|
|
|
Measuring Inclusions in Aluminum Melts
10-31-02
Non-metallic inclusions are an unwanted by-product of molten aluminum processing, and significant effort is made to remove them by filtration techniques. Knowing the effectiveness of these techniques, on-line and in real time, is a goal for inclusion measurement in both the cast house and foundry environments. What are the options?
First, let's take a quick look at the need. Cost-effectively improving the metal quality of aluminum is an ongoing pressure on commercial organizations, whether the companies are producing ingots for subsequent processing or shaped castings in the foundry. But what is metal quality? Essentially metal quality is the degree to which an aluminum alloy is free from alkali metal contaminants, non-metallic inclusions, and dissolved hydrogen. The extent to which an aluminum melt contains these undesirable features depends on the source of the metal (smelter vs. remelt) and processing, while the degree to which they need to be reduced depends on the final product performance requirements. A complete discussion of this complex situation is beyond the scope of this short article, so let's focus on the area of non-metallic inclusions.
Inclusions in aluminum can take a range of forms including films and particles
of oxides, carbides, borides, nitrides, intermetallics, and other compounds. This
was reviewed by researchers at Worcester Polytechnic Institute's Metal Processing
Institute. This reference suggests that the number of inclusions in aluminum melts
can range between hundreds and thousands per kilogram and their size from a few
microns to up to half a millimeter. They can affect both the mechanical and surface
properties of the alloy, and substantial effort has been devoted to efficient
and effective methods for the removal of inclusions such as deep bed filters,
ceramic foam filters, and other methods.
Measurement of inclusions in the aluminum melt is vital both from an R&D perspective and, perhaps more importantly, for process monitoring and control in production. A wide variety of methods have been developed and in some cases commercially implemented, and each has its own plusses and minuses. As usual, these are known by a variety of trade names and acronyms, and the following brief summary of these techniques is intended to help identify and better understand the options available.
- LiMCA and LiMCA II (Liquid Metal Cleanliness Analyzer): an in-line device utilizing electrical resistivity measurement method for detecting inclusions in the 20-300 micron range. Measures both amount and size distribution of inclusions. Also detects gas bubbles from degassing as particles, which complicates analysis. Primarily used in large cast houses due to need for continuous access to molten metal in the transfer troughs and relatively higher instrument cost. Excellent for small inclusion measurement.
- PoDFA and PoDFA-f (Porous Disc Filtration Analysis): off-line technique in which a fixed volume of metal is forced through a filter, concentrating the inclusions on the surface of the filter. Subsequent metallographic evaluation is needed to measure the amount and type of inclusions present. While inexpensive to collect samples, the time and expertise required to acquire the measurements is a negative.
- LAIS (Liquid Aluminum Inclusion Sampler): similar to PoDFA except that the sample is obtained in the melt, after preheating of the sampling device, using vacuum pressure. Metallographic analysis of resulting sample is required.
- Prefil-Footprinter: A real time measurement using pressure filtration of a known quantity of metal and measurement of the weight gain of the filter using a load cell. By comparing the results with benchmark levels for the particular alloy and process, relative cleanliness can be assessed. While no specific information on the inclusions is obtained, a sample for subsequent metallographic analysis is available.
- Acoustic/ultrasonic detection methods: These methods, such as the 4M (Mansfield Molten Metal Monitor) technique, use the pulse-echo technique in the melt. Found to be suitable only for large inclusions.
- Electromagnetic separation coupled with image analysis: This technique, developed and patented by the group at the Metal Processing Institute, utilizes Lorentz force to push the inclusions to the melt surface where they are imaged by an optical image analysis system. Under development.
- Fracture test methods, such as the K-Mold test, developed by Kitaoka of Nippon Light Metal Ltd.: A foundry floor-level test in which the melt is cast into a mold containing notches and then bent to expose a fracture surface. The visual observation of inclusions on the fracture is used to determine a K-value for the melt and compared to a preset standard. Suitable for large inclusions and inclusion clusters.
Obviously, the choice of a particular technique depends greatly on the specific product, casting environment (e.g. ingot cast house vs. foundry), and production level. As a case study, Dawid Smith of Cast Doctor, Inc. (www.castdoctor.com), provided the following evaluation table for some of the inclusion detection methods that might be considered for use in a small foundry not employing metal transfer troughs.
| Characteristic |
Importance
(Weight) |
PoDFA-f |
LAIS |
Prefil |
LiMCA |
| rating |
product |
rating |
product |
rating |
product |
rating |
product |
| Cost of Equipment |
4 |
License |
2 |
8 |
4 |
16 |
3 |
12 |
1 |
4 |
| Outside |
5 |
20 |
| Cost of consumables |
3 |
Self |
3 |
9 |
2 |
6 |
5 |
15 |
3 |
9 |
| Outside |
2 |
6 |
1 |
3 |
2 |
6 |
| Amount of labor involved |
5 |
Self |
2 |
10 |
1 |
5 |
5 |
25 |
2 |
10 |
| Outside |
4 |
20 |
3 |
15 |
2 |
10 |
Accuracy or Repeatability
of results |
3 |
3 |
9 |
3 |
9 |
2 |
6 |
5 |
15 |
| 4 |
12 |
| "Immediacy" of results |
5 |
2 |
10 |
2 |
10 |
4 |
20 |
5 |
25 |
| 3 |
15 |
Distinguishing
inclusion types |
3 |
Graph |
5 |
15 |
5 |
15 |
1 |
3 |
1 |
3 |
| Both |
5 |
15 |
| Ease of access |
4 |
|
5 |
20 |
3 |
12 |
5 |
20 |
1 |
4 |
| TOTAL: |
Podfa |
81 or 100 |
LAIS |
73 or 79 |
Prefil |
101 or 90 |
LiMCA |
70 |
For this situation, the method with the highest score is the Prefil-Footprinter, due primarily to the ability to rapidly obtain results with moderate equipment cost and labor requirement. Note that the ratings provided depend to some degree on whether the analysis work is done in house or on an outside contract basis. The equipment suppliers as well as consultants and test laboratories provide the latter service.
(PoDFA-f, Prefil and LiMCA are trademarks of ABB Bomem, www.abb.com)
Article courtesy of Secat, Inc. - Research Resource for the Aluminum Industry
www.secat.net
|
|
|
