Printer Friendly

Mandating IHTC through casting simulation.

Inside This Story:

* Interfacial heat transfer coefficients play a pivotal role in predicting permanent mold casting cooling rates.

* Casting modeling software assists in these predictions by simulating fluidity and thermal activity.

* This article details how comparisons of actual castings with simulation software can simplify IHTC calculations of complicated designs.

For the simulation of permanent mold casting, the interfacial heat transfer coefficient (IHTC) is the most important factor in determining the cooling rate. High quality castings can be achieved through directional solidification, thus emphasizing the role of the IHTC to predict freezing patterns. The solidification analysis during filling is crucial for thin sections to avoid premature freezing of the metal and the resultant defects.

Of the many investigations on mold-metal interface heat transfer, IHTC values have fallen in a wide range of 500-16,000 W/[m.sup.2]K, with many of these studies focusing on simple castings. Although these values have shed some light on the behavior of heat transfer at the mold-metal interface, they can't represent the whole metalcasting industry, which includes a lot of complicated castings.

Recent investigations focused on determining the IHTC value of an intricate commercial hubcap casting. A number of experimental casting trials were conducted with temperature measurements in both the casting and the mold, and these trials were compared with the results of casting process modeling simulation trials. An evaluation of the IHTC value then was made based on the comparisons.

Examining Real Castings

Multiple trial experiments creating real castings were conducted on a 30-1b (13.6-kg) A356 permanent mold hubcap component. Temperatures were monitored at seven locations, using two Nanmac thermocouples at the mold interface and five K-type thermocouples in the mold (Fig. 1). A coating was applied to the mold (but removed from the thermocouple surfaces), and the molten metal temperature was monitored in the casting ladle immediately before pouring. Pouring and mold temperatures averaged 1,342.4F (728C) and 500F (260C) and were controlled for each experiment. The average pouring time was 9 sec. The hubcap was cast a minimum of 15 cycles for each trial. The trials also had average mold dwell times of 120 sec., and thermal data was collected every 0.5 sec.

[FIGURE 1 OMITTED]

In one of the trials, in which 25 cycles were conducted, the cycle time throughout the trial was very consistent, and good casting quality was achieved (Fig. 2). It can be seen that except for cycle 17, which had a longer cycle time due to metal leaking out the back of the mold, the cycle time throughout the trial was consistent. The temperatures represent the average of two sets of data in the mold cavity.

[FIGURE 2 OMITTED]

The hubcap castings were cast using different cycle times in each trial to determine which heat transfer rates do not affect the casting quality and how cycle times can be reduced. It was found the cycle time as well as mold temperature had an impact on the quality (Table 1). An average cycle time of 227 sec., including a dwell time of 120 sec., provided the highest casting quality. "Good" in Table 1 means no defects were observed related to filling and solidification, whereas "Poor" means either an incomplete fill or distortion was observed in the casting.

Because of the complexity of the casting and the concerns for filling the hubcap's thin center section, initial experiments were conducted to determine the optimum mold temperature so all parts could be filled. In the later trials, a constant dwell time of 120 sec. was maintained regardless of the mold temperature. It took 16 cycles for the mold to reach a steady state ideal for proper casting procedures (Fig. 3), and the starting mold temperature was 700F (371C). It was believed that metal temperature, either too high or too low, would not guarantee the casting quality.

[FIGURE 3 OMITTED]

After some preliminary experiments, the metal temperature was set at 1,350F (732C) for all cycles throughout the trial. The maximum casting temperature measured reached a steady state after 15 cycles (Fig. 4), similar to the mold temperature changes in the same trial.

[FIGURE 4 OMITTED]

Because mold coatings always are used in permanent mold casting of aluminum alloys and the formation of an air gap is inevitable during solidification due to the shrinkage tendencies of cast aluminum, no set of data exists that accurately describes the heat transfer process from casting to mold. Here, the mold coating and air gap influence on the rate of heat transfer at the interface was reflected in the value of the IHTC, which was determined by comparing the modeling with experimental trials.

