Printer Friendly

New refractory test duplicates aluminum furnace conditions.

New Refractory Test Duplicates Aluminum Furnace Conditions

Aluminum producers and casters have long been dissatisfied with the standard benchmark tests used to evaluate refractory materials designed for their use. In a furnace environment, refractories face conditions far more severe than traditional cup tests and suspended immersion tests can provide.

Nearly any furnace engineer can tell horror stories about materials that earned excellent test ratings, but failed miserably in the field. As a result, many engineers have stayed with long-established refractories, and passed up the opportunity to discover truly superior materials.

Cup tests and suspended immersion tests do not duplicate temperatures, metal pressures and alloy chemistries of the furnace environment. However, a new test has been developed that makes it practical to refractories under actual operating conditions. Called the aluminum resistance test (ART), this procedure allows furnace engineers to proceed with greater confidence in newly developed, high-resistance refractories.


Historically, cup and immersion tests have provided adequate data to discriminate between good products and-poor performers. Researchers could determine corrosion and aluminum penetration by visual analysis of samples subjected to these tests, and, for instance, could readily discriminate the performance advantages of higher cost alumina refractories over the conventional alumina-silica types.

However, a gap remained between test conditions and the severity of the operating environment, and this gap made further advances difficult.

In the standard cup test, a small container composed of the refractory material being tested holds 50 grams of 7075 aluminum alloy at 1500F for a 72-hour period. Of course, this small quantity of aluminum cannot come close to duplicating the metal pressures experienced in the furnace, and pressure influences rate of penetration.

In addition, the 50-gram sample cannot maintain alloy chemistry over the course of the test since certain alloying elements, such as magnesium, readily volatize from so small a quantity of molten metal. This loss of magnesium significantly affects refractory corrosion results and seriously impairs the test.

Attempts to raise the temperature above 1500F only increase the rate of loss. As a result, more realistic simulations of furnace heat cause less realistic estimates of refractory corrosion.

The other standard test, involving suspended immersion of a test refractory in molten aluminum, is primarily designed to check thermal shock characteristics. The refractory samples are suspended for 12 hr in 30-50 lb of 7075 alloy at 1500F, followed by 1000F heated air. This process is repeated seven times.

Here, one encounters the same problems with inadequate temperature that are experienced in the cup test. In addition, because the sample is suspended, not submerged, the process does not test for the effect of metal pressure. Finally, the limited continuous exposure to molten aluminum reduces the extent of penetration over the course of the test.

In 1983, Alcoa developed a severe test employing a furnace to evaluate refractories under actual operating conditions. While this offered the opportunity for discriminating metal resistance in real-world conditions, the test procedure was generally considered to be beyond the economic capabilities of individual refractory companies. As a result, the test relied on the willingness of a major aluminum producer to fund the program.

Each test required the complete construction of a production-scale aluminum melter with 2000 lb of molten metal capacity, and the ability to contain more than 50 samples. This was beyond the scale of most refractory companies, and was incompatible with a laboratory setting.

The time commitement included the construction process, scheduling of vendor samples, 45 days of test operation and a lengthy period for sample analysis. The total time often exceeded nine months, and, if for some reason the samples did not perform as expected, the entire procedure had to be repeated.


Deficiencies of the various alternatives spurred development of a test in which the sample would be exposed to the same environment as seen in the Alcoa test furnace, but without the expense of builing an actual furnace. The setup for this procedure is shown in Fig. 1.

Here, a silicon carbide crucible rests inside a gas-fired, 100 lb capacity "pot furnace."

For the test, the crucible holds approximately 75 lb of 7075 alloy, with an additional 3.5% magnesium, to bring the total magnesium content above 6%. The resulting composition is then approximately 6% Mg, 6% Zn, 1.6% Cu, 0.25% Fe and 0.08% Si, with the balance consisting of aluminum and minor impurities.

This modified alloy combines the high zinc level of the traditional cup test with the high magnesium typical 5XXX series alloys used for can end-stock. Since both elements contribute to refractory corrosion, this test is more aggressive than using the high zinc, but lower magnesium, unmodified 7075 alloy.

Test samples are cast or cut to standard dimensions (1.5 x 1.5 x 4.5 in.) and prefired to the desired temperature. Before placing the prefired samples in the test pot, test operators heat them to 1000F to reduce thermal shock and prevent steam explosions.

The quantity of samples in the test is limited to ten to maintain a constant relationship of metal to refractory, equivalent to that encountered previously in the Alcoa test furnace. This relationship works out to approximately 0.25 lb of metal per square inch of exposed refractory surface.

In this way, the ART assures realistic metal pressure conditions. However, for the ART to duplicate the furnace environment, it must also achieve realistic temperatures while maintaining alloy chemistry. To accomplish the first part of this goal, the ART brings metal temperature to 1500F, then rapidly increases it to 1700F and holds that temperature for the seven-day duration of the test.

