- THE MAGAZINE
In the 1980s, the glass manufacturing situation began to change. R&D departments became smaller and employed fewer highly trained scientists, engineers and technicians, and companies reduced the amount of time and resources dedicated to maintaining, developing or building their own instruments for measuring the physical properties of glass. As a result, the remaining technologists were faced with performing more functions with fewer and less qualified personnel.
Today, the situation is even more challenging. Continued consolidation in the glass industry has forced plants to streamline their R&D and quality control operations even further; in some cases, a single QC department might be in charge of collecting and analyzing data for several manufacturing operations. The original instruments have become outdated and, in some cases, obsolete. Yet companies have little time and personnel available to maintain their existing systems, much less develop new instruments using current technologies.
Fortunately, industry suppliers have responded by developing automated instruments that require minimal operator attention and are based on the existing ASTM procedures. By using the latest technology and software, today's glass technologists can collect and analyze data faster, easier, more accurately and more reproducibly than ever before.
*ISO procedures, which are commonly used outside the U.S., rely on the same laws of physics as ASTM procedures, so it is likely that modern instruments meet the ISO procedures or can be modified to meet them. However, the discussion in this article is limited to ASTM procedures.
Then vs. NowSix ASTM standards specifically address temperature points or glass viscosity measurement:
- ASTM C-336: Annealing & Strain Points via Weighted Fiber Elongation
- ASTM C-338: Softening Point via Fiber Elongation
- ASTM C-598: Annealing & Strain Points via Beam Bending Viscometer
- ASTM C-965: Range of Viscosities via Molten Glass Viscometer
- ASTM C1350M: Range of Viscosities via Beam Bending Viscometer
- ASTM C-1351M: Range of Viscosities via Parallel Plate Viscometer
As mentioned previously, these standards originated decades ago and were based on manual procedures that required a skilled technician to conduct the test and collect and analyze the data. For example, ASTM C-336 is a complex procedure that was originally published in 1954 and was based on papers by Littleton in 1920 and Lillie in 1931 and 1954.1,2 This test is performed on a vertical fiber that elongates under an applied weight during a cool down and requires simultaneous furnace control and data acquisition, plus several data analysis routines. The results are single temperature points rather than viscosity points, and changes in the annealing and strain point temperatures are indications of chemistry changes. These measurements are widely used production control parameters.
The original test required a manually controlled heating rate in a special furnace, while an operator observed and manually recorded the amount of fiber elongation, time and temperature at 60-second intervals. During the heating period, the operator calculated the elongation rate and changed the heating cycle to a controlled cooling cycle when a threshold elongation rate was observed. After the test was complete, the operator plotted the amount of elongation vs. time on graph paper, calculated the rate of elongation, and plotted a curve showing the log rate of elongation vs. temperature on three-cycle semi-log graph paper. The operator then drew a straight line through the best data points. From prior calibration tests, the operator selected the temperature at which the target rate of elongation occurred to identify the annealing point temperature. The operator extended the line on the graph paper and extrapolated the temperature that corresponded to a rate 0.0316 times that observed at the annealing point. This extrapolated temperature was the strain point temperature.
While the ASTM standard hasn't changed, modern technology has significantly reduced the amount of time and operator involvement required to perform the analyses. The operator simply places the sample in the instrument and pushes a button. A PID controller automatically controls the thermal cycle in the furnace (instead of a manual rheostat), while a linear variable differential transformer (LVDT) system monitors the elongation of the fiber (instead of a manual telemicroscope and scale), and a thermocouple connected to a digital device monitors the temperature (instead of a manual potentiometer with an ice bath). The temperature and elongation signals are sent to a personal computer and are stored in a computer file (instead of manually recording readings on a sheet of paper). During heat up, the computer software calculates the rate of fiber elongation at 1- or 2-second intervals-a significant improvement over the 60-second-interval calculations done previously. When the target turnover rate of elongation is reached, the software sends a signal to the controller to begin cooling the furnace and continues storing the elongation and temperature signals.
At the conclusion of the test, the software automatically performs the linear regression on the log rate of elongation vs. temperature data, determines the annealing point temperature from the previously stored data table of calibrated annealing point rates, and determines the strain point data from the extrapolated linear regression curve. Data collection is powerful, fast, accurate and repeatable, and the potential for human error and bias during the manual curve fitting and extrapolation have been eliminated.
Analogous improvements in speed, accuracy and reproducibility with minimal operator attention are possible for the other ASTM procedures using LVDTs (ASTM C-598, C-1350M and C-1351M), lasers (ASTM C-338) and torque devices (ASTM C-965).
Maximized ResultsThe viscosity testing procedures of today are still based on the work of the scientists and technologists of the 1920s to the 1980s and rely on the phenomena they described. Fortunately, glass manufacturers are no longer restricted to the time-consuming manual tests and data acquisition methods used decades ago to obtain viscosity and temperature point information. Modern temperature control products (PID controllers) and analog output sensors (LVDTs, torque sensors and lasers) have improved the speed, accuracy and overall quality of the data generated using standard ASTM test procedures. But the most powerful enhancement to these procedures has been the incorporation of low-cost computers and powerful software, which have freed the operator from performing the measurements, eliminated human bias and error, and made collecting and processing data faster, easier, more accurate and more reproducible.
By incorporating modern instrumentation into their R&D and QC operations, glass manufacturers can minimize the number of skilled technicians required to perform viscosity testing procedures while maximizing their results. Likewise, glass technologists can spend their time resolving glass problems instead of trying to duplicate instruments that are commercially available.
For more information about glass viscosity measurement, contact Reed Slevin at the Orton Ceramic Foundation, 6991 S. Old 3C Hwy., Westerville, OH 43082-9026; (614) 895-2663, ext. 31; fax (614) 895-5610; e-mail firstname.lastname@example.org; or visit http://www.ortonceramic.com.
References1. Littleton, J. T., and Roberts, E. H., "A Method for Determining the Annealing Temperature of Glass," Journal of the Optical Society of America, Vol. 4, 1920, p. 224.
2. Lillie, H. R., "Viscosity of Glass Between the Strain Point and Melting Temperature," Journal of the American Ceramic Society, Vol. 14, 1931, p. 502; "Re-Evaluation of Glass Viscosities at Annealing and Strain Points," Journal of the American Ceramic Society, Vol. 37, 1954, p. 111.