Determining Material Response for Polyvinyl Butyral (PVB) in Blast Loading Situations

Abstract

Protecting structures from the effect of blast loads requires the careful design of all building components. In this context, the mechanical properties of Polyvinyl Butyral (PVB) are of interest to designers as the membrane behaviour will affect the performance of laminated glass glazing when loaded by explosion pressure waves. This polymer behaves in a complex manner and is difficult to model over the wide range of strain rates relevant to blast analysis. In this study, data from experimental tests conducted at strain rates from 0.01 s −1 to 400 s −1 were used to develop material models accounting for the rate dependency of the material. Firstly, two models were derived assuming Prony series formulations. A reduced polynomial spring and a spring derived from the model proposed by Hoo Fatt and Ouyang were used. Two fits were produced for each of these models, one for low rate cases, up to 8 s , and one for high rate cases, from 20 s −1 . Afterwards, a single model representing all rates was produced using a finite deformation viscoelastic model. This assumed two hyperelastic springs in parallel, one of which was in series with a non-linear damper. The results were compared with the experimental results, assessing the quality of the fits in the strain range of interest for blast loading situations. This should provide designers with the information to choose between the available models depending on their design needs.

Viscoelastic Constitutive Artificial Neural Networks (vCANNs) – A Framework for Data-Driven Anisotropic Nonlinear Finite Viscoelasticity

“…Large-scale simulations of these structures require simple material models that capture the mechanical behavior over a wide range of strain rates. Del Linz et al [104] conducted high-stretch rate experiments on PVB, with stretch rates between 0.01 s −1 and 400 s −1 . The viscous properties of PVB likely vary within such a wide range of strain rates.…”

“…Ideally, the constitutive model should be able to represent this transition accurately. The authors of [104] proposed an FNLV model and used the strain-dependent viscosity function by [47]. The model describes the experimental data well at high stretch rates, although it cannot accurately resolve the peak stress and subsequent softening at 𝜆 ≈ 1.2.…”

“…At low strain rates, the fit quality is quantitatively unsatisfactory. In [104], the authors also fitted a standard generalized Maxwell model with constant relaxation coefficients and times for comparison. The model comprised six Maxwell elements whose relaxation times were chosen to be uniformly distributed on the logarithmic scale and kept fixed during the parameter identification of the relaxation coefficients.…”

On the bonding between glass and PVB in laminated glass

“…The material properties used are summarised in Table 2. For the glass a linear elastic material model was used and for the PVB a viscoelastic material model developed by Del-Linz was used [25].…”

Issues on the Vibration Analysis of In-Service Laminated Glass Structures: Analytical, Experimental and Numerical Investigations on Delaminated Beams

“…From a structural point of view, critical design conditions can include explosive events [3][4][5] and impacts [6][7][8], natural hazards and earthquakes [9,10], and dynamic loads in general (including moving loads, in the case of pedestrian systems [11][12][13][14][15][16]), whose effects need dedicated calculation methods.…”