Thismanual presents in detail design of hollow core slabs for gravity and lateral loads. A complete revision of the diaphragm chapter includes the latest design information. The Third Edition of the Manual has added Hollow Core Wall Panels and is updated to ACI 318-11, PCI 7th Edition Design Handbook and 2012 IBC.
-126-15
I have to use hollow core slab in my project, but I am having difficulites in choosing the right parameters (see picture). I am not going to design the hollow core slab, I just want to use for its own weight and because it will take some of the lateral load from wind. Do I have to check mark the option "Disregard stiffness of transvers bending", as the width of the hollow core slab is only 1,2m. When I choose the "Model", should I choose "Deck Slab one-way", "Slab-Flexible diaphram" or "curtian wall"? I have also attached the Robot File, where one hollow core slab is in x-direction and another one in y-direction.
Thanks!
The point is how you want to transfer these loads from the slab to supports and this is the choice you have while deciding on the panel calculation model. Using FE option this is done based on stiffness of the elements of the model. For the triangular and trapezoidal method the distribution is based on the tributary areas assigned to each of the supporting elements.
This is what is usually assumed for hand calculations yet you may also take into account the fact that such slab works in other direction too (with much less stiffness). Reading your answer I assume that you want to keep the disregard the stiffness in the secondary direction marked.
Thank you very much for your detailed description of the stuff that I have to take into consideration.
The hollow core slab in practice will be supported on bottom flange of I-profile (with welded plate on bottom) in one side and another by RHS beam (with welded plate on bottom). As it is complicated to do that in Robot, I will just use the default setting in Robot, i.e. the hollow core slab will be supported in the center axis of beam. I think, the hollow core slabs' ends will be reinforced in the support, hence I will consider it as pinned or rigid.
On the tranvers side, there will be not supported by anything, therefore I thought, I would check mark the option. But you have pointed some good points, that my assumptions are based on hand calculation. If you know by any chance any good site or literature about that area, I would be thankful, if you could give me, because I am newbee in FEM design and computer-based calcution.
I have modelled a test model (see attached file), but I am having difficulites in respect to span direction. I have changed the panel local system in accordance with the global coordinate system and defined the hollow core to span in Y-direction, but the result shows the it spans in X-direction.
Someone suggested that I could also define "material orthorooy" instead of hollow core slab by defining one of the "n=0.0001". Which option do you prefer?
Thanks
I have modelled a test model (see attached file), but I am having difficulites in respect to span direction. I have changed the panel local system in accordance with the global coordinate system and defined the hollow core to span in Y-direction, but the result shows the it spans in X-direction.
Thank you very much for your answer. As you can see, I have shear forces on the two beams, which are parallel to the slab and they should not carry the load of the slab. I want only the the two beams, which are perpendicular to the Y-axes to carry the slab load. And the slab on the those two beam should be supported as pinned, but I am having difficulites to define such a situation. I am really sorry for all these newbie questions
Looking at the screenshot, the load distribution is "two way". It seems to me that despite the load distribution is two way but ticking the"disregard the stiffness..." will forced the distribution to be "one way", is this correct?
An example with the model of a precast hollow core slab used for ceiling construction. The model was analyzed in the IDEA StatiCa Beam app using a composite section. Comprehensive calculation with different material characteristics for each part of the section and including construction phases.
When I was trying out the 'structural precast extension' for Revit, I tried creating a hollow-core slab. The devision was fine and all, the split just didn't start where I wanted it to start. It started with a complete slab at the top in the drawing, but I want it to start at the bottom.
Since there is a symbol present of the shifting direction, I thought clicking it would have an option next to it to turn around the arrow and with that, the order in which the split was performed, but there is no such option.
Since I didn't start from a rectangle, like in your example, but by drawing and picking lines, I figured out it might have to do something with clockwise/counter-clockwise things, like with walls etc.
Abstract: The former FIP Commission Prefabrication drafted the FIP Recommendations on the design of Precast prestressed hollow core floors, published by FIP in 1988 (Telford, London, ISBN 0-7277-1375-2). That document was highly appreciated by designers and public authorities because of the lack of guidance available elsewhere, especially with respect to some specific features of the product, for example the absence of transverse reinforcement. It has also served as a reference guide for national standards and especially for the CEN product standard on prestressed hollow core slabs.
During the production of that report it was felt that some design rules were incomplete or missing. In addition, research carried out since has resulted in complementary knowledge on the behaviour of hollow core floors, for example in combination with slender floor beams.
The present guide to good practice is intended to complement the existing recommendations. The research for the different items was carried out at Chalmers University of Technology (Sweden), Politecnico di Torino (Italy), VTT (Finland), University of Nottingham (United Kingdom), Building Research Institute (Poland), and the University of Rome (Italy).
Actually I am thinking to model precast pretensioned hollow core slab in Structural Bridge Design. Can the software help to calculate the prestress losses calculation and check design based on Eurocode 2 ?
Structural Bridge Design is intended to be used for the structural analysis and design of bridges to BS5400, Eurocode 2, AASHTO and AS5100, but the methods are generally very similar to those for buildings so with a little bit of "engineering judgment" it may be possible to use the software to carry out the calculations for hollow core slabs used in building construction.
The standard beams used in the software are standard bridge beams, but, any section shape and arrangement of prestress tendons can be used by using "user defined" sections and manually defining tendon locations.
The design can be carried out to Eurocode 2 and full consideration of prestress losses is taken into account, based upon the casting, prestressing and construction methods used for bridges. I'm not familiar with the manufacturing methods used for standard hollow core slabs for buildings, especially with regards to curing temperatures and how and when transfer of the prestress into the concrete occurs, but I'm sure that this can be accommodated by adjusting the many parameters available within the software.
Founded in 1904 and headquartered in Farmington Hills, Michigan, USA, the American Concrete Institute is a leading authority and resource worldwide for the development, dissemination, and adoption of its consensus-based standards, technical resources, educational programs, and proven expertise for individuals and organizations involved in concrete design, construction, and materials, who share a commitment to pursuing the best use of concrete.
Abstract:
This study proposes a practical design approach to estimate theweb-shear strength of deep prestressed hollow-core slabs (PHCS).It explores the effects of critical factors such as the shear stressdistribution, biaxial tensile strength, and the reduction in effectivecompressive stress in concrete, quantifying their impact onweb-shear strength. A data set of 85 entries is used to undertakea comparative assessment, demonstrating the improved safetyand accuracy of the proposed methodology against current designprovisions and previous proposals. Moreover, it is shown thatneglecting the beneficial effect of the prestressing force in thetransfer region leads to a conservative estimation of the web-shearstrength. Furthermore, the study introduces three modified designexpressions based on ACI 318-19, fib Model Code 2010, andCSA A23.3-14 standards. The proposed methodology has practicalimplications for enhancing the safe and cost-effective use of deepPHCS in construction practice.
You can choose from a predefined list of hollow core slab cross-sections or create your own. If you have a CAD drawing of your section, you can use the import function and the program will find the hollow-core profile and automatically generate the section for you. You will need to configure the prestressing wires/strands though using the Strand Pattern Manager.
Add topping to your prefab element and make a complete calculation taking into account the cast-in-place concrete layer. You can add reinforcement to the topping layer and have it to work together with the prestressed hollowcore. You can also perform a fire calculation.
It is very easy to add prestress tendons/wires/strands and standard reinforcement to your element. Applying the bars can be done manually or through the pattern manager and you can customise and save any layout for future projects.
3a8082e126