21 2.3.2 Are Special Fits with Matching Products Required? Often, certain dimensions of a product are specified with unnecessary close tolerances, when all the designer wanted to convey is that the product should fit suitably on another product (tightly or loosely), typically, a container and a matching lid. This requirement must be clear. Especially, when molding plastics with high shrinkage factors (e.g., PP or PE), it can be difficult to arrive at the proper “steel” dimensions, and some experimenting may be required to achieve the required fit. Specifying the matching diameters with standard, loose tolerances may yield pieces correct in size, but wrong because the fit is not as desired. The alternative – providing closer tolerances – could be unreasonable, because the dimension of the molded product depend not solely on the steel dimensions of the stack parts but also on the molding parameters. In such cases, it is of advantage to complete the more complicated mold first and test it in actual molding conditions until the best cycle time is established. The critical mold parts of the matching product (e.g., the lid) should be finish-machined only after having established what the actual molded container dimensions are. This could require completing the lid mold with only one cavity, using assumed suitable dimensions, testing the un- finished mold until the best cycle is achieved, and then adjusting the assumed dimensions so that the proper fit can be achieved. All lid mold parts can then be finished. For more information on this subject see [5]. 2.3.3 Tolerances for the Filling Volume This applies specifically – but is not restricted – to containers into which a more or less viscous product will be filled by volume to within closely specified limits (typically, containers for margarine, paint, etc.). In their end use, it is important for the seller that a minimum amount must be filled into the package without shortchanging the buyer, but also they should not be over- filled, which would mean a loss for the seller. There should be clearly defined fill lines (usually inside the container) to mark the minimum and maximum volumes. This can be a problem with plastics with large shrinkage factors such as PE and PP. It requires special consideration when dimensioning the cavity and core because of the unavoidable variations in shrinkage values, as the plastic flows away from the gate and slowly cools and as the injection pressure within the mold decreases. The same considerations apply to measuring cups or vials which have the various levels (or volumes) indicated by lines on the sides of the product. It may be necessary to first test the mold to find the best cycle times, and then establish the location of the measuring lines. Prototyping is often used to verify the required dimensions or fits of a part after shrinkage 2.3 Accuracy and Tolerances Required 1281han02.pmd 28.11.2005, 10:4821 Previous Page 22 2 The Plastic Product 2.3.4 Stacking of Products and Free Dispensing Any product stacked for shipping must have a clearly defined stacking height, which is usually created by resting the outside or the bottom of one piece on the inside stacking provision of the following piece. These provisions for stacking can be “stacking lugs”, or clearly defined steps in the product. The purpose of these lugs (or steps) is  The products must not jam when pushed together, which would make it difficult to separate them where required by the user, and  They will ensure a total stack height of a certain, specified number (e.g., 20, 25, 40, etc.) of the products when stacked. The stack height should be suitable for the size of boxes or containers (preferably, standard size cartons) in which stacks will be shipped.  If special cartons are to be provided, it may be necessary to investigate if their size will suitable for standard rail or sea shipping containers, for best use of the available space inside these containers. Stacking is more difficult if the angle of the sidewalls is small. Obviously, a cylindrical container (0° draft) cannot be stacked at all. A typical disposable drinking cup has approx. a 7° angle. Larger angles stack easily. Problems can also arise when parts are used in an assembly line or in a dispensing mechanism (e.g., vending machine) where it is important that the parts will release easily, without fail, from the stack, i.e., not being “hung up” by vacuum or by friction because the gap between two stacked con- tainers is too small, even though they are properly stacked as designed. When the gap between two sidewalls is very close, static electric charges may also prevent the lowest part from falling from the stack when desired. Some dispensers have mechanical separators and don’t depend