6 Causes of Molded-Part Variation: Processing Process variables cause significant mold shrinkage and warpage effects Melt temperature, injection pressure and speed, holding pressures and time, molding temperature, and demolding temperature contribute to successful molding conditions As is a recurrent theme in this book, the effects of these process variables are interactive not only with each other but also with materials, part design, and mold design variables The processing of plastics takes place under specific molding conditions set by the particular variables A framework for understanding the molding conditions and process variables that most affect shrinkage and warpage are examined in depth in this chapter 6.1 Molding Conditions Plastic materials have positive coefficients of thermal expansion and are compressible in the molten state As a result, the volume that a given mass of material occupies will change with both temperature and pressure Some general molding-condition considerations are applicable to volume change: The lowest possible plastic melt-temperature that permits good molded parts will tend to produce less shrinkage The smaller the temperature range between the molten plastic as it enters the mold and room temperature, the smaller the amount of thermal contraction and the less time available for crystallization Because plastic is compressible, the amount of holding pressure (after the cavity is filled) affects the shrinkage of the plastic part Note that the duration and effectiveness of packing is dependent, to a great extent, on the size and design of the gate and the runner system After the gate or runner freezes, no further benefit can be gained by continued application of packing pressure The higher the pressure in the cavity when the gate freezes, the greater the mass of plastic that is trapped in the mold and the lower the total shrink of the molded part Higher packing and holding pressures generally lead to a global reduction in mold shrinkage, while lower pressures increase shrinkage Unfortunately, pressures in the cavity vary from a maximum at the gate to a minimum at © Plastics Design Library the end of the flow, due to melt compressibility and viscosity The pressure differential over the length of the cavity can be very significant, particularly for longer flow lengths or thinner-walled parts This pressure-history differential, which occurs over the course of the cycle, results in mold shrinkage values that tend to be greater towards the end of the cavity compared to shrinkage closer to the gate area Differential mold-shrinkage due to cavity-pressure history differences can also lead to dimensional distortion or warpage of the molding Longer holding times during the cooling portion of the cycle cause the plastic to stretch a little in the mold, thus reducing apparent shrinkage The core or other details of the mold restrain shrinkage as long as the part is trapped in the mold This causes the plastic part to stretch and yield somewhat when the molding cycles are long The mold itself acts as a cooling fixture Hot molds increase mold shrinkage but reduce post-mold shrinkage Cold molds have the opposite effect Cold molds (cooling the plastic as rapidly as possible) reduce shrinkage, especially when molding crystalline materials However, they freeze-in some stresses that may be relieved later with time and exposure to elevated temperature Within some limits, a semicrystalline plastic will try to crystallize further, especially if it is exposed to elevated temperatures Semicrystalline thermoplastics are particularly influenced by the cooling rate The polymer chains in the melt are in a disorganized state (from a crystallization standpoint), and in solidification they form a dense structure With increasing crystallinity, the density and the shrinkage of the structure increases Extra cooling is required on core pins and inside corners of plastic parts to encourage the plastic to cool evenly Core pins and external corners of mold cores have more surface area exposed to heat per unit volume than other areas of the mold This causes greater heat loads on core pins and external corners of mold cores Ch 6: Causes of Molded-Part Variation: Processing 80 Where accurate parts are necessary, a molding machine in top-notch shape with well-calibrated temperature and pressure controls will give the most consistent parts The effects of various changes in molding conditions on molded parts are organized into a series of graphs labelled (a) through (o) in Fig 6.1.[31] Figure 6.1 Injection-molding machine settings can affect properties of thermoplastics.[31] (Reprinted by permission of HanserGardner.) Ch 6: Causes of Molded-Part Variation: Processing © Plastics Design Library 81 Graph (a) illustrates the effect on molecular orientation in response to changes in several other variables As mold temperature or cavity thickness increases, more time is available for stretched and oriented molecules to relax and reorient before the melt solidifies and freezes the molecular orientation As injection pressure and packing time increase, more stress and stretching of the molecules occur and are maintained closer to the freezing time This increases orientation in two ways The higher stress causes higher orientation in the first place The longer packing time maintains a low level of flow into the mold for a longer time, which maintains more orientation Initially, as melt temperature increases, the individual molecules have more freedom to align themselves with the flow of the material However, as the melt temperature rises further, the time the part remains molten increases after the part is filled out This allows more time for stress relief and molecular disorganization Graph (b) shows how pressure loss through the gate decreases as melt temperature increases because the plastic becomes less viscous as it is heated and is easier to push into the mold At low temperatures, it is hard to supply enough pressure to fill the cavity As stated elsewhere in this book, low shrinkage is associated with low injection temperatures As temperatures increase, so does shrinkage; so while the cavity is easier to fill at higher temperatures and filling and packing pressures are more effective at higher temperatures, eventually, the higher shrinkage associated with high melt-temperatures tends to overcome the filling and packing pressures, leading to higher shrink In graph (c), the falling weight (f.w.) impact strength of a part is increased as melt temperature increases The higher temperature allows for filling with lower stress and longer time for stress relief, both of which yield lower molded-in stresses and higher impact strength In graph (d), the effect of mold temperature on flow and cross-flow shrinkage is shown Higher temperatures allow more time for disorganization of the molecules and thus more stress relief In semicrystalline materials, higher mold temperature allows more time for crystallization and more shrinkage Graph (e) shows that part-weight increases with increasing packing or holding time, up to the point where the gate freezes After that time, more holding or packing time does not affect the part weight Higher melt and mold temperatures, as shown in graph (f ), allow more time for the material to conform perfectly to the mold surface, and to achieve a higher level of gloss © Plastics Design Library Higher melt temperature, graph (g), can affect the IZOD impact-strength in two ways First, longer exposure to higher melt-temperatures increases the heat history of the plastic and typically causes material degradation and a reduction of property values Under certain circumstances, higher melt-temperature causes increased molecular orientation (and molded-in stress) This means there is less give in the molecular structure before the molecules reach their breaking point Therefore the part is more brittle High IZOD impact-strength in polypropylene correlates well with successful application of the living hinge If a part has a low IZOD impact-strength, that implies low elongation before rupture It should be obvious that a great deal of elongation is necessary for a living or integral hinge application to be successful As the cavity thickness increases, graph (h), there is more time for the plastic to relax internal stresses A greater percentage of the thickness of the part will be in tension More time is available for crystallization of semicrystalline materials Thus there is more shrink Graph (i) shows that increasing packing time and pressure increases cooling time More material is forced into the cavity by increasing the packing time and pressure Since there is more material in the cavity, there is a slight increase in cooling time due to the increased mass of material to be cooled As mold temperature increases, graph (j), there is more time for crystallization to take place That means that, all other things being equal, the plastic will be denser because of the greater degree of crystallization Higher melt temperature, graph (k), reduces the stress on the material as it flows into the mold because the viscosity of the material is lower Also, the higher melt-temperature allows more time for any stresses to relax as the material cools Lower moldedin stress levels mean the material can withstand higher temperatures before it distorts In other words, higher melt-temperature equates to higher heat-distortion temperature The more material that is packed into the cavity before the gate freezes, the less the shrinkage will be Graph (l ) shows that as long as the gate is fluid, increasing packing time reduces shrink Once the gate freezes, more packing time is of no value The graph shows these effects for a restricted gate and for an open gate It also shows the effects of small (restricted) gates versus larger (open) gates The larger the gate, the easier it is for more material to flow into the cavity Thus, larger gates lead to less shrinkage even with very short packing time In addition, the larger gates will stay fluid Ch 6: Causes of Molded-Part Variation: Processing 82 longer, allowing more material to be packed into the mold before the gate freezes, which also reduces shrinkage Graph (m) shows that thick parts or short flowpaths fill easier and at lower pressure than thin parts or long flow-paths The lower the pressure required to fill the cavity, the less the clamp pressure required to hold the mold shut Conversely, long flow-paths or thin parts, requiring high injection-pressures, demand higher clamping pressure to prevent flashing at the parting line Graph (n) shows that as the distance from the gate increases, the density of the plastic decreases This is caused by two phenomena First, the plastic furthest from the gate is cooler than that at the gate; therefore it does not have as much time to crystallize Secondly, the pressure is higher at the gate than it is anywhere else Lower pressure away from the gate leads to lower density also Weld tensile strength increases with increasing injection pressure, as shown in graph (o), because the strength of the weld line is proportional to the force with which the two flow fronts are forced together Also, higher injection-pressure implies more shear at the gate and a higher melt-temperature when the flow fronts meet, allowing more time and better (higher temperature) conditions for some molecular migration across the front There are upper and lower limits for mold temperature, melt temperature, and injection pressure that can produce a fully filled part with no flash Figure 6.2 represents a typical upper and lower limit of injection pressure and mold temperature for a given melt temperature.[31] The area inside the curve is called the molding window Higher temperatures or pressures cause flash around the part Lower temperatures or pressures result in a short shot, that is, an incomplete part A more accurate representation of the molding window is a three-dimensional graph of conditions that permit the mold to fill without flashing Figure 6.3 shows such a three-dimensional window.[31] The size and shape of the window will vary with the design of the part, the mold, and the plastic being molded The rate at which the cavity is filled has some influence on the size and shape of this window, however the influence is relatively small It behooves the mold designer and the part designer to maximize the size of the molding window The larger the molding window, the more flexibility the molder has to take action to control shrinkage and warpage Figure 6.2 A molding “window” at a given melt temperature.[31] (Reprinted by permission of HanserGardner.) Figure 6.3 A three-dimensional representation of the “molding window.”[31] (Reprinted by permission of HanserGardner.) Ch 6: Causes of Molded-Part Variation: Processing © Plastics Design Library 83 6.2 Injection Melt Temperature The U-shaped curve in Fig 6.4 shows that shrinkage is higher at both high and low melt temperatures At low melt temperatures, the plastic barely fills the cavity before the gate freezes The pressure gradient from gate to end of flow is high and there is no significant time to pack-out the cavity The pressure at the end of flow is low, so the shrinkage is high At high melt temperatures, a lot of shrinkage is inherent as a result of temperature change The melt core is hotter when the gate freezes than it is at lower melt temperatures (unless the gate is the same thickness as the part) At very high melt temperatures, the holding time may end before the gate freezes This can happen when melt temperature is raised without increasing the holding-pressure time Both the high meltcore temperatures and the likelihood that the gate stays open past the holding time cause increased shrinkage At some midpoint, the melt viscosity is such that a good balance of pressure exists across the cavity with good cavity-packing when the gate freezes At this point, the shrinkage due to melt temperature is at a minimum Melt temperatures range from a low of about 350°F up to 700°F or more, depending on the plastic being molded 6.3 Injection Rate and Pressure The injection rate and the injection pressure are interrelated in the injection-molding process On older molding machines, a flow-control valve controlled the maximum rate of injection, but the minimum rate was determined by the injection-pressure setting The injection rate has a twofold effect: First, a slow rate of fill allows a thicker wall to build up as the material flows into the mold, thereby raising the pressure requirements to fill the mold A slow fill-rate results in cooler material at the end of the fill cycle A thicker, cooled wall causes a smaller flowchannel in which makeup resin flows, and a greater pressure-drop across the part during the holding phase of the molding cycle Cooler plastic can cause premature freezing at the gate, less effective packing, greater orientation, and more shrinkage Secondly, there is a certain amount of friction heating that occurs at the gate caused by the pressure drop across the gate Higher fill-rates raise the melt temperature in the cavity Higher melt temperature allows the injection pressure to be more effective in filling the cavity and all fine details within it Excessively high fill rates can cause plastic degradation and flash The density of crystalline polymers is inversely proportional to the distance from the gate, because pressures and temperatures near the gate are higher than they are at locations remote from the gate Higher pressure and temperature near the gate allow more time for crystallization and more packing time High injection-pressure forces the two advancing plastic flow-fronts, downstream of a core where the flow is divided around the core, into more intimate contact, which helps create a better bond at the weld line Inadequate cavity pressure can fail to hold the plastic against the walls of the cavity tightly enough to form a smooth skin Surface wrinkles are more commonly associated with low injection-pressure and low injection-rate The low rates allow a thin skin to form along the cavity walls This thin skin sometimes moves slightly with the flowing plastic causing a wrinkle The lower the cavity pressure and injection speed, the more pronounced the appearance of the wrinkle 6.3.1 Figure 6.4 The relationship between melt temperature and shrinkage © Plastics Design Library Injection Speed Injection rate, or speed, influences secondary heating of the plastic as it moves through the gate and into the mold Frictional heat is generated at the gate restriction, and between the flowing material in the center of the part and the material against the walls of the part that have already solidified There can be a significant temperature rise, primarily at the gate, depending on pressure and injection speed One study shows a peak temperature at the interface between the frozen plastic against the mold wall and the moving molten plastic in the center of the thickness of a plastic part.[4] Ch 6: Causes of Molded-Part Variation: Processing 84 An injection rate versus shrinkage curve is as shown in Fig 6.5 Figure 6.6 illustrates that if the fill rate is too slow, the material begins to cool before the cavity is filled and the pressure required to fill the cavity goes up Too low an injection rate inhibits the packing of the cavity because the material cools during the filling phase and the gate will freeze very quickly after the mold is filled This leads to higher shrinkage At the other extreme, if the cavity is filled too quickly, the pressure drop at the gate becomes excessive and the pressure required to fill the part goes up Shrinkage can increase because the temperature of the plastic in the cavity is likely to be significantly higher than the temperature at optimum fill rate The optimum fill rate is found near the minimum filling pressure The optimum fill rate for a part depends on its geometry, the size and location of the gate, the mold temperature, and the melt temperature Rapid fill rates often create a better surface finish; although rapid fill also can cause jetting and/or gate smear Warpage can sometimes be improved or eliminated by careful adjustment of injection speed and mold temperature When molding parts with thick sections and a relatively small gate, it is sometimes helpful to raise mold temperatures and reduce injection rates to delay gate freeze These changes usually sacrifice some surface gloss or finish, but yield an improvement in molded part shrinkage 6.3.2 Injection Pressure may be limited by the clamping capacity of the molding machine because the effective cavity pressure (less than the nozzle pressure) times the projected area of the part must not exceed the clamping pressure Excessive pressure can cause the mold to open and allow parting-line flash or even damage a structurally unsound mold There usually is some variation in cavity pressure with the highest pressure near the gate and the lowest at the last point to fill The variation in pressure decreases with the increase in material thickness, the increase in injection or holding pressure, the decrease in material viscosity, and the increase of injection speed (possibly due to the lower viscosity caused by higher frictional heating at the gate at high injection-speeds) Usually the material furthest from the gate solidifies first The frictional heat generated at the gate usually keeps the material closest to the gate somewhat warmer, causing that area to solidify last A warmer gate area often results in less shrinkage near the gate, which is caused by longer sustained pressure near the gate Differential shrinkage (warpage) is partly the effect of differential cavity pressure Note that molds have been bent by excessive injection pressure and area, combined with inadequate or concave platens on molding machines Molding machines can be damaged by improper setup of small molds on large platens For example, if the platens are deflected into a concave shape by excessive clamping pressure on small molds, the bending stress may be beyond the yield strength of the platens and they can be permanently bent If a large mold is then mounted on these bent platens, the mold can be forced open in Injection pressures must be high enough to fill the cavity, forcing material into the furthest reaches Injection pressures commonly range between 70 and 112 MPa (10–16 kpsi) Higher pressures tend to minimize average mold shrink The maximum injection pressure Figure 6.5 Relationship of injection rate to shrinkage Ch 6: Causes of Molded-Part Variation: Processing Figure 6.6 The relationship between the wall thickness, the injection pressure required for filling the part, and the fill time for the part, where t = wall thickness © Plastics Design Library 85 the center under injection pressure The mold plates then conform to the shape of the platens, bending the mold plates and causing flash in the center of the mold The molder, thinking that he needs more clamping pressure to hold the mold shut, may increase the clamping pressure This does not hold the mold shut because the platens are bent The molder may increase the clamping pressure to the point that the corners of the mold are compressed beyond the yield strength of the mold, hobbing (distorting) the corners of the mold This kind of damage causes the molder and mold builder a great deal of expense and grief One way to monitor the cavity pressure is to place a flow tab in the mold A flow tab is sort of a heavy vent off a runner (or even a part) See Fig 6.7.[26] The flow length that is expected from the material being molded determines the depth of the flow tab Typically the flow tab is about 0.5 mm (0.020 inch) thick at any convenient width and is marked at regular intervals It is cut from the runner to the edge of the mold (to the atmosphere) A flow tab is an excellent runner vent The flow tab must be thick enough to easily see a variation in flow length for any significant variation in normal molding conditions The length of the tab should be such that the plastic will normally flow about half the length of the tab Variations in injection pressure or speed, mold temperature, or melt temperature will affect the length of the plastic flow into the flow tab A flow tab will not give any indication of holding pressure or holding time because it is likely that the flow tab will have frozen before the cavities fill Consistent part-weight is a better indication of holding pressure and time 6.4 Holding Pressure and Time Cavity pressure at the moment the gate freezes is roughly inversely proportional to shrinkage Because plastic is compressible, the greater the cavity pressure, the less the shrinkage As a general rule, the higher the holding pressure, the less the shrink However, if excessively high holding pressures are held long enough (generally this requires a rather large gate), some materials will seem to grow If there is a sufficiently high pressure compressing the plastic in the cavity at the moment of gate freeze, the compression can exceed the shrink When molding conditions reach this point, it is usually difficult to remove the part from the mold because the part is larger than the cavity 6.4.1 Holding Pressure One study determined that holding pressure has a greater effect on shrinkage than any one other variable, when molding polypropylene homopolymers (EXON Escorene® PP 1105) Figure 6.8 illustrates this relationship Figure 6.7 An example of a flow tab.[26] (Courtesy of DuPont.) © Plastics Design Library Figure 6.8 The relationship between cavity pressure (holding pressure) and shrinkage Ch 6: Causes of Molded-Part Variation: Processing 86 According to this study, increasing packing pressure decreases shrinkage The effect of injection rate is small compared to packing pressure Increases in packing time decreased the amount of shrinkage whereas increases in mold temperature did not have any appreciable effect on shrinkage Shrinkage after 168 hours was greater than in-mold shrinkage The shrinkage variation in direction of flow immediately after molding was approximately 1%, and after 168 hours it was approximately 1.3% As seen from Fig 6.9,[32] lower injection velocities (rate) produced less shrinkage immediately after molding, but more shrinkage 168 hours later Shrinkage in the direction of flow was significantly greater than in the transverse direction For Delrin® and other semicrystalline materials, unlike polypropylene, both the packing time and pressure have a great effect on the degree of crystallization, along with other factors The larger the gate and the hotter the mold, the longer the packing pressure can be applied Higher mold temperatures allow more time for crystallization, which causes more in-mold shrink However, high mold temperatures increase longterm stability because there is less post-mold shrink in a part molded in a warmer mold The longer the holding time and the higher the holding pressure, the less apparent the shrink The holding phase is very important for dimensional stability since it helps maintain a uniform and gradual crystallization Figure 6.10 shows the effect on shrinkage of holding pressure at three mold temperatures.[33] There is further discussion of the effect of mold temperature in Sec 6.4 Note that, as discussed in Ch 5, the minimum gate dimension must be at least one-half of the part thickness Thin parts may require gates that are thicker in proportion to the wall thickness An inadequate gate size will cause higher mold shrinkage The holdingpressure time (HPT) effects are shown in Fig 6.13 in Sec 6.4.2,[33] and must be sufficient to hold pressure on the cavity until the gate has frozen The shape of the molded part determines the amount of resistance to shrinkage that the part will experience The greater the restraint, the less the apparent shrinkage Post-mold exposure to time and higher temperatures will encourage post-mold shrinkage Table 6.1 shows shrinkage for some Delrin® grades.[14] Flow patterns and distance from the gate also affect shrinkage Shrinkage far from the gate is typically 0.1% to 0.3% higher than the shrink near the gate Shrinkage in the flow direction is typically about 0.1% higher than the cross-flow shrinkage for Delrin® Holding pressure can be used for small adjustments of part dimensions It has very little effect on post-molding shrinkage Flow patterns during the holding phase can be uneven There is a tendency toward a “river delta” effect: Any area that is slightly warmer than an adjacent area has less resistance to flow; therefore, it is more likely to move and remain warmer than cooler areas nearby The warmer areas (the tributaries) will try to shrink more than the cooler areas This is one cause of high stress in the gate area Figure 6.9 Mold shrinkage in the direction of flow immediately after molding and after 168 hours, vs packing (or holding) pressure for two injection rates.[32] (Courtesy of SPE.) Figure 6.10 The effect of holding pressure on mold shrinkage at three different mold temperatures, for Delrin® 500 HPT: holding-pressure time [33] (Courtesy of DuPont.) Ch 6: Causes of Molded-Part Variation: Processing © Plastics Design Library 87 6.4.2 Holding-Pressure Time Inadequate holding-pressure time (HPT) will allow material to expand out of the cavity into the runner system, if the holding pressure is removed before the gate has frozen Once the gate is solid, plastic can no longer flow into or out of the cavity Additional holding-pressure time after the gate freezes is nonproductive All it does is use energy, add wear-and-tear to the molding machine, and add to the machine oil-cooling load Figure 6.11 shows the time required for a semicrystalline (nylon) material to crystallize from a particular melt temperature.[9] Note that the addition of nucleation agents, reinforcement, or pigments decrease the time required for crystallization Other semicrystalline materials have different crystallization times The HPT must exceed the crystallization time to minimize shrinkage If the gate crystallizes before the part does, the effective HPT is reduced by virtue of the frozen gate Figure 6.12 shows that as holding-pressure time increases, there is an initial drastic reduction in the shrink rate As holding-pressure time increases further, the rate of reduction in shrinkage decreases until there is no further reduction in shrink Figure 6.13 shows the influence of holding pressure on Hytrel® 55 to 80 shore D materials.[34] These are softer, more rubbery materials than Delrin®, but the influence of HPT (also called screw-forward time) is readily apparent As usual, longer holding time results in less shrinkage Table 6.1 Average Mold Shrinkage for Various Grades of Delrin® Average Mold Shrinkage ® Delrin grade In-flow (% ± 0.2%) Transverse (% ± 0.2%) 100, 100 P 2.1 1.9 500, 500 P 2.1 2.0 511 P, 911 P 1.9 1.8 900 P 2.1 2.0 1700 P 1.9 1.8 Colors* 1.8–2.1 1.7–2.0 500 T 1.8 1.7 100 ST 1.3 1.4 500 AF 2.1 1.5 500 CL 1.9 1.9 570, 577 1.2 2.1 *Depends on the pigments © Plastics Design Library Figure 6.11 Crystallization time for several nylon grades The parts are mm in thickness, molded at typical mold temperature with a hold pressure of 85 MPa The melt temperature is 290°C.[9] (Courtesy of DuPont.) Figure 6.12 The approximate relationship between holdingpressure time (HPT) and shrinkage Ch 6: Causes of Molded-Part Variation: Processing 88 Figure 6.13 The influence of HPT (screw-forward time) on the shrinkage of Hytrel ® 55 to 80 shore D materials for 3.2-mm ( 1/8-inch) thick samples.[34] (Courtesy of DuPont.) Figure 6.14 The effect of holding-pressure time (HPT) on mold shrinkage of Delrin® 500 P for three different wall thicknesses.[33] (Courtesy of DuPont.) Figure 6.14 shows the HPT for three different wall thicknesses of Delrin® 500 P.[33] The drastic reduction in shrinkage as the holding time increases is readily apparent For any given part, and considering only the change in holding pressure time, the minimum shrinkage occurs when the HPT lasts until the gate freezes Another conclusion that can be drawn from this figure is that thicker walls cause slower cooling, which increases crystallization Finally, the thicker walls remain melted longer than thin walls, allowing more time for thorough packing of the cavity, provided the gate is large enough, which results in less shrinkage Too short a packing time can also cause porosity, voids, warpage, sink marks, lower mechanical properties, and surface pits or blemishes structure The more time available, the larger and more numerous the structures and the more the material shrinks Likewise, amorphous plastics relax internal molecular stresses when cooled slowly, and the increased order and relaxed stresses result in greater material density and shrinkage Therefore, rapid cooling reduces shrinkage However, rapidly cooled parts are more prone to post-mold shrinkage and warpage with the passage of time and exposure to heat When parts are exposed to higher temperatures in their service life than the mold temperature at which they were manufactured, they may exhibit unusually high and possibly unacceptable post-mold shrink Higher mold temperatures increase cycle times and the time available for molecular stress relaxation in amorphous materials In semicrystalline materials, the longer cycle times also allow more time for crystallization to occur In both cases, with rare exceptions (see the Zenite® LCP aromatic polyester resins), short-term shrinkage increases Post-mold shrinkage, however, decreases Within limits, higher mold temperatures improve long-term stability and minimize post mold shrink and creep 6.5 Mold Temperature Mold temperature affects the cooling rate The faster the plastic part cools, the less time the individual molecules have to order themselves and the less the molded part shrinks Crystalline plastics require some time to rearrange their molecules into the crystalline Ch 6: Causes of Molded-Part Variation: Processing © Plastics Design Library 89 6.5.1 Predicting Mold Temperature Effects Figure 6.15 compares the shrinkage effects of cold molding and hot molding.[35] In some cases, the size of a part after time and stress relief will be the same with a hot mold as with a cold mold Often, though, parts that are measured immediately after molding appear just right, but after a day, week, or a month, some of the molded-in stresses are relieved and the part is smaller, perhaps too small While cycle times are longer when molding with a hot mold, the end result is often a better, more durable, and more stable part Figure 6.16 shows the effects of mold temperature on the initial shrinkage of a 3-mm (1/8 in.) thick part molded of Zytel ® 101 L.[9] As the mold temperature rises, the shrink rate also increases In other cases, parts made from Zytel® 101 L show little difference in shrinkage, including annealing shrinkage, between hot- and cold-mold processing, as seen in Fig 6.17.[35] The shrinkage shown for a hot mold is slightly higher, but the quality of the end product is also higher For maximum stability, especially when used at elevated temperatures, plastic parts should be annealed Annealing promotes stress relaxation and, for semicrystalline plastics, it encourages more complete crystallization Even if the molded part is not annealed, semicrystalline plastic slowly, over time, strives to crystallize as much as possible Thus, over time the molded Figure 6.15 The effect of mold temperature on molded part shrinkage and shrinkage after stress relief.[35] (Courtesy of DuPont.) © Plastics Design Library Figure 6.16 Shrinkage vs mold temperature for a 3-mm thick part of Zytel® 101 L.[9] (Courtesy of DuPont.) Figure 6.17 The total shrinkage for Zytel® 101 NC 10, including shrinkage caused by annealing, is little different for parts molded in hot or cold molds [35] (Courtesy of DuPont.) Ch 6: Causes of Molded-Part Variation: Processing 90 semicrystalline part will continue to shrink because of increased crystallization In nylon parts, the absorption of water may balance the shrink due to post-mold shrinkage and annealing Material suppliers can recommend an appropriate annealing temperature and time The temperature should be above the expected use temperature and below the heat distortion temperature to minimize undesirable warpage Fixturing may be necessary to prevent warpage The time required for annealing can be determined by checking the part for size change periodically When there is no size change between checks, the part is adequately annealed The graph in Fig 6.18 augments Fig 6.17, showing that there is less annealing size change for material molded in a hot mold than there is for a part molded in a cold mold.[35] The quality of the part molded in a hot mold is generally higher and the part has less moldedin stress; therefore the part is tougher and more stable 6.5.2 Relationship Between Mold Temperature and Shrinkage As mold temperature is adjusted upwards, the molded part cools more slowly and the cycle time must increase to allow the same degree of cooling before the part is ejected Slower cooling promotes more stress relaxation and more shrinkage in amorphous and in semicrystalline molded plastic parts (see Fig 6.19) Slower cooling encourages a greater degree of crystallization in semicrystalline parts, which leads to higher shrinkage, if all other variables remain constant Figure 6.18 Shrinkage during annealing vs mold temperature for Zytel® 101 NC 10.[35] (Courtesy of DuPont.) Ch 6: Causes of Molded-Part Variation: Processing Warmer molds typically produce better quality parts, with better surface finishes, better physical properties, lower stress levels, and higher shrink rates 6.5.3 Relationship Between Wall Thickness and Shrinkage When the wall thickness is increased, more time is required to cool the center of the thicker wall Since the plastic cools more slowly, there is more time for crystallization and stress relaxation Hence, thicker walls lead to longer cycles, lower stress, higher crystallization, and higher shrinkage Plastic is a poor conductor of heat Thicker walls cool more slowly and are generally much warmer when the gate freezes than thinner walls This causes a greater amount of cooling after gate freeze and, for semicrystalline materials, more time for crystallization to occur The higher the percentage of crystals in semicrystalline parts, the higher the shrinkage The change in shrink as a result of wall thickness change may be curved (as was shown in Fig 3.1), or linear as shown in Fig 6.20 Thicker walls allow more effective packing because the molten material can flow into the cavity for a longer time and the pressure is transferred more readily to the areas of the mold furthest from the gate Figure 6.20 shows the predicted shrink rate for Zytel® 101 L versus part thickness.[9] It indicates that the shrink rate nearly doubles with a fourfold increase in thickness Other materials have similar shrink changes with respect to part thickness This data can be obtained from the material manufacturer Figure 6.19 The relationship between mold temperature and shrinkage © Plastics Design Library 91 Figure 6.20 Shrinkage of Zytel® 101 L as a function of part thickness for a mold temperature of 70ºC and a hold pressure of 90 MPa.[9] (Courtesy of DuPont.) Figure 6.21 The average mold shrinkage vs thickness for various Delrin® compositions.[14] (Courtesy of DuPont.) Figure 6.21 shows average shrinkage versus wall thickness for several Delrin® resins.[14] The mold temperature for the standard grades was 90°C, while the mold temperature for the toughened grades was 50°C The lower mold temperature for the toughened grades did not lead to high post-mold shrinkage For parts with uniform wall thickness, the mold shrink is relatively uniform If the part varies in thickness, relatively constant shrinkage can be obtained if the part is gated in the thickest area and the gate is of adequate size to maintain a sufficiently long HPT Where these molding conditions are not met, warpage increases because shrinkage is not uniform The greatest shrink is in the heaviest sections Pits, sinks, and voids are likely to occur, and mechanical properties are lowered is probably being demolded too soon Alternatively, there may not be a sufficient number of ejectors or they may be placed incorrectly Figure 6.22 shows a typical relationship between demolding temperature and cycle time for three different wall thicknesses.[4] It is obvious that the demolding temperature has a great effect on the total cycle time In most cases, molders leave the molded part in the mold far longer than necessary in order to “be safe.” This leads to a much longer cycle than necessary and cuts the profit margin considerably Figure 6.23 shows a typical temperature profile across a molded part when it is demolded The maximum temperature is at the center of the part; the plastic against the walls is much cooler Plastic suppliers frequently publish the recommended mean demolding temperature, which is shown as the straight line across the temperature profile in Fig 6.23.[4] The mean temperature is the temperature at which there is an equal volume of plastic with temperatures above and below the mean Some recommended mean demolding temperatures are given in Table 6.2.[4] If no recommended demolding temperature is published by the supplier, the supplier may provide a shear modulus curve One can examine the shear modulus curve for the temperature at which there is a sharp 6.6 Demolding Temperature The demolding temperature is the temperature that must be attained in the hottest region of the molded part (usually the center of the heaviest wall section) before the part can safely be ejected from the mold without risk of distortion due to lack of rigidity or loadbearing capability If ejector pins leave blush or distortion marks on the opposite side of the part, the part © Plastics Design Library Ch 6: Causes of Molded-Part Variation: Processing 92 drop in the shear modulus Above that temperature, the plastic is not able to withstand any significant load Below that temperature, there is a sharp increase in the load-bearing capability of the plastic If neither of the above is available, find the demolding temperature by examining the PVT curves for the plastic and find the Tg for amorphous materials or the Tc for semicrystalline materials Figure 6.22 The relationship of demolding temperature and cooling time.