Volume 07 - Powder Metal Technologies and Applications Part 11 docx

hf = tkr In this equation, t is the part thickness, kr is the compression ratio, and kr = q/qa, where q is the part required compaction density and qa is the apparent density of the loose powder If the fill height is greater than the maximum fill height that can be accommodated in the press selected on the basis of the compacting load required, a larger capacity machine should be selected, which has the required fill height capacity Powder Metallurgy Presses and Tooling Revised by John Porter, Cincinnati Incorporated Tooling Systems High-production P/M compacting presses are available as standard production machines in a wide range of pressing capacities and production rate capabilities Presses are designed to produce parts of a specific classification, as discussed previously Single-action tooling systems generally are limited to production of class I parts During the compacting cycle, the die, core rod, and one of the punches (usually the lower punch) remain stationary Compacting is performed by the moving punch, which is driven by the action of the press One or more core rods may form any through holes in the part During ejection, the upper punch moves away from the formed part, and the part is ejected from the die by the lower punch The core rod (Fig 7) is stationary, and the part is ejected from the die and core rod simultaneously On some presses, the core rod is arranged so that it is free to move upward (float) with the part as it is ejected The compacted part experiences slight elastic expansion on ejection from the die, which causes the part to free itself from the core rod The core rod is then free to move downward to the fill position This floating core rod arrangement reduces ejection forces and core rod wear Fig Compacting sequence utilizing single-action tooling Dashed line indicates motion of lower punch Double-action tooling systems primarily are used to produce class I and II parts Force is applied to the top and bottom of the part simultaneously, because the punches have the same travel rate The die and core rod are stationary Densification takes place from the top and bottom, with the lowest density region near the center of the part Although the core rod is fixed in this system, it can be arranged in a floating position Figure shows the compacting sequence of a double-action tooling system Fig Compacting sequence utilizing double-action tooling Dashed line indicates motion of component parts Floating die tooling systems are similar to double-action arrangements As shown in Fig 9, the die is mounted on a yielding mechanism (springs) However, pneumatic or hydraulic cylinders usually are used, because they offer an easily adjustable resisting force As the upper punch enters the die and starts to compact the powder, friction between the powder and die wall causes the die to move down This has the same effect as an upward-moving lower punch After pressing, the die moves upward to the fill position, and the upward-moving lower punch ejects the part The core rod can be fixed or floating Fig Compacting sequence utilizing floating die tooling Dashed lines indicate motion of component parts Withdrawal tooling systems use the floating die principle, except that the punch forming the bottommost level of the part remains stationary and that the die motion is press activated rather than friction activated The die and other lower tooling members, including auxiliary lower punches and core rods, move downward from the time pressing begins until ejection is complete Figure 10 shows the compacting sequence in a multiple-motion withdrawal tooling system During compaction, all elements of the tooling system except the stationary punch move downward The die is mounted on the top press member of the platen and is supported by pneumatic or hydraulic cylinders Auxiliary punches are mounted on additional platens, which are similarly supported and have positive pressing stops The stops control the finished length of each of the levels within the compacted part Before ejection, these stops are released or disengaged so that the platens can be moved further downward During ejection, the upper punch moves upward, away from the compact, while the die and lower punches move sequentially downward until all tool members are level with the top of the stationary punch The compact is fully supported by the tooling members during ejection, resting on the stationary punch as the die and lower punches are lowered to release it Fig 10 Compacting sequence utilizing floating die withdrawal double-action tooling Dashed lines indicate motion of component parts The core rod can be provided with pressing position stops to allow a part to be produced with blind or counterbored holes The core rod is held stationary until the part is free of all other tooling members before moving downward to the ejection position At this point in the machine cycle, the feeder moves across the die, pushing the compacted part from the die area and covering the die cavity The die and auxiliary lower punches move upward to their respective fill positions The core rod then moves upward, displacing the excess powder into the partially empty feed shoe The feeder retracts, wipes the top fill level, and readies the press for the next cycle Powder Metallurgy Presses and Tooling Revised by John Porter, Cincinnati Incorporated Types of Presses Anvil presses generally are limited to compaction of class I parts in a single direction Anvil presses not have an upper punch; a moveable, solid, flat block seals the top of the die Compacting is done by the lower punch, which, after the anvil is released and moved, moves farther to eject the compact from the die Anvil presses are available with pressing capacities ranging from 6.7 to 310 kN (0.75 to 35 tons), with maximum depth of fill ranging from to 75 mm (0.040 to in.) Multiple-cavity pressing frequently is used in anvil presses, with possible production rates of >100,000 pieces per hour Some anvil presses can be converted to double action, using an upper punch entry system Anvil presses usually are mechanically driven Figure 11 shows a schematic of an anvil press operation Fig 11 Compacting sequence utilizing sliding anvil single-action tooling Dashed line indicates motion of component parts Rotary presses generally are limited to compaction of single-level class II parts, although some class III parts, such as flanged bushings, are produced Rotary machines are available with pressing capacities ranging from 36 to 590 kN (4 to 66 tons), with a depth of fill up to 75 mm (3 in.) Production rates of >60,000 pieces per hour are possible, depending on machine size and the number of tooling stations Rotary presses are mechanically driven Single-Punch Opposing Ram Presses Like rotary presses, these machines are limited to production of class II and some class III P/M parts These presses are available in top- and bottom-drive models, with pressing capacities ranging from 36 to 980 kN (4 to 100 tons) and with a maximum depth of fill up to 100 mm (4 in.) Production rates of up to 3000 parts per hour are possible using mechanical presses with single-cavity tooling, although production rates of 900 to 1800 pieces per hour are more common Hydraulic presses produce 900 pieces per hour Ejection of the part is accomplished by the lower punch moving upward Mechanical and hydraulic presses are available Single-punch withdrawal presses have essentially the same partmaking capabilities as the single-punch opposing ram system in terms of pressing capacity, depth of fill, and production rate The major difference is that floating dies are used to achieve top and bottom pressing The die is moved downward to eject the part Multiple-motion die set presses can be designed to produce the most complex P/M parts These presses use floating die and withdrawal tooling methods Machines are available with either bottom- or top-drive arrangements Pressing capacities range from 27 to 7830 kN (3 to 880 tons), with a maximum depth fill of 180 mm (7 in.) Production rates vary from more than 6000 pieces per hour on smaller machines to 1800 pieces per hour for 1960 kN (220 ton) presses In addition to producing complex parts, the removable die set (tool holder) minimizes press downtime for part changeover if the die set for the next part to be produced is set up outside the press and is ready for installation Pressing position for each level being produced by a separate tooling member is controlled by fixed-height tooling blocks (stop blocks), which usually are ground to the proper height to produce a given dimension on the part A small adjustment in the block mounting member allows for minor changes to part dimension Full range adjustments are available on more recent presses Multiple-motion adjustable stop presses have the same partmaking capability as multiple-motion die set presses and use the same tooling methods Pressing capacities range from 980 to 7340 kN (110 to 825 tons), with a maximum depth of fill of 150 mm (6 in.) These presses not incorporate removable die sets; however, press stop positions are adjustable, and a change in any dimension of the part in the direction of pressing is easily accomplished Powder Metallurgy Presses and Tooling Revised by John Porter, Cincinnati Incorporated Advanced Tool Motions A common limitation of some rigid tooling systems is that part features not perpendicular to the direction of pressing cannot be compacted and stripped Frequently, it is cost effective to form features such as cross holes and threads by machining Other nonperpendicular features, notably helix shapes and hidden flanges, can be formed using complex tool motions Another type of advanced tooling system permits production of complex shapes with magnetic orientation of the microstructure Helical shapes, typically helical spur gears, are produced in rigid compaction tool sets with punch rotation capability In a simple system, a helical form lower punch is engaged in a die with a matching gear form In such a system, the lower punch remains engaged in the die at all times, as is common practice for all rigid tool systems, so that indexing rotation of the punch to the die is avoided The die acts as a guide Rotation is carried out on a thrust bearing, which rests on the punch platen that supports the lower punch An upper punch is not required, because the top of the die cavity is closed by an upper anvil, which does not enter the die cavity Central core rods, with or without additional features such as splines and key forms, are commonly operated in this helical tool system Helical gears made in this manner are limited to helix angles of 25° and a thickness of 32 mm (1 in.) due to fill limitations along the helix tooth form More complex helical gear tooling systems have been developed for routine production using helical upper punches, driven by follower cams for indexed die entry, with inner and outer lower helical punches for stepped helical gears Split Die Systems Another rigid tooling system that avoids some through-cavity limitations is known as the split die, or "double die," system It enables the compaction of parts with completely asymmetric upper and lower sections in the pressing direction Figure 12 shows typical tool motions in split die compaction This system requires two die-holding platens to carry the upper and lower die Each platen is controlled and moved independently Fig 12 Split die compaction sequence Wet magnetic compaction (Fig 13) has enjoyed wide usage in the production of magnetically oriented ferrite shapes In this production process, a feed shoe is not required Instead, the die cavity is injected with an aqueous slip (slurry) that has a high concentration of ferrite powder, with the addition of green binders as required Typically, the die filling pressure is 35 MPa (5000 psi) By using an aqueous slip, many of the gravity die fill problems, such as attainment of uniform powder density and filling the areas that are difficult for the powder to reach, are avoided Fig 13 Wet magnetic compaction (a) Force-time diagram for magnet presses (b) Schematic of press tool for chamber-filling method designed for withdrawal operation Following die fill injection, an orienting magnetic field is applied to the slip, resulting in magnetic polarization of the individual ferrite particles, which remain mobile at this point The optimal orientation of the ferrite particles directly determines the quality of the finished permanent magnet After magnetic orientation, the main pressing load is applied, densifying the ferrite mass and causing the suspending aqueous carrier to be expelled through drainage ports The compact is imparted with the precision shape and dimensions of both the upper and lower dies, plus any core rods that may be inserted The cycle is completed by separation of the press platens and ejection of the compacted ferrite shape Powder Metallurgy Presses and Tooling Revised by John Porter, Cincinnati Incorporated Tooling Design Traditionally, P/M tooling was designed on the basis of production experience In simple parts, such as single-level class I and II parts, these determinations proved successful As state-of-the-art materials and presses advanced to the production of complex, multilevel parts, the "cut-and-try" method of tool design became obsolete The high cost of complex tooling and adapters, plus downtime to redesign and rebuild tooling, requires the partmaking system, including the press, to be carefully analyzed in terms of load, stress, and deflection Tooling layout is required to design a suitable set of tools and to determine the physical dimensions (length and thickness) of tooling members A preliminary layout helps to determine fill, pressing, and ejection positions and to eliminate interference at these positions The die space drawing supplied with every compacting press, which usually starts with the ejection position, is the basis of the tooling assembly layout Generally, tooling members are never closer than in the ejection position, which constitutes the minimum space available to contain all components and their adapters Die Design Dies are commonly constructed by using inserts that are held in the die case by shrink fitting The amount of interference between the insert and the die case depends on the inside and outside diameter of each member and on the compacting pressure used The powder can be considered a fluid in a closed container that transmits the compacting pressure in all directions; therefore, the die must be designed as though it were a pressure vessel with internal pressure In actual practice, radial pressure on the die walls due to compacting rarely exceeds 50% of the compacting pressure The interference fit of the die case and die insert should be such that the stress on the insert always remains in compression for round dies However, for shaped dies such as gears, cams, and levers, the use of finite element analysis is the best method for accurately determining stress and deflection In P/M tooling, the die normally controls the outer peripheral shape and size of the piece part Typically, it is constructed from materials such as tungsten carbide or high alloy tool steels, such as T15, D2, CPM-10V, or CPM-15V with high hardness and good wear resistance Dies are usually constructed in one or more sections and compressed into a retaining ring made of a low-alloy steel, such as AISI 4340 or 6150 Considerations in die design and material selection include initial tool cost, shear strength of the die material, and die shape A large die may require tungsten carbide, which costs ten times as much as tool steel materials Tungsten carbide may be the best material for a set of gear tools with a relatively steep helical angle Sectional die construction may be required for specific shapes such as sharp corners or projections into the die cavity Die Wall Thickness An exact calculation of the stress on die walls is almost impossible from a practical point of view because stress distributions in the compact are extremely complicated and include variables such as part shape, particle size distribution, and other factors that affect transmission of compressive stress in the lateral direction (Ref 2) The vertical axial load can exert a horizontal force after a certain degree of consolidation has been attained For example, when a simple shape is compacted at 400 MPa, as much as 120 MPa pressure can be exerted radially against the die walls If for purposes of simplification, the internal pressure is considered strictly hydrostatic in nature and the confined material is an incompressible liquid, then the die wall thickness for a cylindrical die could be determined by using Lame's formula: where S is the maximum allowable fiber stress for the material of the die, D is the outer diameter of the die, d is the compact diameter, and p is the radial stress acting on the die wall This is a simplification because during metal powder compaction the pressure is not hydrostatic and the material is not incompressible Initially, the powder is compressed with a consequent reduction in the vertical height of the space filled by the powder The compressed material begins to resemble a solid after a certain degree of compaction has been reached Poisson's ratio is 0.3 for fully dense and isotropic steel While this wrought form value cannot apply to powder metal, it is assumed to be applicable in the fully compacted condition Thus, the Poisson's ratio is introduced into the previous equation, and the following modified Lame's formula is used for estimating the die wall thickness for metal powder compaction where = Poisson's ratio = 0.3 This formula, however, does not take into consideration that the internal pressure acting over the length of the compact is balanced by the strength of the die having a larger length The formula does address the friction at the tooling/powder interfaces resulting in nonuniform pressure distribution in the compact Generally speaking, the formula produces more conservative results than are necessary The interference fit between the shrink ring and the die insert should be such that the stress on the insert always remains compressive for round dies For shaped dies such as those used for production of gears and cams, the use of finite element analysis is the best method for accurately determining the stress and deflection Core Rods Basically, the core rod is an extension of the die that controls the inner peripheral shape and size of the piece part Tungsten carbide and M2 or M4 high-speed steels are the most common materials used for core rods Primary factors in materials selection include wear resistance and hardness, which enable the core rod to resist the high compressive force exerted during compaction and the abrasive action sustained during part ejection Core rods >25 mm (1 in.) in diameter or area are held to a base by mechanical means, such as a screw, while smaller core rods are held by means of silver solder or braze Punches can perform the function of a die or a core rod and carry the full load of the compressive force required to compact the P/M part Wear resistance and toughness are the most important factors in materials selection The most commonly used materials are A2, D2, S7, and H13 tool steels Dimensional control, especially in areas such as concentricity and hole-to-hole location, depends on the amount of clearance that can be maintained between the punches, die, and core rods Clearance should be calculated for each specific range and size of part It is important to note that thermal size changes occur during operation, primarily because of the friction created by stripping the compacted part and the speed of the pressing cycle Punch Component Stress Compacting powder causes compressive stress in the punch This stress must be below the yield strength of the punch material Calculation of buckling stability should be made for long, thin-walled punches Figure 14 shows the effect of axial compressive force on a tubular punch A tubular punch is subjected to internal pressure during compacting of multilevel parts In this case, the resulting circumferential tensile stress in the punch wall should be calculated If the stress and accompanying deflection is excessive, tooling clearances should be designed so that when the outer punch wall expands, it is supported by the die wall before the stress reaches the yield limit (Fig 15) Fig 14 Effect of compressive stress on tubular punch Fig 15 Tensile stresses in a tubular punch during compacting Large arrows indicate action of powder on walls of punch During ejection, the punch is subjected to compressive stresses by resisting the stripping action of the die and to tensile stresses from the stripping action of punch These stresses normally are lower than compacting stresses Components of the punch subjected to stress include the punch clamp ring and bolts, which should resist the ejection of the punch without permanent deformation Punch adapters are subjected to bending loads that create a tensile stress around the center hole during compacting This stress should not exceed the fatigue limit of the adapter material stringent This would apply to air-conditioner compressor components and some hydraulic pump parts The process stages are: • • • • Draw vacuum in process tanks to remove air from pores Transfer sealant from storage tank to process tank and submerge parts Release vacuum and pressure process tank with compressed air pressure Remove parts, wash, and cure Fig Dry-vacuum-pressure impregnation In pressure injection the part is fixtured to seal passageways to internal cavities (Fig 5a) Then it is pressurized with a sealant that flows through the porosity network until emerging on the exterior (Fig 5b) This treatment is applied on automotive engine cylinder liners and other components in which pressure retention of fluids and gases is required Fig Pressure injection impregnation (a) Part is fixtured to direct resin flow through the part (b) Pressurized sealant flows through porosity network Impregnation resins are mainly combinations of high-boiling-point methacrylate monomers formulated to meet the physical and chemical property demands of the impregnation process The materials are low-viscosity fluids, which are cured within the porosity to form a polymer solid The solid is achieved by formulating the monomer blend so that the resulting polymer has a cross-linked structure that is inherently inert and resistant to temperature, pressure, and process fluids The properties of the polymer are controlled by the blend of monomers used and the inclusion of minor ingredients such as corrosion inhibitor and antioxidants Key monomer properties are: • • • Low vapor pressure for vacuum methods Low viscosity for penetration (7 to 10 cP) Low surface tension for wetting and adhesion Important properties of the polymer after curing are: • • • Temperature resistance Chemical resistance Shrinkage on polymerization • • • Adhesive strength Tensile strength Hardness Temperature and chemical resistance are key factors Typically, the temperature ranges of modern organic impregnation sealants are -55 to 200 °C (-65 to 400 °F) Pressure retention upper limits are governed by component designs Many wP/M components used in hydraulic applications are subject to constant conditions of more than 65 MPa (10,000 psi) with elevated operating temperatures Low shrinkage is required for good sealing performance in single impregnation, while good adhesive strength between sealant and metal contribute to performance under conditions such as temperature cycling and physical flexing As previously noted, there are four types of materials used in P/M impregnation Sodium silicate is rarely employed, polyester resin uses toxic elements, and heat-cure resins are complex and difficult to handle The most common resins are anaerobic sealants Sodium Silicate Also known as waterglass, this inorganic material was the first widely used impregnation sealant Although still found in some applications, particularly where extremely high temperatures are encountered, waterglass has been largely abandoned for most porosity-sealing procedures Problems include slow processing, difficult cleanup of parts, and lengthened solidification of sealant, allowing parts to leak long after impregnation Because of these difficulties, sodium silicate often requires multiple impregnation cycles Polyester Resin (Styrene) In use since the 1940s, this organic material provides reliable sealing, has good chemical resistance, and can operate well at temperatures up to 200 °C (400 °F) However, substantial disadvantages related to costs, productivity, and environmental health and safety, along with waste disposal issues, have virtually disqualified these resins from most industrial facilities Heat-Cure (Elevated-Temperature-Cure) Resins This modern approach to porosity sealing uses various forms of organic methacrylate or polyester resins cured at temperatures below the boiling point of water Sealants in this category generally are lower in viscosity than other styrene resins and often are applied with much simpler impregnation equipment Basically, hot-water curing at 90 °C (195 °F) satisfies the requirements of manufacturers who need porosity sealing more effective than spray sealing without the maintenance requirements of an anaerobic system All heat cure systems require a sealant catalyst to render the sealant reactive These additives typically are hazardous and must be measured and mixed with care in order to achieve desired curing Many heat-cured materials also require a strong detergent solution for satisfactory washing of excess sealant from parts Some bleedout