DC ProcessTechnologies

Turn Research into Action Business Solutions Based on NADCA Research
PROCESS TECHNOLOGIES
2008
NORTH AMERICAN DIE CASTING ASSOCIATION
Leak Testing for Die Castings

J. Brevick & C. Adams, Ohio State University
Project Objectives: The topic of leak testing can be a confusing one for the individual whose specialty is in another eld such as the die casting practitioner. The questions of How do I measure a specied leakage rate? and How is a maximum allowable leakage rate assured in a production setting? lead to a myriad of other questions for which the answers can be ambiguous or difcult to obtain. One objective of this project was to review standard leak testing technology as applied to die castings to provide the industry with a place to begin for understanding leak testing for die castings. The determination of maximum allowable leakage rates for die castings is based on functional requirements and is made by the designer or customer of the die casting. In many situations casting designers and customers are uncertain about which method of leak testing is appropriate for their die casting applications. More than one leak test method is sometimes even used for the same casting in different production facilities. Leak testing in production is usually an accelerated test using a uid (gas) other than that (liquid) used in service. Castings may demonstrate measurable leakage in an accelerated test, but not leak in service. Ideally, the level of leakage acceptable in the accelerated production test should have a known correlation to the leak rate in service. This correlation is typically done via empirical testing of the casting under conditions that the casting encounters in its service life. Another objective of this project was to generate a mathematical model which would directly correlate the gaseous ow or leakage through a die casting during accelerated production testing with the actual uid leakage which would occur during its service life. Approach: The literature on leak testing was explored and the most common methods for leak testing die castings were reviewed. A report was subsequently written, which documents many of the technical aspects of leak testing of die castings and provides empirical data used in industry to correlate accelerated leak testing with actual leaks in the eld. Specically, the following technical areas relevant to leak testing were summarized: Physical behavior of gases. o Physics of gases. o Fluid dynamics of gases. Gaseous ow through leaks. Leak testing technology as applied to die castings. Methods of leak testing die castings.
Results: The project resulted in a comprehensive summary of the technologies relevant to leak testing of die castings, including the physical behavior of gases, gaseous ow through leaks, and leak testing technology. Additionally, a description of and a procedure for the most common methods for leak testing a die casting were presented. The methods addressed were: Pressure Decay and Flow Rate Testing, Bubble Emission Leak Testing, Mass Spectrometer Leak Testing, and Halogen Tracer Gas Leak Detection. These efforts were intended to provide die casting practitioners with a place to begin with assurance of leakage rates for die castings. The original proposal from NADCA for this project was to investigate the possibility of analytically correlating the ow of various liquids encountered in the application of castings (e.g. transmission uid) where leakage is a concern, to the ow of gaseous testing media (e.g. air). It was the intent of the project to provide for the die casting industry a means to escape the empirical nature of determining appropriate gaseous leakage rates to which to test a casting to assure acceptable leakage rates of liquid media in service. This objective was not accomplished. Analytically correlating liquid ow to gaseous ow would be a very difcult task. Before one could attempt such a correlation, analytical modeling of each type of ow would have to be completed. Although fairly detailed modeling of gaseous ow is detailed in the project report, application of these models to liquid ow is applicable under very limited conditions at best. To complicate issues further, a variety of operating liquids with widely varied viscosities and included additives are encountered in use with die castings, as well as a myriad of operating temperatures and pressures and varied uid leakage mechanisms. The table below does describe the maximum production test air leak rates that have been reported for some die cast products and the corresponding eld leak rates. Implementation Strategy: The results of this project, contained in the nal report, provide comprehensive discussions concerning the physics of leak testing, detailed explanations of the various types of leak testing systems, and input into how each of these technologies can be used effectively. Those who wish to better understand the leak testing requirements for die casting products or wish to adopt effective specications for leak testing components should read and understand the nal report on this project.
Casting / Fluid Contained Transmission Case / Automatic Transmission Fluid Carburetor Assembly / Gasoline Master Cylinder / Brake uid Engine Block / Water/Glycol/Oil Cooling Systems/Water Oil (various) Diesel Fuel Gasoline Air Conditioning/Refrigerant Electrical Housings/Connectors
Typical Test Pressure Range bars (psig) 0.2 to 0.4 (3 to 5) 0.4 to 0.7 (5 to 10) 1.4 to 7.0 (20 to 100) 0.3 to1.4 (4 to 20) 1 to 2 (15 to 30) 0.1 to 7 (1.5 to 100) 1 to 10 (15 to 145) 1 to 5 (15 to 75) 2 to 20 (30 to 290) 0.1 to 1 (1.5 to 15)
Typical Leakage Rate - cc/min (air) 10 to 20 1 to 5 5 to 20 10 to 30 4 to 7 6 to 15 0.1 to 15 3 to 15 5 to 15 gm/year (tracer gas method) 0.01 to 1
Monitoring Task Force: Process Technology Task Force Sponsored by: NADCA This research is featured in more detail in the following transaction Transaction: T04-023
For further information, contact:

North American Die Casting Association 847.279.0001 phone 847.279.0002 facsimile www.diecasting.org/research research@diecasting.org
North American Die Casting Association

241 Holbrook Drive Wheeling, IL 60090 Email: research@diecasting.org
www.diecasting.org/research
ENERGY AND TECHNOLOGY ASSESSMENT FOR DIE CASTING PLANTS

S. Udvardy and J. Wilkey, NADCA
Project Objectives: The purpose of this project was to: Review recent NADCA research project results to identify the new technologies that can be utilized to enhance the energy efciency of die casting plants. Develop an audit form to assist in conducting an energy and technology assessment of two die casting plants to assist the plants in improving its use of energy and in reducing costs associated with the use of energy. Approach: The initial task of the project was to identify die casting technologies that have the capability of providing energy benets to the die casting industry. Based on a review of completed and mature R&D projects, several technological results that were directly transferable to the die casting shop oor were identied. In some cases, projects had been implemented by a die casting industry partner that participated in a specic project in order to validate the laboratory test results. In these instances, the successes noted by an industry partner were captured and formed the basis for the estimated amount of energy savings improvements. The results of each project identied as having viable results were summarized on a single PowerPoint slide. The projects with successes noted by an industry partner were included as case studies on the same slide or on additional slides. The case studies were intended to provide the industry at large with tangible evidence of the true opportunities that exist for energy savings and technological advancement. PowerPoint was selected as the format for compiling the results for ease of presenting viable technologies to the plants for which the assessment was performed and to have a presentation prepared for future die casting plant assessments. The second task of the project was to establish an audit form to be used as a tool in performing an energy and technology assessment of die casting plants. Hence, subsequent to the identication and compilation of viable energy related technological results or new technologies, key questions were developed, targeted at determining whether die casting plants were aware of or were using the technologies. The questions were made specic enough to determine the awareness or use of a technology but general enough to not have too many questions. These questions were used to develop the technology related portion of the audit form and were categorized in ve topical areas Cast Materials, Die Materials, Design Aids, Sensors & Controls, and Process Technologies.
The third task of the project was to utilize the audit form and perform a plant-wide energy and technology assessment of two the plants. Initially, a copy of the audit form was sent to each facility. The form was completed by staff members of each facility and submitted back to NADCA for review. The no responses on each of the audit forms were viewed as potential opportunities for identifying energy savings improvements for each respective facility. Next, a walk-through audit was conducted at each facility and interviews were conducted with members of each staff to obtain information pertinent to the facility, including the current manufacturing processes and programs in use, as well as past and current efforts in conserving energy and controlling costs. Results: An assessment tool has been developed to assist die casters in determining where opportunities exist to apply new technologies. The application of these technologies will assist in enhancing the energy efciency of plants through improved design, process and operational efciencies. A presentation to aid in reporting/presenting ndings to plants has been established. The presentation can be easily tailored to address specic improvement opportunities sited through the use of the assessment tool. The information in the presentation has been formulated to assist in encouraging the implementation and use of new die casting technologies. The assessment tool was utilized to perform audits of two die casting facilities. Over 40 improvement opportunities were identied. The combined potential improvement in energy usage is anticipated to be a 1020% reduction in energy use for the facilities. Projected savings may be realized through a 10% increase in productivity in die casting operations, a 30-40% improvement in die life, and a 10-15% reduction in scrap. Implementation Strategy: Each die casting facility should utilize the developed technology presentation and audit form to assess the degree to which NADCA research results are being utilized. These can be utilized in several ways: 1. Secure the audit form from NADCA and conduct a self-audit of the plant, which will identify areas which offer potential energy efciency improvement. 2. Secure the presentation on the implementation of R&D results to be sure plant personnel are aware of the latest research ndings.
3. Request an on-site plant audit by technical experts from NADCA to develop a list of potential areas for improved energy efciency. 4. Once an initial audit has been performed, request a follow-up audit to quantify the savings realized and identify new opportunities resulting from new research results.
Monitoring Task Force: NADCA Staff Sponsored by: U.S. Department of Energy and NADCA


