Adhesion quality control of laminated safety glass using ultrasonic velocity and attenuation measurements

Suchy, Thomas

Permalink

https://hdl.handle.net/2142/29455 Copy

Description

Title

Adhesion quality control of laminated safety glass using ultrasonic velocity and attenuation measurements

Author(s)

Suchy, Thomas

Issue Date

2012-02-01T00:47:03Z

Director of Research (if dissertation) or Advisor (if thesis)

Reis, Henrique M.

Department of Study

Industrial&Enterprise Sys Eng

Discipline

Systems & Entrepreneurial Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Laminated Safety Glass
• Adhesive Bond
• Ultrasonic Testing
• Energy Velocity
• Attenuation
• Dispersion

Abstract

Laminated safety glass (LSG) specimens were prepared by Ceralink Inc. using a novel FastfuseTM radio frequency (RF) lamination technology in lieu of conventional autoclaving for the final stage in the lamination process. Two groups of LSG specimens were provided for experimental ultrasonic testing (UT) with the following layup: glass/copolymer/glass. Group 1 specimens contained an automotive grade polyvinyl butyral (PVB) interlayer and float glass outer layers. Six specimens with various combinations of RF lamination time and applied pressure were prepared for Group 1. Group 2 specimens contained identical float glass as Group 1, but an architectural grade PVB interlayer with a higher stiffness was used. The RF lamination time and cooling time under a constant pressure were varied between four specimens in Group 2. An analytical model was developed using a quasi-static spring model (QSM) to simulate guided wave behavior in LSG for different levels of adhesion. Energy velocity and attenuation dispersion curves were traced to complement experimental UT. An analytical sensitivity analysis was performed to observe the effect glass and PVB stiffness have on the guided wave behavior in LSG. Ultrasonic energy velocity and attenuation measurements were carried out to characterize material layer properties and to estimate the adhesive bond strength in each LSG specimen. Preliminary energy velocity measurements were successful at evaluating the Rayleigh velocity of the laminated specimens, which was found to be directly related to the stiffness of the glass layers. Additional energy velocity and attenuation measurements were performed, although conclusions were somewhat limited as many assumptions were made in the analytical models about material properties and surface roughness of each laminate constituent. Energy velocity measurements from Group 1 specimens exhibited similar trends and were all estimated to hold relatively low adhesion levels of approximately pummel number 3. Destructive pummel tests were performed and revealed actual adhesion levels between pummel number 1 and pummel number 2 for all specimens. Skewed assumptions from the analytical dispersion models likely led to the overestimates in the adhesion levels predicted from energy velocity measurements. All of Group 1 specimens exhibited similar attenuation measurements, although higher adhesion levels were predicted, between pummel number 5 and pummel number 6. In addition to assumptions made in the analytical mode, increased error in adhesion level predictions from attenuation measurements was likely associated with additional modes of energy loss and general testing setup. Energy velocity and attenuation measurements performed on Group 2 specimens predicted a slight increase in adhesion from Group 1 specimens, although definitive conclusions could not be supported, as pummel testing was not performed and material properties were not disclosed. Overall, it was encouraging to find that UT was successful at predicting similar adhesion levels for all laminates in Group 1, which was supported with pummel testing results. Additional testing is recommended on a set of laminates with higher levels of adhesion and known material properties to investigate the integrity of this UT approach further. In general, the adhesive bond strength of each specimen from Group 1 and 2 was found to be relatively low on the pummel scale. It is believed that greater RF lamination pressure is needed to allow adequate flow characteristics in the PVB interlayer to induce proper interfacial bonding. Proper bonding will allow the adhesive bond strength to increase accordingly. Once proper lamination parameters are discovered, FastfuseTM RF lamination shows great potential for the final stage in the lamination process for LSG. Lamination time is reduced from many hours to a few minutes, and up to 95% energy savings can be expected. Additional analytical and experimental work is recommended using this technology to help characterize the adhesive bond in LSG.

Graduation Semester

2011-12

Permalink

http://hdl.handle.net/2142/29455

Copyright and License Information

Copyright 2011 Thomas Suchy

Owning Collections