High sensitivity multi - specimen device for rapid measurement of linear thermal expansion


P. Gaal1

1Anter Corporation, Pittsburgh PA, USA

Keywords: measuring device
property: thermal expansion
material: very low expansion materials

This paper describes the evolution and construction of an instrument for the purpose of measuring the linear thermal expansion of very low expansion materials in a rapid fashion, and over a limited temperature range.   Certain industrial applications (such as aluminum refining, steel production, etc.) require highly precise knowledge of the thermal expansion characteristics of carbon anode and cathode stock and manufactured ingots.  Deviations from design parameters usually carry substantial cost increases in use, and as a consequence, frequent and large number of tests are usually performed in design, engineering and production control phases of use.  For these applications, it is desirous to have as low expansion materials as practicable.  From the testing aspect, low expansion when measured over a limited temperature range, require a combination of high accuracy and high resolution instrumentation, and comparatively long specimens, beyond what conventional dilatometers can provide.  Further complicating matters is the need to process a large number of specimens, without employing large multiples of such instruments.

The device that was developed to satisfy all of these somewhat diverging requirements is able to house 36 specimens on an indexed structure to be tested sequentially, each suspended in a reflective cavity of its own.  The specimens are fixed in length, but are variable in cross section.  The indexed structure brings each cavity into the testing position, where the specimen is heated at a high rate for a programmed nominal 100 oC or 200 oC temperature excursion using infrared heaters.  The expansion is measured with an interferometer.  A test is completed in 60 to 180 seconds, depending on the specimen’s size and the temperature range selected.

Specimen temperature is measured at an internal location and on the surface.  The temperature distribution within the specimen was modeled to establish correspondence with these two discrete measurement locations.  The accuracy of the model was verified using several highly instrumented specimens in a static mode.

A full description of the device, its operation, and representative data obtained in sequential test mode are presented.


     Official Sponsors

uni Anter_logo Tziolas_logo_ linseis netzsch