Draft is the taper applied to vertical walls so a part can release from the mold. It costs nothing to add during design and is expensive to add afterward, which makes it one of the most consequential details on a molded part drawing.<\/p>\n\n\n\n
It is also the detail most frequently missed, usually because CAD models are drawn with vertical walls by default and nobody revisits them before release. The result is a DFM comment at quoting stage, a design revision, and a delay that was entirely avoidable.<\/p>\n\n\n\n
This guide covers how much draft is needed, what changes the requirement, and what happens when there is not enough.<\/p>\n\n\n\n
Why Draft Is Necessary<\/strong><\/p>\n\n\n\n Plastic shrinks onto the core as it cools. That shrinkage grips the mold steel, and the part must be pushed off against that grip.<\/p>\n\n\n\n A perfectly vertical wall slides against the steel along its entire length during ejection. Friction is high, and the part resists. Adding even a small taper means the wall separates from the steel immediately as ejection begins, and the remaining travel is free.<\/p>\n\n\n\n The effect is not marginal. A small angle converts a difficult ejection into an easy one, because the contact is released at the start of the stroke rather than maintained throughout it.<\/p>\n\n\n\n The General Guidance<\/strong><\/p>\n\n\n\n As a starting point for unfilled thermoplastics with a standard machined finish, around one degree per side is a reasonable baseline for moderate depths.<\/p>\n\n\n\n That figure is a starting point, not a rule. Several factors move it substantially, and the rest of this guide covers them.<\/p>\n\n\n\n Where a design can accommodate more draft without functional consequence, more is better. Two or three degrees ejects more reliably than one and costs nothing if the geometry allows it. Designers frequently specify the minimum they think they can get away with, when the generous figure would have caused no problem at all.<\/p>\n\n\n\n What Increases the Requirement<\/strong><\/p>\n\n\n\n Texture<\/strong><\/p>\n\n\n\n The largest single factor, and the one most often overlooked.<\/p>\n\n\n\n A textured surface is a series of small undercuts. Every peak and valley in the texture grips the part as it ejects, and the deeper the texture, the more grip.<\/p>\n\n\n\n Textured surfaces need considerably more draft than smooth ones. A rough guideline used in the industry is roughly one to one and a half degrees of additional draft per 0.025 mm of texture depth, on top of the base requirement. Deep textures can therefore require several degrees.<\/p>\n\n\n\n Texture suppliers publish draft recommendations for their specific texture standards, whether SPI, VDI or a proprietary specification. Consult these rather than estimating, because the variation between textures is large.<\/p>\n\n\n\n The practical failure here is specifying a texture late in the design process, after draft was set for a smooth finish. The part then drags, scuffs and shows ejection marks, and correcting it means either reducing texture depth or modifying the tool.<\/p>\n\n\n\n \u6df1\u3055<\/strong><\/p>\n\n\n\n Deeper features need more draft, for a simple geometric reason: the same angle produces more separation over a longer wall, but the total contact area is also greater.<\/p>\n\n\n\n Tall ribs, deep pockets and long cores are the usual problem areas. Deep cores in particular are difficult, because they have high surface area, are surrounded by shrinking plastic, and are often thin enough that the steel itself is fragile.<\/p>\n\n\n\n For deep features, increase draft beyond the baseline. Where the depth-to-width ratio is high, discuss the feature specifically at DFM stage rather than applying a general figure.<\/p>\n\n\n\n \u7d20\u6750<\/strong><\/p>\n\n\n\n Materials behave differently on ejection.<\/p>\n\n\n\n Higher-shrink materials<\/strong> grip the core harder as they cool and generally need more draft. Semi-crystalline materials such as polypropylene, polyethylene and nylon shrink more than amorphous ones.<\/p>\n\n\n\n Stiffer, less forgiving materials<\/strong> are less able to flex slightly during ejection, so they need adequate draft rather than tolerating marginal draft.<\/p>\n\n\n\n Filled materials<\/strong>, particularly glass-filled grades, are both stiffer and more abrasive, and generally need more draft than the unfilled equivalent.