on gravity, but it is preferable not to depend on having such separators (added costs). It is highly recommended to make sure that any stacking height dimensions are carefully checked before beginning to build a mold. If they are wrong, the mold has to be changed after finishing, or the packaging (carton size) has to be redesigned after the height of the stack was not as originally planned. Occasionally, a mold maker may decide to make slots in the mold cavity for the stacking lugs by EDM into the core only after the mold is finished, rather than do it before and then have to increase the height later. The disadvantage of this method is that the mold has to be dismantled to be able to machine the cores (costs!). The advantage is that a minimum stacking height can be achieved. Also refer to Appendix 12 for more advice for mold designers. Figure 2.24 View of stacked lids Figure 2.25 View of stacked products Figure 2.26 View of products stacked on lugs 1281han02.pmd 28.11.2005, 10:4822 23 2.3.5 Mismatch (Deliberate) Trying to produce a “perfect match” between two surfaces is not only difficult to achieve but also very costly. Designers often create deliberate mismatches for ease of manufacturing. There are two areas of “deliberate mismatch” to consider, and two typical examples are shown. There are many variations of matching parting lines, or between lids and covers, but the basic principle applies to all of them. Mismatch at the Parting Line, Between Cavity and Core First, it must be clarified, whether a rounded edge is really necessary for the product. In many cases, the product designer may not be aware of the possible additional cost involved to produce a round edge as in Fig. 2.27, (a) or (c), and will often agree that a simple, “sharp” edge (b) or (d) would be just as acceptable for the application. Figure 2.28 shows just one of several designs of a round edge, with the ideal case (1) having a perfect match at the parting line. However, due to the build- up of manufacturing tolerances of the mold parts, such ideal case is not practical. In reality, the nominal diameter D of the cavity, or of the core, will be either larger or smaller than the matching one, and create either a hook (2), which is generally not tolerable, or small step (3), which in most cases is perfectly acceptable. Note that the actual differences caused by the tolerances of the diameters are small, usually less than 0.1 mm (0.004 in.), so that a step would not be more than about half this amount. However, a step is much less noticeable than a hook. In fact, a mismatch can be corrected by very time consuming handwork, by grinding or stoning (polishing), but this should be avoided because of the high cost. The suggested proper (and most economical) approach is to dimension the matching diameters so that there is always a step, as shown in Fig. 2.28, item (3), of a magnitude between 0 and 0.1 mm (0–0.004 in.). Mismatch Between Two Matching Pieces, such as Box and Lid The conditions are similar when designing and building molds for “matching” boxes and lids. Here, deliberate mismatch (2) is even more important, because the products may come from different cavities and even molds, made under varying molding conditions, and the mismatch due to build up of many tolerances (in cavities for both products) could be much larger. Figure 2.29 shows the ideal condition (1), which is difficult to achieve, and a way to minimize the effect of a mismatch between matching parts (2). There is also another way shown by adding a “decorative” band to the larger part (3). Always consider: 1. Is the rounded edge really necessary? 2. Is the sharp edge really necessary? 2.3 Accuracy and Tolerances Required 1281han02.pmd 28.11.2005, 10:4823 24 2 The Plastic Product Figure 2.28 Round edge: ideal (1), with “hook” (2), and with “step” (3) Figure 2.27 Typical round edges where a “sharp” edge could be considered Figure 2.29 Mismatch avoidance between box and lid 1281han02.pmd 28.11.2005, 10:4824 25 2.4 Tolerances, Mold Alignment, and Mold Costs The relationship between: (1) product tolerances, (2) machining tolerances of mold parts, (3) resulting requirements for alignment of the mold halves (cavities and cores), and (4) the mold cost could be the subject of another book. Here, we will try to condense the subject, by outlining some major points when making the decision of which method of alignment to select. The main reason for any alignment method between cavity and core is to keep the centerlines of cavities and cores in line. Any deviation from the actual centerlines of cavity and core from the “true” centerline will result in thickness variations of the sidewalls of the product. This is true for any cup- shaped product. With flat products, including