03086_03a

8/13/2019 03086_03a

http://slidepdf.com/reader/full/0308603a 1/21

8/13/2019 03086_03a


50 3 Cost Factors Affecting Productivity

Another example: a mold and machine worked perfectly, but occasionally,for several hours, produced pieces with surface blemishes that looked like

blisters. Investigation showed that it happened only on very humid days.This particular operation required a rather long mold open cycle. Onhumid days, the water in the air condensed in tiny droplets on the coldmold cores during the few seconds the mold was open and the cavitiesand cores were exposed to the shop air; the droplets appeared as blisterson the surface of the product. After slightly increasing the cooling watertemperature to bring it above the dew point the problem disappeared.The “penalty” was a slightly longer cycle time, but it ensured continuous

production of quality products.

Corrosion Prevention

It is important to decide how the mold will be protected from corrosion if itis evident that the mold is operated and stored in a humid environment.This can affect the mold cost. A common approach in many shops is to protectthe molding surfaces before the mold is put into storage by using siliconspray (“Mold Saver”) or to just apply plain, clean machine oil. Many shopspaint the outside of the mold shoe with a permanent oil paint to protect theoutside of the mold against corrosion.

Another approach is to flash chrome plate the stack parts or to make themfrom stainless steels; both methods will of course add to the mold cost. Forthe mold shoe (the mold plates) itself, instead of using oil paint, it can beprotected against corrosion with electro-less nickel plating (ENP), which hasthe additional advantage that it also protects some of the inner surfaces of the mold shoe, which would not normally be covered when the mold is justpainted on the outside. ENP also enters the cooling channels to some extentand protects them against corrosion caused by the coolant, but the pene-tration is limited and does not cover the walls of the channels deep insidethe plates. ENP is hard (70Rc) but thin and not resistant to scratches and wear.

The best method may be to make the entire mold shoe from stainless steel(SS). The basic cost of SS is higher than the cost of mild steels or pre-hardenedmachinery steels. However, when SS is bought in large quantities, the cost

difference can be much less. When molds are expected to run for a longtime, the advantage of SS over other steels can justify the higher cost. Wemust not forget that chrome plating or ENP also cost money. We must alsoconsider the costs of transportation to and from the nickel or chrome plater,the additional time required for these operations, the lack of control overthe transport, and the dependence on an outside supplier.

Another problem with chrome plating is that any change (requiring re-machining) of a chromed surface requires that the chrome must first be

removed from the steel part. This requires shipping the part to the plater forremoving the coating by a process similar to plating. After re-machining, thechanged part must again be shipped to be plated. This is an expensive andtime-consuming procedure.

Mold shoe material options: Pre-hardened plate steel, painted Plate steel with ENP Stainless steel

Always consider the total costs whencomparing mold material costs

8/13/2019 03086_03a


513.2 Coolant Supply

Note that corrosive plastics such as rigid PVC always require chrome platingor, better yet, SS for the stack parts.

The use of full-hardened (or pre-hardened) SS for cavities, cores, and insertsis quite common today, even though the steel cost is higher. When consideringthe expenses and risks with chrome plating of mold steels and the time saved,the total cost could be more than using SS.

Another solution for all these issues is to provide the molding plant and themold storage facilities with air conditioning or at least with controlled, low humidity air. Some modern molding plants have this equipment, althoughthis means added expense and may not be needed or cannot be justified

economically unless in cases where delicate products are mass-produced.Occasionally it can be useful to surround the machine with a shroud to keepthe environment immediately around the mold and machine at a desiredlow humidity with a portable dehumidifier.

3.2 Coolant Supply

The available cooling water supply (quantity, quality, and pressure of thecoolant) must also be considered. Also, remember, for water-cooling to beeffective, the water must flow fast enough to establish turbulent flow.Turbulent flow removes significantly more heat per liter (or gallon) and canbe calculated (see [5], Chapter 13).

3.2.1 Is the Coolant Supply Large Enough for thePlanned Mold?

There is no point to design a mold with an expensive, elaborate cooling systemif there is not enough coolant flow and pressure available to take full advantageof it. I have seen some mold plants that developed from only a few to a highnumber of machines, but neglected to increase the cooling water supply togrow with the rest of the operation. This resulted in the molds running much

slower than they could if the cooling water supply had been sufficient.Good cooling of a mold depends not only the coolant temperature but alsoon the volume of coolant that flows through the mold, measured in liters orgallons per minute. This volume depends essentially on the pressure differen-tial between IN and OUT of the cooling channels in the mold and on themethod of distribution through the mold (see [5], Chapter 13).

3.2.2 Is the Cooling Water Clean?Cooling water must be clean, i.e., free from contaminants and/or oxidizers,which corrode the inside of the cooling channels. This is where stainless steel

Table 3.1 Calculating Chiller Requirements

Resin Chiller lb/h/ton

HDPE 30

LDPE 35

PMMA 35

PP 35

PA 40

PPE 40

ABS 50

PS 50

Acetal 50

Tons required =

Resin lb/h/ton × lb/h of resin consumed

For highest productivity ensure thatthe cooling channels in the mold arefree from sediments (lime, rust, etc.)

8/13/2019 03086_03a



is of great advantage. The coolant must also be free from lime and dirt, whichwill gradually settle in corners of the cooling system and plug the coolingchannels, especially if the channels are small and in elaborate circuits, as isoften required in high production molds to cool small mold parts. Undersuch bad conditions, a mold will probably run satisfactorily and produce asplanned for the first few months, but because of buildup of dirt in the coolingchannels, the mold will gradually lose its cooling efficiency and run slowerthan it could with good, clean coolant. Dirt in the water will also requiremore mold maintenance, as the channels will have to be cleaned from timeto time. Such mostly unnecessary costs are often overlooked while worryingabout the high initial mold cost.

Rust is an insulator and will eventually slow the molding cycle as it buildsup.

3.3 Power Supply

Electric power supply is not always as stable as required, especially outsidethe larger industrial areas of North America and Europe. In many parts of the world, especially in developing countries, there are often considerablevoltage fluctuations because of weak and overloaded power lines; moldersexperience occasional, and sometimes even daily, “brownouts” (periods of lower voltage) and are often plagued with complete power failures (blackouts)lasting anywhere from just minutes to many hours. To say the least, thesestoppages are annoying, but they can also be very expensive if a mold stopsfrequently just because of failure of the machine controls.

Voltage fluctuations affect molding operations for two main reasons.

Logic controls are sensitive to voltage fluctuations and may requirevoltage stabilizers. Although this is a machine requirement, it needs tobe pointed out. Every time the machine stops, the mold also stopsproducing. In general, electronics are quite sensitive to high ambienttemperature.

Melt temperature. Virtually all heaters in molds and molding machinestoday are electric resistance heaters. The heat output of a resistance heateris proportional to the square of the voltage applied. A drop of just 10% involtage will reduce the heat output by 20%. While the barrel heaters of the extruder are always thermostatically controlled, a transformer, withoutfeedback, often controls the machine nozzle heaters. With heat controls,any reduction in voltage (and temperature) will be automatically com-pensated by having the heaters ON for longer time periods. In hot runner

molds, the hot runner manifold heaters are always equipped withthermocouples; however, because of the high initial costs (in the mold,and for the associated external controls required) many molds do nothave heat controls on the nozzle tip heaters and can therefore experience

A drop of 10% in voltage will reduceheat output by 20% if not thermo-statically controlled

Figure 3.3 Rusted mold components

8/13/2019 03086_03a


53

major temperature variations as the voltage varies. This will lead to troublein the mold’s performance. Even so, today, about 80% of the high-production hot runner molds are equipped with thermostatically con-trolled nozzles as the added costs can be easily justified with the increasedproductivity.

Cold runner molds: With such molds, power interruptions, while annoying,are not serious. If an interruption is only of short duration – in the order of a few minutes – the plastic in the injection unit is probably still hot enoughso that production can resume immediately, without causing problems. If the interruption takes longer, it will take again the time necessary to heat up

the injection unit before resuming production after purging.

Hot runner molds: With these molds, power interruptions can be more serious.Short interruptions of a minute or two can be tolerated without problems,but any longer stoppage will cause the plastic in the manifold and the hotrunner nozzle

to degrade, especially heat-sensitive plastics in the still hot manifold, and

it will freeze sooner, because the masses of the manifold are much smallerthan the masses of the extruder. It takes time to heat up the whole systemto operating temperatures, and the plastic both within the injection unitand in the hot runner system must be first purged before resumingoperation.

Note that well designed and built hot runner systems require less time forrestarting than poorly designed systems. A good hot runner system shouldbe ready for resuming production in about 10–15 minutes after any inter-ruption.

These details are important to understand before deciding on the kind of runner system to select for the mold. A hot runner system may be moresuitable than a cold runner mold for a certain application, but may causeendless grief if the power supply is poor. All the well-known and provenadvantages of a hot runner system can be lost because of the frequentstoppages due to power supply problems.

3.4 Will the Mold Run in a Variety

of Machines or a Single Machine?

The mold will often be required to operate in different models of molding

machines. This may result in quite some complications in the mold layoutand will certainly increase the mold cost. In particular, different locations of the machine ejectors can affect the ejection and the cooling layout of themold and the overall size of the mold.

3.4 Will the Mold Run in a Variety of Machines?

8/13/2019 03086_03a



The mold must be equipped with all features that are compatible withthese various (existing or planned future) machines. This applies to several

areas of the mold:

Shut height

Any downstream automation

Mold mounting (including any systems for quick mold changing)

Locating ring size

Sprue bushing size and shape Machine ejector pin locations

Cooling- and air-circuits

Hydraulic functions

Electrical connectors

If a mold is to be designed for one machine only, in one location only, it canresult in a simpler mold. For example, there would be no need to provide forvarious sizes of locating rings and the ejector mechanism and the moldmounting provisions could be designed for the pattern of the selectedmachine only.

3.5 Is the Mold Planned to Run in a NewlyCreated Operation?

It is a very desirable condition for the mold designer when a mold (or aseries of molds) are planned to be operated in a new factory (or in a separatesection of an existing factory), because it creates an opportunity for closecooperation of the mold designer with the planning of the whole project. Itprovides an opportunity to participate in the selection of the most suitable

machine for the product to be made, but also to take part in the plant layout,power distribution, cooling water system, and so forth.

This is also a good time to introduce standardization of many of the moldelements and mold sizes, of mold mountings (including quick mold changes),power and cooling connections, and any other feature that will affect notonly the mold(s) now under consideration, but also future molds for thislocation.

Standardization of mold components, molding machines, and ancillary equipment will not be further discussed here, but they are an importantfield where savings both in investment (costs of equipment) and increase inproductivity can be made.

8/13/2019 03086_03a


55

3.6 Projected Requirements

How many pieces of the product will be made from the planned mold? Thiscould be the most important question to ask before deciding on the type of mold required for any job. But this is also often the most difficult question toanswer, particularly if the product is new on the market. It is nearly impossibleto foresee if the product will find the hoped-for acceptance and increase insales, or if the product will not be accepted as expected. Also, assuming atotal quantity is known, what is the time frame when these quantities arerequired?

If 1,000,000 pieces of a new product are to be molded, the question is:

Is this a limited production run, say within four months (usually as soonas possible) or

Is this quantity needed every year, for a unspecified number of years, or

Is this quantity needed over the expected life of the product, e.g., 5 years,in which case the annual requirement is only 200,000 pieces.

3.6.1 Making Prototype or Experimental Molds

3.6.1.1 Prototype Molds

Prototype molds are required to make samples of a new product for evalua-tion of a newly developed shape, to see how the product appeals to the eyeand/or to the touch. Molded samples can be subjected to the expected stresses

and wear and the results are better than testing a hand made (machined, orassembled) model. The result also could be more accurate (and possibly cheaper) than a computer simulation. Because it is only important to moldthe overall shape of the product, without worrying about productivity of the mold, shortcuts can be taken everywhere: mold materials such as mildsteel, aluminum, even plastics (epoxy, etc.) can be selected, as long as they are sufficiently strong and resistant to the heat and the pressure of the injectedplastic. Working to close tolerances is usually not necessary. Generally, therewill be no need to worry about surface appearance (polish, engraving, evenflashing). There is no need for cooling channels; it will take just a little longerto cool the plastic before being able to remove the molded sample from themold. In many cases there is also no need for an ejector mechanism. An air

jet directed against the edge of the product at the parting line, or a few simpleejector pins that can be manually pushed to eject, may be all that is required.

Other features of the product, such as internal or external threads, can beproduced by using loose inserts in the mold that can be ejected with the

product and then unscrewed by hand. Loose inserts can also be used for oddshapes in the sides of the product, which would otherwise require side cores.Round holes or simple openings in the sidewalls could be machined afterthe molded piece is cold. These are just some of the mold features that can


Figure 3.4 Typical prototype mold for a lid,capable of 4 in to 8 in lid prototyping(Courtesy: Husky)

Figure 3.5 Single-cavity prototype moldfor production 2×2 system. In this case, theprototype stack was used as a spare in theproduction mold (Courtesy: Husky)

Projecting the number of moldedpieces is often the most importantand difficult question to answer

8/13/2019 03086_03a



be omitted to simplify the stack and to reduce the cost of the prototype mold.If prototypes are frequently required, the stacks could be mounted in acommon mold shoe, thus saving even more costs. The runner system wouldnormally consist of a simple sprue gate directly into the product or a sprueand short runner could be used for edge gating. The gate will then be cutmanually.

3.6.1.2 Experimental Mold

This type of mold is different from the prototype mold: it will be used mostly to establish the behavior of the plastic in a newly developed product during

injection. Some of the above cited shortcuts to save costs can be used, but ingeneral, the mold would be closer to a simple, single-cavity production mold.The gate should be located as planned for the production mold. The moldcould also be used to establish the most suitable location of the gate and themethod of gating for the product. Such a mold would normally require theproper finished appearance of the product. Note that especially in thin walledproducts, the finish affects the flow of plastic through the cavity space. Coolingefficiency is not as important as in a production mold, but some cooling

should be provided to maintain a stable mold temperature. Because thequality of a molded piece depends very much on the accurate repetitivenessof cycle time, an ejector mechanism should be provided rather than manualproduct removal to eliminate any operator-created variations in ejection (andcycle) time. An important feature of an experimental mold is often the facility with which some stack parts can be changed. This adds costs but will makeexperimentation easier. Experiments with such molds can also determinethe effect on molding cycles when areas of the mold are not cooled, littlecooled, or well cooled. Such information can be valuable before an expensive,multi-cavity production mold is designed. The difference between “ordinary”and “exceptional” cooling could mean much in engineering the productionmold. Reduction in cycle time achieved by exceptional cooling could beinsignificant and not worth the additional costs and complications to themold.

3.6.1.3 Combination of Prototype and Experimental Mold

This applies when an inexpensive mold is required to establish the shape of the product, but at the same time it is planned to explore market acceptanceof such product by manufacturing a few hundred or even thousand of samplesfor field testing. Typically, such molds should run “fully automatic,” but thereis no need to achieve maximum efficiency in molding, as with better cooling,better runner system, etc., and without special finish or most engravings.Such molds can also be used to establish shrinkage conditions.

I remember a case where a client wanted a very simple prototype mold tosee how a newly designed LDPE cover would fit as a shield over a metalproduct he had been selling for years. The prototype mold was supposed

Figure 3.6 Typical 4-cavity experimentalmold that will emulate the behavior of theproduction mold (48 up to 144 cavities)(Courtesy: Husky)

8/13/2019 03086_03a


57

to produce about 100 samples. We made a mold with a very simple cavity and core, all mild steel, with a few ejectors and a simple through-shooting

gate right into the center of the product; some cooling channels, no polish,no engraving. There was hardly a simpler mold possible. The clientpromised that if the new idea was accepted in the field, he would buy aproduction mold. After a few months, I called to ask him how the ideatook on, and he told me that the mold has already produced severalthousand pieces and was still in perfect condition, and that he wont needanother mold. A properly designed production mold would surely runfaster – i.e. produce more pieces per hour – but with really small quantitiesthis is not worth the extra cost.