[4] (Courtesy of Bayer.) Figure 6.23 The temperature profile across the wall of a molded part at the time of demolding The demolding temperature is measured where the part is hottest, θE (at the center) The temperature of the sides of the wall, θW, assumes equal temperature on both surfaces of the mold.[4] (Courtesy of Bayer.) Ch 6: Causes of Molded-Part Variation: Processing Table 6.2 Recommended Demolding Temperatures Abbreviation Thermoplastic Mean demolding temperature guide (°C) PC-HT Apec® HT 150 PC-ABS Bayblend® 110 PA Durethan ® 100 PC Makrolon ® 130 ABS Novodur® 90 PBT Pocan® 130 Failing all other sources, use the Vicat temperature (heat deflection temperature) as a demolding temperature This is listed on almost any physical data sheet for a plastic material The heat deflection temperature is the temperature at which a plastic bar of a specific dimension will deflect a certain amount under a given load On the shop floor, the mean temperature of a part can be determined by measuring the surface temperature of a molded part at its hottest or thickest point with a non-contact temperature-measuring device Immediately out of the mold, the surface temperature will rise as indicated in Fig 6.24,[4] because the core temperature is considerably higher than the surface temperature The part tries to stabilize by cooling the core and heating the surface Soon the temperature across the thickness of the part is essentially uniform When the temperature reaches a maximum and then begins to drop, that maximum temperature is a very good approximation of the mean temperature of the molded part when demolded If there are significant variations in the surface temperature of the part, when its temperature stops rising, that is an indication that the cooling system may be inadequate If there are significant thickness variations, then it may be impossible to maintain a uniform rate of cooling 6.7 Molded-In Stresses Every molded part has some molded-in stress The lower the stress level, the lower the post-mold shrinkage and warpage There is less post-mold shrink when © Plastics Design Library 93 running a hotter mold The more gentle cooling process in a hot mold requires longer cycles and allows for some stress relief as the part cools Larger gates, runners, sprue, and the machine nozzle maintain pressure in the cavity longer, resulting in less shrink and lower shear-stress at the gate An isolated thick section within a thinner section can cause tensile stresses at the edges of the thick section because it shrinks more than the neighboring thin section If the thickness change is too abrupt, such as a square corner between the thick and thin section, it is possible for stresses at the square corner to be high enough to cause cracking This is especially true if environmental stress cracking agents (such as aromatic hydrocarbons in the presence of polycarbonate) are present The following steps will minimize stresses • Keep wall thicknesses as uniform as possible • Gate into the thickest section • Increase cycle time while increasing mold temperature • Anneal after molding (jig to avoid distortion) • Preheat molded-in metal inserts • Avoid excessive packing time and pressure It is generally agreed that using lower injection pressure and a shorter injection time can reduce internal stresses in molded parts Reduced molded-in stress reduces part warpage Thermofil has a line of processing aids that helps fill a part with lower injection pressures and in a shorter time Their Thermolube® material improves the flow characteristics of a plastic, leading to improved surface appearance.[36] Parts with heavy walls or variable wall-thickness can be selectively foamed (see Sec 6.8.2.1) with agents such as Thermofil’s Thermofoam® concentrate This will counteract the inherent tendency of high shrinkage in thick sections that are encountered where ribs and bosses intersect the main body of molded parts.[36] 6.8 Other Molding Processes 6.8.1 Lost Core Process A general discussion of post-mold annealing of plastic parts is contained in Ch 8, Sec 8.5.7, “PostMold Fixturing/Annealing.” The “lost-core” process provides at least partial in-mold annealing The lost-core process of molding involves two steps First, a low-temperature metal alloy is molded into the shape of the interior of the desired part This metal-alloy core has extensions that can be gripped by a second mold so that it is precisely positioned within the second mold The low-temperature core actually melts at a temperature that is far below the melting and injection temperature of the plastic The reason the low-temperature core does not melt during the molding cycle is that it takes some time for the metal-core temperature to rise to the melting point of the core Once the plastic is molded around the core, the plastic part and core are gently warmed, sometimes in boiling water, to melt the low-temperature metal core without melting the plastic part Once the metal core is melted, it is drained from the plastic part The plastic part and the metal are cooled The metal can then be reused in a new core The plastic part, when cooled, is then complete The lost-core process of molding hollow parts has an annealing effect by virtue of the exposure of the plastic to temperatures considerably above ambient The temperatures necessary to melt the low-temperature alloy are high enough to perform at least partial annealing of the plastic part 6.8.2 Figure 6.24 Temperatures at various distances from the surface of a molded part vs time Demolding is at time equals 50 sec.