Sensors for Die Casting

C. Smith, G. Allgood & S. Viswanathan, Oak Ridge National Laboratory
Project Objectives: The objective of this project was to investigate the use of a vibration sensor for die casting process diagnostics. During the project, the effect of three different process variables on the vibration signal obtained during die casting was analyzed. Approach: Vibration sensors, or accelerometers, detect motion or acceleration, and translate mechanical vibrations into electronic signals. Typically, an accelerometer contains a piezoelectric crystal, which is congured such that vibrations cause the crystal to be compressed. The compression of the crystal causes it to produce an electronic signal that is proportional to the amplitude and frequency of the vibration. This electronic signal can then be amplied, ltered, recorded, and analyzed. For use in die casting, vibrations arising from the movement of the die casting machine and from the ow of metal are converted into electrical signals by a vibration sensor attached to the die. These signals are electronically conditioned and digitized. Mathematical algorithms analyze the timing, frequency, and amplitude of the vibrations and correlate them with process variables to form a vibration signature. This vibration signature is then compared against a reference vibration signature that is known to have resulted in an acceptable quality part. The comparison of specic signature characteristics between the test vibration signature and reference vibration signatures can be used to determine the quality of the part as well as operational characteristics of the die casting machine. For this project, 10 castings were made at each of 5 test conditions, for a total of 50 castings. The parts were produced on a 125 ton cold chamber die casting machine with 380 aluminum alloy at a nominal pouring temperature of 1280 Fahrenheit. A visual examination criterion and rating system was developed. Each of the 50 castings was examined visually and rated accordingly. Four castings from each sample were selected for detailed analysis. These 20 castings were analyzed by x-ray radioscopy to determine their internal soundness. An x-ray rating criteria was developed and utilized. During production of the test castings, vibration readings were obtained through the use of a low-cost commercially available Endevco Model 7701-100 accelerometer mounted to one side of the die block. A simple vibration signature was formed by calculating the band limited Root Mean Square of the vibration signal from the accelerometer. A more complex vibration signature was formed by calculating a vibration spectrum over a period of time that corresponds to a specic length of ram travel. The effect of different shot proles on the
vibration signature and vibration spectrum was studied by varying the intermediate and fast portions of the shot prole by approximately 50 percent. The slow portion of the shot was held constant. Results: The following results were obtained from the experimentation in this project: Each step in the die casting process, such as door closing, injection, and part extraction can be related to a characteristic feature in the vibration signature. Each shot prole can be identied by its vibration signature. In particular, increasing or decreasing the fast shot (or intermediate shot) signicantly changed the vibration signature. The frequency based segment analysis demonstrated that a frequency spectrum containing broad peaks indicates atomized metal ow, while a frequency spectrum dominated by harmonic resonance indicates laminar (or non-atomized) ow of the molten metal. The vibration signature also displays unique characteristics during the solidication of the casting. The results of the project have demonstrated that vibration signatures can identify shot prole parameters and provide insight into the ow characteristics of the metal. The identication of ow characteristics is the rst step in correlating the vibration signature with part quality. As renements are made in the analysis of the vibration signatures, it is expected that the vibration characteristics will correlate with part quality. Once these characteristics have been identied, they can be incorporated into real-time machine diagnostics for use in production to instantly differentiate between good and bad parts. Implementation Strategy: The results of this project showed the potential for using vibration sensors to monitor the die casting process. Vibration sensors have been used for many years by industrial maintenance personnel to locate eminent failures of drives, motors and bearings. This same technology could possibly be used provide immediate feedback from the die casting process as well. This project was a preliminary study on using vibration sensors to monitor the die casting process. Signicant additional work is required to correlate vibration signatures to part quality. However, for die casters already familiar with the use of vibration signatures for maintenance purposes, their application to the die casting process could provide additional useful data.
Figure 1. Location of vibration sensors.
Figure 2. Vibration signature showing events in the casting cycle.
Monitoring Task Force: Process Technology Task Force Sponsored by: U.S. Department of Energy and NADCA This research is featured in more detail in the following transaction Transaction: T02-055


Understanding the Relationship Between Filling Pattern and Part Quality in Die Casting
J. Brevick & A. Miller, Ohio State University
Project Objectives: The overall objective of this project was to investigate phenomena involved in the lling of die cavities with molten alloy in the cold chamber die casting process. It has long been recognized that the lling pattern of the molten metal entering a die cavity inuences the quality of the resulting die cast parts. However, the relationship between ll parameters and part quality is not completely understood. Fill parameters typically include the velocity of the molten alloy through the gate(s) at the start of cavity lling, molten alloy temperature, die temperature, time to ll the cavity, intensication pressure, and gate location, size and geometry. Approach: The methods available for assessing lling patterns for high pressure die casting are rather limited because the process occurs in an optically opaque steel die. The water analog technique and computer simulation are two commonly used techniques to study lling patterns. Observation of lling patterns in sand, lost foam and permanent mold casting has been done with radiography, and in this project, real-time radiographic observation of die casting lling patterns was evaluated. This method is considerably more challenging in die casting because of the short time (milliseconds) and high pressure involved. Work during the project was focused on water analog testing, computer modeling, development of real-time radiographic testing, and corresponding casting trials. Results: The water analog method involves the use of a high-speed camera to record water or low melting point alloy ow in transparent dies. The water is typically colored using dyes to make the ow patterns easily visible. Room temperature water has pertinent uid properties that are very similar to molten aluminum. One drawback of the water analog approach is that it is isothermal, no heat is lost in the system and therefore no solidication or changes in viscosity occur as they would in a real die casting operation. A variety of water analog tests were conducted using two different geometries, various gate velocities and pre-ll percentages. Pre-lling is a departure from NADCA recommendations that many die casters use successfully. Pre-lling involves partially lling the casting volume at a very low gate velocity, then quickly ramping up to high gate velocity to complete lling before the alloy solidies. The results of the water analog work demonstrated that pre-lling certainly does inuence the ll pattern observed. The complex at plate geometry showed a signicant lling pattern improvement using various percentages of die cavity pre-ll. The simple geometry showed virtually no improvement in lling pattern using pre-ll. However, pre-ll is only a
feasible option when the casting geometry is massive enough (thick wall sections) to allow longer total cavity lling times. Otherwise the molten alloy would solidify before the cavity could be lled. The objective of the computer modeling work was to conduct computer simulations of lling patterns for the simple and complex geometries using commercially available software. The computer simulations were generated to compare with the water analog simulation ll patterns and with the ll patterns from the radiographic experiments. In order to accomplish the various computer simulations, the parametric solid models for both geometries and the gating systems were constructed in Unigraphics. The models were then exported from Unigraphics as STL models and loaded into the commercial simulation software packages. The commercial software programs utilized were: CastView, MAGMASOFT, NovaCast, Flow3D, and dieCAS. Generally, the water analog and computer lling pattern simulations, using the same geometries and lling parameters, yielded results that were strikingly similar. Certainly, there are minor local differences that can be seen from one simulation to the next, so the level of precision is still an issue for discussion. From one point of view the similarity is surprising because according to current wisdom, the ow at the gate exit into the cavity is supposed to be atomized. Yet none of the computer simulation tools was capable of modeling atomized ow. However, as demonstrated in this project via comparison with actual castings, on a macroscopic scale, computer simulations of lling patterns have the ability to predict ow related defects in castings. For the radiographic testing, it was necessary to design and build an experimental apparatus suitable for realtime radiography experiments to observe die cavity lling patterns and the molten metal ow regime from the point of exiting the gate. Information about ow regime at the gate as a function of gate velocity and time during cavity lling could be very useful for die casters. Fundamental static and dynamic experiments were conducted and showed that such a technique is feasible. An electric motor was used to spin step blocks of lead and zinc through the x-ray beam at various velocities in the die casting range. The lead sample step block can clearly be observed through 2 inches of aluminum with a resolution of 0.4 milliseconds. Likewise, the zinc can be seen easily at 1.3 milliseconds and with some difculty at 0.4 milliseconds. Based on these tests, an aluminum die was chosen for use with both lead and zinc casting alloys. An experimental apparatus for conducting the
Figure 1. Water Model vs. Computer Model Results
radiographic experiments was designed and was nearing completion at the conclusion of the project. Implementation Strategy: The results of this project showed that water analog simulations and computer simulations can predict defects found in actual die castings. Die casters should utilize these tools to better understand die lling and optimize die casting parameters.
Monitoring Task Force: Process Technologies & Computer Modeling Task Forces Sponsored by: U.S. Department of Energy and NADCA This research is featured in more detail in the following transaction Transaction: T06-123


