关 键 词：

注塑 模具设计

资源描述：

塑盆注塑模具设计,注塑,模具设计

内容简介：

注塑模具的设计和热分析S.H. Tang , Y.M. Kong, S.M. Sapuan, R. Samin, S. SulaimanDepartment of Mechanical and Manufacturing Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia摘 要本文介绍塑料注塑模具在生产翘曲测试样品和进行模具内模具热残余应力的影响的热分析方面的设计。对在注塑模具设计中所需的技术，理论，方法以及应该考虑的因素进行描述。使用商业计算机辅助版本为13.0 Unigraphics设计软件进行模具设计。由于试样的不均匀冷却，用于进行残余热应力分析模型使用被称为LUSAS分析师，版本为13.5的商用有限元分析软件进行建立和解决。该软件通过塑料注塑成型周期为模型提供温度分布的曲线图和温度变化程度，这个注塑成型周期通过绘制时间响应曲线得到。结果表明，相比其他区域来说，收缩可能发生在冷却通道的附近。这种模具在不同区域的不平衡冷却效果归咎于翘曲.2005爱思唯尔B.V.版权所有。关键词：塑料注塑模具设计;热分析1 前言塑料工业是世界上增长速度最快的行业之一，位居少量的十亿美元的产业之中。在日常生活中几乎每一个产品都涉及到塑料的使用，并且大多数这些产品可以通过塑料注塑成型方法1进行生产。注塑成型工艺以利用较低的成本2创造具有各种外观和复杂的几何形状的产品而著称。注塑成型过程是一个循环的过程。在这个过程中有四个重要的阶段。这些阶段是：加料，注射成型，冷却和推出。塑料注塑成型过程开始于将树脂和合适的添加剂从注射机的料斗送到注塑机的注射系统3中。这是加料阶段，在这个阶段热聚合物熔体在注射温度下充满型腔。在聚合物充满型腔之后，在注射成型阶段，额外的聚合物熔体在较高压力作用下注入型腔，以补偿由于熔体凝固而可能引起的收缩。紧接着的是冷却阶段”在这阶段，模具被冷却直到塑件有足够的刚性被推出。最后一步是“推出阶段”在此阶段模具被打开，塑料制件被推出，之后模具关闭重新开始下一个周期4。设计和制造具有理想特性的注塑成型塑料制件一个昂贵的过程，这主要靠经验包括反复修改实际的工具。在模具设计任务中，设计通常在核心部位加一个特定几何形状想凸起和凹坑的模具是非常复杂，为了设计一套模具，许多重要的设计因素必须考虑。这些因素是模具大小，模腔的数量，腔的布局，热流道系统，浇注系统，补缩和推出系统6。在模具的热分析中，其主要目的是分析热残余应力或模压对产品尺寸的影响。热诱导应力主要发生在注射的冷却阶段，主要是低的热导率和熔融聚合物与模具之间的温度差异所导致的。在冷却过程中在产品内存在不均匀的温度场。在冷却过程中，靠近冷却管道的地方比远离冷却冷却管道的地方的受的冷却程大一些。 这种温度的不同，使得材料存在不同的收缩而引起热应力。显著的热应力可以导致翘曲问题。因此，在冷却阶段对注塑成型塑件的残余热应力进行模拟是很重要的8。通过了解热残余应力的分布特点，能对其引起的变形进行预测。在本文中塑料注塑模具在生产翘曲测试样品和进行模具内部模具的热残余应力 影响的分析的设计得以描述。 2 方法论2.1 翘曲测试样本的设计本节说明翘曲测试样的设计将用于注塑模具。很明显翘曲是一些薄的壳形件中存在的主要问题。因此，产品发展的主要目的是设计出一种塑料制件以确定出薄的壳形的注塑成型塑件翘曲问题的有效因素。翘曲测试样品来源于薄的壳形的塑料制品。样本的外形尺寸长120毫米，宽50毫米和厚度为1毫米。 用于生产翘曲测试试样材料是丙烯腈丁二烯苯乙烯（ABS），注射温度，时间和压力分别为210C，3秒和60兆帕。图1显示了翘曲测试试样的生产。图1 生产的翘曲测试样品2.2 翘曲测试样本塑料注塑模具设计本节介绍的设计方面和其他与设计生产翘曲测试样品模具相关的因素。用于生产用于产生翘曲测试样品的塑料注塑模具的材料是AISI 1050碳钢。在设计模具 过程中四中设计方案被考虑包括：一，三板式（方案1）双分型面一模一腔模具。由于成本高，不适用。二，双板式（方案2）单分型面一模一腔无浇注系统模具，生产效率低，不适用。三，双板式（方案3）单分型面一模二腔有浇注系统和推出机构，由于塑件过于薄，推杆可能会损坏产品。四，双板式（方案4）单分型面一模二腔有浇注系统，利用浇口拉出器扮演推杆的角色避免在推出过程中产品受损。在设计得到翘曲测试样品的模具中，第四种方案被应用，不用的设计因素在设计中已经被采用。首先，模具的设计是基于注射机活动模板尺寸，所用的注射机是杆（BOY22D）。这种注射机有一个限制，那就是活动模板的最大面积通过两根拉杆之间的距离决定，注射机的拉杆之间的距离是254毫米。因此，模具板最大宽度不能超过这个距离。此外，为保证模具的开启和运转，两根拉杆与模具之间保留4毫米的距离，这使最终模具的最大宽度为250毫米。模具标准模架采用250毫米250毫米。模架通过压板从模架或模板的右上角和左下角与注射机进行配合，其他与模板相关的尺寸如表1中所示。模具的设计要使得锁模压力产生的锁模力高于内部型腔的压力（反压力）以避免发生溢料。基于模具标准件库提供的尺寸，型芯的宽度和高度分别为200和250毫米，这些尺寸使在型芯上设计两腔时两腔的位置能够均衡的放置，这样在凹模是空的时候模具有足够的空间去容纳塑料熔体注入腔内用到的浇道衬套，因此，唯一的一个标准分型面设计在产品表面。表1 模具板尺寸(宽度高厚)顶部夹紧板25025025凹模20025040型芯20025040侧板/支撑板3725070推杆固定板12025015推板板12025020底部夹紧板25025025在模具打开时，该产品和凝料在分型面被推出来。模具设有主浇口和侧浇口。为了顺利浇注，浇口的底部设计成倾斜角为20度，厚度为0.5毫米。并且为了塑料熔体能够进入型腔，浇口设计成宽4mm,厚0.5mm。在模具设计中主流道,选用抛物线截面类型，这种截面优点是在一套模具中可以使注射机简化一半，在这种情况下，它就是型芯。然而这种类型的流道与圆截面的类型相比存在有更多的热量损失和废料的缺点。这可能会是塑料熔体凝固的更快。这个问题可以通过设计一个短和大的直径的流道来缓解，流道的直径是6mm。设计的流道要在同一时间，同一温度和同等的压力下将材料或者塑料熔体分流到凹模内这一点是很重要的。由于这个原因，型腔布局设计成对称形式。另一个应该被考虑的设计因素就是通风孔的设计，型芯和凹模的配合面是经过非常的精细的加工以防止溢料的发生。然而这会导致当模具关闭时候空气的进入并导致短射或有缺陷。为保证进入的空气被释放从而避免成型不完全，需要设计足够的通风孔。冷却系统沿着模腔开通，并且均衡的布置在模具中以确保均匀的冷却。这些冷却管道在凹模和型芯的两侧可设。它在湍流的情况下对模具进行充分的冷却。图2所示为通风孔和冷却管道在型腔上的布局。图2 通风孔和冷却通道在型芯上的布局在模具设计中，推出机构仅仅有推杆固定板，推杆和推板组成。推杆位于型芯的中间位置，它不仅仅是作为一根推杆准确的固定塑料制品。同时在脱模阶段，当模具打开时它也充当推出器将塑料制品推出模具。在型腔内没有使用和放置额外的推出装置因为生产的产品非常的薄，即1mm。在型腔中加额外的推出装置可能会使产品出现小孔并且在推出过程中损坏产品。最后，足够的尺寸公差给予考虑以补偿材料的收缩。图3显示了应用Unigraphics开发的模具的三维实体造型以及线框造型。 图3 模具的三维造型和线框模型3 结果与讨论3.1 产品的生产和修改结果在审核过程中，设计和制造的模具生产出来的翘曲测试样品存在一些缺陷。这些缺陷只要是短射，溢料和翘曲。短射可以通过随后在凹模的拐角处加工额外的通风孔以让更多的进入里面的空气溢出而得以消除。同时，溢料可以通过减小注射机的成型压力而减少。翘曲可以通过控制各种参数比如注射时间，注射温度和熔化温度来控制。对这些进行修正之后，模具可以通过浇口以较低的价格， 较少的工序生产出高质量的翘曲测试样本。图4为经修改够加工了额外的通风孔以消除短射的模具。图4 额外的通风孔消除短射3.2 模具和产品的详细分析模具和产品开发后，分析模具和产品开展了。在塑料注射成型过程中，熔融的ABS在210C通过凹模上浇道衬套被注入，直接进入型腔中，冷却开始后，该产品就形成了。一个产品的周期大约需要35s包括20s的冷却时间。用于生产生产翘曲测试试样的材料是ABS，注射温度，时间和压力分别为210C，3秒和60兆帕。模具采用AISI 1050碳钢作为材料。这些材料的性质是非常重要的，它决定了模具中的温度分布，可以利用温度分布进行有限元分析。表2为ABS和AISI 1050碳钢材料的性质。表2 模具和产品的材料属性属性模具(碳钢AISI 1050)产品（ABS聚合物）密度()7860 kg/m31050 kg/m3杨氏模量(E)208 Gpa2.519 Gpa泊松比()0.2970.4屈服强度(SY)365.4Mpa65 Mpa拉伸强度(SUTS)636Mpa636Mpa热膨胀系数()6510-6K-111.6510-6K-1电导率(K)0.135 W/(m K)49.4 W/(m K)比热(C)1250 J/(kg K)477 J/(kg K) 对模具最关键的分析部分是在凹模和型芯上，因为它们是产品成型的部位。因此热分析来研究温度的分布和在不同时间的温度是使用商业的有限元分析软件来完成，这款软件叫做LUSAS Analyst版本为13.5.为了研究热残余应力对模具不同部位的影响，进行一次二维的热分析。由于对称性，热分析通过凹模和型芯的垂直截面和侧面的上半部分的这个模型来实现，凹模和型芯在注射阶段就固定在一起了。图5为无规律的配合分析的热分析模型。图5 热分析模型该模型的建模还涉及分配属性 ，工序或模型的周期。这需要有限元分析求解器来分析模具建模和绘出时间响应图以便显示在某一段时间不同区域的温度变化。 针对产品分析利用版本为13.5LUSAS分析师进行一个二维的拉伸强度的分析。基本上述样品处于一端受拉另一端固定的状态。增加载荷使模型达到塑性变形。图6为受载荷作用的模型的分析。图6 为分析产品所用的受载荷模型3.3 模具和产品分析的结果和讨论 针对模具分析，对在不同时间间隔内的热分布进行了观察。图7为塑料注塑成型在一个完整的周期不同的时间间隔内热或热量分布的等高线图二维分析。 图7 不同时间间隔内的热分布等高曲线对模具的二维分析，绘制了时间响应图表以分析热残余应力对产品的影响。图8为为绘制时间响应图标而选取的点。图9-17为在图8中所示不同点温度分布曲线。从图9-17中绘制的温度分布中可以看出，图中的每一个点的温度都在上升，即从环境温度到某一高于环境温度的温度然后在某一段时间内保持这个温度不变。这种温度的升高是由于产品熔体塑料进入型腔所致。在一段时间之后温度继续上升至最高温度并且保持在最高温度，这个温度的升高与成型阶段高的压力有关。这个温度一直持续到冷却阶段的开始，冷却使得模具的温度到一个较低的值且保持在这个水平。绘制的曲线并不光滑，这是因为缺少输入塑料熔体填充和冷却水冷却速度的功能。图表只显示在这个周期内温度能够达到的最大值。在热残余应力时至关重要的阶段是冷却阶段。这是因为冷却阶段使材料从高于玻璃化转变温度到低于这个温度。材料应成分收缩而产生热应力。热应力有可能造成翘曲。图8 在产品内部选择的点作的时间响应图图9 第284点的温度分布图图10 第213点的温度分布图图11 第302点的温度分布图图12 第290个点的温度分布图图13 第278点的温度分布图图14 第1838点的温度分布图图15 第1904点的温度分布图图16 第1853点的温度分布图图17 第1866点的温度分布图从图9-17所示的冷却阶段之后的温度可以看出，靠近冷却管道的区域由于温度降低快而冷却程度大一些，而远离冷却管道的区域冷却程度要小一些。冷却速度快，冷却程度大意味着在这部分区域发生更多的收缩。然而最远的点，第284点由于热量散失到环境中而冷却程度更大。结果，由于冷却管道位于型腔的中间位置而使中间部分的温度差异比其余的位置要大。由于收缩量大使得塑件中间部分产生压应力。不均匀收缩导致翘曲。但是，冷却之后各个点的温度差异小了，翘曲变形就不会那么显著了。对于一个设计师，设计一套具有高效冷却系统和较小热残余应力的模具是很重要的。对于产品分析，从开展研究的第一步到分析注塑成型产品整个过程，产品在不同载荷因子作用下的应力分布都进行了二维观察分析。图18-21为在不同载荷作用下对应的应力等高线图。一个至关重要的点127，这点处塑料制品收到最好的拉应力作用，这一点的应力-应变关系以及受力情况与应力的关系曲线都绘制在图22和图23中。从图23中受力情况与应力关系的曲线来看，制品所受的拉力一直增加直到达到图23中的载荷因子的值，这个值是1150N。这就意味着制品可以承受1150N的拉力。高于这个值的载荷会导致制品破坏。在图23中，这种破坏更可能发生在试样的固定端所在的区域，这个最大应力为3.27x107Pa.试样应力分析由于产品是为了进行翘曲测试并且与拉力分析无关而使得其解释非常有限的信息。然而在将来，研究者指出产品的所处状态应该定下来以便于在不同的载荷作用下试样的变形展开更深入的研究。图18 载荷增量1作用下的相对应力图 图19 载荷增量14作用下的相对应力图图20 载荷增量16作用下的相对应力图图21 载荷增量23作用下的相对应力图 4 结论设计出来的模具使生产高质量的翘曲测试样品成为可能。这种成为是样品可以决定出影响翘曲的参数。这种测试样品以较低的价格生产并且只需要比较少的加工。塑料注塑成型模具的热分析使得热残余应力对产品的变形形状的影响被理解，产品的拉伸应力分析能够成功的预测翘曲测试样品能够承受的最大拉伸应力。 图22 ABS材料的应力应变关系曲线图23 ABS材料的应力载荷关系曲线致谢作者要感谢马来西亚博特拉大学工程学院推出的本出版物。 参考文献1 R.J. Crawford, Rubber and Plastic Engineering Design and Applica-tion, Applied Publisher Ltd., 1987, p. 110.2 B.H. Min, A study on quality monitoring of injection-molded parts,J. Mater. Process. Technol. 136 (2002) 1.3 K.F. Pun, I.K. Hui, W.G. Lewis, H.C.W. Lau, A multiple-criteria envi-ronmental impact assessment for the plastic injection molding process:a methodology, J. Cleaner Prod. 11 (2002) 41.4The Determination of Injection Moulding Parameters of ThermoplasticMaterials: EX-PIMM, J. Mater. Process. Technol. 128 (2002) 113.5 M.R. Cutkosky, J.M. Tenenbaum, CAD/CAM Integration ThroughConcurrent Process and Product Design, Longman. Eng. Ltd., 1987,p. 83.6 G. Menges, P. Mohren, How to Make Injection Molds, second ed.,Hanser Publishers, New York, 1993, p 129.7 K.H. Huebner, E.A. Thornton, T.G. Byrom, The Finite ElementMethod for Engineers, fourth ed., Wisley, 2001, p. 1.8 X. Chen, Y.C. Lam, D.Q. Li, Analysis of thermal residual stressin plastic injection molding, J. Mater. Process. Technol. 101 (1999)275.Journal of Materials Processing Technology 171 (2006) 259267Design and thermal analysis of plastic injection mouldS.H. Tang, Y.M. Kong, S.M. Sapuan, R. Samin, S. SulaimanDepartment of Mechanical and Manufacturing Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, MalaysiaReceived 3 September 2004; accepted 21 June 2005AbstractThis paper presents the design of a plastic injection mould for producing warpage testing specimen and performing thermal analysis forthe mould to access on the effect of thermal residual stress in the mould. The technique, theory, methods as well as consideration neededin designing of plastic injection mould are presented. Design of mould was carried out using commercial computer aided design softwareUnigraphics, Version 13.0. The model for thermal residual stress analysis due to uneven cooling of the specimen was developed and solvedusing a commercial finite element analysis software called LUSAS Analyst, Version 13.5. The software provides contour plot of temperaturedistribution for the model and also temperature variation through the plastic injection molding cycle by plotting time response curves. Theresults show that shrinkage is likely to occur in the region near the cooling channels as compared to other regions. This uneven cooling effectat different regions of mould contributed to warpage. 2005 Elsevier B.V. All rights reserved.Keywords: Plastic Injection mould; Design; Thermal analysis1. IntroductionPlastic industry is one of the worlds fastest growingindustries, ranked as one of the few billion-dollar industries.Almost every product that is used in daily life involves theusage of plastic and most of these products can be producedby plastic injection molding method 1. Plastic injectionmolding process is well known as the manufacturing processtocreateproductswithvariousshapesandcomplexgeometryat low cost 2.The plastic injection molding process is a cyclic process.There are four significant stages in the process. These stagesare filling, packing, cooling and ejection. The plastic injec-tion molding process begins with feeding the resin and theappropriateadditivesfromthehoppertotheheating/injectionsystemoftheinjectionplasticinjectionmoldingmachine3.This is the “filling stage” in which the mould cavity is filledwithhotpolymermeltatinjectiontemperature.Afterthecav-ityisfilled,inthe“packingstage”,additionalpolymermeltispacked into the cavity at a higher pressure to compensate theexpectedshrinkageasthepolymersolidifies.ThisisfollowedCorresponding author.E-mail address: .my (S.H. Tang).by “cooling stage” where the mould is cooled until the part issufficiently rigid to be ejected. The last step is the “ejectionstage” in which the mould is opened and the part is ejected,after which the mould is closed again to begin the next cycle4.The design and manufacture of injection molded poly-meric parts with desired properties is a costly process domi-nated by empiricism, including the repeated modification ofactual tooling. Among the task of mould design, designingthe mould specific supplementary geometry, usually on thecore side, is quite complicated by the inclusion of projectionand depression 5.In order to design a mould, many important designingfactors must be taken into consideration. These factors aremouldsize,numberofcavity,cavitylayouts,runnersystems,gating systems, shrinkage and ejection system 6.In thermal analysis of the mould, the main objective isto analyze the effect of thermal residual stress or molded-instresses on product dimension. Thermally induced stressesdevelop principally during the cooling stage of an injectionmolded part, mainly as a consequence of its low thermalconductivity and the difference in temperature between themolten resin and the mould. An uneven temperature fieldexists around product cavity during cooling 7.0924-0136/$ see front matter 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.jmatprotec.2005.06.075260S.H. Tang et al. / Journal of Materials Processing Technology 171 (2006) 259267During cooling, location near the cooling channel experi-ences more cooling than location far away from the coolingchannel. This different temperature causes the material toexperience differential shrinkage causing thermal stresses.Significantthermalstresscancausewarpageproblem.There-fore,itisimportanttosimulatethethermalresidualstressfieldof the injection-molded part during the cooling stage 8. Byunderstanding the characteristics of thermal stress distribu-tion, deformation caused by the thermal residual stress canbe predicted.In this paper the design of a plastic injection mould forproducingwarpagetestingspecimenandforperformingther-mal analysis for the mould to access on the effect of thermalresidual stress in the mould is presented.2. Methodology2.1. Design of warpage testing specimenThis section illustrates the design of the warpage testingspecimen to be used in plastic injection mould. It is clearthat warpage is the main problem that exists in product withthin shell feature. Therefore, the main purpose of the prod-uct development is to design a plastic part for determiningthe effective factors in the warpage problem of an injection-moulded part with a thin shell.The warpage testing specimen is developed from thinshell plastics. The overall dimensions of the specimen were120mminlength,50mminwidthand1mminthickness.Thematerial used for producing the warpage testing specimenwas acrylonitrile butadiene stylene (ABS) and the injectiontemperature, time and pressure were 210C, 3s and 60MPa,respectively. Fig. 1 shows the warpage testing specimen pro-duced.2.2. Design of plastic injection mould for warpagetesting specimenThissectiondescribesthedesignaspectsandotherconsid-erationsinvolvedindesigningthemouldtoproducewarpagetestingspecimen.ThematerialusedforproducingtheplasticFig. 1. Warpage testing specimen produced.injectionmouldforwarpagetestingspecimenwasAISI1050carbon steel.Four design concepts had been considered in designing ofthe mould including:i. Three-plate mould (Concept 1) having two parting linewith single cavity. Not applicable due to high cost.ii. Two-platemould(Concept2)havingonepartinglinewithsingle cavity without gating system. Not applicable dueto low production quantity per injection.iii. Two-plate mould (Concept 3) having one parting linewithdoublecavitieswithgatingandejectionsystem.Notapplicable as ejector pins might damage the product asthe product is too thin.iv. Two-platemould(Concept4)havingonepartinglinewithdoublecavitieswithgatingsystem,onlyusedspruepulleract as ejector to avoid product damage during ejection.In designing of the mould for the warpage testing spec-imen, the fourth design concept had been applied. Variousdesign considerations had been applied in the design.Firstly,themouldwasdesignedbasedontheplatendimen-sion of the plastic injection machine used (BOY 22D). Thereis a limitation of the machine, which is the maximum area ofmachine platen is given by the distance between two tie bars.The distance between tie bars of the machine is 254mm.Therefore, the maximum width of the mould plate shouldnot exceed this distance. Furthermore, 4mm space had beenreserved between the two tie bars and the mould for mouldsetting-up and handling purposes. This gives the final max-imum width of the mould as 250mm. The standard mouldbasewith250mm250mmisemployed.ThemouldbaseisfittedtothemachineusingMatexclampattheupperrightandlower left corner of the mould base or mould platen. Dimen-sions of other related mould plates are shown in Table 1.The mould had been designed with clamping pressurehaving clamping force higher than the internal cavity force(reaction force) to avoid flashing from happening.Based on the dimensions provided by standard mould set,thewidthandtheheightofthecoreplateare200and250mm,respectively.Thesedimensionsenableddesignoftwocavitieson core plate to be placed horizontally as there is enoughspace while the cavity plate is left empty and it is only fixedwithspruebushingforthepurposeoffeedingmoltenplastics.Therefore,itisonlyonestandardpartinglinewasdesignedatTable 1Mould plates dimensions.ComponentsSize (mm)widthheightthicknessTop clamping plate25025025Cavity plate20025040Core plate20025040Side plate/support plate3725070Ejector-retainer plate12025015Ejector plate12025020Bottom clamping plate25025025S.H. Tang et al. / Journal of Materials Processing Technology 171 (2006) 259267261the surface of the product. The product and the runner werereleased in a plane through the parting line during mouldopening.Standardorsidegatewasdesignedforthismould.Thegateis located between the runner and the product. The bottomland of the gate was designed to have 20slanting and hasonly 0.5mm thickness for easy de-gating purpose. The gatewas also designed to have 4mm width and 0.5mm thicknessfor the entrance of molten plastic.In the mould design, the parabolic cross section type ofrunnerwasselectedasithastheadvantageofsimplermachin-ing in one mould half only, which is the core plate in thiscase. However, this type of runner has disadvantages such asmoreheatlossandscrapcomparedwithcircularcrosssectiontype. This might cause the molten plastic to solidify faster.This problem was reduced by designing in such a way thatthe runner is short and has larger diameter, which is 6mm indiameter.Itisimportantthattherunnerdesigneddistributesmaterialor molten plastic into cavities at the same time under thesame pressure and with the same temperature. Due to this,the cavity layout had been designed in symmetrical form.Another design aspect that is taken into consideration wasair vent design. The mating surface between the core plateand the cavity plate has very fine finishing in order to preventflashingfromtakingplace.However,thiscancauseairtotrapin the cavity when the mould is closed and cause short shotor incomplete part. Sufficient air vent was designed to ensurethat air trap can be released to avoid incomplete part fromoccurring.The cooling system was drilled along the length of thecavities and was located horizontally to the mould to alloweven cooling. These cooling channels were drilled on bothcavity and core plates. The cooling channels provided suffi-cientcoolingofthemouldinthecaseofturbulentflow.Fig.2shows cavity layout with air vents and cooling channels oncore plate.In this mould design, the ejection system only consists ofthe ejector retainer plate, sprue puller and also the ejectorFig. 2. Cavity layout with air vents and cooling channels.plate. The sprue puller located at the center of core plate notonly functions as the puller to hold the product in positionwhen the mould is opened but it also acts as ejector to pushthe product out of the mould during ejection stage. No addi-tional ejector is used or located at product cavities becausethe product produced is very thin, i.e. 1mm. Additional ejec-tor in the product cavity area might create hole and damageto the product during ejection.Finally, enough tolerance of dimensions is given consid-eration to compensate for shrinkage of materials.Fig. 3 shows 3D solid modeling as well as the wireframemodeling of the mould developed using Unigraphics.3. Results and discussion3.1. Results of product production and modificationFrom the mould designed and fabricated, the warpagetesting specimens produced have some defects during trialrun. The defects are short shot, flashing and warpage. Theshortshotissubsequentlyeliminatedbymillingofadditionalair vents at corners of the cavities to allow air trapped toFig. 3. 3D solid modeling and wireframe modeling of the mould.262S.H. Tang et al. / Journal of Materials Processing Technology 171 (2006) 259267Fig. 4. Extra air vents to avoid short shot.escape. Meanwhile, flashing was reduced by reducing thepacking pressure of the machine. Warpage can be controlledby controlling various parameters such as the injection time,injection temperature and melting temperature.After these modifications, the mould produced high qual-ity warpage testing specimen with low cost and requiredlittle finishing by de-gating. Fig. 4 shows modifications ofthe mould, which is machining of extra air vents that caneliminate short shot.3.2. Detail analysis of mould and productAfterthemouldandproductsweredeveloped,theanalysisofmouldandtheproductwascarriedout.Intheplasticinjec-tionmouldingprocess,moltenABSat210Cisinjectedintothe mould through the sprue bushing on the cavity plate anddirected into the product cavity. After cooling takes place,the product is formed. One cycle of the product takes about35s including 20s of cooling time.The material used for producing warpage testing speci-men was ABS and the injection temperature, time and pres-sure were 210C, 3s and 60MPa respectively. The materialselected for the mould was AISI 1050 carbon steel.Properties of these materials were important in determin-ing temperature distribution in the mould carried out usingfinite element analysis. Table 2 shows the properties for ABSand AISI 1050 carbon steel.The critical part of analysis for mould is on the cavity andcore plate because these are the place where the product isformed. Therefore, thermal analysis to study the temperatureFig. 