Graduate Thesis Or Dissertation
 

Three Methods for Analyzing Charring Behavior in CLT Panels Exposed to ASTM E119 Standard Fire Test Condition

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/p8418w12b

Descriptions

Attribute NameValues
Creator
Abstract
  • This thesis consists of three manuscripts related to the methodology of evaluating charring behavior in cross laminated lumber panel specimens subjected to fire performance tests on cross-laminated lumber panel specimens. The topic of the first manuscript is the measurement of char rates of unprotected CLT layups fabricated with ponderosa pine subjected to the standard ASTM E119 temperature curve. There is a common assumption shared by several studies conducted at OSU since 2017 that ponderosa pine can be successfully utilized in CLT panels for low- and medium-rise structures. Fire tests for CLT elements are required before manufacturers will consider using ponderosa pine CLT in certain building types and specific applications. The general approach was to measure the char rates in eight unloaded 5-ply (1219 mm x 1829 mm) ponderosa pine CLT panel specimens representing two adhesive systems (MUF, and PUR HBX) instrumented with thermocouples, following modified standard ASTM E119 test protocol for 150 minutes. Statistical analysis of char rates recorded at the same distance from the exposed surface shown no significant differences regardless of installation positions and adhesives, which allowed pooling data from all tested panels for further analysis. Except for the first 5 mm from fire exposed surface, the mean char rate in ponderosa pine CLT (0.75 mm/min or 1.77 in./hr.) was significantly faster than the design benchmarks published National Design Standard (1.5 in./hr. [or 0.62 mm/min] in NDS) or in EN 1995-1-2 (0.65 mm/min). The second manuscript is concerned with the variability and spatial distribution of the char depth in full-scale unprotected CLT assemblies fabricated with SPF and Douglas fir-Larch subjected to standard fire tests. Charring rate, or the progress of char depth in a unit of time, is a crucial parameter in numerical modelling of fire performance of structures and in standard calculations of fire rating for mass-timber structural elements. The general approach was to use charred specimens and data from previous research, in which two sets of 175 mm thick 5-ply CLT assemblies were tested for fire performance in loaded condition following the ASTEM E119 protocol, one in vertical configuration (3.0 m x 3.0 m walls) and the other in horizontal configuration (4.3 m x 5.5 m floors). Each set included three panels that differed by the combination of wood species and adhesive systems. Residual char surface topography immediately after the 2-hour fire exposure was documented on series of digital images used later for creating char surface maps. The assemblies were subsequently cut into 305 mm x 305 mm blocks to measure the residual depth at eight points on the perimeter and to create a final char front map for each assembly. The char base maps combined with char surface topography maps allowed determination of the residual char depth distribution for each assembly. The charring rates calculated using recorded thermocouples data varied between the nine locations of TC clusters (COV up to 47% for floors and up to 30% for walls), but there was no clear pattern related to the location on the panel. The char base maps allowed the analysis of correlation between char rates calculated based on the residual depth with char rates measured with the nine TC clusters and analysis of potential correlations between the char rates and the residual char depths at various locations. While the shape of char base line might be affected by a variety of non-homogeneities already lost to the fire it was observed that one of the factors was the local shape of annual rings. The third manuscript is based on a hypothesis that the char fall-off events during standard fire performance tests following curated temperature curves (like ASTM E119) may be detected with reasonable accuracy based on the analysis of the microdisturbances of the furnace temperature signal. The most desired behavior for flammable structural materials like CLT panels in fire events is a gradual transformation of flaming combustion into smoldering combustion until the process is naturally self-extinguished. Such behavior is observed when the charred wood on the surface provides layer of insulation for the core material. However, a large proportion of CLT products is bonded with an early variety of PUR HBE adhesive system that at some point of fire event lets the char layer fall off creating additional source of heat in the interior and exposing unburned surface of the next layer to fire. The severity of the effect of such event on the fire dynamic may depend on the precise timing compared to the progress of the char front through the panel section. This manuscript presents the theoretical background and a proof of concept for a practical application of a method of detection of char fall off events based on the analysis of the microdisturbances of the furnace temperature signal. The general approach is to distinguish between the disturbances of the temperature change of the furnace being controlled to satisfy the ASTM E119 standard furnace temperature curve, and the disturbance in furnace temperature signal triggered by the second flashover, the standard deviation of rolling average (SDRA) of the furnace temperature was calculated and analyzed. The SDRA and the integral moving standard deviation (IMSD), which is the cumulative value of the standard deviation of rolling average (SDRA) over time allowed quantification of the amount of temperature disturbances over a certain period of time. In summary, this thesis presents methods allowing efficient utilization of data generated from standard fire performance tests, such as ASTM E119, for structural mass timber panels, which were developed to determine the char rate of CLT panels, char rate variability across depths, and timing of CFO determined through analysis of furnace temperature data collected from empirical full-scale tests. The char rates in unprotected CLT layups fabricated by Ponderosa lumber pine subjected to the standard ASTM E119 temperature curve are successfully determined. The variability and spatial distribution of the residual char depth in full-scale unprotected CLT assemblies subjected to standard fire tests in horizontal and vertical positions, and the effect on the local charring rates determined from the residual unburned panel depth are determined. The hypothesis that the char fall-off events can be detected with reasonable accuracy based on micro-perturbation analysis of furnace temperature signals is verified.
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Rights Statement
Publisher
Peer Reviewed
Language

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Hong_Seunghyun_2022.pdf 2022-08-01 Public Download