Printer Friendly

How to measure really, really hot stuff.

How to Measure Really, Really Hot Stuff If you want to measure the temperature of something that's at 1000 C or less, you pretty much rely on conventional methods, such as standard thermocouples or optical pyrometers. Unfortunately, these methods start developing problems as you go higher than 1000 C.

Conventional iron-constantan or chromel alumel thermocouples, for example, begin to lose accuracy and reliability. Some materials, such as aluminum oxide, yttria, or magnesium oxide, that are used to electrically insulate thermocouples start conducting at temperatures greater than 1000 C.

And the emissivity of many materials, which is needed to make conventional pyrometric measurements, varies with wavelength and requires complex correction factors at high temperatures, and sometimes the emissivities aren't even known.

Many industrial areas can benefit from improvements in high-temperature sensors, including steel processing and automotive engine development. The driving force for research in new sensors, however, has been in jet and rocket enging development programs.

A variety of innovative high-temperature sensors have resulted, including thin-film thermocouples, multiwavelength pyrometers, phosphor thermographics, Johnson-noise systems, Rayleigh-scattering diagnostics, and laser-induced fluorescene systems.

Researchers at NASA's Lewis Research Center, Cleveland, have applied Pt/Pt-Rh thin-film thermocouples to gas turbine components for several years. These sputter-deposited devices, about 2-[mu]m thick, are used because they are nonintrusive to the turbine's gas stream and can tolerate repeated temperature cycles to about 1100 C.

"We'd like an outside firm to supply these sensors for whoever needs them," says William Niederding, deputy chief of the Instrumentation and Controls Div., LRC." But each sensor is so highly tailored and specialized that it's difficult to find someone willing to make them.

"In the space shuttle's main engine, for example, the blade material of the high-pressure turbo pump was changed to obtain longer life," he says. "To apply a thin-film thermocouple to this part, we had to come up with a whole new sequence of coatings (thermal barrier, insulator, and overcoat), so it turns out to be as much a materials program as it is a sensor program."

Gas turbines currently being developed with ceramic turbine blades are expected to see temperatures between 1200 and 1700 C, which precludes the use of conventional Pt/Pt-Rh thermocouples because of rhodium's low oxidation resistance and platinum's low melting point (1772 C).

For this case, a thin-film resistance thermometer consisting of 2- to 4-[mu]m-thick depositions of [MoSi.sub.2] has been developed by R.F. Bunshah, professor of materials science at Univ. of California, Los Angeles.

The high melting point Mo[Si.sub.2] develops an adherent slow-growing [SiO.sub.2] protective layer in an oxidizing environment, according to Bunshah.

Basic criteria for fabricating these thin-film devices are: they must adhere to the high speed blade; they must have a linear, repeatable resistivity as a function of temperature; and they must have negligible drift in the measured values.

Along with the pyrometric emisivity problem mentioned before, external radiation sources used to heat ceramic materials can reflect heat off of your specimen and into a pyrometric detector, thus erroneously affecting the output reading of the pyrometer.

"We've addressed these two problems with a mutlwavelength pyrometry system," says Dan Williams, chief of the research sensor technology branch at LRC.

"Measuring radiation over a broad spectrum eliminates the reflected radiation problem," he says. "And by measuring long wavelengths, eigth to 12 [mu], where the emissivity of ceramics is relatively constant, we solve the emissivity problem."

Researchers at Oak Ridge (TN) National Laboratory (ORNL) also have developed a system that isolates the background radiation problem in conventional pyrometric methods.

"We put phosphors on different surfaces, stimulate them with ultraviolet radiation, and then monitor temperature remotely using optical techniques," says Michael Cates, a researcher in the Applied Technology Div. at ORNL. "The emissions are very narrow band, and you can use filtering techniques to isolate the emissions from the rest of the blackbody spectrum."

A family of these phosphors have been developed that can be used from liquid helium temperatures (4 K) to more than 1700 C (see accompanying table).

Temperatures are indicated by measuring the decay time of the stimulated phosphor or by measuring the absolute strength of the emitted signal.

"If you stimulate the phosphor with a narrow pulse from a high-intensity laser, where the pulse is only a few nsec wide, the phosphor emissions will decay in a few msec or [mu]sec," says Cates. "This time discrimination often can reduce blackbody background effects by a factor of [10.sup.6].

"The technical challenges boil down to finding the phosphors (for a particular temperature range), calibrating them, and then learning how to put them on surfaces so they stay there during the time of interest," he continues. "They must be intimately associated with the sample surface, so their temperatures correspond directly to the surface temperature."

The phosphors can be applied with plasma sprays, electron-beam deposition, or UV-transparent binders. "Sometimes you even can just rub this stuff onto a roughend surface and it will work," says Cates.

Many gas turbine manufacturers already are involved with ORNL's phosphor development. Other applications involve development of automobile engines, where piston temperatures in conventional pyrometry systems often are obscured by swirling 2000 C gases in the combustion chamber.

ORNL researchers also have applied "puffs" of phosphors to steel products in processing mills so the surface temperatures can be monitored during different stages.

ORNL researchers also are working on Johnson-noise thermometers for measuring high tempertures. A Johnson-noise system measures the small electrical noise produced in a resistor that is solely a function of its temperature, not of a current running through it.

An optical equivalent to the electrical Johnson-noise effect also is being investigated. "The optical effect is probably minute to the point of 12 orders of magnitude lower than what we are now able to detect," says Robert Shepard, head of the Instrument and Control Div. at ORNL.

This system has potential for monitoring and controlling the growth of innovative space-processed materials, such as niobium glass, where the emissivity of the grown material would be totally unknown and unmeasurable.

Rocket and jet engine development, such as that for the National Aerospace Space Plane's (NASP's) scramjets, is driving development of still other temperature measuring systems.

Researchers at LRC have developed a Rayleigh-scattering diagnostic technique] that they use to measure the characteristics of rocket exhaust plumes (see accompanying photo).

"This technique measures the gas velocity, temperature, and species concentration all at the same time," says Daniel Lesco, chief of LRC's Optical Measurement Systems Branch.

The spectrum of argon-ion or doubled Nd:YAG laser light scattered by gas molecules in the exhaust plume is measured by a scanning Fabry-Perot interferometer. Gas temperature is determined from special width measurements, while gas density is determined from the total scattered power, and velocity is measured as a Doppler shift.

Another system being developed by researchers at Physical Sciences, Andover, MA, determines the high-temperature distributions in the NASP's propulsion system.

This system combines laser-induced fluorescence with a Raman system. It is in the early stages of development and is very complex. More than $500,000 in excimer and Nd:YAG lasers and intensified solid-state cameras are used in it.
COPYRIGHT 1991 Advantage Business Media
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:measuring temperatures over 1000 C
Author:Studt, Tim
Publication:R & D
Date:Apr 1, 1991
Previous Article:Conducting polymers could simplify circuit fabrication.
Next Article:Manage your data better with a LIMS system.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters