How is the deflection of aluminum frame structures influenced by the stiffness of corner cleats and the fitting tolerance?
Explanation:
An ( orthogonal ) corner where two (extruded) aluminum profiles meet takes bending moments (to a certain degree).
(1)
However , system manufacturers are not able to explain the bending stiffness of corner cleats (around the strong and weak axis) which connect these profiles in the corner.
Usually, is is assumed that corners are rigid.
However, changing the rotation stiffness (N/rad) (within a reasonable range) can increase deflections significantly.
(2)
Further, I assume that a fitting tolerance is always present (between corner cleat and extruded aluminum shaft), hence the corner cleat gets activated (for bending) once a initial rotation is exceeded.
Considering (1) and (2) , the deflection can be higher than usually predicted.
As I am not able to find any related research on this topic nor can I get an explanation from system manufacturers, I wonder if my assumptions are valid?
In order to have tight tolerance for mitering and corner joints, design your frame structure so that it snaps together without gaps that would allow moisture penetration. Aluminum extruded components can be designed for quick, easy snap-together construction with tight fitting joints. Definitely when your structure is flexible you will face more deflection and less stiffness, therefore the dimensions of the cross-section of the member should be considered somehow having more moment of inertia.
There can be a considerable effect. I suggest that this paper is a good starting point:
http://www.newsteelconstruction.com/wp/wp-content/uploads/TechPaper/TechApr15.pdf
Kabir Sadeghi thank you for your answer.
(1) I have to think more about tight tolerances, which might NOT be the problem. (2) Further, you wrote: "Definitely when your structure is flexible you will face more deflection..." - In manufacturer specifications, the mitered corner , connected by a corner cleat is assumed to be rigid. Let's assume a stiff corner cleat (which I am not convinced about), what about a local plastic hinge?
(3) Further, I will need to think about the rigidness of the corner cleat itself.
Phil Francis Thank you for the provided link. The described problem is related to my question and it confirms my concern.
(4) The shown eaves connection (following the link under the sub-chapter " Connection flexibility ") is a rigid connection. However, it is written " Because the channel sections are thin, (and the plate between may be thin) there is considerable deformation ". This is astounding since the connection detail looks rigid and gives the hint that a aluminum profile connection with t~1-2 mm is not rigid. I have to investigate this in more depth.
You are welcome.
In the hing formation phase, large deflections and curvatures appears, and concentration of micro cracks will be on the critical section and there an event of closing the micro cracks in other sections. In this phase the traditional double integration of curvature underestimates the deflections that should be paid attention to it and an elastic-plastic method should be applied.
I want to add some insights even though the question is almost one year old:
(i)
It was observed (during testing) that the rotational stiffness is almost constant for an internally structurally overdetermined system but for an internally structurally determined system (e.g. L-connection), the rotational stiffness decreases in a linear manner as a function of the rotation. Hence, no plastic stress redistribution is possible (often, at max ULS loads, the structural design is required to utilize plastic stresses)
(ii)
It was also observed, that the fitting tolerance is not relevant as connections are executed in a crammed/screwed/glued manner at the same time.
(iii)
Further, for different configurations, a very ductile failure was observed - between the onset of failure and the loss of rudimentary structural resistance, a load increase factor of 3-5 can be assigned.
(iv)
Further, it was observed that the way the load is introduced - eccentric or centric - does influence the stiffness greatly which is often not considered in a global member analysis (!) The torsional rotation of the profile from eccentric load introduction (e.g. IGU load onto glass supports...) can lead to extensive torsional deformation.
- Badges
- Science topic
- Similar topics
- Computer Science
- Research Topics