lids, usually we do not have this concern; in such cases, alignment of the mold halves using only the machine tie bars could be sufficient, even without leader pins. But don’t forget: leader pins (even if not used for alignment) on the core side are also meant to protect the (projecting) cores from damage. They should always be higher than the cores. There are, basically, four methods of alignment used: 1. Use only the machine tie bars to align cavity and core. This can be done in some cases where the alignment between cavity and core is not very important; it can be used for experimental and prototype molds, or even for limited-production molds. This case will not further be discussed here. 2. Alignment of the mold plates with leader pins. This is the oldest and most common method used, for any size of mold, and for any number of cavities. This is the lowest cost method of alignment. 3. Alignment with taper pins between mold plates, and occasionally between cavities and cores, and taper locks between the individual sets of mold stacks, whether in single- or multi-cavity molds. This method usually also requires at least two or more “loosely” fitting leader pins (with or without bushings), not for real alignment purposes, but to protect the core(s) from damage and to facilitate handling of the mold outside of the machine. This method is more expensive than leader pin alignment. With taper locks, we also have to chose between (a) round tapers (less expensive), or (b) wedges (adjustable) 4. There are also combinations of these two methods of alignment, such as where the mold plates are “loosely” aligned with usually 2 (sometimes 3, rarely 4) leader pins, but the final alignment is achieved with tapers between each cavity and core stack, in single- or in multi-cavity molds. Figure 2.30 1+1 cavity mold requires only leader pin alignment to keep mismatch to an acceptable level (Courtesy: Stackteck) Figure 2.31 This lid mold has leader pins and round taper lock alignment, while the modules have no alignment mechanism. This works well for shallow parts (Courtesy: Husky) Figure 2.32 Lid stack module with flat (no) alignment on the stack. Mold alignment is typically accomplished with round taper locks on plates 2.4 Tolerances, Mold Alignment, and Mold Costs 1281han02.pmd 28.11.2005, 10:4825 26 2 The Plastic Product Figure 2.33 shows an example of a modular mold for a container with circular alignment tapers (A). Note that the cavity (B) is set into the cavity retainer plate (not shown), while the core (C) is mounted on top of the core backing plate (not shown) to ensure proper alignment. Note the absence of a stripper ring: this product is air-ejected from the core, making for a much simpler mold. For best cooling efficiency, there is a beryllium-copper alloy (BeCu) core cap (D), and a BeCu gate insert (E) in the cavity bottom plate (F). Note the intricate venting channels in both cavity and core to ensure fast filling of the cavities. This type of stack usually produces at 6.0 s or less. Figure 2.34 shows an 8-cavity modular mold for rectangular containers. The cavities (A) are set into the hot runner plate (B), the cores (C) are mounted with float on the core backing plate (D). Each core is aligned with its cavity with wedges (E); the containers are ejected by air. Figure 2.34 8-cavity mold for rectangular containers (Photo: Courtesy Dollins Tool Corp., USA) A A B C D E F Figure 2.33 Modular mold for container (round taper interlock) (Courtesy: Husky) 1281han02.pmd 28.11.2005, 10:4826 27 Figure 2.35 shows a modular single-cavity mold for a very large, thin-wall container, with wedge-lock alignment. This mold too operates with air ejection only, and is simple in construction. For the most effective cooling, there is a BeCu core cap (A) and a BeCu gate insert (B). The jaws (C) for the wedge lock are easily adjustable. The two leader pins (D) are basically only for mold handling and for protection of the cores and fit only loosely in the leader pin bushings (E). If the sidewall tolerances are large, which is often the case with heavy-walled products, a possible misalignment between cavity and core is usually insignifi- cant, and alignment with leader pins is perfectly viable. The average clearance between leader pins and leader pin bushings (standard hardware) is about 0.04 mm (0.001 in.). If, e.g., a wall is 1.5 mm thick (0.060 in.) and the tolerance is ± 0.1 mm (± 0.004 in.), any misalignment falls within the permissible limits, and leader pins are perfectly acceptable for the mold. Note: In theory, only two leader pins are ever required to ensure proper alignment. The fact that many molds use 4 pins is mainly to protect the cores during servicing the molds. If the walls are thinner than in the