5. Model for thermal analysis.distribution and temperature at through different times areperformedusingcommercialfiniteelementanalysissoftwarecalled LUSAS Analyst, Version 13.5. A two-dimensional(2D) thermal analysis is carried out for to study the effectof thermal residual stress on the mould at different regions.Due to symmetry, the thermal analysis was performed bymodeling only the top half of the vertical cross section orside view of both the cavity and core plate that were clampedtogether during injection. Fig. 5 shows the model of thermalanalysis analyzed with irregular meshing.Modelingforthemodelalsoinvolvesassigningpropertiesandprocessorcycletimetothemodel.Thisallowedthefiniteelement solver to analyze the mould modeled and plot timeresponse graphs to show temperature variation over a certainduration and at different regions.For the product analysis, a two dimensional tensile stressanalysis was carried using LUSAS Analyst, Version 13.5.Basically the product was loaded in tension on one end whilethe other end is clamped. Load increments were applied untilthe model reaches plasticity. Fig. 6 shows loaded model ofthe analysis.3.3. Result and discussion for mould and productanalysisFor mould analysis, the thermal distribution at differenttime intervals was observed. Fig. 7 shows the 2D analysisTable 2Material properties for mould and productCarbon Steel (AISI 1050), mouldABS Polymer, productDensity, 7860kg/m3Density, 1050kg/m3Youngs modulus, E208GPaYoungs modulus, E2.519GPaPoissons ratio, 0.297Poissons ratio, 0.4Yield strength, SY365.4MPaYield strength, SY65MPaTensile strength, SUTS636MPaThermal expansion, 65106K1Thermal expansion, 11.65106K1Conductivity, k0.135W/(mK)Conductivity, k49.4W/(mK)Specific heat, c1250J/(kgK)Specific heat, c477J/(kgK)S.H. Tang et al. / Journal of Materials Processing Technology 171 (2006) 259267263Fig. 6. Loaded model for analysis of product.contour plots of thermal or heat distribution at different timeintervals in one complete cycle of plastic injection molding.For the 2D analysis of the mould, time response graphsare plotted to analyze the effect of thermal residual stress onthe products. Fig. 8 shows nodes selected for plotting timeresponse graphs.Figs. 917 show temperature distribution curves for dif-ferent nodes as indicated in Fig. 8.From the temperature distribution graphs plotted inFigs. 917, it is clear that every node selected for the graphplotted experiencing increased in temperature, i.e. from theambient temperature to a certain temperature higher thanthe ambient temperature and then remained constant at thistemperatureforacertainperiodoftime.Thisincreaseintem-perature was caused by the injection of molten plastic intothe cavity of the product.After a certain period of time, the temperature is thenfurther increased to achieve the highest temperature andremained constant at that temperature. Increase in temper-ature was due to packing stages that involved high pressure,Fig. 7. Contour plots of heat distribution at different time intervals.264S.H. Tang et al. / Journal of Materials Processing Technology 171 (2006) 259267Fig. 8. Selected nodals near product region for time response graph plots.Fig. 9. Temperature distribution graph for Node 284.Fig. 10. Temperature distribution graph for Node 213.Fig. 11. Temperature distribution