To maintin the severe alloy chemistry, the ART offers two controls. First, it starts with a relatively large quantity of alloy, compared to the traditional benchmark tests. Second, the metal is stirred daily to redistribute alloying elements, and the metal chemistry is analyzed to determine change in composition. Test operators then add elements as needed to bring the critical zinc and magnesium contents back to original values. In addition, the operators replace alloy skim or oxide buildup with fresh alloy.

Another issue to be settled in designing the ART was test duration. Experimenters considered three days, seven days and 21 days as alternative test periods. Seven days produced substantially more reaction than three days, but 21 days gave identical results to the week-long test. As a result, the seven-day period was selected.

Finally, at the end of the immersion period, one must evaluate the condition of the refractory samples. To assure more thorough analysis of reaction to the alloy, the ART calls for both visual and microscopic inspection. Test operators allow the samples to cool, then section them lengthwise, polish the metal-refractory interface, inspect refractory conditions and evaluate the samples on the following scale:

* POOR: metal reacted thoroughly;

* FAIR: incomplete but severe metal reaction;

* GOOD: edge reaction only;

* EXCELLENT: no reaction as verified microscopically.

Because of the severity of the ART, many samples dissolve completely and receive no rating. To have a sample survive this test and get any rating at all is commendable.

To verify that the ART produced equivalent results to an actual furnace test, the test designers compared refractory samples from several Alcoa test furnaces to ART samples. In all cases, the visual and microscopic results were identical.

Test Results

Figures 2-4 show sample ART results, in most cases with the accompanying results for the 72-hour cup test. Table 1 summarizes the cup test and ART results for all samples used in developing the ART.

It is important to realize that all these products are materials currently in use. The bauxite-based, phosphate-bonded brick and plastic are typical of products used in the field since the early 1970s and which continue to be used in metal operations today.

The other materials are state-of-the-art developments employing a variety of nonwetting technologies. Among these, the most notable point is that the higher alumina products, although finely penetrated with metal, retain their integrity better than the lower alumina materials. The majority of the samples have reacted extensively, and would, in all likelihood, eventually support the attachment of corundum growht.


Traditional benchmark refractory tests have not matched the severity of the furnace environment. Neither the 72-hour cup test nor the suspended-immersion test achieve the temperatures, metal pressures or alloy chemistries found in an aluminum furnace, and this disparity had led to costly failures of refractory materials that performed well under test conditions.

The ideal test environment is an actual furnace, but such a test is both expensive and time consuming. The ART allows refractory manufacturers to duplicate furnace conditions using equipment that is either already in their possession or readily available to them. By providing a more severe test environment, the ART distinguishes reaction resistance more precisely than previous tests, so that refractory vendors and users may proceed with more confidence in the direction of new metal-resistance tecnology.

The ART, however, does not test for certain conditions, most notably the metal-line environment and the presence of salt fluxes. In the future, tests similar to the ART may allow researchers to evaluate refractory performance under these conditions.

It is essential that a new generation of products be developed in accordance with the toughest criteria demanded by their applications. These criteria must be met before products are applied in the field, where the risk of failure affects not only the credibility of the vendor, but the safety and profitability of the end user.

As one might anticipate, the experimenters did not stop with developing this test, but have utilized the procedure to distinguish a direction for new metal-resisting technology. The results for Beta 1.0, 2.0 and 3.0 indicate what that direction might be.

As a result of its ART performance, Beta 1.0 has been put to work in furnace applications, where it continues to make "Excellent" ratings after a year in the field. Beta 2.0 has done equally well, though with less field experience to date, and field results are not yet available for Beta 3.0.

What Next?

Of course, even the ART does not test for all conditions that a refractory may be subjected to in application. For instance, the metal line represents a highly severe environment, unlike anything in the test. Here, normal temperatures range up to 2600F, with hot excursions above 3000F at points where the magnesium has thermited in the skim layer.

Also, the ART does not evaluate for the effect of a salt cover on refractory materials. Tests based on the ART that do not allow for these extreme conditions are currently under development.
COPYRIGHT 1990 American Foundry Society, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dulberg, Jed
Publication:Modern Casting
Date:Jan 1, 1990
Previous Article:Bager Alloys modernizes to provide full service.
Next Article:CAD/CAM spells success for Michigan patternmaker.

Related Articles
Product showcase.
Madison-Kipp "molds" an innovative ceramic fiber solution.
Metal saturation and finning problem can be avoided: controlling metal saturation is a function of knowing refractory porosity, permeability.
Controlling refractory erosion in induction furnaces: the linkage between equipment, alloy system and slag environment affects refractory...
Optimizing induction furnace refractories.
A troubleshooting guide to silica dry ram refractories.
Tips for maximizing crucible life in your aluminum foundry.
Casting answers & advice.
Improve profitability by reducing aluminum melting costs.
A melt performance comparison: stack melter vs. reverberatory furnace.

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