Designing a Composite Member 1

[Gousey]

Hi,

I am trying to design a composite floor member that will be subjected to the typical floor forces.

The idea is to create a composite floor element that uses Cross-Laminated-Timber (CLT) and Autoclaved Aerated Concrete (AAC).
The reason for this is to increase the Fire Rating of the floor system.

The AAC is to be fixed on top of the CLT using screws - each screw has a pull-through capacity of over 1kN.

How do I ensure the connection between the two materials are composite? Is there a way to calculate the minimum spacing and number of screws required?

Thanks

 

[MotorCity]

Since you are connecting the AAC with screws, I assume that the AAC is some sort of thin, pre-fabricated sheet material (perhaps like cement board)? If that's the case, are you really relying on it to act compositely w/ the CLT? Seems to me that there would be strain compatibility issues between the two materials. Since the main purpose of the AAC is fire proofing, I would have a hard time being convinced that you could resolve the compatibility issues and come up with a connection that would result in any meaningful composite action. But, if you wanted to go the theoretical route and design for composite action, design the connectors based on VQ/I.

 

[Nor Cal SE]

Echoing MotorCity, I believe your best route is to design the CLT to take all the stresses, not counting on any composite action from the AAC. To design as a composite member, it seems you would have to do a detailed analysis in which you use the ratio of the elastic modulus between each material to create a transformed section, to ensure the strain compatibility. Additionally, you would need to know the allowable stresses of the AAC, and hopefully there aren't any scenarios where the top of your floor is the tension face and you have bending, because I'm guessing the AAC will have very little tension capacity. Assuming you do all that correctly, you could probably check your screws to resist the shear flow at that location of the transformed section.

This seems like more hassle than it's worth.

 

[Gousey]

The AAC topping is actually a prefabricated 75mm thick steel reinforced panel. This will add fire protection to the CLT as well as boost the acoustic properties of the floor. My client wants to adopt this option without simply placing additional loading on the CLT flooring, but rather have a composite floor system.
I am not sure how to determine the strain capatability of the two elements, however as the AAC element is reinforced, I would have thought the only check required would be the shear flow at the junction of the two materials and ensure the screws can withstand that force.
The shear capacity of the screws in the AAC panels (just before it starts ripping through the AAC) is 0.85kN per screw.
Can I simply use this to ensure I meet the shear flow requirement?

 

[KootK]

I believe that a few projects on the west coast of Canada have been completed using the concrete toppings compositely. Early versions used screws projecting into the concrete at 45 degree angles. Later versions have used strips of metal lath like stiff mounted into reglets sawn into the CLT and fastened with adhesive.

Some project specific testing has definitely been done and associated design methods developed. I'm not sure as to whether or not provisions have been adopted into any major codes. I'd recommend contacting FPInnovations. They're at the forefront of the research associated with these technologies and often partner with the consulting firms designing the projects. If anybody knows where the state of the art is at, it'll be them.

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.

 

[BAretired]

The panels may add fire protection from the top but they don't do anything for the bottom, so they do not provide a fire rating recognized by any code I've seen.

Without knowing the strain compatibility between the two materials, you can't determine shear flow. If the Young's Modulus of both materials is known, a transformed section may be found from which the shear flow may be calculated.

BA

 

[Gousey]

@[KootK]
Thank you for the insight, I will start looking into this and will also see what FPInnovations have done. As I am using prefabricated elements (not possible to do with in-situ AAC, since it would all need to be cast in an Autoclave), I am trying to get a screw option to work. Preferably vertically installed screws as I know the shear capacity through the AAC.

@[BAretired]
I see your point made regarding the fire protection from the top of the CLT - not too sure why this is the case, however I will ask the question.
I am slightly confused at this stage as my understanding of shear flow (as of this today once I have started to look deeper into it) is that q = VQ/I

Does the youngs modulus come into effect where I calculate Q[=A'Ybar']?

Would that mean A' would be equal to the transformed area of AAC to CLT? as opposed to A' being the area of AAC not transformed?

Seems to make sense to me now that I have written this - no other areas in the calculation requires the need for youngs modulus from what I can see.

 

[BAretired]

Would that mean A' would be equal to the transformed area of AAC to CLT? as opposed to A' being the area of AAC not transformed?

I believe that is correct. Also, to find 'y', you need the center of gravity of the transformed section.

BA

 

[Gousey]

@[BAretired]

Thank you very much! It's all clear as crystal now

 

[KootK]

So the AAC would be prefabbed before being attached to the CLT? If so, I change my answer: don't use composite action for this. I don't feel that interface slip/stiffness can be adequately addressed with vertical, post installed screws. And there's no testing precedent to lean on that I'm aware of.

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.

 

[Gousey]

@[KootK]
There has been independant testing that has determined the shear capacity of screws in reinforced AAC panels. The recorded results included characteristic values at both the initial shear slip as well as ultimate shear failure.

For the purpose of my calculations, I was only going to use the initial slip values to ensure a conservative design.

Is there anything that states I cant do this? Am I looking at this incorrectly?

 

[BAretired]

Would there be some type of adhesive that would work in addition to screws?

BA

 

[KootK]

There has been independent testing that has determined the shear capacity of screws in reinforced AAC panels. The recorded results included characteristic values at both the initial shear slip as well as ultimate shear failure.

Neat. Can you direct us to this online? Was the testing done on AAC fastened to wood?

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.

 

[Gousey]

@[BAretired]
No, the concept was to only use screws. This is to ensure the install time is kept efficient and the costing can be kept to a minimum. However this may be considered should it be found that the screw-only option is not viable based on the calculations.

@[KootK]
Unfortunately as this assessment has been completed using a specific brand of AAC I cannot disclose any documentation at this stage unless approved by the document owners.
Initial testing was completed using AAC screwed into timber, however as it was quickly determined that the AAC panels governed the design; the rig setup was changed to eliminate another variable in the testing.
The testing rig setup was updated to using a steel angle connected to a hydraulic ram which had a screw passing through the angle and into the AAC panels. The ram would then be loaded which produced a shearforce vs deflection curve. Hence we could obtain initial shear-slip and ultimate shear capacity.

 

[KootK]

Thanks for that Gousey. Unfortunately for your purposes, I suspect that you're the expert on this technology, at least among this crowd. Your installation would have the screws installed in the field, from the top, right? And one way span? Here are a few complicating factors that come to mind although I fully admit that some may be of little practical significance.

A. Initially, the AAC will have to be drawn down to match the self weight deflection of the CLT. That will lock in some non-composite stresses into both the CLT and the AAC. I doubt it would amount to much but it would be an issue to consider if it cracked the AAC.

B. The concrete and the CLT will both creep under load, at different rates.

C. Only include every other, parallel to stress CLT lamination in your VQiT calcs. That will shift the neutral axis closer to the concrete and increase fastener demand.

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.

 

[MotorCity]

This is an interesting concept, and I applaud you and your client for the willingness to stick their neck out on these outside the box solutions. Heck, if we just kept building things the same way, we would all be living in grass huts. This is pioneer type stuff compared to the everyday steel and concrete assemblies we are used to. Rest assured, we will be eagerly following to see if there are any code acceptance issues or testing results

 

[Agent666]

Link

Similar application from a few years ago here in NZ? I've never seen it used though, no idea if it considered composite action or not.