DIE CAVITY INSTRUMENTATION

C. Mobley and J. Brevick, Ohio State University
Business Benefit: This project provides the die caster with information on the use of monitoring equipment to better understand the die casting process and improve the quality of parts. The monitoring observed the relationships between casting weight, volume and density, with die temperature, mean cavity pressure during solidification and intensification stroke length. Project Objectives: The primary objective of this project was to evaluate the performance characteristics and usefulness of near-cavity temperature, liquid pressure, and gas flow rate sensors for improved monitoring and control of die casting processes. Three types of near cavity sensors were evaluated: 1. A multi-thermocouple probe used for determining the surface and near-cavity thermal history of the die. 2. A commercially available direct cavity pressure sensor for measuring the pressure history of the liquid and solidifying alloy in the die cavity. 3. A vent gas flow sensor for monitoring whether gas exits the cavity vent during cavity filling. Approach: As part of the evaluation of the near or in-cavity process sensors, die casting experiments were conducted at the Manufacturing Laboratory of the Ohio State Universitys Engineering Research Center for Net Shape Manufacturing. A beta site die casting campaign was also performed at the GM Casting Advanced Development Center (CADC), at Bedford, IN. One hundred and fifty six rear axle aluminum alloy transmission cases were die cast during the GM CADC beta site experimental campaign. Detailed shot profile, die temperature, and cavity pressure data were collected for each shot. The cycle time for each of the 156 shots in the campaign was 121 seconds and there were no operating delays or breaks during the campaign. The volumes, densities and weights of the individual castings were determined and correlated with the measured casting variables. Results: Each of the three types of sensors provided valuable information.
results from computer programs (BINORM and MagmaSoft) for the experimental die and casting cycle. The experiments with the thermal probes were short duration tests accomplished on an experimental die, so the longevity of the thermal sensors is unknown. Longer duration Beta-site tests were conducted with multiple thermocouple sensors located throughout the die, but not at or near the cavity surface. The "composite" die temperature based on these temperature sensors was found to be one of the principal process variables needed to predict the volume and weight of the die castings produced.
In-Cavity Direct Pressure Sensors: The in-cavity direct pressure sensors evaluated in this project were manufactured by Kistler Instrument Corporation (Kistler Instrument Corporation part #6175A2). The Kistler direct pressure sensors demonstrated sufficient response time capability to measure the metal pressure in its local cavity region during cavity filling and intensification. However, once a skin solidified on the surface of the pressure sensor, the reported sensor pressure decreased in comparison with the metal pressure inferred from the hydraulic records.
The direct pressure sensors provide complimentary data to the machine hydraulic record. Specifically, the hydraulic pressure sensors on die casting machines are of limited value for measuring the inertial pressure spike that occurs in the cavity at the end of filling. However, the direct pressure sensors were able to monitor the pressure during the final stages of cavity filling, the pressure spike at the end of filling, and the pressure during the onset of intensification. These data are very useful during initial trials with a new die to determine the preferred injection system parameters as part of machine set-up to produce the desired casting quality without flashing the die. Data from the156 shots at the GM CADC beta site experimental campaign were collected using the Kistler direct pressure sensors with no reliability problems with the sensors or the associated data acquisition equipment.
Thermal Probes: Thermal sensors located at the die cavity surface demonstrated sufficient response time to successfully measure gate freezing time as well as near-surface heat fluxes from molten alloy entering the cavity and from spray cooling of the die surfaces. These data can be extremely useful to die designers, in terms of validating their approach for thermal management of the die. Data gathered from the near-cavity temperature sensors were consistent with thermal analysis
Vent Gas Flow Sensors: OSU experiments indicated that thermocouples and microphone transducers are adequate to determine if a die cavity vent is open or closed during a given shot. However, the need exists to develop robust sensors and/ or systems that can be used for multiple sequential determinations. Access to the vent exhaust location is a major problem in attempting to utilize the vent gas flow sensors on dies currently in use. If vent gas sensors are to be used to monitor
the gas flow from the cavity during filling, their placement should be considered in the early design of the die vent and means provided for the sensor placement and replacement. Additional effort is required to demonstrate the robustness of the thermocouple or microphone transducer type vent flow sensors. Implementation Strategy: Die casters can use the information to correlate near-cavity sensor data with casting properties. The volume, density and weight of the trimmed castings produced during the GM CADC beta site campaign were determined using the Archimedes method. All three characteristics (volume, density and weight) were determined to five significant figures.
intensification stroke length with shot number for the GM CADC campaign is shown in Figure 2. Linear regression analysis indicated that the volumes of the GM beta site castings were also given by the relation: Vc = 281.911 0.052327 T + 14.6116 [2] where T is the composite die temperature (in degrees Fahrenheit) and is the intensification stroke length (in inches), and the casting volume is in units of cubic inches. The R2 correlation coefficient for the Equation [2] fit was 0.973.
Volume: The volumes of the aluminum alloy rear axle transmission cases die cast during the GM CADC campaign are plotted as a function of the shot number in Figure 1. It was anticipated that the volume of the casting is primarily a function of die cavity temperature and the liquid pressure in the die cavity. The casting volume is unique or specific to a given machine and die, as it depends on the dimensional stability/response of the die casting system. For the GM CADC die and machine system used in the beta site campaign, the volume of the die castings, Vc, was related to the die casting conditions by the expression:
Vc = 281.088 0.0524 T+ 0.0003862 Pmean + 6x10-8 Pmean2 [1]
Density: The measured densities of the GM CADC castings as a function of shot number are shown in Figure 3. Based on the model presented in the 1993 NADCA Transactions paper entitled "Equations for Predicting the Percent Porosity in Die Castings", it was anticipated that the density of the castings would depend primarily on the quantity of gas contained in the casting, the pressure applied during solidification, and the amount of liquid fed to the die cavity during intensification. For the GM CDCA beta site castings, the density of the individual trimmed cases was fit by the expression:
= 0.0988 - 4.0345/Pmean + 0.4470 0.19/Vc [4] All of the variables are expressed in English units; in pounds per cubic inch; Pmean in pounds per square inch; intensification stroke length () in inches, and casting volume (Vc )in cubic inches. The R2 correlation coefficient for the Equation [4] fit was 0.933. Figure 2. Intensification Stroke Length versus Shot Number For GM Beta-site Campaign
where T is the composite die temperature (in degrees Fahrenheit) at the start of a given shot and Pmean is the mean or average pressure in the liquid during solidification (in pounds per square inch, psi). The R2 correlation coefficient for the relation given in Equation [1] is 0.970. Figure 1. Casting Volume versus Shot Number for GM Beta-site Campaign
Figure 3. Casting Density versus Shot Number for GM Beta-site Campaign
The die temperature significantly influences the volume of the die-casting. This is consistent with the observation that casting volumes and weights differ significantly during startup and during or following production stoppage or delays. The volume of the casting is greater when the die is cold and decreases with increasing die temperature. The volume of the castings correlated well with the intensification stroke length or plunger travel past impact. The variation of
Weight: The weight of the GM CDCA beta site castings as a function of the shot number is shown in Figure 4.
Figure 4. Casting Weight versus Shot Number for GM Beta-site Campaign
(and/or the intensification stroke length), and the density depends on the mean pressure and intensification stoke length, the weight of the casting (for a completely filled and non-flashed cavity) is expected to depend primarily on the die temperature, mean applied pressure, and intensification stroke length. A regression analysis for the trimmed casting weight gave the best-fit relation: W = 28.303 0.00515 T 1713.5/Pmean + 1.4962 [6]
All the variables in Equation [6] are expressed in English units. The R2 correlation coefficient for Equation [6] was 0.985. The observed relations between casting weight, volume, and density with die temperature, mean cavity pressure during solidification, and intensification stroke length clearly show the value of using existing and near cavity measurements to better monitor and understand the die casting process and resultant parts. The observed dependence of casting weight with the selected process variables also suggests the monitoring and use of weight data as part of a process and product quality control procedure as the weighing of castings does not require complex equipment and/or long time measurement procedures. I
The weight of the casting is the product of its density and volume, that is: W = Vc [5]
As the volume of the casting depends on the die temperature and mean pressure applied to the liquid during solidification
Monitoring Task Force: Process Technologies Task Force Sponsored by: U.S. Department of Energy and NADCA


