Photoelastic Study of NonRiveted Lap Joints - PowerPoint PPT Presentation

1 / 23
About This Presentation
Title:

Photoelastic Study of NonRiveted Lap Joints

Description:

Photoelastic Study of. Non-Riveted Lap Joints. Deonna ... a) Glued scarf joint in timber. b) Butt-weld in metal. c) Welded Lap joint. d) Riveted Lap joint ... – PowerPoint PPT presentation

Number of Views:344
Avg rating:3.0/5.0
Slides: 24
Provided by: ITS8355
Category:

less

Transcript and Presenter's Notes

Title: Photoelastic Study of NonRiveted Lap Joints


1
Photoelastic Study of Non-Riveted Lap Joints
  • Deonna Woolard and William Pluim
  • Randolph-Macon College
  • CSAAPT
  • April 21, 2001

2
Types of Lap Joints
a) Glued scarf joint in timber b) Butt-weld in
metal c) Welded Lap joint d) Riveted Lap joint
(Structures by J.E. Gordon, pages 136-138)
3
Modes of Failure
a) Failure by shearing the rivets b) Failure by
tearing the rivets out of the plate c) Failure
by tearing the plate
(Structures by J.E. Gordon, pages 142)
4
Non-Riveted Lap Joints
  • Modern adhesives offer an alternative to joints
    made with rivets
  • Adhesives permit similar or dissimilar
    materials to be jointed without harming the
    materials themselves
  • Using an adhesive on the full overlapped area
    distributes the load across the entire area of
    the bond reducing stress concentrations
  • The geometry of structures can get more complex
    with non-riveted lap joints

5
Photoelasticity and Lap Joints
Although the lap joint is commonly used in
design, properties and reactions of non-riveted
lap joints under load are still widely unknown.
Photoelasticity
6
Photoelastic Effect
Polarizer
Analyzer
Material
7
Photoelasticity by Reflection
Polarizer 1/4 Wave Plate
Light Source
Photoelastic Coating
Test Part
1/4 Wave Plate Polarizer
observer
8
Isoclinic vs. Isochromatic
Linear Polarization
Circular Polarization
9
Non-Riveted Lap Joints
The lap joints were built from aluminum and
acrylic strips, and bonded with two department
store adhesives, Weldit and 5-Minute Fast Drying
Epoxy, and the PC-1 adhesive from The
Measurements Group. The acrylic is photoelastic
by itself, but also clear so a reflective paint
was applied to the back of the joints. The
aluminum samples were reflective, but a
photoelastic coating was required to allow
photoelastic analysis.
10
Preparation, Problems, and Solutions
Problem
Solution
  • The initial samples did not take more than 25 lbs
    of load from the frame.
  • The acrylic laps broke due to chips in the cut
    edges. The reflective paint aggravated the chips
    into cracks.
  • The Weldit cement did not hold even after
    roughing up the surfaces of the laps.
  • Sanding the surfaces that were bonded made them
    hold up to greater loads.
  • A local glass company cut laps and polished the
    edges so there were no chips to crack.
  • Use of the Weldit cement was discontinued due to
    lack of results.

11
Experimental Setup
12
Lap Joint Failure Data
1) Al .05cm 5-Min 2) Al .1 cm 5-Min 3) Al .1 cm
PC-1 4) Al .3 cm 5-Min 5) Al .3 cm PC-1 6) Al .3
cm Weldit 7) Ac .3 cm 5-Min 8) Ac .3 cm PC-1
13
Characteristic Fringes in Aluminum Samples
14
Characteristic Fringes in Acrylic Joints
15
Aluminum Fringe
Fringe Movement on Aluminum Joint with PC-1
Adhesive
16
Fringe Velocity
Fringe velocity for Al sample. Load dependent
fringe with load increasing with time.
17
Finite Element Analysis (FEA)
  • To better understand the mechanisms behind a
    lap joint under load.
  • FEA is used for situation where there is no
    closed mathematical solution to the problem.
  • A commercial software package, COSMOS/M, was
    used.
  • It allowed us to build the model in a
    three-dimensional wireframe, apply boundary
    conditions to the model, and analyze the behavior
    of the lap joint under controlled loads and
    conditions.

18
COSMOS/M Model
Wireframe model of a Non-Riveted Lap Joint
19
Explanation of Sample Results
FEA view of the top strip of a lap joint
FEA view of an internal section of the bond
20
Proposed theory
Proposed theory for deformation as load increases
from top to bottom
It was theorized that an out of plane deformation
was causing the joint to bend, and the crease
created by the bend caused the fringe to move as
the load increased.
21
FEA Deformation
Lower Limit Response Max. Def 1.13E-6
Upper Limit Response Max Def 1.13E-5
22
Conclusions
  • Photoelasticity not effective in monitoring lap
    joint health
  • Surface was inactive under photoelastic analysis
  • The bond took the majority of the load
  • The load did not transmit to the surface
  • Fringe movement on the sample due to out of plane
    deformation

23
Suggestions for Future Work
  • Determine elastic modulus of the adhesives
  • Further refine COSMOS/M model
  • Further study the out of plane deformation of the
    lap joint
Write a Comment
User Comments (0)
About PowerShow.com