Trials with Modeling

Commercial casting process modeling software was used in the analysis of filling and solidification for a number of trials. In order to increase speed without sacrificing accuracy, different approaches to meshment were used for the mold, the casting and the thin sections, such as the inlet. The resulting model contained more than 6 million control volumes, of which 537,240 are in the casting.

The casting process modeling trials conducted thermal simulations through at least 15 cycles (the experimental trial number to reach mold temperature equilibrium) in each run so temperature profiles obtained could be compared with experimentally measured profiles.

In the first few simulation runs, the initial temperature used was 1,342.4F (728C), with the mold temperature at 500F (260C). The heat transfer coefficient used was from the software supplier's database, the profile of which is shown in Fig. 5. A total of 25 cycles were modeled in the first version without filling analyses. In the second run, five cycles were carried out, and the filling analysis was conducted at the fifth cycle.

[FIGURE 5 OMITTED]

The casting temperatures obtained in the modeling always were found to be higher by 50F (10C) than those in the experimental trials (Fig. 6). One of the reasons for this discrepancy was believed to be the improper selection of initial metal and mold temperature in the simulation settings. According to the software supplier, the initial temperature should be that of the molten metal that first comes into the inlet, which is lower than the pouring temperature measured.

[FIGURE 6 OMITTED]

The issue of mold temperature is more complex because it depends on whether a filling analysis is conducted in the "warm-up" cycles before the mold reaches a steady state. When "do filling" (the point at which all metal in the mold will be the same temperature) is switched off, the subsequent solidification simulation will start with a uniform temperature. Otherwise, it will start with the temperature profile obtained from the filling analysis, which provides a non-uniform temperature distribution throughout the mold. In this investigation, the results showed solidification simulation could better describe the heat transfer behavior at the mold-metal interface when the filling analysis was included.

Another possibility for the discrepancy was the IHTC profile was not a good description of what really happened at the interface. After some trials, a good IHTC for the computer simulations was determined by the 35th iteration of the simulation modeling (Figs 7 and 8), and the initial metal and mold temperatures were adjusted, as well. The good IHTC was determined by matching the modeling to the actual casting curves, and it portrayed a good estimate of the overall heat transfer behavior at the metal-mold interface. Therefore, such an IHTC was considered to be accurate to represent the value for the hubcap casting in the present permanent mold casting process of aluminum alloy A356.

[FIGURES 7&8 OMITTED]

Mandating IHTC

The findings in these investigations helped prove the necessity of a proper IHTC in permanent mold casting. Although a complicated casting was chosen here. it seems possible the IHTC could be estimated in a fairly easy manner. The evaluations of some of the complex issues like the mold coating thickness, the air gap forming between the casting and the mold. and casting geometry could be avoided in such a process. In addition to comparisons between the experimental and modeling cooling curves, more modeling efforts can be focused on finding appropriate simulation setups so the virtual temperature distributions can be as close to the experimental ones as possible.

Additional investigations are being carried out to examine response time issues with the thermocouple measurements and the temperature gradient within the mold. These include placing the thermocouples at different positions and determining the direction of heat transfer, but a final theory on the best practice has yet to be determined.

Good starting IHTCs can save in simulation time, and an approach of trial and error, although sometimes time-consuming, almost always can lead to a reasonably good estimate of an IHTC or set of IHTCs. Table 2 gives the simulation setups under which Figs. 6 and 8 have been obtained. The IHTC function plays a significant role in reaching a good agreement between experiment and simulation.

With the exponential increase of computing power and the availability of commercial software, the procedures used in this work can easily and relatively efficiently achieve the objective of determining suitable values for the IHTC. Further, the procedure utilized to determine the interracial heat transfer coefficient can be applied to other casting processes.

Related Article: IHTC equals rate of conduction activity.