<\/p>\n\n\n\n Softer elastomeric materials<\/strong> can sometimes tolerate less draft because they flex, though they may grip more.<\/p>\n\n\n\n Where a material is unusual or the part is demanding, ask the material supplier or your molder for a recommendation specific to the grade rather than applying a general figure.<\/p>\n\n\n\n \u8868\u9762\u4ed5\u4e0a\u3052\u306e\u54c1\u8cea<\/strong><\/p>\n\n\n\n Counter-intuitively, very highly polished surfaces can sometimes need more draft rather than less, because a highly polished surface can create a vacuum effect as the part separates. This is geometry-dependent and worth discussing where large flat polished surfaces are involved.<\/p>\n\n\n\n What Happens With Insufficient Draft<\/strong><\/p>\n\n\n\n Drag marks and scuffing<\/strong> along the walls where the part scrapes the steel during ejection. Cosmetically obvious and often the first sign.<\/p>\n\n\n\n Ejector pin marks and push-through<\/strong>, where ejector pins mark or penetrate the part because the force required exceeds what the wall thickness can take locally.<\/p>\n\n\n\n Part distortion<\/strong> as ejection forces deform the part.<\/p>\n\n\n\n Stress whitening<\/strong> in some materials where local stress from ejection is visible.<\/p>\n\n\n\n Longer cycle times<\/strong>, since more cooling is needed before the part is rigid enough to eject without damage.<\/p>\n\n\n\n \u5de5\u5177\u6469\u8017<\/strong>, as repeated dragging polishes and eventually wears the cavity surfaces.<\/p>\n\n\n\n Sticking<\/strong>, in the worst case, requiring intervention to remove the part and halting production.<\/p>\n\n\n\n Note that most of these appear as production problems rather than design problems, which is why they often get attributed to process settings and troubleshooting effort gets spent in the wrong place. Insufficient draft cannot be fixed by adjusting the machine.<\/p>\n\n\n\n Where Draft Is Applied<\/strong><\/p>\n\n\n\n Draft is needed on any surface parallel to the direction of mold opening. That includes:<\/p>\n\n\n\n Surfaces perpendicular to the opening direction do not need draft, and neither do surfaces formed by a side action in the direction that action moves, though those need draft relative to their own withdrawal direction.<\/p>\n\n\n\n Parting line position matters. Draft is measured from the direction of pull, so if the parting line moves, the draft requirement changes. Agreeing the parting line early, as part of DFM review<\/a>, lets draft be applied correctly the first time.<\/p>\n\n\n\n Applying Draft in CAD<\/strong><\/p>\n\n\n\n Two approaches, and the difference matters dimensionally.<\/p>\n\n\n\n Drafting from the large end<\/strong> keeps the largest dimension nominal and reduces material as the wall recedes.<\/p>\n\n\n\n Drafting from the small end<\/strong> keeps the smallest dimension nominal and adds material.<\/p>\n\n\n\n Which you choose depends on which dimension is functionally critical. For a boss receiving a screw, the inner diameter at the entry matters. For a wall that must fit inside another part, the outer dimension at the widest point matters.<\/p>\n\n\n\n Decide deliberately rather than accepting the CAD default, and specify which dimension is nominal on the drawing.<\/p>\n\n\n\n Note also that adding draft changes wall thickness, since a wall tapering over its height is thicker at one end. On deep features this can be significant enough to affect filling and shrinkage, which links draft to wall thickness considerations.<\/p>\n\n\n\n Zero-Draft Features<\/strong><\/p>\n\n\n\n Some features genuinely cannot have draft. Sealing surfaces, precision bores and mating interfaces sometimes must be parallel.<\/p>\n\n\n\n These are possible but they add cost, typically through additional tooling actions or through accepting a more difficult ejection with specific measures to manage it.<\/p>\n\n\n\n Where a zero-draft feature is genuinely necessary, flag it explicitly and early. A toolmaker who learns about it during DFM can design around it. One who discovers it after the tool is designed has a problem.<\/p>\n\n\n\n The important discipline is distinguishing features that genuinely require zero draft from those simply drawn without it. In most designs, the majority of the second category can take draft with no functional consequence at all.<\/p>\n\n\n\n A Practical Checklist<\/strong><\/p>\n\n\n\n Before releasing a molded part design:<\/p>\n\n\n\n Running this before quoting saves a DFM revision cycle, which typically costs more in schedule than the review costs in time.