[4] (Courtesy of Bayer.) © Plastics Design Library Gas Assist Processing Gas-assist or gas-injection molding describes at least three different concepts: full-shot, partial- or shortshot, and hollow-injection molding The full-shot process involves filling the part completely and then introducing gas under high pressure to fill out the heavier sections as the material cools and shrinks The object Ch 6: Causes of Molded-Part Variation: Processing 94 of the full-shot process is to eliminate sinks and minimize molded-in stresses resulting from differential shrink in thick and thin sections The part must be designed in such a manner that the gas can reach and flow into all thick sections Sometimes this requires making some sections heavier intentionally or creating internal runners through which the gas can move Short-shot, gas-assisted injection-molded parts are those in which the mold is filled to 90–95% capacity, and then a gas (usually nitrogen) is injected into a thick section The gas is under high enough pressure to force the plastic to finish filling the cavity as if it were filled by normal injection-molding processes The gas displaces still-molten plastic in the thicker sections, such as the intersection of ribs and designed-in internal runners, to facilitate mold filling This results in intentional voids in the heavier sections The gas must be vented from the mold before the mold opens or else the plastic part is likely to explode or balloon as a result of the high-pressure gas it contains Hollow-injection molding can be considered as a special form of short-shot injection molding.[37] In this case, the short shot may be more like 50% of the actual solid-part volume followed by gas Take the example of an automotive door arm rest This part can be an inch or more in maximum cross section If plastic is introduced near one end and designed to flow across the gas-injection nozzle near the gate of the part, the percent of fill can sometimes be less than 50% As soon as the plastic short-shot is completed, gas is introduced under high pressure to force the plastic ahead of the gas to flow toward the far end of the mold As the plastic flows, it coats the side walls of the cavity and the gas pressure forces the plastic to conform to the side wall There is somewhat less molecular fiber orientation in this process as compared to normal injection molding Thus, the shrink is more isotropic and there is less molded-in stress Furthermore, as the gas expands into the cavity, it cools significantly and helps cool the plastic part—from the inside out Clamp-tonnage requirements for gas-assist molding are usually considerably lower than they are for an injection-molded part.[37] Gas-assist parts usually have thicker walls and flow channels, making it much easier to move the viscous plastic into and across the mold The enormously high pressures required to pack out a thin-walled plastic part no longer apply The usual ruleof-thumb for clamp tonnage for an injection-molded part is about 2.5 tons per square inch of projected area For single-nozzle gas assist, the clamping force is typically from one-half ton to one ton per square inch In large multi-nozzle gas-assist moldings, the clamping Ch 6: Causes of Molded-Part Variation: Processing pressure can be below one-quarter ton per square inch The lower required clamping force can lead to making much larger parts on smaller machines, thus saving on machine cost and resulting hourly rates Also, there are potential savings in making the mold lighter because it does not have to withstand as high a pressure as in thin-wall injection molding In addition, by moving walls further apart (with gas-assist voids between), the stiffness for a given weight of material can be increased An emerging technology in the gas assist field is to add a small amount of water to the incoming gas in such a way that the water is in a mist form The water aids in cooling because the water vaporizes (becomes steam) and absorbs significantly more heat (the heat of vaporization) than the gas alone can absorb Thus the molding cycle is shortened Another developing variation on this theme is to introduce water under pressure commensurate with the gas-assist process The water, having a very low viscosity, flows through the mold in much the same way that the gas would and absorbs much more heat than the gas so the part cools even more rapidly There is some indication that the wall thickness around the water channels is more uniform using the water assist process than the gas-assist process The primary problems with the water-assist process are the design of an appropriate needle with which to introduce the water and the difficulty in removing all of the water after the cycle is complete Any water residue remaining can leak as the part is ejected and if it stays in the molding area, or if any water leaks from the needle before the plastic passes the needle will cause a water splay defect on the surface of the next molded part The obvious conclusion is that the water needle must be at the lowest point in the expected water channel and it may be necessary to apply a vacuum briefly to vaporize any residual water in the part The needle may have to approximate the design of a hot nozzle with a shut off valve 6.8.2.1 Foaming Foaming can also be considered a form of gasassisted processing A foaming agent is mixed with the resin which causes the material to expand, aiding in filling the mold Foaming agents tend to randomize the orientation of fiber fillers and molecular strands This leads to a more isotropic shrink pattern, even with anisotropic materials When foam is used, there is lower pressure in the mold and the foam helps fill the cavity, almost as © Plastics Design Library 95 if the foam acts as a lubricant The lower pressure minimizes molded-in stresses because of the lower pressure in the cavity Most foamed parts have a rougher surface and the surface contains streaks and swirls that may be considered either blemishes or decoration, depending on the customer’s view There are techniques available that will give a smooth surface with a foamed core This requires two injection units wherein the mold is partially filled with unfoamed material, then additional foamed material follows along to form a foamed core with the unfoamed material forming the outer skin 6.8.2.2 Shrinkage of Gas-Assisted Processes Some testing[38] has been done comparing shrinkage, heat-deflection resistance, and tensile strength of © Plastics Design Library solid injection-molded parts, 10% foam-injected parts, and 10% gas-assisted injection-molded parts The foamed parts were filled with a volume of plastic sufficient to fill 90–95% of the cavity, and a small amount of foaming agent was used to expand the material and complete the filling Shrinkage sinks were eliminated in both the foamed and gas-assisted parts However, both strength and heat-deflection resistance were reduced by the addition of gas (in the form of gas-assist or foam) In both cases where gas replaced resin in the heavy sections, there was a reduction of strength and stiffness because some of the resin was displaced with gas The effective area to resist direct load was reduced when resin was replaced by gas The moment of inertia to resist bending was also reduced as a result of replacing some of the volume of resin with gas Ch 6: Causes of Molded-Part Variation: Processing