ASSESSMENT OF THE INFLUENCE OF SLOW TO FAST SHOT TRANSITION POINT ON CAVITY FILLING PATTERNS IN COLD CHAMBER DIE CASTING
J. Brevick, Ohio State University
Business Benefit: This project provides the die caster with information about using flow modeling to minimize porosity related scrap. This knowledge may also preclude the need for costly additional equipment, such as a vacuum system or casting impregnation equipment, to reduce the amount of air entrapped., Project Objectives: The purpose of this project was to evaluate the influence of cavity pre-fill percentage, slow shot (prefill) velocity, and transition time from slow to fast shot velocity on filling patterns and gas porosity in die castings. In cold chamber die casting, the acceptability of castings is often dependent upon the location, size distribution and total volume of contained gas porosity. These attributes of gas porosity are influenced by the approach chosen to fill the cavity with molten alloy. Specifically, metal injection parameters such as slow shot velocity, acceleration from slow to fast shot velocity, slow to fast shot transition point (amount of cavity pre-fill) and fast shot velocity can all influence the size, location and amount of gas porosity in castings. With respect to slow to fast shot transition point, the most common approach among die casters is to start the fast shot when the metal is just arriving at the gate(s). An alternative method is the practice of cavity pre-filling, where the transition to fast shot is delayed until the casting cavity is partially filled at the slow shot velocity. The practice of pre-filling has been shown to produce castings having equal or superior quality in terms of porosity and surface finish. However, the best approach for "engineering" the shot profile to obtain cavity fill patterns that minimize contained gas casting defects is not known. Approach: Two approaches were taken; computer flow modeling and physical simulation using the water analog approach. Both approaches assumed an isothermal system; no heat flow was considered in either approach. Several casting geometries, ranging from simple to complex, were selected for evaluation using these methods. The computer flow modeling was accomplished using the finite element based software Flow3D. Filling pattern analyses were conducted using experimental design approach. The factors investigated were: 1. Cavity pre-fill percentage 2. Slow shot (pre-fill) velocity 3. Transition time from slow to fast shot velocity 4. Complexity of cavity geometry.
The measured response was the gas volume (air) trapped in the casting cavity at the moment the metal filling the cavity sealed off the vents. The outcome of the designed experiment was a regression model demonstrating the influence of the factors investigated on the volume of gas entrapped in castings. The computer generated cavity fill patterns were also very useful in identifying the probable locations (distribution) of trapped gas in the castings. Interim results of the computer flow modeling work have been presented at the 1999 NADCA Technical Congress in Cleveland, OH, and at the 2000 NADCA Technical Congress. The results of the computer flow analyses were compared with results from water analog physical simulations. Water analog experiments were conducted using a Prince test stand with a VisiTrak Worldwide shot controller/monitor. Experimental dies having geometries matching the computer models were machined from transparent acrylic blocks. Injection parameters were controlled to match the computer model using the VisiTrak equipment. Cavity flow patterns were captured using a high-speed camera capable of 4,500 frames per second. Filling patterns were recorded on S-VHS tapes and reviewed in slow motion. Results: Three significant finding were discovered during the testing: 1. The cavity filling patterns predicted using computer flow models were in excellent agreement with the filling patterns observed with high speed video camera recordings of the water analog physical simulation experiments. 2. It was observed that filling patterns in die cavities were significantly different when employing cavity pre-fill. A major benefit when employing cavity pre-fill was the unilateral expansion of entering metal flow in die cavities. Compared to the fill patterns that develop when inertia dominated jet flow occurs at the gates, in some cases pre-fill helped create more uniform cavity fill pattern. In those cases, the amount of air entrapped could be significantly reduced using cavity pre-fill. 3. Major factors influencing air entrapment in die cavities turned out to be: the geometry of the die cavity itself, along with the locations of associated gate(s), and vents; and the percentage of cavity pre-fill and associated interactions with slow shot (pre-fill) velocity and transition time from slow to fast shot velocity. Therefore, for a given cavity geometry an appropriate combination of pre-fill (if any) and other injection parameters should be employed to minimize air entrapment. This result could possibly explain why the influence of cavity pre-fill
(by itself) on porosity in die castings has been somewhat controversial in industry. Implementation Strategy: Die casters should consider the filling strategy early in the die and process design phases. Computer flow models can be invaluable aids in determining where air will be trapped in the cavity. For a given casting geometry, the pre-fill percentage, slow to fast shot transition time, gate geometry and location, and vent location need to be considered together to achieve filling patterns that will minimize air entrapment during metal injection. It should be noted that employing the cavity pre-fill approach will increase cavity fill time. Therefore, the die caster should be careful to calculate the available fill time for the casting to avoid filling problems due to heat loss before considering the use of cavity pre-fill. Typically, pre-fill can be used with diecastings that are relatively thick in wall section, such as automotive parts, which allow longer filling times. Also, in thick wall castings employing the pre-fill approach, thicker gates may advantageous because they can be used to feed solidification shrinkage during intensification. Cavity pre-fill can also increase die life on components located close to the gates, where impinging metal may cause rapid erosion of the die. Also, design (location) of gates and vents are very important factors influencing air entrapment, which can be reduced by cavity pre-fill. Accordingly, to minimize air entrapment and maximize the benefit of using cavity pre-fill, the cavity pre-fill approach should be considered not only in the process design phase but also in the die design phase. In the die and process design phases, it is also important to know the capabilities of the die casting machine to be used in the production of the casting. Important machine capabilities include the degree of shot velocity control, and acceleration rate from slow to fast shot velocity. Using flow modeling in conjunction with knowledge about die casting machine shot end capabilities, filling patterns can be designed to minimize, or change the distribution of, entrapped air in die castings. Casting scrap attributed to visual porosity on machined surfaces, or pressurized leak test failures can be minimized. The results of this research demonstrate that in many cases casting porosity can minimized by designing the shot profile while concurrently considering die geometry and die casting machine capabilities. This approach can minimize porosity related scrap and may obviate the need for costly additional equipment to reduce the amount of air entrapped, such as a vacuum system or casting impregnation equipment. Also, the
time required for pre-production try-out, scrap generated during start-up due to trial and error, and shot parameter setup time can be significantly reduced. The use of pre-fill can also potentially reduce tool repair expenses by reducing premature erosion failures near the gate(s).
Case Study
Problem: The problem was a high scrap rate situation due to visual porosity on machined surfaces for an aluminum cold chamber die-casting (see Figure 1). Figure 1. Near-surface entrapped air porosity in subject A380 casting (100X)
Implementation/Action: The project results suggested performing a computer flow analysis to identify a combination of slow to fast shot transition point, pre-fill percentage, and transition time from slow to fast shot velocity that would minimize entrapped air volume. In this case, the die and gate geometry, and gate and vent locations were already fixed. Also, the transition time from slow to fast shot velocity in this case could not be changed because of limitations in the die-casting machine hydraulic and control systems. Computer flow analysis suggested that changing the slow to fast shot transition point from existing practice (13.5 inches) to a low pre-fill percentage (greater than 16 inches) would reduce entrapped air porosity during filling. Results: An experimental casting campaign was conducted where the slow to fast shot transition point was varied between 13.5 and 18 inches. The castings made during the experiment were then tested by the Archimedes method for total percent porosity. The results of the experiment are shown in Figure 2.
Figure 2. Casting porosity as a function of slow to fast shot transition point
Castings made with a slow to fast shot transition point of 16 inches or less had an average porosity of 3.5 %. Castings made with a slow to fast shot transition point of 16.5 to 18 inches had an average porosity of less than 1%. Also, the variation in porosity among the experimental castings was much lower when the slow to fast shot transition point was greater then 16 inches. The die-casting machine was programmed
with a new shot profile having a slow to fast shot transition point of 17.5 inches. Subsequent to this change, porosity related scrap was significantly reduced. I


DEFECTS CAUSING REJECTION OF DIE-CASTINGS AFTER MACHINING (PHASE 2)

Business Benefit: This project provides die casters with information on ways to achieve savings by integrating the part, tooling and process design. Incorporating these strategies can reduce lead time as well as scrap. Project Objectives: The Defense Logistics Agency (DLA) objective of this project was to investigate manufacturing processes that could potentially reduce procurement costs for military replacement parts. A major thrust of the program was for Army Materiel Command (AMC) participants to identify components that were produced by methods other than casting (fabrications, forgings, and machined parts) and to assist the DLA in evaluating the potential benefits of changing the manufacturing method to casting. A secondary goal of the DLA was to conduct studies that would demonstrate methods to improve the quality and reduce the costs of castings. In Phase 2, the project objective was to select military die castings experiencing defects and apply process engineering methods to demonstrate the application of these techniques to reduce the frequency of defects. Approach: In Phase 2, a group of military die castings experiencing a high scrap rate subsequent to machining were analyzed for defects. Process engineering techniques such as PQ2, engineering of the slow shot profile, and evaluation of the casting filling pattern and hot spots using computer simulation were employed to minimize or eliminate the defects. Results: In Phase 2, the primary defects discovered in military die castings were visual and structural problems such as visual porosity on machined surfaces, and cold shuts. The defects were caused by undesirable filling patterns and hot spots related primarily to the existing gate and overflow designs. Machine injection parameters were also not appropriate in some cases to provide the best opportunity to fill the cavity without entrapped air porosity, or to fill the cavity when the die was cold. Since most military die castings are made in production runs of less than 200 castings, the scrap castings made during the start-up of each run can be a significant percentage. For example, if 20 warm-up shots are required for a run of 200 shots, that is a 10% in-house scrap rate. Therefore, it is important to reduce the number of shots required to make the first acceptable casting. Implementation Strategy: Die casters should employ Design for Manufacturing (DFM) techniques whenever possible. Concurrently, they should consider the impact of the design of casting geometry, gates and overflows, as well as machine capabilities and injection parameters on the filling patterns
and solidification times of critical portions of castings. Computer models showing filling patterns and hot spots can be invaluable in helping designers and engineers identify potential problems early in the design of a casting and process. Since most military die castings are "short-run", the gating and machine injection parameters should be designed to generate a desirable filling pattern even when the die is cold during start-up. Designing injection parameters to minimize air entrapment in the shot sleeve and during mold filling can also minimize visual porosity on machined surfaces. Also, die casters should monitor metal quality and practice melting methods to ensure that oxides and sludge are minimized or eliminated. Clean metal will minimize unexpected filling pattern problems due to poor fluidity and also minimize the occurrence of hard spots during machining. Integration of the part, tooling, and process design will minimize the overall design time (administrative lead-time). The time required to manufacture tooling, try out tooling and cast acceptable parts will also be reduced. Casting defect problems can also be identified and eliminated early in the design process. Die-casters employing these strategies can reduce scrap during start-up as well as scrap in regular production. Startup times can also be minimized, thus increasing machine utilization and labor efficiency. Savings in broken tooling and downtime during subsequent casting machining operations can be significant.
Case Study #1
Problem: A Weapon Efficiency Recorder Housing (see Figure 1) for the F-18 Hornet aircraft was being cast by a custom diecaster in a cold chamber machine. This thin-walled aluminum alloy casting was experiencing an unacceptable level of misrun and flow line defects particularly during start-up (see Figure 2). Figure 1. Weapon Efficiency Recorder Housing with gating and overflows shown.
Figure 2. Photomicrograph showing a flow defect (100x, A380 aluminum).
experiencing drill and tap breakage during machining of two bosses on the aluminum alloy part. Action/Implementation: The plan called for conducting radiographic and optical microscopic evaluation of the problem bosses, and conducting computer simulations of filling and hot spot analysis. Also, the existing injection profile on the diecasting machine was evaluated. Radiographic and optical microscopy revealed the existence of solidification shrinkage in bosses, causing small diameter drills and taps to wander and break during machining (Figure 6). Computer solidification simulations were conducted and showed inadequate gating and overflows in the boss areas. Overflows connected to bosses were enlarged and gates were added to boss the areas to improve the solidification pattern (Figures 7 & 8). The metal injection profile was also modified to improve filling. Results: The changes resulted in reduction in the amount of shrinkage in the bosses and reduced drill and tap breakage. I Figure 5. Weapon Efficiency Recorder Bracket with gating and overflows shown.
Action/Implementation: Since no previous flow analysis had been conducted during the design of the gating system, a computer filling analysis using the existing gating system was performed. Corrective revisions to the gating system were evaluated so as to obtain an acceptable filling pattern even with a cold die. The gating inlet areas were revised by changing the geometry of the gates - transitioning more gradually from gate to casting using 45 degree angles rather than existing single 90 degree angle. Results: The computer flow analysis demonstrated better flow patterns, even with cold dies. Misruns and visible flow lines were minimized over a wider range of die temperature, hence, fewer shots were required to obtain acceptable castings during start-up. Figures 3 and 4 illustrate the filling patterns predicted by computer simulation for the Weapon Efficiency Recorder Housing for the old and new gating, respectively. Figure 4. PEI 2640 Revised Figure 3. PEI 2640 Old Gating and Filling Pattern Gating and Filling Pattern
Figure 6. Photomicrograph showing boss shrinkage defects (50x, A380).
Case Study #2
Problem: A Weapon Efficiency Recorder Bracket (see Figure 5) for the F-18 Hornet aircraft was being produced by a custom die-caster in a cold chamber machine. The company was
Figure 7. Thermal results with prior overflow design
Figure 8. Thermal results with revised overflow design
Monitoring Task Force: Process Technologies Task Force Sponsored by: U.S. Department of Defense (Defense Logistics Agency) and NADCA, Through the American Metalcasting Consortium


GUIDELINES FOR SHORT RUN DIE CASTING

A. Miller, V. Shah, and K. Wee Chau, Ohio State University
Business Benefit: This project provides the designer and die caster with guidelines to determine when short-run die casting may be a cost-effective alternative to other production methods, increasing the potential market for die cast products. Project Objectives: The objective of this project was to investigate the reasons why short-run die casting is not economical. Based on a literature review and an industry survey, dies are the biggest cost factor in short-run casting. One of the ways to make short runs economical is by making the tool cheaper while maintaining the quality. If improvements can be made in the biggest cost factors, the cost of the die will go down and short run die casting will be more economical. Another aspect of the research was analyzing the biggest variables adding to the lead time for dies. Approach: Die casting is a near net shape manufacturing (NSM) process that produces components that are widely used in many industries, including automotive, aerospace, computer, telecommunication and consumer appliances. Their use has been increasing due to their quality, low cost and low weight. Typical reasons for the selection of the die casting process include: High production volume and rate with good repeatability from part to part Part accuracy and dimensional stability Long die life Several small components with different shapes can be combined into a single die casting Elimination of machining operations and few or no secondary operations Recyclability Die castings made today have complicated features and maintain very high dimensional accuracy. Significant technological advances such as controlling the process and equipment, and by producing precise dies have helped achieve these requirements. However, expensive equipment and dies have made die casting a capital oriented business. To reduce cost per part, the production volume of die cast parts usually exceeds 10,000 units because of the high die cost. Therefore, die casting is usually considered only when high part volume is required. However, there is demand for low volume die cast parts as functional prototypes or low volume production parts. The production of these castings is described as short-run die casting production. In this research, short-run die casting is defined and classified as follows:
Type 1 3,000 parts or less are produced in total for the life of the product Type 2 Demand calls for 200-500 parts periodically, but the product life extends over several years
Type 1 short-run die casting is used for prototyping or situations where die castings are required for functional reasons but the total number of parts required is small. Type 2 short-run die casting typically is not envisioned at the start of the project. However, it turns out that it is very common for military and other applications. A weapon system or component may stay in inventory for long periods of time but few are destroyed or damaged. The demand for new components or replacement parts is low and castings are required only periodically. The die will sit on the casters shelf for extended periods and be infrequently placed in production for short production runs. This type of casting operation has its own special problems.
The objectives of the research were to understand the characteristics of the die casting industry as it relates to short-run production and to determine how short-run die casting can be made more economical. Part of the question was to determine the technical and business barriers. A two-pronged approach was utilized. The process structure and cost structure of the industry were mapped and causes of cost and lead time leakage were analyzed. A case study was carried out using a military part as the test case to determine the feasibility and cost of short-run production. The second aspect of the work was to determine the characteristics of the industry and the current nature of short run practice. This was accomplished by a survey and plant visits. Results: The research has shown that "economy of scale" is not the only way to make die casting economically viable. A case study was conducted to carefully review the die making process. A die casting die was designed and manufactured at The Ohio State University and 500 parts were produced on the laboratory machine. This case study demonstrated that time and cost can be cut from the machining of the die, and as long as die tryout is not excessive, the costs are very competitive. Extensive use was made of computer simulation to optimize the design and a soft insert was used to minimize the material delivery lead time and the machining time. Measurements taken from the parts showed relatively little die wear over the 500 parts but there were some difficulties with die soldering due to the process.
The case study demonstrated that Type 1 short-run production can be very competitive with machining even with production runs as low as a few hundred parts. The survey, however, clearly showed that most die casters avoid both types of shortrun production. Very few casters devote more than 10% of their machine utilization to these runs. Even though a premium is usually paid for short-run castings, the casters believe that the risk of problems outweighs the benefits and they prefer conventional business when it is available.
The results of this project identified two opportunities for die casters: 1. The techniques identified in the project which reduced both the cost and lead time for die construction can provide savings to die casters engaged in both short-run and conventional part production. 2. Opportunities exist for the production of more short-run die castings for those die casters willing to accept the risks involved. I
Monitoring Task Force: Computer Modeling Sponsored by: U.S. Department of Defense and NADCA


ENERGY CONSUMPTION OF DIE CASTING OPERATIONS

J. Brevick, C. Mount-Campbell & C. Mobley at Ohio State University
Business Benefit: This project provides the die caster with information on methods to monitor and reduce energy consumption in order to reduce costs. Project Objectives: The primary objective of this research project was to develop models for die-casting operations that can be used to assess the influence of equipment or process changes on energy consumption. Molten metal processing is inherently energy intensive and roughly 25% of the cost of die-cast products can be traced to some form of energy consumption. The obvious major energy requirements are for melting and holding molten alloy in preparation for casting. The proper selection and maintenance of melting and holding equipment are clearly important factors in minimizing energy consumption in die-casting operations. In addition to energy consumption, furnace selection also influences metal loss due to oxidation, metal quality and maintenance requirements. Other important factors influencing energy consumption in a die-casting facility include geographic location; alloy(s) cast; starting form of alloy (solid or liquid); overall process flow; casting yield; scrap rate; cycle rates; number of shifts per day; days of operation per month; type and size of die-casting machines; related equipment (robots, trim presses); and downstream processing (machining, plating, assembly, etc.). Each of these factors also may influence the casting quality and productivity of a die casting enterprise. In a die casting enterprise, decisions regarding these issues are made frequently and are based on a large number of factors. Therefore, it is not surprising that energy consumption can vary significantly from one die casting enterprise to the next, and within a single enterprise as a function of time. The influence of local decisions within a die casting enterprise on energy consumption is often difficult to determine because of the scale and complexity of die casting operations. A change made in one aspect of the system may not have the degree of impact anticipated on the entire die casting system. In addition, individual components of the die casting system are not often metered for energy consumption, making it very difficult to assess the actual influence of enterprise decisions on energy consumption. Approach: The general approach of this project was to conduct a literature review regarding energy use in die casting, and then create and distribute a survey regarding energy consumption to North American Die Casting Association (NADCA) corporate members. The survey responses were then collected and evaluated. The goal of these activities was to establish
an accurate flow chart capable of mapping energy inputs for the die casting industry. Also, these data were used to determine the relative importance of various energy-consuming operations in die-casting, and to determine the amount and quality of energy data available in the industry. In addition to energy survey data, selected energy audits of die-casting operations at The Ohio State University (OSU) die casting laboratory and at industry sites were conducted. The purpose of these audits was to establish the relative amount of energy required by various die-casting operations, such as alloy melting, alloy holding and the die casting operation itself. Based on the information derived from the energy survey and on-site energy audits, computer-based models were developed that allow the energy "journey" in die-casting operations to be assessed. Results: A literature review regarding energy use in die casting was conducted and yielded pertinent information regarding energy consumption of melting and holding furnaces used for die casting. An energy survey instrument was developed, distributed to North American Die Casting Association (NADCA) corporate members, and the data collected and analyzed. It was determined that the amount and quality of energy data available in the die casting industry is generally poor. Using data from the literature and survey an accurate flow chart for mapping energy inputs for the die-casting industry was developed. The relative importance of various energy-consuming operations in die casting, such as alloy melting, alloy holding and die casting, were determined via energy audits conducted at the OSU die casting laboratory and at industry sites. Based on the information derived from the energy survey and on-site energy audits, the computer-based models TEAM and iThink were developed. The Energy Assessment Model (TEAM) is based on Absorbing State Markov Chains (ASMC). A dynamic energy model was created using iThink software. These models allow the energy "journey" in die-casting operations to be assessed. Both the TEAM ASMC and iThink models can be applied to individual plants or industry aggregates although some analysis is needed to see if aggregation introduces any significant systematic bias into the estimates that come from the models. The ASMC model does not directly capture the dynamic effects of time as will the I-Think model. However, once the dynamic effects are well understood, it may be possible to incorporate them in the spreadsheet with the TEAM model.
Implementation Strategy: Die casters should give more attention to the energy consumption and energy costs of their operations. Many operations do not have the necessary metering or accounting techniques to properly identify energy costs. The following approaches could provide significant benefit to die casting operations: Quantify the cost per pound of metal melting and holding equipment Measure the energy cost per pound of finish product shipped and set goals for improvement
Utilize the models demonstrated in this project to help make equipment and operational decisions based on their energy impact and cost Maintain knowledge about new energy-efficient equipment that could provide savings in energy costs I


EFFECTS OF EXTERNALLY SOLIDIFIED PRODUCT ON WAVE CELERITY AND QUALITY

T. Liang, C. Mobley, The Ohio State University
Business Benefit: This project provides the die caster with information to analyze die casting conditions, such as extremely long shot delay times, that affect the microstructural features of cast products. Project Objectives: The objective of this project was to determine the effects of externally solidified product (ESP), also referred to as "cold flakes", on the quality of die cast aluminum components. It was anticipated that the amount of ESP formed in the shot sleeve would increase with shot delay time, and that this increased amount of ESP would influence the microstructures and associated tensile and axial fatigue properties of the resulting die castings. Approach: Modified 383 aluminum alloy die castings were produced using selected shot delay times and amounts of plunger lubricants. The die castings were produced at Briggs & Stratton Corporation during June, 2001. The total shot weight (total quantity of alloy poured) for the sump cover casting produced was about 5.25 pound, and the resultant weight of the casting and overflows was 4.0 pounds. Twenty or more die castings were produced at each of three shot delay times (0.5, 6.5, and 13 seconds) and two amounts of plunger lubricant (1.3 and 2.6 grams), providing a 3 by 2 matrix for a total of six processing conditions. The other machine setup conditions were left as constants. The castings were cut at selected locations in the runners, near the main gate, at a point two-thirds of the way through the metal flow path, and near the casting overflows. The selected dimensions of the tensile and fatigue samples were consistent with the recommended dimensions set forth in ASTM Standard for Tensile of Metallic Materials E-8). The normal thickness of the tensile and fatigue samples was 0.114 inches. After machining, selected samples were visually examined and subjected to X-ray radiography to assess their porosity and related defect occurrence. The accepted samples were then subjected to either uniaxial tension or axial fatigue testing at room temperature. Following completion of the tension or fatigue tests, the fracture surfaces of selected samples were examined with the naked eye, low magnification optical microscope, and scanning electron microscope (SEM) equipped with energy dispersive X-ray analysis (EDAX). Results: The tensile data indicate no statistically significant difference between the 0.5- and 6.5-second shot delay conditions. The tensile fracture stress and fracture strains are significantly lower with the 13 second shot delay. There is no
significant effect of the two levels of plunger lubricant at any given shot delay time. The lower fracture stress and strain values associated with the 13-second shot delay condition are attributed to: 1. An anticipated larger volume fraction of ESP. 2. An increased porosity level (decreased density) 3. A greater oxide content 4. An increased occurrence of "cold shuts" The NADCA Specification values for the yield strength and fatigue strength of 383 aluminum alloy are 22 and 19 ksi, respectively. However, according to the fatigue data cited in the NADCA Specification, the number of cycles to failure for the samples should have been in excess of ten million at the stress level used (12 ksi). Only one sample of all those tested survived more than 4 million cycles. The average number of cycles to failure was 537,900 or less for all of the conditions tested. The significantly lower fatigue lives is in part attributed to the rectangular cross section of the samples and the associated stress concentration effect of the corners. The lower number of cycles to failure for the 13 second shot delay samples is attributed primarily to the increased occurrence of cold shuts and surface flow lines observed for those samples. All of the tensile samples and most of the axial fatigue samples contain at least one feature on the fracture surface that can be detected with the naked eye. The presence of those features is extremely detrimental to the tensile and fatigue properties. These features are classified into four categories: cold shuts, surface flow marks, ESP or oxide films, and macroporosity. Cold shuts are the dominant feature observed in the 13-second shot delay time samples for both lubrication conditions. Surface flow marks are the dominant feature in the 6.5-second shot delay samples and the ESP/oxide film feature is dominant in the 0.5-second shot delay samples. Within those four features, cold shuts reduce the tensile strength and ductility the most, the macroporosity is next most influential, and the surface flow marks and ESP/oxide films have nearly equal, and lesser, effect on the tensile properties. Implementation Strategy: The results of this study support the premise that the tensile and fatigue properties of die castings are strongly dependent on the microstructural features present in those products. The microstructural features observed (cold shuts, surface flow marks, ESP/oxide films, and macroporosity) influence the tensile and fatigue properties of the modified 383 aluminum alloy die castings. The extent of occurrence of the four microstructural features was dependent
on the die casting conditions, with cold shuts occurring most frequently with extremely long shot delay times. Accordingly, die casters should minimize the shot delay time to optimize casting mechanical properties. I


INCREASING PRODUCTIVITY & REDUCING EMISSIONS THROUGH LUBRICANT CONTROL

R. Eppich, Eppich Technologies
Business Benefit: This project provides the die caster with information to improve die casting operations by maximizing the dilution ratio of the die lubricant, minimizing the spray time for the die lubricant, and increasing the cycle rate of the process to reduce the energy required for each casting produced. Project Objectives: The overall objective of the project was to characterize the particulate levels and materials emitted in die casting. Secondary objectives included the use of "cutting edge" die lubrication formulations and lubrication application technology in order to reduce emissions, reduce scrap and reduce energy consumption. Several industry partners actively participated in this project. They included: Spartan Light Metal Products, Inc., Sparta, IL Chem-Trend, Howell, MI RMT, Inc., Brookfield, WI Rimrock Corporation, Columbus, OH Metaldyne, Niles, IL Approach: Total capture for emission measurements on smaller machines has been previously accomplished by "building a house" around the machine. This house-approach has resulted in good, but unpublished, information on a 600ton machine. Other emission measurement efforts have been reported using a partial hood that hangs over the machine, but this arrangement does not guarantee total capture. The enclosure designed for this project consisted of a sealed canopy top of sheet metal for the top of the die casting machine. Machine surfaces and sliding doors provided isolation around the diesides of the machine. The capture system was designed on the basis of 200cfm per square foot of opening. On this basis, an 18-inch duct was used in conjunction with a 30 HP blower that was operated at approximately 6500 cfm. A 1,400-ton cold chamber machine was chosen as the "test bed" for two phases of tests. The machine produced an aluminum two cavity casting with a total shot weight of 16.0 pounds. The individual castings are transmission components each weighing 4.68 pounds. The Phase I tests consisted of three one-hour runs during which emissions samples were gathered during normal production. Two hundred and seventy shots were made during the test periods, including 92 shots made while the system was reaching stabilization before and between tests. The cycle time of the machine, which ran on automatic, was 67 seconds. Phase II consisted of five twohour tests with a total of 480 shots made on the machine. For Phase I, approximately 55 gallons of lubricant at a 46:1
dilution ratio were used for the 270 shots. Based on these measurements, approximately 0.204 gallons of diluted spray were used per casting. All liquid in the gutter of the machine base was captured. This captured liquid amounted to 6.83 gallons. This information suggests that approximately 88% of the applied lubricant was released as air emissions and only 12% became over-spray in the sump. Results: Analysis of samples taken from the die casting machine suggest that very little, if any, of the residue is exhausted out into the environment. This low level of particulate matter may suggest that the die casting machine cell is absorbing some of the die lubricant residue much like a sponge rather than releasing it as a vapor or mist. Another explanation may be that a limited amount of the die spray is emitted to the environment since these products are designed to withstand high temperature environments. The die spray may coalesce quickly, perhaps faster than the solvents typically found in air quality tests, and results may show that there is a maximum level that can be emitted. A completed mass balance test must be applied to fully determine how much material is emitted into the environment. The tests also showed that a simple filter designed to collect the aerosolized droplets of water-based die spray should be sufficient to eliminate much of the particulate emissions. Designing a hood collection assembly capable of holding all of the mist for the entire casting cycle will assist in providing the most efficient method of reducing the emissions further. It was hoped that doubling the collection time from the Phase I tests to the Phase II tests would double the quantity collected and provide more accurate data, but this was not the case. This information suggests that longer air monitoring test times between 5-8 hours will be necessary to gain a better understanding of the total material emitted to the air. Several improvements to the process were accomplished during the testing. The lubricant dilution ratio, which started at 43 parts water to 1 part concentrated lubricant, was ultimately increased to 140:1, resulting in only one-third the amount of concentrate per gallon of diluted lubricant. The lubricant spray time was also reduced by 25%, from 8 seconds to 6 seconds, which helped to reduce the overall cycle time of the process from 67 seconds to 62 seconds. This demonstrates how quickly a die lubricant can impact the process and reduce the overall energy consumption. Since the initial process was optimized for the current lubricant, other process changes, including improved machine timing (robots, extractors, dwell time, etc.)
had to be adjusted to eliminate the time required to get 10% more quality castings per hour. Emission factors should be lower with increased production rates. The 10%increase in parts per hour should reduce consumption of energy, the same level or slightly more based on initial calculations with a total energy savings of around 1.5 times the cycle time savings. Further work is needed to fully characterize the savings in energy with improved productivity. Implementation Strategy: The results of this project suggest that there is very little, if any, residue that is exhausted into the environment from aluminum die casting machines in operation.
The emission factors and emission rates were two magnitudes below those for typical iron foundries. A simple filter design should be sufficient to control emissions. The project also demonstrated that significant improvements in die casting operations can be made by maximizing the dilution ratio of the die lubricant, minimizing the spray time for the die lubricant, and increasing the cycle rate of the process to reduce the energy required for each casting produced. I
Monitoring Task Force: Process Technology Task Force Sponsored by: U.S. Department of Energy (NICE3) and NADCA


INVESTIGATION OF SPRAY LUBRICANTS

Business Benefit: This project provides die casters with information about techniques that can be used to control the cooling rate provided by the die spray and other techniques, such as pulsing the spray, which can further increase the cooling rate. Project Objectives: Since the early 1990s General Motors Power Train Division (GMPT) pursued the development of a reliable off-line testing method to evaluate the performance of commercially available spray lubricants. A Lubricant Testing Apparatus (LTA) was constructed for this purpose and since 1995, the LTA has been utilized at The Ohio State University (OSU). This project continues the research conducted with the LTA at OSU. The first primary objective of this research was to investigate the influence of the spray system on the thermal cooling performance of water and lubricants. Specifically, the first goal was to evaluate the cooling performance of internal versus external style atomization nozzles at various flow rates and pressures. A second goal was to evaluate the cooling performance of commercially available spray tip patterns. A third goal was to evaluate the influence of pulsating spray on cooling performance. A second primary objective was to investigate the potential of employing a melt spinning technique for measuring the heat transfer coefficient of commercially available lubricants on H13 die steel. Approach: In an effort to optimize the performance of the spray system, studies were undertaken to link operating conditions with the thermal performance. In this study, different nozzle designs and spray patterns were evaluated. Pressure and flow rates of the lubricant and air supply were varied to understand the role of spray droplet momentum on heat transfer of the lubricant droplets at the surface of the plate. De-mineralized double distilled water was used as the cooling agent in all these studies to prevent the type of lubricant to influence or skew the results. Specifically, the following studies were conducted: Investigation of internal versus external spray nozzles - Air-mist (or pneumatic) atomization nozzle designs - Effect of operating parameters Investigation of the cooling performance of commercial spray tip patterns Investigation of pulsating spray on cooling performance
The thermo-physical properties of the die lubricants and the interactions between the lubricant and the die and liquid cast alloy are critical to developing stable system operation to produce acceptable quality die-castings. However, the LTA is not capable of characterizing the influence of deposited lubricants on the heat transfer coefficient between molten aluminum and H13 die steel. In a second part of this research, the chill block melt spinning (CBMS) process was evaluated as a possible laboratory method to assess the heat transfer coefficient between the die and the solidifying alloy, with and without die lubricants. A CBMS apparatus was constructed and used to assess the heat transfer coefficient between an aluminumsilicon eutectic alloy melt stream and an H13 tool steel wheel. The influence of two die casting lubricants on heat transfer coefficient was evaluated and reported. Results: The results regarding the influence of the spray system on the lubricant performance were as follows: 1. The Externally Mixed Nozzle provides a faster cooling than the Internally Mixed Nozzle. Externally Mixed Nozzles are characterized by small size droplets in a close spray. The heat transfer is better with this light spray in the film boiling regime to break down the vapor barrier. The design also provides for independent control of air and lubricant line pressures for control of the droplet size and velocity. 2. Increasing lubricant line pressure with constant air pressure leads to a smaller SMD for the droplets and also adds kinetic energy to the droplets. A light spray with fast moving droplets gives a better cooling performance than a dense spray. 3. Heat flux goes on increasing with increase in lubricant mass flow rate for both Internally and Externally Mixed Nozzles. Flow rate is more important a factor than the Sauter mean diameter (SMD) because of the interaction of the droplets downstream which is difficult to capture with an equation. 4. Study of spray patterns introduced the concept of spray flux (spray flow over a unit area). A tight solid cone spray would be helpful in cooling hot spots effectively because of concentration of spray over a small impingement area. Widening cone spray design can cover a large area and is useful to remove heat from bulk of the die. 5. Pulsating spray is an interesting method of spray lubrication. Experiments on the LTA did prove its merit. Pulsations in the range of 360 pulsations per minute were capable of removing heat from the plate at temperature of 315 deg C (i.e. 600 deg F) at a faster rate as compared to a continuous spray. Pulsation rate was correlated to the time required for vapor barrier break down and the rebound of heat in the plate (when
barrier break down and the rebound of heat in the plate (when spray is stopped which depends on the heat conductivity of the plate material). The chill block melt spinning (CBMS) process was investigated as a possible viable technique to evaluate the thermal influence of die lubricant coatings in the die casting process. A CBMS apparatus using a molten Al-12.6%Si eutectic alloy stream impinging on an H13 tool steel chill block wheel was successfully developed to closely emulate the die casting process. Important experimental CBMS parameters were established and reported. Analytic methods for estimating the heat transfer coefficient and freezing time based on experimental measures were established and the results compared with data available in the literature and with a one-dimensional heat flow computer simulation (BINORM). The CBMS process was shown to be capable of distinguishing differences in heat transfer coefficient, freezing time, and resulting filament dimensions for the various experimental conditions used in this research. To different degrees, the die lubricants applied
Actual photograph of the CBMS in operation.
to the H13 chill block spinning wheel reduced the heat transfer coefficient, increased freezing time, and produced thicker and wider filaments. Specifically, die lubricant #12 demonstrated greater insulating properties than die lubricant #16. Implementation Strategy: Die casters should utilize the results of this project to optimize the use of die spray. The results indicate techniques that can be used to control the cooling rate provided by the die spray and other techniques, such as pulsing the spray, which can further increase the cooling rate. The chill block melt spinning process shows potential for measuring the effectiveness of die lubricants but much more research is required on this technique. I
CBMS and Die Casting Process Comparison
Fused Silica Tube Pressurized head of the molten metal Wheel acts as one sided mold Dwell time proportional to freezing time
Monitoring Task Force: Process Technologies Task Force Sponsored by: NADCA

Schematic of CBMS


DC ProcessTechnologies

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DC ProcessTechnologies

Uploaded by

Copyright:

Available Formats

Turn Research into Action Business Solutions Based on NADCA Research

NORTH AMERICAN DIE CASTING ASSOCIATION

Leak Testing for Die Castings

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

ENERGY AND TECHNOLOGY ASSESSMENT FOR DIE CASTING PLANTS

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

Sensors for Die Casting

Turn Research into Action Business Solutions Based on NADCA Research

Figure 1. Location of vibration sensors.

Figure 2. Vibration signature showing events in the casting cycle.

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

Figure 1. Water Model vs. Computer Model Results

For further information, contact:

North American Die Casting Association

DIE CAVITY INSTRUMENTATION

Turn Research into Action Business Solutions Based on NADCA Research

Figure 3. Casting Density versus Shot Number for GM Beta-site Campaign

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

Turn Research into Action Business Solutions Based on NADCA Research

Turn Research into Action Business Solutions Based on NADCA Research

Figure 2. Casting porosity as a function of slow to fast shot transition point

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

DEFECTS CAUSING REJECTION OF DIE-CASTINGS AFTER MACHINING (PHASE 2)

Turn Research into Action Business Solutions Based on NADCA Research

Figure 2. Photomicrograph showing a flow defect (100x, A380 aluminum).

Figure 6. Photomicrograph showing boss shrinkage defects (50x, A380).

Turn Research into Action Business Solutions Based on NADCA Research

Figure 7. Thermal results with prior overflow design

Figure 8. Thermal results with revised overflow design

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

GUIDELINES FOR SHORT RUN DIE CASTING

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

ENERGY CONSUMPTION OF DIE CASTING OPERATIONS

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

EFFECTS OF EXTERNALLY SOLIDIFIED PRODUCT ON WAVE CELERITY AND QUALITY

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

INCREASING PRODUCTIVITY & REDUCING EMISSIONS THROUGH LUBRICANT CONTROL

Turn Research into Action Business Solutions Based on NADCA Research

For further information, contact:

North American Die Casting Association

Turn Research into Action Business Solutions Based on NADCA Research

INVESTIGATION OF SPRAY LUBRICANTS