of sealant almost always occurs as parts heat up and the impregnation material expands during cure, with a resultant reduction in sealing integrity and possible fouling of surfaces Also, an operator must be present to ensure that parts remain in the heated water long enough to reach proper temperature to completely cure the sealant Failure to so can result in parts leaving the sealing operation with inadequate curing If this occurs the polymer will not fully cross link, and sealant performance is affected adversely Self-Curing Anaerobics Anaerobic impregnation has consistently proven to be the most effective porosity sealing system These organic materials self-cure at room temperature, eliminating the need for a hot-water treatment This precludes sealant bleedout, fouling of surfaces, rework of parts, and the need for operators at cure stations A highly cross-linked product, it will not liquefy and weep out of porosity Many manufacturers rely on the superior sealing capability and consistent performance of anaerobics to ensure that all parts reaching the assembly line are leak free, with no testing before or after impregnation When impregnated into a part, the anaerobic no longer has a source of stabilizing air and, thus deprived of its oxygen supply, begins to solidify chemically Prior to that, polymerization is prevented by light aeration at constant temperatures in the sealant storage vessel Contact with metal promotes curing, as does application of heat and chemical activators, but none of these are essential to the process They regulate the speed of cure by controlling the rate of reaction, allowing the sealant system to be tuned for optimal results in each application A unique activator rinse gives anaerobics much greater versatility in sealing a wide range of pore sizes, especially larger ones Anaerobic impregnation is generally recommended for volume manufacturing where the highest-quality control standards are required for sealing parts The unique, self-curing capability of anaerobic sealants, along with the ability to regulate the rate of cure, has made this material the overwhelming choice for impregnating P/M parts There are several anaerobic impregnation techniques such as wet vacuum, wet vacuum/ pressure, dry vacuum, pressure injection, and spray sealing Resin Impregnation of Powder Metal Parts Charles M Muisener, Research, Development & Engineering Group, Loctite Corporation Performance Plastic impregnation generally has little or no effect on tensile strength and ductility, and often resin sealants for impregnated parts have very similar physical properties Therefore, sealants are often considered roughly equivalent in terms of component strength and machinability However, there are important performance differences that have arisen with further developments and evaluation For example, developments in improved sealing performance have coincided with adhesive strength improvement from to 50 N/cm2 (Ref 1) Tests of the radial crushing strengths for unimpregnated and impregnated components using different impregnants also show changes in both strength and ductility (Table 1, Fig 6) Resin has a 10% improvement in yield strength, while resin shows improved ductility The resin that improves the strength has good adhesive properties, whereas the resin improving ductility forms the harder polymer The harder polymer is more brittle, yet the components impregnated with resin exhibit more pronounced brittle failure (Ref 1) Table Radial crushing strength (K) comparison Sample Maximum load,N Unimpregnated Resin Resin 6504 7178 6768 Source: Ref Strength (K), MPa 258 285 269 Extension to maximum load, mm 0.382 0.376 0.352 Minimum deformation, mm 0.432 0.414 0.472 Fig Radial crushing strength of impregnated and unimpregnated P/M part Source: Ref Corrosion Resistance Corrosion and surface blemishes are a chronic problem in P/M parts Pits, blisters, stains, and other imperfections break out because corrosives and industrial solvents are absorbed into pores Even after surfaces have been treated with protective coatings and platings, the effects of internal corrosion may not appear until well after other such treatments as tumbling, spraying, painting, polishing, cleaning, and anodizing have been completed (Fig 7) Fig Surface condition of P/M part after plating Without a protective thermoset plastic impregnant, corrosive solvents soak into pores, eventually seeping upward and damaging the most hardened and lustrous surfaces The fluids involved in P/M applications are numerous, including: • • • • • • • • • • • • Ethylene glycol Water Lubrication oil Jet fuels Carbon removal compounds Motor oils Hydrocarbon fluids Hydraulic fluids Machining fluids Acids Gasoline Alkaline cleaners Machining Benefits One of the more remarkable developments is the demonstrable improvements in machinability with impregnated metal parts versus those that remain untreated The precise reason for this result is unclear, but the consensus points to a natural lubricity present in thermoset plastic resins used in most P/M impregnation processes These cured polymers minimize tool chatter or vibration, heat buildup, and the interrupted cut associated with the machining of P/M parts They also reduce chip thickness and adhesion, improve surface finishes, help achieve consistent finish dimensions, and improve dimensional control of parts By filling voids in the porous metal, the impregnating material promotes better chip formation and separation by the cutting tool The plastic fill cushions the tool as it passes through metal, giving the tool edge an uninterrupted feel, thus extending its service life through a reduction in cutting force Many manufacturers impregnate parts solely to derive machining benefits For example, a major equipment manufacturer verified benefits from impregnating parts with thermoset resins including higher-quality electroplating, pressure-tight sealing of hydraulic and pneumatic parts, and major improvements and cost savings in machinability Tool life doubled in tapping and turning operations and tripled on parts with high hardness Overall, perishable tooling costs dropped about 50% and, in some cases, the company is machining almost ten times as many pieces per tool Drilling is the most common machining operation used on P/M parts, and several studies (Ref 2, 3, 4, 5, and 6) confirm lower drilling forces and longer tool life with impregnated parts Some deterioration