above example, say, in the order of 1.0 mm or less, and the tolerances are tighter, alignment with leader pins may not be good enough to ensure that the variations fall between the allowable limits. In these cases, individual ”taper locks” (of various designs) are required. Round tapers are relatively easy to manufacture, but require high accuracy to ensure concentricity with the center of the cavity, and to ensure that the proper preload is achieved. The basic requirement of any taper fit is the preload between the matching faces. A B C D E Figure 2.35 Modular single-cavity mold for large thin-walled container (square lock alignment) (Courtesy: Husky) The tolerances of the product decide which method of alignment to use Without preload, a taper is useless for alignment 2.4 Tolerances, Mold Alignment, and Mold Costs 1281han02.pmd 28.11.2005, 10:4827 28 2 The Plastic Product The importance of preload is discussed in detail in [5], Chapter 30. The biggest problem with round taper fits is that the tapers wear with time, and need to be reset or replaced, which is often quite expensive. But it is still the most economical method of alignment. The “wedge lock” method is a very good, efficient method, used mainly for molds where accuracy is very important and the higher cost can be easily justified over long periods of use. It consists of two opposing pairs of matching wedges, at 90 degrees The advantage is that the wedges are easily accessible and can be adjusted (by shimming and/or grinding) or replaced with little cost. The main disadvantage of the wedge lock design is that more space is required to accommodate the wedge lock than the space required for a round taper lock, thereby making the mold larger and more expensive. 2.5 Heat Expansion, Alignment, and Mold Cost Heat expansion [6, Chapter 14], is another area that must be taken into account. It is always necessary to have both mold halves at the same tempera- ture; particularly the mold plates carrying the alignment elements. The plates on the cavity side in a hot runner mold can easily become hotter than the plates on the core side of the mold. For example, a temperature difference of 20 °C between two plates, on a distance of 400 mm, causes an expansion of 0.091 mm (0.004 in). This can result in a serious misalignment. If we depend on leader pins for alignment, they will deflect and/or wear rapidly, as will the bushings. If taper elements are used, they too will wear out rapidly and lose their usefulness. There are basically only two ways to avoid misalignment caused by heat and/ or manufacturing variations 1. Make sure that the cooling channels are laid out so that the temperatures of the plates are kept the same; this has little effect on the mold cost. 2. For molds with more than one cavity, allow the cores to “float”: the cavity side consists usually of a ”cavity retainer plate” into which the individual cavities are set in. These locations are fixed but subject to manufacturing variation (tolerances). The mold can be designed so that in the individual stacks, the cores (with their taper alignment) can ”float” on their mounting surface (plate) to “find” the matching taper in the cavity. There are two methods commonly used to achieve this: – The cores are screw-mounted to the backing plate, with the screws accessible from the parting line. The mold is assembled completely, Usually, molds are designed with fixed cavities and floating cores 1281han02.pmd 28.11.2005, 10:4828 29 but these screws are, at first, not tightened fully so that the mold, the first time it is closed, will push the cores into proper relation to the cavities. After the mold is opened again, the screws can be fully tightened to be ready for production. This method is satisfactory as long as the temperature difference between the two mold halves is kept low, at about 5 °C or less. – A better, but more expensive method is to make the cores really floating, regardless of the temperature differences, as shown in [6]. Note that the amount of float is limited and only in the order of 0.1 mm (0.004 in.) 2.6 Surface Finish The finish of the mold parts, the molding surfaces, and the fitting surfaces where mold parts meet, are important cost factors. The finer the machining finish, and the more hand finishing is required, the higher is the mold cost. This appears to be obvious but is often overlooked or neglected. The relationship between surface finish and costs and the relationship between tolerances and costs (as shown in Fig. 2.23) are very similar and apply here too. 2.6.1 Finish of Molding Surfaces Molding surfaces (the areas in contact with the plastic product) are finished 1. To provide the required appearance or function of the product 2. To ensure that the product