<\/p>\n\n\n\n Getting It Reviewed Before Tooling<\/strong><\/p>\n\n\n\n Draft problems are cheap to fix in CAD and expensive to fix in steel. A tool built with insufficient draft either produces marked parts for its whole life or requires modification, and modifying a hardened tool is difficult.<\/p>\n\n\n\n Elite Mold Tech provides DFM review before quoting on \u751f\u7523\u6210\u5f62<\/a> and prototype tooling, covering draft alongside wall thickness, gating and ejection strategy. Sending a part file with the intended material and any texture requirement produces specific comments on where draft is insufficient rather than a general note.<\/p>\n\n\n\n \u3088\u304f\u3042\u308b\u8cea\u554f<\/strong><\/p>\n\n\n\n Q: How much draft angle does an injection molded part need?<\/strong><\/p>\n\n\n\n A: Around one degree per side is a common starting point for unfilled materials with a smooth finish at moderate depth. Texture, depth, material and finish all increase the requirement substantially.<\/p>\n\n\n\n Q: Why do textured surfaces need more draft?<\/strong><\/p>\n\n\n\n A: Texture is effectively a series of small undercuts that grip the part during ejection. Deeper textures need more additional draft, and texture suppliers publish specific recommendations for their standards.<\/p>\n\n\n\n Q: What happens if a part has no draft?<\/strong><\/p>\n\n\n\n A: Ejection becomes difficult, producing drag marks, ejector pin damage, distortion, longer cycles and eventually tool wear. These appear as production problems but cannot be resolved by adjusting machine settings.<\/p>\n\n\n\n Q: Can a molded part have zero draft anywhere?<\/strong><\/p>\n\n\n\n A: Yes, where genuinely required for sealing or precision fits, but it adds tooling cost and complexity. Flag such features early so the toolmaker can design around them rather than discovering them later.<\/p>\n\n\n\n Q: Does adding draft change my part dimensions?<\/strong><\/p>\n\n\n\n A: Yes, since a drafted wall is thicker at one end. Decide which dimension is functionally critical and specify it as nominal, and check that the resulting thickness variation does not affect filling.Draft is the taper applied to vertical walls so a part can release from the mold. It costs nothing to add during design and is expensive to add afterward, which makes it one of the most consequential details on a molded part drawing.<\/p>\n\n\n\n It is also the detail most frequently missed, usually because CAD models are drawn with vertical walls by default and nobody revisits them before release. The result is a DFM comment at quoting stage, a design revision, and a delay that was entirely avoidable.<\/p>\n\n\n\n This guide covers how much draft is needed, what changes the requirement, and what happens when there is not enough.<\/p>\n\n\n\n Why Draft Is Necessary<\/strong><\/p>\n\n\n\n Plastic shrinks onto the core as it cools. That shrinkage grips the mold steel, and the part must be pushed off against that grip.<\/p>\n\n\n\n A perfectly vertical wall slides against the steel along its entire length during ejection. Friction is high, and the part resists. Adding even a small taper means the wall separates from the steel immediately as ejection begins, and the remaining travel is free.<\/p>\n\n\n\n The effect is not marginal. A small angle converts a difficult ejection into an easy one, because the contact is released at the start of the stroke rather than maintained throughout it.<\/p>\n\n\n\n The General Guidance<\/strong><\/p>\n\n\n\n As a starting point for unfilled thermoplastics with a standard machined finish, around one degree per side is a reasonable baseline for moderate depths.<\/p>\n\n\n\n That figure is a starting point, not a rule. Several factors move it substantially, and the rest of this guide covers them.<\/p>\n\n\n\n Where a design can accommodate more draft without functional consequence, more is better. Two or three degrees ejects more reliably than one and costs nothing if the geometry allows it. Designers frequently specify the minimum they think they can get away with, when the generous figure would have caused no problem at all.<\/p>\n\n\n\n What Increases the Requirement<\/strong><\/p>\n\n\n\n Texture<\/strong><\/p>\n\n\n\n The largest single factor, and the one most often overlooked.<\/p>\n\n\n\n A textured surface is a series of small undercuts. Every peak and valley in the texture grips the part as it ejects, and the deeper the texture, the more grip.