in surface roughness occurs (Ref 6), presumably due to melting of the impregnated resin Nonetheless, several research programs definitively support resin impregnation as a method to significantly improve machinability In one major program, exhaustive tests performed on various P/M alloys found that resins reduced drilling forces up to 75% in some cases (Fig 8, 9, 10, and 11) Fig Machining chart of P/M iron UI, unimpregnated; OI, oil impregnated; RIR, resin-impregnated resinol Fig Machining chart of P/M 304 stainless steel UI, unimpregnated; RIR, resin-impregnated resinol Fig 10 Effect of resin impregnation on drilling UI, unimpregnated; IR, impregnated resin Efficiency of improved machinability: 370% = 3.28 + 0.89 × 100 Test condition Density Cutting tool Value 7.0 g/cm3 Carbide, P20 Depth of cut, mm Feed rate, mm/rev Cutting speed, m/min 1.0 0.1 100 Fig 11 Effect of resin impregnation on turning Atomized iron powder SMF1015 impregnated with resinol UI, unimpregnated; IR, impregnated resinol References cited in this section M Potts and G Randell, Resin Impregnation of P/M Parts, Advances in P/M Parts Production, MPR Publishing, 1990, p 26.1-26.14 L.G Roy, G L'Espe'rance, P Lambert, and L.F Pease, Prealloyed Manganese Sulfide Powders for Improved Machinability in P/M Parts, Progress in Powder Metallurgy, Vol 43, Metal Powder Industries Federation, 1987, p 489-498 K.S Chopra, Manganese Sulfide in Machining Grade Ferrous P/M Alloys, Modern Developments in Powder Metallurgy, Vol 21, 1988, Metal Powder Industries Federation, p 361-380 S Nigarura, G L'Espe'rance, L.G Roy, A DeRege, and L.F Pease III, The Influence of Powder Processing on the Nature of Inclusions and Its Reaction to the Machinability of MnS Prealloyed P/M Parts, Advances in Powder Metallurgy & Particulate Materials, Vol 4, 1992, p 223-243 R.A Chernenkoff, D.W Hall, S Mocarski, and M Gagne, "Material Characterization of Powder-Forged Copper Steels," Technical Paper 901055, SAE International Congress and Exposition (Detroit, MI), Society of Automotive Engineers, March 1991 I Shareef, Machinability of Resin Impregnated Sintered Steel, Machining of Composites II, ASM International, 1993, p 159-169 Resin Impregnation of Powder Metal Parts Charles M Muisener, Research, Development & Engineering Group, Loctite Corporation References M Potts and G Randell, Resin Impregnation of P/M Parts, Advances in P/M Parts Production, MPR Publishing, 1990, p 26.1-26.14 L.G Roy, G L'Espe'rance, P Lambert, and L.F Pease, Prealloyed Manganese Sulfide Powders for Improved Machinability in P/M Parts, Progress in Powder Metallurgy, Vol 43, Metal Powder Industries Federation, 1987, p 489-498 K.S Chopra, Manganese Sulfide in Machining Grade Ferrous P/M Alloys, Modern Developments in Powder Metallurgy, Vol 21, 1988, Metal Powder Industries Federation, p 361-380 S Nigarura, G L'Espe'rance, L.G Roy, A DeRege, and L.F Pease III, The Influence of Powder Processing on the Nature of Inclusions and Its Reaction to the Machinability of MnS Prealloyed P/M Parts, Advances in Powder Metallurgy & Particulate Materials, Vol 4, 1992, p 223-243 R.A Chernenkoff, D.W Hall, S Mocarski, and M Gagne, "Material Characterization of Powder-Forged Copper Steels," Technical Paper 901055, SAE International Congress and Exposition (Detroit, MI), Society of Automotive Engineers, March 1991 I Shareef, Machinability of Resin Impregnated Sintered Steel, Machining of Composites II, ASM International, 1993, p 159-169 Planning and Quality Control of Powder Metallurgy Parts Production Jack R Bonsky, Cleveland State University, Advanced Manufacturing Center Introduction QUALITY CONTROL involves several basic principles, where the planning of a quality control program requires an understanding of process variables and the reality of statistical variation in any process or measurement Therefore, this article attempts to first summarize the basic concepts of statistical process control as a tool to measure, quantify, and analyze inherent variations This understanding is considered essential in present-day industrial manufacturing In addition, the unique aspects of quality control planning in P/M parts production are also discussed in this article Quality control for P/M parts production involves several process factors such as powder properties, press settings, tooling designs, and furnace conditions These factors are discussed in terms of their impact on process variations and quality improvement This article also discusses the methods that address P/M quality control in terms of: • • Defect control by prevention or detection Dimensional control for form and fit requirements Defect and dimensional control are two major quality measures for P/M production, and each has a unique set of attributes for inspection and quality assessments Finally, it should also be noted that quality control also depends on an effective work environment that facilitates involvement and communication This basic factor is often exemplified by the 14 Points for Management by W Edward Deming, the eminent statistician who championed the role of statistical process control for industrial application His 14 Points are: • • • • • • • • • • • • • • Create constancy of purpose for the improvement of product or service Adopt the new philosophy of process control and variation reduction Cease dependence on mass inspection for quality control End the practice of awarding business on the basis of price Improve constantly and forever the system of production and service in order to improve quality and productivity and thus continuously decrease costs Institute thorough and better job-related trainings Institutionalize leaderships Drive out fears, so that everyone may work effectively for the company Break down barriers between departments Eliminate slogans, exhortations, and targets for the workforce that ask for zero defects and new levels of productivity Eliminate work standards on the factory floor Remove the barriers that rob employees at all levels in the company of their right to pride of workmanship Institute a vigorous program of education and self-improvement Put everybody in the organization to work to accomplish the transformation The 14 Points define essential institutional elements of quality and productivity improvement through statistical thinking and methods It should also be noted that in the spirit of his 14 Points, Deming refines and improves these tenets Acknowledgements Portions