graph for Node 302.Fig. 12. Temperature distribution graph for Node 290.which caused the temperature to increase. This temperatureremains constant until the cooling stage starts, which causesreduction in mould temperature to a lower value and remainsat this value. The graphs plotted were not smooth due to theabsence of function of inputting filling rate of the moltenplastic as well as the cooling rate of the coolant. The graphsplotted only show maximum value of temperature that canbe achieved in the cycle.The most critical stage in the thermal residual stress anal-ysis is during the cooling stage. This is because the coolingFig. 13. Temperature distribution graph for Node 278.S.H. Tang et al. / Journal of Materials Processing Technology 171 (2006) 259267265Fig. 14. Temperature distribution graph for Node 1838.Fig. 15. Temperature distribution graph for Node 1904.stage causes the material to cool from above to below theglasstransitiontemperature.Thematerialexperiencesdiffer-ential shrinkage that causes thermal stress that might resultin warpage.From the temperature after the cooling stage as shown inFigs. 917, it is clear that the area (node) located near thecooling channel experienced more cooling effect due to fur-Fig. 16. Temperature distribution graph for Node 1853.Fig. 17. Temperature distribution graph for Node 1866.ther decreasing in temperature and the region away from thecooling channel experienced less cooling effect. More cool-ing effect with quite fast cooling rate means more shrinkageisoccurringattheregion.However,thefarthestregion,Node284experiencemorecoolingalthoughfarawayfromcoolingchannel due to heat loss to environment.Asaresult,thecoolingchannellocatedatthecenteroftheproduct cavity caused the temperature difference around themiddle of the part higher than other locations. Compressivestress was developed at the middle area of the part due tomoreshrinkageandcausedwarpageduetounevenshrinkagethat happened. However, the temperature differences aftercooling for different nodes are small and the warpage effectisnotverysignificant.Itisimportantforadesignertodesignamouldthathaslessthermalresidualstresseffectwithefficientcooling system.Fortheproductanalysis,fromthestepsbeingcarriedouttoanalyze the plastic injection product, the stress distributionon product at different load factor is observed in the twodimensional analysis. Figs. 1821 show the contour plots ofequivalent stress at different load increments.A critical point, Node 127, where the product experiencesmaximum tensile stress was selected for analysis. The stressversus strain curve and the load case versus stress curves atthis point were plotted in Figs. 22 and 23.Fromtheloadcaseversusstresscurvesatthispointplottedin Fig. 23, it is clear that the product experiencing increasedin tensile load until it reached the load factor of 23, whichis 1150N. This means that the product can withstand tensileload until 1150N. Load higher than this value causes failuretotheproduct.BasedonFig.23,thefailureislikelytooccurattheregionneartothefixedendoftheproductwithmaximumstress of 3.27107Pa.The product stress analysis reveals very limited informa-tion since the product produced was for warpage testingpurposes and had no relation with tensile loading analy-sis. In future, however, it is suggested that the product ser-vice conditi