<\/p>\n\n\n\n Textured surfaces need considerably more draft than smooth ones. A rough guideline used in the industry is roughly one to one and a half degrees of additional draft per 0.025 mm of texture depth, on top of the base requirement. Deep textures can therefore require several degrees.<\/p>\n\n\n\n Texture suppliers publish draft recommendations for their specific texture standards, whether SPI, VDI or a proprietary specification. Consult these rather than estimating, because the variation between textures is large.<\/p>\n\n\n\n The practical failure here is specifying a texture late in the design process, after draft was set for a smooth finish. The part then drags, scuffs and shows ejection marks, and correcting it means either reducing texture depth or modifying the tool.<\/p>\n\n\n\n \u6df1\u3055<\/strong><\/p>\n\n\n\n Deeper features need more draft, for a simple geometric reason: the same angle produces more separation over a longer wall, but the total contact area is also greater.<\/p>\n\n\n\n Tall ribs, deep pockets and long cores are the usual problem areas. Deep cores in particular are difficult, because they have high surface area, are surrounded by shrinking plastic, and are often thin enough that the steel itself is fragile.<\/p>\n\n\n\n For deep features, increase draft beyond the baseline. Where the depth-to-width ratio is high, discuss the feature specifically at DFM stage rather than applying a general figure.<\/p>\n\n\n\n \u7d20\u6750<\/strong><\/p>\n\n\n\n Materials behave differently on ejection.<\/p>\n\n\n\n Higher-shrink materials<\/strong> grip the core harder as they cool and generally need more draft. Semi-crystalline materials such as polypropylene, polyethylene and nylon shrink more than amorphous ones.<\/p>\n\n\n\n Stiffer, less forgiving materials<\/strong> are less able to flex slightly during ejection, so they need adequate draft rather than tolerating marginal draft.<\/p>\n\n\n\n Filled materials<\/strong>, particularly glass-filled grades, are both stiffer and more abrasive, and generally need more draft than the unfilled equivalent.<\/p>\n\n\n\n Softer elastomeric materials<\/strong> can sometimes tolerate less draft because they flex, though they may grip more.<\/p>\n\n\n\n Where a material is unusual or the part is demanding, ask the material supplier or your molder for a recommendation specific to the grade rather than applying a general figure.<\/p>\n\n\n\n \u8868\u9762\u4ed5\u4e0a\u3052\u306e\u54c1\u8cea<\/strong><\/p>\n\n\n\n Counter-intuitively, very highly polished surfaces can sometimes need more draft rather than less, because a highly polished surface can create a vacuum effect as the part separates. This is geometry-dependent and worth discussing where large flat polished surfaces are involved.<\/p>\n\n\n\n What Happens With Insufficient Draft<\/strong><\/p>\n\n\n\n Drag marks and scuffing<\/strong> along the walls where the part scrapes the steel during ejection. Cosmetically obvious and often the first sign.<\/p>\n\n\n\n Ejector pin marks and push-through<\/strong>, where ejector pins mark or penetrate the part because the force required exceeds what the wall thickness can take locally.<\/p>\n\n\n\n Part distortion<\/strong> as ejection forces deform the part.<\/p>\n\n\n\n Stress whitening<\/strong> in some materials where local stress from ejection is visible.<\/p>\n\n\n\n Longer cycle times<\/strong>, since more cooling is needed before the part is rigid enough to eject without damage.<\/p>\n\n\n\n \u5de5\u5177\u6469\u8017<\/strong>, as repeated dragging polishes and eventually wears the cavity surfaces.<\/p>\n\n\n\n Sticking<\/strong>, in the worst case, requiring intervention to remove the part and halting production.<\/p>\n\n\n\n Note that most of these appear as production problems rather than design problems, which is why they often get attributed to process settings and troubleshooting effort gets spent in the wrong place. Insufficient draft cannot be fixed by adjusting the machine.<\/p>\n\n\n\n Where Draft Is Applied<\/strong><\/p>\n\n\n\n Draft is needed on any surface parallel to the direction of mold opening. That includes:<\/p>\n\n\n\n Surfaces perpendicular to the opening direction do not need draft, and neither do surfaces formed by a side action in the direction that action moves, though those need draft relative to their own withdrawal direction.<\/p>\n\n\n\n\n
\n
\n