of this article are adapted from the following articles: • • Eric Rasmussen and David Zenger, Implementing Statistical Process Control for Powder Component Production, Characterization, Testing and Quality Control, Advances in Powder Metallurgy and Particulate Materials 1994, Vol 2, Metal Powder Industries Federation, p 252 Richard DeVor and Tsong-how Chang, Statistical Quality Design and Control, Nondestructive Evaluation and Quality Control, Vol 17, ASM Handbook, 1989, p 719-753 Planning and Quality Control of Powder Metallurgy Parts Production Jack R Bonsky, Cleveland State University, Advanced Manufacturing Center Statistical Process Control Concepts Statistical process control (SPC) is simply a method for monitoring the statistical variations of a process, where the objective is to control or reduce variations between upper and lower process limits Control charts (Fig 1) are used to monitor the statistical variations, but this is just one part of the overall SPC method For the techniques of SPC (specifically Shewhart control charts such as the schematic in Fig 1) to be successfully employed as an off-line problem identification and problemsolving tool, it is essential to keep in mind that it is a three-step process, as follows: Use statistical signals to find improvement opportunities through the identification of process faults Use experience, technical expertise, and fault diagnosis methods to find the root cause of the fault that has been identified Develop an action plan to correct the fault in a manner that will enable any gains that are realized to be held This three-step process can be explained by using the classical feedback control system perspective, as shown in Fig There are five distinct stages in the generic control loop (Fig 2), which facilitate the three-step process in the following way: Use of statistical signals (observation and evaluation) Fault diagnosis (diagnosis) Action plan (decision and implementation) Fig Impact of having process initially in a state of statistical control versus improvement resulting from a breakthrough in performance Fig Classical feedback control system view of SPC implementation However, bringing a process into a state of statistical control does not necessarily mean that a fundamental improvement has been achieved Clearly, a bad situation has been rectified by bringing the process into control, and quality and productivity are enhanced However, bringing a process into control simply means that the process is back to where it should have been to begin with At this point, it is then possible to begin to assess the present ability of the process to realize the potential it was initially intended to have It may be failing to realize this potential because the implementation of the process is flawed or because the design of the process itself is flawed In either case, the root cause(s) of the chronic common cause problem must be identified and removed at the system level This constitutes a breakthrough in performance; that is, an improvement in the process has taken place The results of the essential steps leading to such a breakthrough are shown in Fig This article focuses primarily on the key statistical concepts and definitions that are essential for appreciation of the SPC methods There are several important concepts that should be understood, including the requirements of rational sampling and the definition of measurements by attributes (or defects) Once these basic definitions are established, the next step in the three-step process is an appreciation of the root cause or fault diagnosis for variations in P/M processing The factors related to fault diagnosis for P/M quality control and planning are discussed in the section "P/M Process Planning" in this article Shewhart Control Chart A succession of parts emanating from a process under statistical control will exhibit variability in their measurements because of a constant set of common causes These variable measurements tend to collect into a predictable pattern of variation that can be easily described by a few simple statistical measures, namely, a mean ( ), a standard deviation ( ), and a frequency distribution (normal distribution, Poisson distribution, etc.) These measures stand as a model that predicts statistical behavior if the process is subject only to a constant set of common causes Many years ago, Dr Walter Shewhart showed how data from a manufacturing process could be developed and interpreted through the use of very simple but useful statistical methods Given a sufficient number of samples, the output of any process operating solely under a set of common causes can be shown to exhibit a normal distribution Therefore, when taking a series of samples from a process that is in control, one may expect the observed values to fall within ±3 standard deviations of the process mean in 997 out of 1000 cases By combining this logic with a simple system of charting measurement data versus time, an operator may easily identify the occurrences of special causes If a point appears outside the standard deviation limits, it is very likely that a special cause has had some effect on the process The charts generated by this system are known as Shewhart control charts Each control chart actually consists of two charts for each characteristic being tracked Figure shows an example set of control charts Each point on the charts represents a group of parts, called a sample subgroup, or just "sample" for short Samples typically consists of three to six parts, though all samples for a given chart are the same size The upper chart, called the (X-bar) chart, shows the history of the process mean The lower chart, called the range chart, or R chart for short, shows the history of variation within a sample By using the and R charts, significant shifts in both the process mean and spread can be detected as an indication of a process out of control Fig Control charts for diameter measurements of a part (a) -chart (b) R-chart Data are for k = 20, n = The lines marked UCL and LCL represent the upper and lower control limits, respectively These control limits are based on the mean value of the variable being charted ±3 standard deviations In addition, the standard deviation used for calculating the control limits is based on variation within sample subgroups and not on variation within the entire population of parts sampled Because each point appearing on an chart is really the mean of several individual measurements, it would be improper to compare it to the limits for the entire population For this reason, it is essential that an R accompany the chart, so the R values for each sample subgroup can be compared Once statistics of and R are assembled for a number of sample subgroups, then charts for and is the average range and can also be assembled, where is the average of from several sample groups, When calculating initial control limits, a large group of samples, consisting of at least 25 sample subgroups, is measured and the means and control limits calculated If any points fall outside the control limits for the R chart, and they can be attributed to some special cause, then these points are eliminated from the chart and the control limits for both charts are recalculated Next, this process is repeated with the chart Once the effects of all known special causes have been eliminated, the resulting control limits are considered to be the standard for future samples Thus, the means and control limits are not recalculated with the addition of each new sample to the control charts However, if the process experiences some improvement that causes a sustained shift on either the or R chart, then the control limits should be recalculated based on the newer data In addition to the test for points appearing outside the control limits, there are a number of other items that can be applied to determine if a process is exhibiting normal random variation Two common additional tests are the run test and the trend test If there are eight or more consecutive points on one side of the mean for either chart, it is considered a run If a run is detected, it is likely that there has been a shift in the process mean or variability (Ref 1) If six or more consecutive points exhibit a continuing upward or downward trend on either chart, it is considered a trend If a trend is detected, it is likely that the process is drifting (Ref 2) Calculations for Setting Up and R Control Charts Once the statistical basis for Shewhart control charts has been established, the first step in setting up and R control charts is the selection of the samples As previously noted, it is important that all samples be rational samples so that variation is attributable only to one constant system of common cause Sampling from different machines, sampling over extended periods of time, and sampling from product combined from several sources are all nonrational sampling methods and must be avoided Rational samples are discussed in more detail later in this section As a rule of thumb, 25 to 50 samples should be selected to provide a solid basis for the initiation of the control charts This helps to ensure more precise estimation of the process mean and standard deviation The sample/subgroup size should be relatively small (between n = and n = 6) With k rational samples of n each, the following steps can be used as a guide when constructing , R control charts with limits (the appropriate values for d2, A2, D3, and D4 in these calculations are obtained from Table 1): Calculate the sample mean and sample range for each sample using = X/n and R = Xmax - Xmin Calculate the grand mean of the n sample means and the average range using = /k and = R/k Calculate the control limits for the R-chart Although the true distribution of sample ranges is not normal and not symmetric, the symmetric limits are conventionally used for the R-charts With assumed normal distribution for the individual measurements, the following formulas can be used for the calculation of the control limits: UCLR = D4 and LCLR = D3 Calculate the control limits for the -chart Although the required standard deviation (the standard deviation of the sample mean, ) for setting the limits is X, this value is conveniently estimated by /(d2 3/(d2 ), where d2 is a function of n For limits, one uses a factor called A2, which is equal to ) and can be found in Table Thus, the control limits are calculated by: UCL X = + A2 and LCLX = - A2 Table Factors for Sample size, n and R control chart limits Factors for control limits R-chart -chart, D3 D4 A 1.880 1.023 0.729 0.577 0.483 0.419 0.373 0.337 0.308 0.285 0.266 0.249 0.235 3.472 0.212 0.203 0.194 0.187 0.180 10 11 12 13 14 15 16 17 18 19 20 0 0 0.076 0.136 0.184 0.223 0.256 0.283 0.307 0.328 0.223 0.363 0.378 0.391 0.403 0.415 3.267 2.573 2.282 2.114 2.004 1.924 1.864 1.816 1.777 1.744 1.717 1.693 1.672 0.347 1.637 1.622 1.608 1.597 1.585 Factor for calculating x from range (R), d2 1.128 1.693 2.059 2.326 2.534 2.704 2.847 2.970 3.078 3.173 3.258 3.336 3.407 1.653 3.532 3.588 3.640 3.699 3.735 Comparison of Tolerances and Control Limits It is important to clearly differentiate between specification limits and control limits The specification limits or tolerances of a part are: • • • • Characteristic of the part/item in question Based on functional considerations Related to/compared with an individual part measurement Used to establish the conformability of a part The control limits on a control chart are: • • • • Characteristic of the process in question Based on process variability Dependent on sampling parameters, namely, sample size Used to identify presence/absence of special cause variation in the process Control limits and tolerances must never be compared numerically and should not appear together on the same graph Tolerances are limits on individual measurements and as such can be compared against the process as a whole as represented by many individual measurements collected in the form of a statistical distribution Process Capability Indices (Cp and Cpk) Once a process is in statistical control, it is common to measure process capability in units of standard deviation for the process There are actually two measures, Cp and Cpk, both called the process ... Fe-50Ni Fe-2Ni-0.5C Fe-3Si Fe-29Ni-17Co Fe-16Cr-4Ni-4Cu Fe-17Cr-12Ni-2Mo-2Mn Fe-13Cr-1Mn-1Si Co-28Cr-4W-3Ni-1C Fe-32Ni-5Co Ti-6Al-4V Fe-6W-5Mo-4Cr-2V-1C W-4Ni-1Fe 93 97 95 94 96 94 99 98 96 96... Fe-14Cr-1Si-1Mn-0.3C Fe-18Cr-1Si-1Mn-1C Fe-5Cr-1.5Mo-1Si-1V-0.4Mn-0.4C Fe-6W-5Mo-4Cr-2V-0.3Mn-0.8C Fe-1.7Cr-0.8Mn-0.5Mo-0.4V-0.35C WC-10Co Hardness, HRC 50 57 50 61 30 80 Suggested applications Corrosion-resistant... 1 7-4 PH (HT) Stainless 316L Stainless 420 (HT) Stellite Super Invar Ti- 6-4 Tool steel Tungsten heavy alloy Fe-1Cr-0.4C Fe-2Ni-1Mo-0.4C Fe-2Cu-0.8C Fe-1Cr-0.5C Fe-50Ni Fe-2Ni-0.5C Fe-3Si Fe-29Ni-17Co