, that eases the pain somewhat. That said, I do think it's worth pressing on when all of it can be explained away with a basic physics principal.

Again, thanks for your reply.

 

[phamENG]

I disagree that the net effect on the assembly is suction. The net effect is a pressure from the inside of the building, putting the roof assembly into negative bending.

Fair enough. I consider that hair duly split. Let me know how your assault on vacuum cleaner marketing campaigns goes! (I am just kidding - I get what you're saying. It's a lot like centrifugal force. We know what it is. We can feel it. But from a purely physics standpoint, it doesn't exist. To say that suction is just pressure pushing from below rather than pulling from above is to say that suction doesn't exist, it is merely an imperfect description of the result of pressure from below. In an abstract sense, I get it and I think I agree with you...but good luck convincing most people.)

The issue I see is this: very few people want to be proven wrong. Even those of us who actively try to improve our knowledge and understanding by seeking out things we don't know get uncomfortable when a seemingly well understood and otherwise immovable phenomenon is questioned intelligently. You start tugging at the parts of the persons brain that are ruled purely be emotion rather than rationality. So if I can come out of a situation where my client, the occupant, the public, and I are no worse off for a person's insistence on being wrong - I just let them have it. I have more important things to do with my time. (I would consider appropriate compensation for the added work and potential downstream production delays being no worse off.)

 

[thewolfofconcrete]

Thanks again, phamENG. I agree, I am going to struggle to convince decision makers on this if they've already made up their mind.

At the least, it's helpful to have feedback that my thinking isn't off base.

 

[RWW0002]

Do you know if the shelter was designed as Partially Enclosed with a GCpi of 0.55? (This very well may be the case due to the extreme venting requirements recommended by ICC-500 to justify an Enclosed Structure) As an "olive branch" so to speak, I wonder if you could offer to design the topping to slab interface for the net wind pressure resulting from the external pressure coefficient only. While this may sacrifice on your "first principle physics" a bit, it may significantly reduce the net uplift, or even potentially eliminate it once the weight of topping is considered.

I would think it would be much easier to argue that the "Internal Pressure" comes solely from the pressure difference and thus acts on the "inside" surface of the roof, thus reducing the interface uplift significantly.

A counter-argument for your "pressure always from the bottom" argument might be the current standard of care for the design of roofing material attachment. As far as I know, membrane roofing is required to be adhered to meet the full 10ft2 Component and Cladding uplift pressures, even though it too is part of a roof assembly that receives surface pressure due to the same combination of wind and pressure difference that your precast roof sees.

 

[XR250]

"The issue I see is this: very few people want to be proven wrong. Even those of us who actively try to improve our knowledge and understanding by seeking out things we don't know get uncomfortable when a seemingly well understood and otherwise immovable phenomenon is questioned intelligently. You start tugging at the parts of the persons brain that are ruled purely be emotion rather than rationality. So if I can come out of a situation where my client, the occupant, the public, and I are no worse off for a person's insistence on being wrong - I just let them have it. I have more important things to do with my time. (I would consider appropriate compensation for the added work and potential downstream production delays being no worse off.)"

So well stated, Pham!

 

[Greenalleycat]

An interesting topic

What's the topping thickness ? Can you do a number to show that the weight is sufficient to prevent uplift anyway?
Or subtract the dead weight from the suction and come up with some tiny interface tension required that is presumably achievable?

 

[thewolfofconcrete]

An interesting topic

What's the topping thickness ? Can you do a number to show that the weight is sufficient to prevent uplift anyway?
Or subtract the dead weight from the suction and come up with some tiny interface tension required that is presumably achievable?

The topping is 4" thick. The uplift values for ICC 500 storm shelters vary depending on site factors, and the uplift is worse around the perimeter and in the corners, but a good ballpark value is +/-300 PSF ultimate. So, we're not even close to cancelling it out with selfweight.