In order to determine the heat transfer coefficient at the metal-mold interface, it is necessary to know the heat flux and the surface temperature of the media at both interfaces. The rate at which heat is transferred by conduction, q, is proportional to the temperature difference, shown as:

q = h([T.sub.cast] - [T.sub.mold])

where h is the interfacial heat transfer coefficient (IHTC) and [T.sub.cast] and [T.sub.mold] refer to casting and mold surface temperatures.

However, experimentally measuring the surface temperatures directly is not feasible. Placing thermocouples as close to the surface as possible is a common practice so the surface temperature could be determined later, using the Inverse Heat Conduction Problem (IHCP). This is not the procedure used to determine the IHTC in this investigation. Rather, a direct method is used, meaning the IHTC is evaluated by finding the best match between the experimental and simulated temperature profiles in the casting. This is becoming a more commonly used approach with the rapid advance of both computer hardware and software.

RELATED ARTICLE: Correlating heat transfer with air gap formation.

The overall interfacial heat transfer coefficient in a gravity permanent mold casting process, in which a gap forms at the metal-mold interface, can be expressed as:

[h.sub.total] = 1/ 1/[h.sub.gc] + [h.sub.gr] + 1/[h.sub.c]

where, [h.sub.gc] = [k.sub.g]/[d.sub.g] is the heat transfer coefficient (HTC)due to the conduction of the gas in the gap, [h.sub.gr] is the HTC due to the radiation across the gap and [h.sub.c] = [k.sub.c]/[d.sub.c] is the HTC due to the conduction of the coating. If the thermal conductivity of the gas and the coating material and a time-dependent record of the air gap [d.sub.g] are known, then finding a correlation between the HTC and the air gap formed upon solidification is straightforward. However, real-time monitoring of an air gap during the solidification process still is a challenging issue. The absence of reliable data on thermal conductivity values of some mold coatings and their variations with temperature further limit the application of this method.

This article was adapted from a paper (05-189) presented at CastExpo '05 in St. Louis.

For More Information

"Using Helium to Increase Heat Transfer at the Metal/ Mold Interface in Permanent Mold Casting," W. Wan and R.D. Pehlke, American Foundry Society Transactions (04-023).
Table 1. Average Cycle Time for Several Experimental Casting Trials

Trial No. Dwell Time Avg. Cycle Time Casting quality
 (sec) (sec)

 1 140 317 Poor
 2 125 295 Poor
 3 120 227 Good
 4 120 218 Good
 5 120 236 Good

Table 2. Simulation Setup Data for Three Modeling Versions

Version IHTC [T.sub.initial] [T.sub.mold] Filing
 No. Function (F) (F) Analysis

 1 IHTC-1 1,342 500 No
 2 IHTC-2 1,342 500 Yes
 35 IHTC-35 1,256 512.6 Yes

Version Cycle Definitions
 No. Cycle # Dwell Time (sec) Cycle Time (sec)

 1 5 120 230
 2 5 120 230
 35 15 120 230


David Moore is the engineering supervisor, and Kip Mohler is a senior engineer at the Hayes Lemmerz International Inc. Technical Center, Ferndale, Mich. Xianhua Wan is a former research scientist, and Robert Pehlke is a retired professor of materials science and engineering at the Univ. of Michigan, Ann Arbor, Mich.
COPYRIGHT 2005 American Foundry Society, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Pehlke, Robert
Publication:Modern Casting
Date:Aug 1, 2005
Words:2105
Previous Article:Building the foundation for green sand.
Next Article:VCT produces solutions for solid castings.
Topics:


Related Articles
Survey provides profile of casting design software use.
Can you justify the simulation investment? Ask Sivyer Steel.
1999 casting simulation software survey.
Walker Die Casting achieves success with simulation software.
Question of the month.
Keeping up with process simulation technology: this casting process modeling software roundup provides metalcasters with details on the latest...
Supercharging BMW, production with casting process modeling.
Leaping microstructure hurdles with process modeling: beyond simply identifying shrink porosity and fill problems, numerical tools can predict more...
Simulation software.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters