• Author / Uploaded
• luisA1923

Citation preview

NORTH AMERICAN DIE CASTING ASSOCIATION

DIE CASTING POROSITY GUIDEBOOK

Publication #513

By: J. Brevick & S. Midson

Although great care has been taken to provide accurate and current information, neither the author(s) nor the publisher, nor anyone else associated with this publication, shall be liable for any loss, damage or liability directly or indirectly caused or alleged to be caused by this book. The material contained herein is not intended to provide specific advice or recommendations for any specific situation. Any opinions expressed by the author(s) are not necessarily those of NADCA. Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification and explanation without intent to infringe nor endorse the product or corporation. © 2019 by North American Die Casting Association, Arlington Heights, Illinois. All Rights Reserved. Neither this book nor any parts may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher.

Die Casting Porosity Guidebook:

Generating Porosity Standards and Measuring the Effect of Porosity on the Properties and Performance of Die Castings

Prepared by: J. Brevick and S. Midson

January 2019

TABLE OF CONTENTS

Introduction

5

Background

7

Porosity Specifications

11

Density

11

X-rays

12

Sectioning

13

CT Scanning

14

The Effect of Porosity on Strength and Ductility

15

General

15

Impact of As-Cast Thickness on Strength

19

Effect of Removal of Surface Skin on Mechanical Properties

19

The Effect of Porosity on Other Mechanical Properties

23

Fatigue Properties

23

Creep

25

Machined Surface Quality

27

Pressure Tightness

29

Impregnation

31

Wet Vacuum Process

32

Dry Vacuum Process

33

Dry Vacuum and Pressure Process

33

Sealants Used with Impregnation

34

Post Casting Heat Treatment

35

Appendix 1

39

References

41

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

3

INTRODUCTION

Due to the inherent physical nature of metals, commercial metal castings of all types tend to contain porosity. Porosity is the presence of voids, or gas-filled pores, in the casting where metal should exist. One common source of porosity in metal castings is solidification shrinkage. Most commercial casting alloys undergo significant volumetric shrinkage (increased density) as they transform from liquid to solid. The outside surfaces of castings solidify first, because that is where the mold is conducting the heat away from the casting. If no additional liquid metal is supplied to the mold cavity as the metal is solidifying (and shrinking), voids will form due to lack of available metal volume in the last portions of the casting to solidify. Voids created by this mechanism in castings are commonly referred to as solidification shrinkage porosity. Another common cause of porosity in metal castings is gas. Many metals contain dissolved gases in the molten state, and those gases are released during solidification and create gas-filled pores distributed throughout the solidified casting. An example is hydrogen gas dissolved in aluminum. In addition, molten metals can also vaporize materials they contact on mold surfaces (such as organic die lubricants), and those vapors can form gas–filled pores at, or just below, the surfaces of castings. Also, as molten metal flows into the mold cavity, it can engulf or entrain the air pre-existing in the die cavity. This can also create gas-filled pores in the solidified casting. High pressure die castings are typically selected when high production rates and volumes are desired, together with complex geometries, good surface finish and dimensional control. However, high pressure die casting presents unique challenges in terms of casting porosity. Because the dies/molds are metal (good thermal conductors), die castings solidify very rapidly. This requires that the molten metal be introduced into the cavity very quickly compared to other casting processes. The high flow rates commonly cause the molten metal to entrain air, which can produce relatively large amounts of gas-filled pores in the castings. The high solidification rate of die castings also severely limits the time available to supply additional molten metal to feed (minimize) solidification shrinkage. As a result, total elimination of all porosity in high pressure die castings is not typically economically feasible. Instead, the strategy normally followed with regards to porosity in high pressure die castings is to acknowledge that porosity will most likely exist in the castings. Furthermore, porosity may, or may not, negatively influence the ability of the casting to perform its functional requirements in service. The overall amount of porosity in a die casting, the type of porosity (shrinkage or gas), the location of porosity in the casting, and the size and distribution of the porosity will all have an influence on the functionality of the castings in service. The amount and location of porosity in die castings can be controlled to a significant

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

5

Introduction

degree via casting design, die design, and processing methods. Ideally, the die casting customer/designer and the die casting producer should communicate during the design phase, and collectively make decisions regarding the following issues: 1. What features/areas of a die casting can tolerate porosity and which areas cannot, and why (in terms of functional requirements). 2. How much porosity can be tolerated in specified regions of the casting, or in the overall casting. 3. Which method will be used to specify the allowable quantity and attributes of porosity in a die casting. 4. How to detect, evaluate, or measure the critical attributes of the porosity specified. The huge number of commercially successful die cast products is a testament to the viability of this general approach. However, it is also true that many die castings go into production with no porosity specification at all (often by customers unfamiliar with die castings), an ill-defined porosity specification (interpretation is unclear, or no defined method to measure or characterize porosity), or a porosity specification that is too strict for the casting service requirements (many truly functional castings are scrapped). All of these situations usually result in unnecessary additional costs for die cast products. Specifying porosity in die castings can be challenging. At this time, there is not a universally accepted and published comprehensive standard for specifying and evaluating porosity in die castings. Many long-time die casting customers have developed their own in-house porosity specifications, based on years of experience with the specific types of die castings they purchase. However, those specifications are not generally accessible, and perhaps not applicable, to all die casting customers. The overall objective of this document, therefore, is to provide some useful examples of methods for specifying and evaluating porosity in die castings. It is meant to be used as a guideline, to enable more effective and productive communication between die casting customers and die casting producers. In addition, as die castings are being used in ever more demanding applications, both with respect to pressure tightness and structural performance, users of die casters need more information regarding the effect of porosity on the mechanical properties and performance of die castings. Therefore, this book will also summarize published information documenting the impact of porosity on properties and performance. It should be emphasized, however, that this book will not review design and processing methods that can be used by part designers and die casters to minimize porosity. NADCA already has several books addressing those subjects, and the reader is directed to References 1-4 to obtain more information on how to minimize porosity levels in die castings. The goal for this book is to suggest methods that can be used by both customers and casters to define porosity levels, and to characterize the impact of this porosity on performance.

6

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

BACKGROUND

The type (shrinkage or gas), amount, location, distribution, size and shape of porosity are characteristics that may be important in a porosity specification. Interestingly there is not a large amount of quantitative information in the die casting literature describing these characteristics. One example, however, is a comprehensive study from 1972 performed by Lindsey and Wallace(5) who estimated that the total porosity content in aluminum die castings ranged between 0.5% and 4.1%, depending upon processing conditions. It is well known that porosity is not uniformly distributed within die castings, but tends to be clustered towards the centerline of the casting, with a relatively porosity-free dense skin at the cast surfaces. Figure 1a illustrates how the cross-section of a typical thin-wall die casting might look when viewed under low power magnification(6). The cast surfaces solidify rapidly, so the microstructure of this “skin” is essentially porosity-free (dense) and very refined (smallgrained). As discussed later in this book, this results in the mechanical properties of the skin of a die casting typically being better than those of the interior. As solidification progresses toward the interior of the casting, solidification shrinkage typically develops near the centerline of the casting. Since die castings solidify under pressure (intensification), when small volumes of gas are present, the gas typically migrates and expands into the relatively low pressure shrinkage voids as they form. As a result, die castings typically contain small, discrete, well-distributed pores near the centerline of the thicker walls of the casting. This porosity is typically a combination of shrinkage and gas, and would be considered normal (not objectionable) for traditional high pressure die castings. Figure 1b shows a thicker wall section of a die casting. The schematic drawing shows that the width of the dense skin is about the same for both the thin-wall and thick-wall castings. Therefore, the central porous region of the thicker walled casting is a significantly larger percentage of the overall volume of the casting (as compared with the thin walled casting) and will, therefore, significantly affect the properties and performance of castings. This will be described in more detail later in this book. The distribution of porosity within an actual die casting is shown in more detail in Figure 2. This is an image obtained using x-ray micro-tomography (also called micro-CT scanning), and shows the distribution of pores larger than 36 μm is a section cut from a die cast tensile bar(7). The dense surface skin is clearly visible, and is generally slightly more than 0.04-inches (1 mm) thick. What is also clear from Figure 2 is that the porosity is not uniformly distributed along the length of the bar, with more porosity in the thicker section of the tensile bar. This is normal in most castings, that solidification shrinkage will collect in the thicker sections of a casting, as the thinner sections will solidify earlier and the shrinkage in these thinner regions is fed from the liquid remaining in the thick sections.

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

7

Background

Figure 1: Illustration showing the characteristics of typical thin (a) and thick (b) die casting wall cross-sections. Thicker wall sections in die castings will be more prone to centerline shrinkage porosity, and will have less skin as a percentage of the total cross-section(6)

Figure 2: Longitudinal view of a micro-CT scan from a section cut from a die cast tensile bar, with the red dots showing the size and distribution of the porosity(7)

With die castings, solidification shrinkage is typically fed by applying high pressure to the biscuit, to feed additional liquid from the biscuit through the runner and gate and into the cavity. However, for this to occur, a feeding path between the biscuit and the cavity must remain open, and once a portion of the casting solidifies (the gate, for example) then all additional feeding is blocked. Simulation can be used to estimate how effective feeding will be in a die casting, and NADCA has a computer program called CastView that is especially useful. CastView works by geometrically “peeling” surface layers from a casting, so that thicker sections can be identified. An example is shown in Figure 3(8), where the pink regions identify the thicker sections of the casting that have remained after the thinner sections have been ‘”peeled” away. Figure 3 shows that the thicker (pink) regions in the castings cannot be fed from the runner, as the gate regions are thinner and so will solidify earlier, and so it is likely that large amounts of solidification shrinkage will exist in the pink regions identified in Figure 3.

8

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

Background

Figure 3: Results from a thickness analysis of a casting performed using CastView, where the pink regions identify the thicker regions of the casting likely to contain shrinkage porosity(8)

All this confirm that porosity will be present in all commercial die castings. However, if too much porosity is present (over and above that shown in Figure 2, for example) it may negatively influence important functional requirements (attributes) of die castings. Porosity specifications are often employed to ensure that die castings maintain a desired level of mechanical properties, pressure tightness, absence of blistering during heat treatment subsequent to casting, or quality of machined casting surfaces. The following sections provide examples of porosity conditions that may affect these attributes of die castings, along with specifications to address these porosity conditions, and methods of evaluating porosity and gas content.

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

9

POROSITY SPECIFICATIONS

For the purpose of ensuring that die castings are maintaining a desired level of mechanical properties performance, there are several methods for evaluating the level and distribution of porosity in die castings. These include the four methods listed below, which are described in more detail in the following sections.  Density  X-ray  Sectioning  CT scanning DENSITY The first method involves specifying a minimum density allowable for the casting. An example specification could be stated as “Casting density to be no less than 2.50 g/cm3.” To evaluate the density of castings in a production setting, one method is to simply weigh the castings on a precision scale. If the assumption is made that all of the castings have the same volume, the weights can be used directly as a means to identify low density castings. Alternately, knowing the overall volume of the casting (from a CAD drawing, or via a fluid displacement test), the density of each casting can be determined via the following equation: m ρc = v c c where: ρc = density of the casting (g/cm3) m = measured weight of the casting (g) v = volume of the casting (cm3) A relatively precise method is to determine the density of a casting via an Archimedes test. First, the volume of the casting is determined by weighing the casting in air, and then submerged in water. The volume of the casting is determined using this equation.

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

11

POROSITY SPECIFICATIONS

(m - m ) v = (ρair - ρwater) water air where: v = volume of the casting mair = weight of casting in air (g) mwater = weight of casting in water (g) ρwater = density of water at room temperature (g/cm3) ρair = density of air at room temperature (required when weighing to the tenth of a gram, otherwise use zero) The density of the casting, ρc can then be calculated using the following equation. ρc =

mair v

If the composition of the casting is known, the theoretical density of the alloy can be calculated using the approach outlined in Appendix 1, and the recent porosity can be calculated based on the difference between the actual and theoretical densities. The advantages of using density to measure porosity is that the measurements are nondestructive in nature, relatively easy to perform, and provide an accurate, quantitative estimate of the overall porosity level. The disadvantage of this process is that it does not provide any information regarding the distribution or the size of the pores within the castings. X-RAYS The second common method for specifying porosity is based on radiographic methods. The American Society for Testing and Materials (ASTM) has published a standard document called ASTM E505 - Standard Reference Radiographs for Inspection of Aluminum and Magnesium Die Castings. This standard consists of defined categories of discontinuities, and associated reference radiographs taken of aluminum and magnesium die castings having various porosity levels, as well as other discontinuities. In the ASTM E505 standard, Category A is dedicated to porosity, and there are numerical values assigned to various levels of porosity. For example, the reference radiograph for category A0 shows very little porosity, A2 has a greater level of porosity, and A4 even greater amounts of porosity. Using the ASTM E505 standard, an example porosity specification for an aluminum or magnesium die casting could simply be, “Porosity specified to be no greater than Category A2 per ASTM E505 reference radiographs”. This type of specification could apply to an entire casting, or just to critical specified areas of a casting. In either case, it would require that all, or a statistical sample, of the die castings produced would have to be radiographed, the images compared to the ASTM standard reference radiograph, and a determination made regarding the acceptability of each casting evaluated. This type of porosity specification would obviously be more costly compared to the simple density/weight specification approach. Furthermore, commercial die castings are typically geometrically complex, which can create challenges in terms of the clear resolution of discontinuities via radiography. There is also another very similar ASTM standard that was developed for aluminum and magnesium castings produced by any casting method called ASTM E155 - Standard Reference

12

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

POROSITY SPECIFICATIONS

Radiographs for Inspection of Aluminum and Magnesium Castings. In ASTM E155, the discontinuities seen in the radiographs are broken-down into specific categories, and actually spelledout in their full terminology, not abbreviated with letters (as is done in ASTM E505). This standard can also be used, and Table 1 shows an example porosity specification for an aluminum die casting that covers both gas porosity and shrinkage porosity limits. E 155 Discontinuity Category

Specified Reference Radiograph

Gas Porosity

#7 or better

Gas Holes

#5 or better

Shrinkage (sponge)

#4 or better

Shrinkage Cavity

#6 or better

Table 1: Example porosity specification using ASTM E155

General descriptions of the ASTM E155 and E505 standards are available online (www. astm.org). However, the actual reference radiographs cannot be reproduced because of copyright laws, and must be purchased from ASTM. ASTM E2422 Standard Digital Reference Images of Inspection of Aluminum Castings is also available from ASTM. The advantages of using x-rays to determine the level of porosity in die castings is that the technique is relatively easy to perform, generates a quantitative number (based on the ASTM standards), provides information about the location of the pores, and is non-destructive. Disadvantages include the fact that, typically, only the largest pores are identified, and so may not provide information about the pressure tightness of a casting. SECTIONING A third technique to estimate porosity in die castings is to cut a casting though a region of interest, polish the sectioned surface, and examine the polished surface using a microscope. Examples are shown in Figure 4. Image analysis software can then be used to measure the porosity on the polished surfaces. A specification could then state that sample castings could contain no more than 1.5% porosity (for example) on a section cut through a critical portion of a casting.

Figure 4: Examples of porosity on polished surfaces sectioned from aluminum die castings

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

13

POROSITY SPECIFICATIONS

Advantages of using sectioning and image analysis software to estimate porosity include the facts that an estimate of the distribution of the porosity is obtained (for example, a measurement of the thickness of the dense skin can be obtained), and the image analysis techniques will provide a measurement of the total porosity content on the sectioned surfaces. The main disadvantage is that an estimate of porosity is only obtained in the region of the die casting that was sampled ­­— no information is obtained regarding the remainder of the casting. In addition, as the technique is destructive in nature, the sampled castings cannot be shipped to the customer. CT SCANNING CT scanning (computed tomography scanning) is a relatively new nondestructive analytical technique that can provide detailed information on the number, size, distribution, and characteristics of porosity within components. CT scanning involves taking a large number of two-dimensional radiographic images as a casting is rotated in front of an x-ray beam, and utilizing digital geometry processing to generate a threedimensional view showing the position and relative size of each pore within a casting. Contrast in is based on differences in the x-ray attenuation throughout the sample volume. An example of a CT scan showing porosity within a die casting is shown in Figure 5(9). CT scanning is starting to be used as a quality control technique for die castings produced by commercial Figure 5: CT scan showing pore distribucompanies. A specification could focus on the total tion within a casting(9 porosity content, the maximum size of pores, or the location of the pores within a casting. The advantages of CT scanning are that it provides detailed quantitative information about the volume, size and distribution of pores within die castings. There are several disadvantages, including the fact that the resolution of conventional CT scanning is limited to pores about one mm in size and larger, and the equipment is very expensive to purchase. It’s worth noting that micro-CT equipment is now available with considerably better resolution (down to 1 μm in pore size), and so can provide significantly more detailed information on pore size and distribution. For example, Figure 2 shows a CT scan of a section cut from a die cast tensile bar casting, clearing showing the distribution of the porosity and the relatively thick dense surface layer (more than 1 mm or 0.040-inches in thickness). A considerable amount of information is available using the micro-CT process. For example, the sample shown in Figure 2 contained 86,696 pores larger in size than 37 μm. The total volume of the porosity was 16.6 mm3, and as the sample weighed 6.6 grams, the micro-CT scannig process estimated that the average porosity content within the sample was 0.7%(7). However, the main drawback of the micro-CT scanning system is that, due to its high resolution, the maximum samples size that can be examined is relatively small, typically a casting that can fit within a one-inch cube.

14

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY The static uniaxial tensile properties of yield strength (YS), ultimate tensile strength (UTS) and elongation are often used in the mechanical design of die castings, and so knowing how porosity affects these properties is important. This section of the book, therefore, will summarize published information documenting the effect of porosity on static mechanical properties (strength and ductility). GENERAL Prof. Campbell in his book entitle Castings(10) described the effect of porosity on the mechanical properties of all types of castings (not just specifically die castings). Campbell suggested that the yield strength of a casting will be only slightly reduced by defects such as porosity. He noted that, as no significant deformation has occurred when the casting starts to yield (begins to deform), then the only effect that porosity will have on yield strength is the reduction in cross-sectional area available to support the applied load. Therefore, if a casting contains 2% porosity, the cross sectional area will be decreased by 2% (i.e., 2% of the cross-sectional area will be holes), and so yield strength will also be decreased by 2%, obviously a very small change for a die casting. Figure 6 shows data from Campbell’s book on Castings(10) showing the impact of second phase particles, including porosity (represented by the open circles in Figure 6), on ductility. Based on these data, Campbell reported that Figure 6 indicates that defects such as porosity will significantly reduce the ductility of all types of castings. Regarding the effect of porosity on ultimate tensile strength, Campbell noted the effect is more complex, but that the ultimate tensile strength of castings will be decreased by the presence of defects such as porosity. The general information from Figure 6: Impact of second phase particles (including porosity) Campbell in his book on Castings(10) on the ductility of pure copper(10)

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

15

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY

is supported by several papers describing the behavior of die castings. For example, Suzzani(11) reported on testing performed to determine the effect of porosity on the mechanical properties of 380-type die castings (AlSi9Cu3) by the Austrian Giesserei-Institute, who found that yield strength was virtually unchanged by the presence of porosity, while elongation decreased considerably as the level of porosity increased. Information published by Adamane et al.(12) shows similar behavior. Their data is reproduced in Figure 7, showing tensile stressstrain curves for 380-type alloys (AlSi9Cu3) at different density levels (obviously as density is reduced, porosity is increased). This shows that the 0.2% yield strength was only slightly reduced by the porosity, while both ultimate tensile strength and elongation were significantly reduced.

Figure 7: Tensile stress-strain cures for three 380-type castings with different density levels (note that as the density decreases, porosity is increased)(12)

Research by Avalle et al.(13) on an aluminum die casting alloy (nominal chemistry of 9.8% Si; 3% Cu) also shown that increasing amounts of porosity generally degrades the static tensile properties. Die cast round (nominal 8 mm diameter gage section) mechanical test specimens were cast using three different gating/venting configurations (referred to as A0, A2, A4). The A0 gating/venting was designed correctly, to create a low porosity casting. The gating/venting systems for the A2 and A4 test castings were purposefully designed incorrectly to create greater amounts of gas porosity. Figure 8 shows scanning electron microscope (SEM) images of the typical porosity levels achieved in their A0, A2 and A4 test specimen castings. Their measured properties at each porosity level are summarized in Table 2.

Figure 8: Fracture surfaces of aluminum alloy mechanical test specimens cast with varying levels of porosity. A0 is a typical, low porosity die casting, A2 has an intermediate level of porosity, and A4 a higher overall amount of porosity, along with larger diameter pores(13)

16

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY

A0

A2

A4

Density (lbs/in3)

0.097

0.092

0.080

Void Fraction (calculated)

0.01

0.07

0.19

Yield Strength (ksi/MPa)

22/150

20/138

16/110

Ultimate Tensile Strength (ksi)/MPa

40/275

36/250

22/150

2.1

1.4

0.85

Elongation (%)

Table 2: Average density, void fraction, ultimate tensile strength, yield strength, and elongation of casting groups A0, A2 and A4(13)

Figure 9 shows a summary plot of UTS, YS and % elongation (data is extracted from Table 2, but is shown here in metric units) as a function of density for the A0, A2 and A4 cast specimen groups. This clearly shows the general decrease in strength and ductility in the castings as the number and size of the pores in the casting increases.

Figure 9: Plot of static mechanical property data reported in Table 1 as a function of the density of casting groups A0, A2 and A4(13). Divide the values in kg/m3 by 1000 to get units of g/cm3

Caceres and Selling(14) suggested, however, that reductions in tensile strength and elongation are not very well related to bulk density (i.e., bulk measurements of porosity), suggesting that it is not a reliable method to predict properties. However, when they measured the volume area of porosity in the fracture surface of their samples, they found a good relationship between the area fraction of defects in the fracture surfaces and tensile strength (Figure 10a) and elongation (Figure 10b). However, the problem obviously is that the area fraction of defects in the fracture surfaces can only be measured after a casting has been destructively tested, and cannot be used to predict the performance of a die casting in commercial application.

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

17

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY

a)

b)

Figure 10: Impact of area fraction of defects in fracture surfaces of tensile samples of an Al-7%Si alloy(14) a) Tensile strength normalized to the strength of the most ductile sample b) Elongation

The effect that porosity can have on the performance of actual die castings is demonstrated in Figure 11(15). The photographs show the difference in failure mode for sections cut from steering wheels die cast from a magnesium alloy. The photograph on the left shows deformation of a steering wheel containing a low level of porosity, where the casting failed by bending. The casting on the right contained a much higher level of porosity, and the steering wheel section failed by fracturing. This difference in behavior is obviously very important for castings used in safety critical applications (such as steering wheels), where it is important that castings bend and absorb energy, rather than fracturing.

a)

b)

Figure 11: Destructive testing of die casting magnesium steering wheels(15) a) Defect-free steering wheel that failed by bending b) Steering wheel containing porosity that failed by fracturing

18

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY

IMPACT OF AS-CAST THICKNESS ON STRENGTH Sequeira and Dunlop(16) reported on a study to analyze the impact of as-cast surface thickness on the mechanical properties of magnesium die castings. Specimens were die cast at three different thicknesses, 1 mm, 2 mm and 6 mm, and Sequeira and Dunlop estimated (using density measurements) that the porosity contents of the castings were 0.67%, 1.17% and 2.34% for thicknesses of 1 mm, 2 mm and 6 mm, respectively. Their results are reproduced in Table 3, showing that as the as-cast thickness increased, the strength dropped significantly, while elongation was observe to increase with cast thickness. Sequeira and Dunlop(16) attributed the drop in strength being due to the dense skin being a significantly lower fraction of the total cast thickness for the thicker samples (as shown schematically in Figure 1). However, they suggested that the low elongation values in the thin samples was due to a higher fraction of the relatively brittle Mg17Al12 phase in the surface layer, rather than the higher porosity level in the center of the thicker samples. As Cast Specimen Thickness (mm)

ASTM E505 Porosity Level

UTS (ksi)

Elong. (%)

1

27

35

1.9

2

23

31

2.4

6

18

26

2.3

Table 3: Impact of cast thickness on the mechanical properties of alloy AZ91D magnesium die castings(16)

EFFECT OF REMOVAL OF SURFACE SKIN ON MECHANICAL PROPERTIES Recently NADCA performed a study to determine the impact upon mechanical properties of removing the as-cast skin for aluminum die castings (i.e., machining tensile bar samples from the porous, central regions of die castings). They machined tensile bars from actual die castings, and compared the mechanical properties of the machined castings to cast-to-size tensile bars of similar compositions. Obviously the consequence of machining tensile bars from die castings is that the dense skin will be removed, and typically specimens are machined from the thickest sections of the die castings, which likely contain the highest amounts of porosity. The cast-to-size samples still contained the dense skin, and it is these types of samples (as defined in ASTM B557, Figure 13) that are used to generate the handbook mechanical property data for die castings (such as published in Reference 6). The results for four alloys are listed in Table 4. This shows that for all four alloys, the yield strength dropped slightly and the ultimate strength dropped significantly when comparing the properties of the machined and the cast-to-size samples. Interestingly for three of the four alloys examined (A380, E380 and B360) the elongations values were essentially unchanged, suggesting that the central porous region are controlling the elongation values for both ascast and machined tensile bars. It is worth noting that, at least in the opinion of one of the current authors (Midson), that the mechanical properties listed in Table 4 for the samples machined from castings do not represent the overall properties of the castings they were machined from. Commercial die castings consist of both dense skin and porous central regions, and testing the properties

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

19

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY

of one or the other does not accurately represent the overall performance of the castings. Clearly, therefore, obtaining representative mechanical properties of commercial die castings is a very difficult proposition, as it will depend upon section thickness (see Table 3) and the thickness of the dense skin layer. Probably the best method, therefore, to estimate the actual performance of commercial die casting would be to perform a functional test, using the entire casting (i.e., not machining away the skin), where the functional test is designed to represent how the casting will be loaded in practice. Condition

0.2% YS (ksi)

UTS (ksi)

Elongation (%)

Cast-to-size

23

47

3.5

Machined from casting

20

31

3.6

Cast-to-size

27

46

3.0

Machined from casting

24

32

2.9

Cast-to-size

23

46

5.0

Machined from casting

24

36

2.6

Cast-to-size

24

47

6.1

Machined from casting

21

36

6.0

Alloy A380

E380

F380

B360 *Handbook data

Table 4: Comparison for four alloys between mechanical properties of cast-to-size tensile bars and specimens machined from die castings As Cast Specimen Thickness (mm)

Yield Strength As-Cast (ksi)

Yield Strength – Skin Removed (ksI)

UTS As-Cast (ksi)

UTS – Skin Removed (ksi)

Elong. As-Cast (%)

Elong. – Skin Removed (%)

1

27

23

35

34

1.9

3.1

2

23

22

31

33

2.4

3.8

6

18

No change

26

No change

2.3

No change

Table 5: Impact upon mechanical properties of magnesium alloy AZ91D die castings of removing the dense skin for specimens machined from die castings (16)

Sequeira and Dunlop(16) performed a similar test for magnesium die castings, where they examined the impact upon mechanical properties of removing the dense skin. For the samples cast at 1mm, 2mm and 6mm thicknesses shown in Table 3, they machined 0.125 mm (0.005-inches) from each side of the samples. Their mechanical property results are shown in Table 5 (which also lists the mechanical properties of the unmachined samples re-

20

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

THE EFFECT OF POROSITY ON STRENGTH AND DUCTILITY

produced from Table 3). As shown in Table 5, the strength values of the 1 mm thick samples dropped considerably, while there was no change in strength for the 6 mm thick samples. Elongation values for both the 1 mm thick and 2 mm thick samples increased after machining away the surface layer, while again there was no change for the 6 mm thick samples. Similar to that noted earlier, Sequeira and Dunlop(16) attributed the higher ductility values to machining away the relatively brittle Mg17Al12 phase present in the surface layer of the magnesium die castings.

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

21

THE EFFECT OF POROSITY ON OTHER MECHANICAL PROPERTIES The objective of this section of the book is to document the effect of porosity on other mechanical properties of die castings. FATIGUE PROPERTIES It is well established that the fatigue life of castings is controlled by the size of the largest defect present. This is illustrated by the data for the A356 foundry alloy reproduced in Figure 12, which shows that cycles to failure during fatigue (i.e., fatigue life) is reduced as pore size increases(17).

Figure 12: Impact of pore size and applied stress on the fatigue life of aluminum castings(17)

This is confirmed for aluminum and magnesium die castings by Mayer et al.(18), who reported that the number of cycles to propagate a crack from an existing pore is much more significant than cycles to initiate a crack. Therefore, when considering fatigue strength, total porosity in the casting is not as important as the maximum pore size (effective diameter). As the size of pores increases, regardless if it is shrinkage or gas porosity, the number of cycles to failure in fatigue at a given cyclic stress level will generally be reduced. Therefore, if fatigue strength is important for a casting, a good approach would be to specify maximum pore size via ASTM E505 or ASTM E155 (similar specification to that shown in Table 1).

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

23

THE EFFECT OF POROSITY ON OTHER MECHANICAL PROPERTIES

Rodrigo et al.(19) noted that the measured fatigue strength of aluminum and magnesium die casting alloys is strongly influenced by the geometry of the fatigue test specimen and the method of testing. Because of the skin effect described earlier, as-cast round fatigue specimens tested in rotational bending fatigue will typically demonstrate the highest fatigue strength. Table 6 shows fatigue strength data for an aluminum die casting alloy (nominal chemistry of 9.8% Si; 3% Cu) obtained with an as-cast 8 mm diameter round specimen in rotational bending fatigue, at three ASTM porosity levels. In comparison, flat specimens (anything with 90° corners) tested in axial or bending fatigue will typically demonstrate lower fatigue strength. Machined fatigue specimens (skin removed, any specimen geometry) will generally exhibit the lowest fatigue strength. ASTM E505 Porosity Level

A0

A2

A4

Mean value of the fatigue strength (ksi)

22

14

11

Table 6: Fatigue strength (at 2 x 106 cycles) for aluminum; 9.8% Si; 3% Cu die casting alloy, Specimens ascast, 8 mm diameter, rotational bending fatigue, at ASTM E505 porosity levels A0, A2 and A4(13)

Borland and Tsumagari(20) examined the effect of as-cast diameter on fatigue life. As noted earlier, as the cast section thickness increases, the as-cast dense skin becomes a smaller faction of the overall thickness of the cast specimens. Their results are summarized in Figure 13, which shows that as the coupon diameter increased, the general trend was that the fatigue life decreased at every stress level. Borland and Tsumagari(20) reported on the fatigue strength at 107 cycles for alloy ADC12 for each sample diameter (estimated at 50% of samples to fail, as evaluated by Weibull statistics). Borland and Tsumagari estimated that the ¼-inch coupons had a fatigue strength of 95.4MPa (13.8ksi), 90.4MPa (13.1ksi) for the 3/8-inch coupons, and 89.2MPa (12.9ksi) for the ½-inch coupons. It is anticipated that an increase in cross- section would increase the likelihood for a defect to be present and initiate fatigue. Larger cross-sections would also be more susceptible to larger defects being present. As with the tensile coupons, reduction in fatigue properties is potentially influenced from the ratio of the skin to the gauge cross-sectional area.

Figure 13: Comparison of fatigue life for alloy ADC12 cast at ¼-inch, 3/8-inch and ½-inch thickness (R = -1)(20)

24

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

THE EFFECT OF POROSITY ON OTHER MECHANICAL PROPERTIES

Borland and Tsumagari(21) also examined the effect on fatigue life of machining aluminum die castings (i.e. removing the dense surface skin). As noted earlier, removing the skin decreases both yield strength and tensile strength. For the ADC12 alloy examined by Borland and Tsumagari(21), they showed that removing the skin decreased median yield strength by 2.07ksi (14.3 MPa) and decreased tensile strength by 2.99ksi (20.6MPa). As shown in Figure 14, Borland and Tsumagari(21) found that removing the as-cast skin decreased fatigue life by approximately an order of a magnitude from the as-cast condition at each stress amplitude tested. This equated to a 38.9% reduction in median fatigue strength from 95.4MPa (13.8ksi) to 58.4MPa (8.46ksi). This means that, similar to that observed with strength, machining fatigue samples from commercial castings will not generate data accurately representing the fatigue behavior of the actual castings, as the machined fatigue samples will not contain the dense surface skin (and so will have a reduced fatigue life). To obtain an accurate representation of the fatigue life of a commercial casting, possibly the best approach again will be to perform a functional fatigue test on the entire casting.

Figure 14: S-N diagram for 0.25-inch diameter alloy ADC12 coupons, both in the as-cast condition and after machining away the dense surface layer(21)

CREEP Most magnesium die casting alloys tend to exhibit low creep resistance in the 150-250°C temperature range. Creep flow rate decreases to a minimum value early in the creep test, and remains relatively constant for a long time thereafter. This is called the minimum creep rate (MCR), and is generally accepted as the characteristic metric of creep resistance. Experimental evidence shows that porosity volume has an influence on the MCR of magnesium alloys(22). Figure 15 shows the trend of higher casting density (lower porosity) decreasing the minimum creep rate for AZ91D Mg alloy. As with static mechanical properties, a porosity specification based on density would be applicable if MCR is of concern as a functional requirement of magnesium die castings. Use of the ASTM E155 or ASTM E505 standard would also be appropriate (similar specification to that shown in Table 1).

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

25

THE EFFECT OF POROSITY ON OTHER MECHANICAL PROPERTIES

Figure 15: Plot showing the relationship between minimum creep rate of AZ91D alloy versus porosity (density) at 150°C and 50 MPa(22). The legend in the upper right corner of the plot area represents various conditions under which the creep specimens were die cast

26

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

MACHINED SURFACE QUALITY

When die casting surfaces are machined, an undesirable side-effect is that the skin of the casting (Figure 1) may be removed. If internal porosity is exposed via machining, the casting may not fulfill the desired functional requirements. For example, porosity can create an aesthetic problem if the machined surface has appearance-related requirements. Exposed porosity on surfaces used for sealing, such as with gaskets or O-rings, may cause sealed surfaces to leak. Tapping internal holes in a die casting can expose porosity that may weaken the threads. Pertinent attributes of porosity exposed by machining include the individual pore size, density or concentration of pores in a defined area, and the minimum spacing between individual pores. A common method for specifying exposed porosity is to define categories or acceptable levels of porosity attributes, an example is shown in Table 7.

Category of Porosity

Maximum Size in any direction Inch (mm)

Minimum Spacing Inch (mm)

Maximum Density Pores in2 (cm2)

Level 1

0.015 (0.4)

0.150 (4.0)

65 (10)

Level 2

0.030 (0.8)

0.100 (2.5)

25 (4)

Level 3

0.060 (1.5)

0.075 (2.0)

20 (3)

Level 4

0.090 (2.3)

0.050 (1.0)

10 (2)

Table 7: Example of defined porosity attributes for porosity exposed on machining

Using this approach, the acceptable porosity for various machined surfaces on a die casting can be called-out on an engineering drawing using the levels defined. For example, the porosity on the machined threads of an internal hole might be specified on the drawing as “Level 1”. A compression gasket sealing surface might be specified as “Level 4”, a piston bore “Level 3, and a static O-ring sealing surface “Level 2”. Another approach is to use a visual comparison system of specification. Figure 10 shows an example of a Level A through H visual guide for porosity on machined surfaces(6). The most common methods of assessing machined surface porosity are by visual inspection, or with systems of digital optical hardware with image analysis software. Regardless of the type of specification employed, when manual visual inspection is used, example photographs (visual standards) of acceptable and unacceptable porosity for critical machined surfaces on the specific casting can be a very effective tool toward the goal of maintaining inspection consistency over time, and between various inspectors.

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

27

MACHINED SURFACE QUALITY

Figure 16: Example of an alphabetic, visual comparison porosity specification system for exposed porosity on machined surfaces in die castings(6)

28

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

PRESSURE TIGHTNESS

Another important attribute of some die castings is that they must be capable of containing fluids under pressure without leaking in service. Examples include engine blocks, transmission cases, brake cylinders, fuel system components, water pumps, thermostat housings, and air conditioning system components. When shrinkage and gas porosity in die castings become interconnected and exposed to casting surfaces, through-wall leaks can occur. Internal porosity is often exposed to surfaces after machining, or by damage caused during the casting process itself when the casting solders onto the die, and severe surface drag occurs during ejection. Other cast-in discontinuities, such as cracks, surface cold flows and inclusions, can also work in conjunction with internal porosity to create unwanted leak paths in die castings. A diagram illustrating a hypothetical leak path in a die casting is shown in Figure 17a(23), while an actual example in a die casting is shown in Figure 17b. Since porosity is known to be a significant contributor to lack of pressure tightness in die castings, some examples of leak testing methods and general specifications are described.

a)

b)

Figure 17: Leak paths in die castings a) Diagram of a typical leak path connected in series through the wall of a die casting(23) b) Photograph of a sectioned die casting, showing a leak path between a high pressure port (HPP) and a screw hole(24)

The most common approach to leak testing die castings is the pressure decay method. The casting to be tested is temporarily sealed and then pressurized to a desired level with either dry air or dry nitrogen gas. Once the desired test pressure is attained, the system is sealed-off, and any loss of pressure is recorded as a function of time. The recorded pressure loss is then converted to a flow or loss rate of air or nitrogen through the casting walls. The results of the pressure decay leak test are commonly expressed in units of std cc/min (standard

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

29

PRESSURE TIGHTNESS

cubic centimeters per minute). Some examples of pressure decay leak test specifications are shown in Table 8.

Casting Type

Fluid Contained

Typical Test Pressure Range Bars / (psig)

Maximum Leakage Rate of Dry Air or Nitrogen (cc/min)

Transmission Case

Automatic Transmission Fluid

0.2 to 0.4 / (3 to 5)

10 to 20

Carburetor Assembly

Gasoline

0.4 to 0.7 / (5 to 10)

1 to 5

Master Cylinder

Brake fluid

1.4 to 7.0 / (20 to 100)

5 to 20

Engine Block

Water/Glycol/Oil

0.3 to 1.4 / (4 to 20)

10 to 30

Cooling Systems

Water/Glycol

1 to 2 / (15 to 30)

4 to 7

Electrical Housings & Connectors

Water

0.1 to 1 / (1.5 to 15)

0.01 to 1

Table 8: Example pressure decay leak test specifications

For castings where desired leak rates are very low, such as for air conditioning components, the use of air or nitrogen is not feasible for leak testing. Instead, helium is used because it has a much smaller atomic radius, and lower molecular mass, allowing it to diffuse and flow through much smaller leak paths and at much higher rates that either air, nitrogen, or refrigerants. While helium can be used with pressure decay leak tests, however, the tracer gas technique is much more common. In the tracer gas test, the casting is internally pressurized with helium. Helium that leaks out of the casting is measured by a sensitive helium detection device. Figure 18 illustrates this method(23).

Figure 18: Schematic diagram showing the helium tracer gas leak testing method(23)

An example leak rate specification for a tracer gas method is shown in Table 9. The 5 to 15 grams per year leak rate of refrigerant would have to be correlated to a maximum allowable helium leak rate in the tracer gas test. Correlation of test gas leakage with other gases at different operating pressures can be accomplished using the following equation.

30

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

PRESSURE TIGHTNESS

Q2 = Q1

( )( η1 η2

P22 - P12 P42 - P32

)

where: Q2 = Test leakage rate Q1 = Operational leakage rate η2 = Viscosity of test gas η1 = Viscosity of operational gas P2, P1 = Absolute pressures on high and low sides at test P4, P3 = Absolute pressures on high and low sides in operation Casting Type

Fluid Contained

Typical Test Pressure Range Bars / (psig)

Air Conditioning

Refrigerant

2 to 20 / (30 to 290)

Example Leakage Rate of CFC Refrigerant 5 to 15 grams/year

Table 9: Example leak rate specification for a die cast air conditioner component

The pressure decay and tracer gas methods of testing are both capable of establishing the existence and rate of leakage. However, the location of the leakage is not determined via these methods. Bubble testing is another method commonly used to determine the location of leaks in die castings. The bubble test is typically performed by pressurizing a casting with air (or other gas), and submerging the casting in a container of water. The water container is usually transparent (acrylic) and placed in a well-lighted area, so that bubbles forming on the outside of the submerged casting can be readily observed, and the leak location(s) noted. When die castings fail leak testing, impregnation sealing is a corrective measure that may be specified for repair. Impregnation sealing generally involves placing castings in a vacuum chamber with the goal of filling the pores with a sealant. Impregnation procedures will be described in detail in the following section. IMPREGNATION Westbrook(25) described impregnation as the process of sealing the porous imperfections in die cast parts with a sealant that cures to a solid during the impregnation process. The sealant is drawn into any pores that are present on the surface of a casting, whether on an as-cast surface or a machined surface. In a series of papers published in the Die Casting Engineer(26-29), Shantz and Versmold described the impregnation process, indicating that, in general, all impregnation process types follow the same steps:  Impregnation of the sealant into the porosity/leak path using vacuum, and possibly pressure (using an autoclave)  Recovery of the excess impregnation sealant from the outside surface of the castings, and from tapped holes and cavities (a recovery station)

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

31

PRESSURE TIGHTNESS

 Removal of sealant from casting surfaces and features where sealant is undesirable (a wash station)  Curing or polymerizing of the sealant that has been impregnated within the castings (a cure station) They noted that there are three general methods for impregnating the sealant into the metallic components, and that these three methods represent 95% of all global applications. The three process are listed here and described below: 1. Wet vacuum 2. Dry vacuum 3. Dry vacuum & pressure Wet Vacuum Process In the wet vacuum process, the castings to be impregnated are immersed directly into sealant contained within the impregnation chamber. Once the castings are covered and the chamber is sealed, a vacuum pump evacuates air from the chamber (and so also evacuates air from the porosity within the castings), allowing the sealant to fill the pores. Following the vacuum cycle, the vacuum is released and the chamber is returned to ambient atmospheric pressure.

Figure 19: Schematic of the wet vacuum process(27)

Due to the hydraulic pressure of the impregnation chamber, castings at the top of the sealant in the chamber experience a higher level of vacuum than do castings in the sealant positioned at the bottom of the chamber. Shantz and Versmold(27) noted, therefore, that if the chamber is 40 inches deep, the castings at the top could experience a vacuum level of 10 milli-bar, while those at the bottom experience a significantly lower level of vacuum (maybe >100 milli-bar), and so a customer may find that castings experience a range of pressure tightness depending upon their position in the chamber. Shantz and Versmold(27) indicated that the advantages of the wet vacuum chamber process involve low cost, ease of use, and that the equipment is economical to purchase. Disadvantages are that, due to the deficiencies listed above, a percentage of the castings will continue to leak after the impregnation process. Shantz and Versmold(27) suggested that the wet vacuum process is best used for powered metal and electrical parts (plastic components) that have open, connected pores, but is typically not used for die castings due to the smaller pores with die castings.

32

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

PRESSURE TIGHTNESS

Dry Vacuum Process Shantz and Versmold(27) indicated that the dry vacuum process can be used for die castings. With this process the casting are placed into the vacuum chamber, and a vacuum is applied to remove air from the chamber and from the pores in the castings (about 5 torr pressure), and so the dry vacuum system creates a uniform vacuum everywhere within the chamber. Once the target vacuum level has been achieved, the sealant is transferred into the chamber, and the negative pressure pulls the sealant into the pores.

Figure 20: Schematic of the dry vacuum process(27)

Shantz and Versmold(27) suggested that the only downside to the dry vacuum process is that it does not use pressure to assist the sealant in penetrating the pores within the castings, but reported that the dry vacuum process is an effective method to seal porosity. Dry Vacuum and Pressure Process This process is similar to the dry vacuum process, but once the sealant has penetrated the pores, a pressure of around 90 psi (6 bar) is applied to the chamber, to assist the sealant in penetrating the pores. Shantz and Versmold(27) reported that the application of pressure can significantly reduce the impregnation time by up to 80%, and also provide an improved impregnation process, and therefore tends to be used for components used in higher performance applications. This process can be used to impregnate fine porosity, such as A/C compressors, high performance engines and transmissions. The disadvantage of this process is higher capital cost due to the need for a pressure chamber, but Shantz and Versmold suggested that this higher capital costs can be quickly recovered though the faster production cycle. Cross(30) and Hynes(31) suggested, however, that pressure is not required for successful impregnation, but that low viscosity sealants can be used with dry systems, and that globally dry vacuum impregnation is the most widely accepted form of impregnation, especially for gearboxes and engine blocks. They suggested that the dry vacuum and pressure process originated from a time when the application of pressure was a necessary step to compensate for the poor characteristics of older sealant technologies (such as silicate and polyester sealants), but that new sealant technology has replaced the need for additional pressure. Gebhard(32) emphasized that it is important for metallurgists, casting engineers and impregnation specialists to collaborate throughout the entire parts-production process, beginning with the early stages of casting design. Just as there are many variables in the casting

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

33

PRESSURE TIGHTNESS

process (mold design, alloy metal ratios, additives, temperature, heat distribution, pressure, etc.), there are also many variables in the impregnation process. Shantz and Versmold(27) suggested that a 95-99% recovery rate should be an excellent benchmark, even achieving higher than 99%, when the process selection and sealant choices are supported by the type of sound analysis described above, which includes pre-impregnation testing, and confirmation through design of experiments testing, as long as the quality of the castings (i.e., the porosity content) is consistent from batch-to-batch.

Figure 21: Schematic of the dry vacuum and pressure process(27)

SEALANTS USED WITH IMPREGNATION Shantz and Versmold(29) noted that all sealants today used for impregnation are based on methacrylates, and as all reputable impregnation sealants are approved according to the MILSTD-I-17563C, different brands typically provide similar performances. Historically sealants were produced from materials such as water glass (sodium silicate), epoxy and polyester resin, but these materials are rarely used today due to occupational health and safety concerns, and because of their ineffectiveness. Water glass is only used for components that have to withstand very high temperatures (above 700ºF, for example). There are typically two types of sealants, both based on the methacrylates monomer — ­ anaerobic vacuum impregnation sealants and thermosetting vacuum impregnation sealants. These differ in the manner in which the monomer is cured or polymerized. Anaerobic sealants cure at room temperature, and as the curing process can take up to 48 hours, the impregnated components cannot be leak tested until after that process. For that reason, most die castings are sealed using thermosetting sealants, and the impregnated castings are heated to about 195ºF (90ºC) by immersing into a bath of hot water, and following this curing treatment (to produce cross-links between the monomers in the raw sealant), the castings can be immediately pressure tested. Westbrook(25) noted that penetration and adhesion are the two most important characteristics of impregnation sealants. Sealants that are low in viscosity penetrate deeply into the smallest pores, while sealants that have higher viscosities adhere vigorously to the inside walls of the porosity. The choice of sealant may depend upon the design of a casting — ­ for example Westbrook noted that the more complex a part, the more porous it is likely to be, and so for a complex part that is large, a low-viscosity sealant may be preferable. Less complex parts might use a higher viscosity sealant.

34

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

POST CASTING HEAT TREATMENT

The final section of this book will address porosity concerns associated with heat treatment. Although the vast majority of die castings are still not heat treated (i.e., used in the as-cast condition), heat treating is becoming more common. Aluminum die castings used for loadcarrying structural members are sometimes heat treated to the T6 condition to increase yield strength. This involves a solution heat treatment, which requires heating the casting into the 500º C (930ºF) temperature range. When gas-filled pores exists in a die casting, they are usually artificially smaller than they would be at atmospheric pressure, because die castings solidify under pressure. When aluminum is heated for solution heat treating, its yield strength drops dramatically (it gets very soft), while due to heating, gas-filled pores in the die casting exert increased pressure on the surrounding aluminum (see Figure 22). As a result, when a die casting contains gas porosity above a threshold level, “blisters” will form on the surface of the casting. Blisters form when the expanding gas inside the casting exceeds the yield strength of the surrounding aluminum (Figure 23). Figure 24 shows an example of blistering.

Figure 22: Schematic drawing showing the impact of increasing temperature on the strength of the aluminum and the gas pressure within a defect(33)

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

35

POST CASTING HEAT TREATMENT

Figure 23: Schematic drawing of a casting, showing the formation of blisters(33) a) As-cast, showing the location of subsurface gas pores b) Blistering occurring during solution heat treatment

Figure 24: Cross-sectional view of a blistered aluminum die casting wall

Blistering during heat treatment is a condition that renders the casting unusable. The amount of porosity, or casting density, is not always directly correlated to blistering. However, the severity of blistering is closely correlated to the total amount of gas volume in the casting. The gas volume contained in a casting can be measured by a vacuum fusion apparatus (Figure 25). A casting (or portion of a casting) of known mass is melted in a sealed vacuum chamber. As the casting melts, any gas contained is released and increases the pressure in the vacuum chamber. The pressure rise is converted into volume using the ideal gas law. The units of measure are cubic centimeters of gas at standard temperature and pressure (STP) per 100 grams of casting alloy (cc/100g@STP). If the gas content measured by the vacuum fusion test is less than approximately 10 cc/100g@STP, aluminum die castings typically will not blister during solution heat treatment.

36

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

POST CASTING HEAT TREATMENT

Vacuum fusion equipment represents a relatively high capital expenditure, and a single test may consume the better part of an hour. Since castings are destroyed during the test, a statistical sampling scheme is required to comply with a specification for contained gas.

Figure 25: Schematic representation of a vacuum fusion apparatus for testing contained gas content of castings

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

37

APPENDIX 1: CALCULATING THE THEORETICAL DENSITY BASED ON COMPOSITION If the composition of a die casting is known, the theoretical density of the casting can be calculated using the equation listed below and the alloying element multipliers provided in Table A1. 100 ρo = ∑ (% x f ) i i i Where ρo = calculated theoretical density (g/cm3) %i = weight percent of each element present in the alloy fi = multiplying factor specified for each of the alloying elements (see Table A1) The multiplying factors specified for each of the alloying elements that are commonly present in commercial cast alloys are listed in Table A1. Element

Factor

Element

Factor

Element

Factor

Al

0.3705

Ni

0.1123

Cd

0.1156

Si

0.4292

Zn

0.1401

Co

0.1130

Fe

0.1271

Ga

0.1693

Li

1.4410

Cu

0.1116

V

0.1639

Pb

0.0882

Mn

0.1346

B

0.4274

Sn

0.1371

Mg

0.5522

Be

0.5411

Zr

0.1541

Cr

0.1391

Bi

0.1020

Ti

0.221

Table A1: Multiplying factors (fi ) for selected alloying elements

The percent porosity in the castings can then be calculated using the following equation: P=

( P P- P )x 100 o

c

o

Where P = calculated porosity (in percent) ρo = calculated theoretical density (g/cm3) ρc = measured density of the castings (g/cm3)

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

39

APPENDIX 1

If the composition of the casting is not known, the nominal density of the alloy can be used (instead of the calculated theoretical density). Values for the nominal densities for various die casting alloys can be found in the NADCA Product Specifications Standards handbook. However, this approach will not be as accurate as calculating density based on the actual composition.

40

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

REFERENCES

1. Alex Monroe, Porosity In Die Castings: An Overview and Analysis, NADCA publication #511, 2011 2. Ed Herman, Porosity Management: Advanced Analysis, NADCA publication #510, 2013 3. W. Walkington & Steve Midson, Gas Porosity - A Guide To Correcting the Problems, NADCA publication #516, 2014 4. Scott Kirkman & Steve Midson, Shrinkage Porosity: Guide to Correcting the Problems, NADCA publication #518, 2015 5. D. Lindsey & J.F. Wallace, “Effect of Vent Size and Design, Lubrication Practice, Metal Degassing, Die Texturing and Filling of Shot Sleeve on Die Casting Soundness”, Proceedings 7th SDCE International Die Casting Congress, 1972, 1-15 6. NADCA Product Specification Standards for Die Castings, NADCA publication #402, 9th Edition, 2015 7. Itamar Brill, Branden Kappes & Stephen Midson, “An Initial Evaluation of CT Scanning for Measuring and Characterizing Porosity in Aluminum Die Castings”, Proc. 2018 NADCA Congress, paper no. ?? 8. R. Allen Miller, “Modeling Success Stories”, Proc. 2006 NADCA Congress, paper no. T06-113 9. Alejandro Golob, “Quality Control Best Practices: Introducing Industrial CT Scanning into Your Quality Assurance Process”, Die Casting Engineer, May 2015, p 14 10. John Campbell, Castings, 2nd Edition, Pub: Elsevier 11. R. Suzzani, “The Influence of Porosity on the Mechanical Properties of Die Castings”, Fonderia, 43, July-Aug 1994, p23 12. Anilchandra R. Adamane, Lars Arnberg, Elena Fiorese, Giulio Timelli, Franco Bonollo, “Influence of Injection Parameters on The Porosity and Tensile Properties of High-Pressure Die Cast Al-Si Alloys: A Review”, International Journal of Metalcasting 9 (1), 2015, 43-53 13. M. Avalle, G. Belingardi, M. Cavatorta, and R. Doglione, “Casting Defects and Fatigue Strength of a Die Cast Aluminum Alloy: A Comparison Between Standard Specimens and Production Components”, International Journal of Fatigue, 24, 2002 1-9 14. C.H. Caceres & B.I. Selling, “Casting Defects and the Tensile Properties of an Al-Si-Mg Alloy”, Material Science and Engineering A220, 1996, 109-116 15. Thomas Pyttel, Adi Sholapurwalla, Sam Scott, Ole Koeser & Elke Lieven, “Reinventing the Wheel”, Die Casting Engineer, January 2007, p 32 16. Winston Sequeira & Gordon Dunlop “Microstructure, Mechanical Properties and Fractography of High Pressure Die Cast Magnesium Alloy AZ91D”, Die Casting Engineer, March 2004, p 62 17. J.F. Major, “Porosity Control and Fatigue Behavior in A356-T61 Aluminum Alloy”, Proc. American Foundry Society, 1997, 901 18. H. Mayer, M. Papakyriacou, B. Zettl & S. Stanzl-Tschegg, “Influence of Porosity on the Fatigue Limit of Die Cast Magnesium and Aluminum Alloys”, International Journal of Fatigue, 25, 2003, 245-256 19. D. Rodrigo, M. Murray, H. Mao, J. Brevick, C. Mobley, V. Chandrasekar & R. Esdaile, “Effects of Section Size and Microstructural Features on the Mechanical Properties of Die Cast AZ91D and AM60B Magnesium Alloy Test Bars”, Journal of Materials and Manufacturing, 108, 1999, 785-789 20. Andrew Borland & Naoyuki Tsumagari, “The Effect of Casting Wall Thickness on the Tensile and Axial Fatigue Properties of ADC12 Alloy Die Castings”, Proc. NADCA Congress, 2005, paper no. T05-111

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

41

Chapter Title

21. Andrew Borland & Naoyuki Tsumagari, “The Significance Of The Die Cast Skin Pertaining To The Fatigue Properties Of ADC12 Aluminum Alloy Die Castings”, Proc. NADCA Congress, 2006, paper no. T06-031 22. E. Gutman, Y. Unigovski, M. Levkovitch & Z. Koren, “Influence of Porosity and Casting Conditions on Creep of Die-Cast Mg Alloy”, Journal of Materials Science Letters, 17, 1998, 1787-1789, 23. C. Jackson et al., Editor, Nondestructive Testing Handbook 3rd Edition, Volume 1 Leak Testing, American Society For Nondestructive Testing, Columbus, OH, 1998 24. Xiaoping Niu, Mathew Anthony and Paul Tuzi, “High Pressure Leak Tightness Improvement by a Unique Local Squeezing Process”, Proc. 2008 NADCA Congress, Paper no. T08-082 25. William L. Westbrook, “The Value of Impregnation to the Die Caster”, Die Casting Engineer, May, 2010, p 30 26. Tom Shantz, “The Basics of Vacuum Impregnation”, Die Casting Engineer, November 2012, p 36 27. Tom Shantz & Ralf Versmold, “Types of Vacuum Impregnation Processes”, Die Casting Engineer, January, 2013, p 41 28. Tom Shantz & Ralf Versmold, “Vacuum Impregnation System Technology”, Die Casting Engineer May, 2013, p24 29. Tom Shantz & Ralf Versmold, “Types of Vacuum Impregnation Sealants – Part 4”, Die Casting Engineer, January, 2014, p 28 30. Mark Cross, “Casting Impregnation: Modern Processes”, Die Casting Engineer May 2013, p 44 31. Stephen Hynes, “Transformation of Casting Impregnation”, Die Casting Engineer July 2016, p 20 32. Peter T. Gebhard, “Die Cast Porosity Control: the Part vs. the Process”, Die Casting Engineer, November 2007, p 38 33. Stephen P. Midson, “Minimizing Blistering during T6 Heat Treating of Semi-Solid Castings”, Die Casting Engineer, November 2011, p 40

42

Die Casting Porosity Guidebook NORTH AMERICAN DIE CASTING ASSOCIATION

3250 N. Arlington Heights Rd., Ste. 101 Arlington Heights, IL 60004 T: 847.279.0001 • F: 847.279.0002 www.diecasting.org