can be easily ejected from the mold, however: – Occasionally, a relatively rough surface in specific areas may be beneficial to keep the product on that side of the mold, from where it will be ejected. – On the other hand, sometimes, a high polish could also be detrimental to easy ejection, depending on the design of the product. In such cases it is the decision of an experienced mold designer to specify the proper finish in these locations (refer to Appendix 16 for list of surface finishes commonly used). Especially with very thin-walled products, the surface finish of the cavity space affects the plastic flow over the molding surfaces. Better finish results in faster filling and shorter cycle time. In some cases (notably with PS), flash chrome plating over a highly polished area can increase the productivity of the mold by up to 10%. 2.6 Surface Finish Figure 2.36 The etched cavity wall gives this tumbler a frosted look The costs rise exponentially with finer finish 1281han02.pmd 28.11.2005, 10:4829 30 2 The Plastic Product Finishing (polishing, etc.) the mold parts is generally an expensive activity in the mold making process because much handwork is required, and should be limited to those areas that really require it. Most mold makers today utilize hand-operated mechanical and some fully automatic methods to finish a surface, but there is still much need for hand finishing wherever the shape of the product does not allow easy access for mechanical or automatic equipment. The purpose of finishing, in general, is to remove the tool marks remaining on the surface of a work piece. In many cases, the rough, “as machined” finish after chip removing operations (turning, milling, etc.) could be quite satisfactory for the appearance of the product, for example on the inside surface of a technical product (enclosures, boxes, television cabinets, etc.), but this may not always be satisfactory for the ejection of the product, because the plastic will not easily (or not at all) slide over too rough a surface. It is also important to consider in which direction the rough machining grooves are lying: to be in line with the ejection could be satisfactory, but across it is usually not acceptable. Also, the draft angle of a wall (or of the sides of a rib) is important. With little draft (a small draft angle), the surface finish must be much better, whereas with a large angle (approx. 5° or more), a much rougher finish, such as “as machined”, could be permissible. With the need to design for less and less mass, the draft angles, especially of ribs, must be kept small, and these walls therefore need a good finish, but not necessarily a polish: a good finish in line with the ejection motion (“draw stoning”) will usually be good enough. If ejectors can be placed under such ribs, the finish becomes even less of a problem. We must always consider what would happen if a piece of plastic breaks off inside a rib: it may save time in the making of the mold but can become expensive later, when the service personnel are frequently required to remove some broken-off bits of plastic from the mold causing severe delays in production. Grinding and electric discharge machining (EDM) leave smaller tool marks on the worked surface; such surfaces may not need any further finish, except polishing where required for appearance. EDM finish can be from rough to very fine, which may not require any polishing at all. Rough finish is the result of high currents and faster cutting speed and therefore requires less time. In addition, with today’s new methods of finish turning and milling hardened surfaces, the achieved finish is often as good as a ground finish and no further polishing is required. 2.6.2 Texturing of Surfaces There are also other surface finishes for appearance, such as texturing, to create leather, basket weave, or other patterns. If it is a deep pattern, it should be clear if any related dimensions apply to the highest point of the pattern or to the base where it is applied to. A rough EDM finish is a good and inexpensive solution for a good-looking, matte surface. Figure 2.39 Polishing area in a shop Be specific as to where dimensions point to; for example, to the peaks and valleys of the finish Figure 2.38 PS tumblers and core/cavity show the highly polished finishes required to achieve the glass-like look of the molded cup Figure 2.37 Typical PS tumblers 1281han02.pmd 28.11.2005, 10:4830 Next Page . surfaces (the areas in contact with the plastic product) are finished 1. To provide the required appearance or function of the product 2. To ensure that the product. in the product. The purpose of these lugs (or steps) is  The products must not jam when pushed together, which would make it difficult to separate them