3.6.2 Production Molds

Production molds are any type of molds other than prototype and experi-mental molds. At this point in the planning for a new mold it becomesnecessary to have basic information on

How many pieces will be required?

What will be the molding cycle?

Once these data are available, there should be not much difficulty to proceed,but both these data are usually difficult to ascertain.

Since the mold type and number of cavities will depend primarily on the

quantities required to be molded, we must first differentiate between thevarious possibilities as they present themselves, before deciding on the kindof mold that will be most appropriate.

3.6.2.1 New Products

The new, untried product is a common case and can be part of a new “invention” or an existing product previously made from a different material.

Will the market accept it as is in its new shape, made from injection-moldedplastics? Will it require modifications after complaints or suggestions fromthe field after it was launched on the market? Or will it be a disappointmentfor the seller, and soon disappear? Unfortunately, the “entrepreneur” takesall the risk when investing in the required mold. Of course it would beconvenient to keep mold cost as low as possible, but we know that this may increase the cost of the products in the long run. The cost of a high cavitationsmold may also affect the timing of the launching of the product. Should a

large production be anticipated, which will require a multi-cavity mold of high quality? In this case, if the product is not accepted in the field, the losscould be substantial. But there is also another, just as serious problem whenlaunching a new product: the investor was overly cautious and is waiting for


Conclusion: There is no clear answerto the above questions. It maydepend on the expected life of theproduct, which is often just asdifficult to estimate. Some products

are seasonal and the demand findsan early saturation point. Someproducts increase in demand untilsome competitive, similar or evenbetter product comes along, in whichcase demand for the original pro-duct could sooner or later disappear.One possible advice is to build a

mold for the initially estimatedvolume, and add 25% for surgedemands, unanticipated stoppage,and some growth

8/13/2019 03086_03a



the acceptance in the field. If the product is a great success, the first mold wasprobably not designed for the unanticipated, high demand. What would bethe best strategy at this point? Make another mold (or even several molds)

similar to the first one and run them side-by-side? This approach may havethe advantage of lower additional investment while providing more flexibility.It is easier to find several smaller machines than larger machines. However, alarger system, using a high-production mold, with more cavities, better runnersystems, better cooling, better ejection, more automation, and therefore higherup time, will result in the lowest cost of the products.

3.6.2.2 Existing Product, Large QuantitiesSome products are “timeless”, meaning that their annual quantities are moreor less constant and known. Their use may vary within seasons and evenwith the economy in general, but they remain essentially unchanged. Thisapplies to many technical articles, as well as to many packaging products,such as food containers and to medical products. In these cases, it is notdifficult to establish annual requirements and a projection for how long theproduct will be in demand. In addition, it is always important to consider

the whole system, i.e., machine, mold, and any after-molding operation(automation, product handling, packaging, assembling, etc.,) that will yieldthe lowest-cost product. With long and high production runs, even high moldcost is insignificant per unit produced and helps lower the product cost,provided it runs faster, longer, and with higher quality products.

3.6.2.3 Limited Quantities

Sometimes, a product is required in a limited quantity or for a one-timeoccasion only. This may be the case where a molded piece is designed for aspecial occasion or application. The quantities are relatively small but usually known. Frequently, a molded piece will be required as a promotional item,such as giveaway items to retail customers. Such promotions are usually limited in time and the requirements are stipulated at the beginning of anadvertising campaign. Usually, such promotion needs fast delivery of themolded pieces, and the total amount of pieces in a very short time span. Adecision will have to be made: Should the order be produced on a large,multi-cavity mold? This will yield the best piece cost but will require a largermachine, which may not always be available at the time the mold is ready forproduction. The mold cost will be higher but the cost per molded piece isprobably insignificant. The problem is that such larger molds will take longerto build and there may not be enough time. Also, it leaves the moldervulnerable, in case of machine or mold breakdowns, in which case therecould be no production at all.

As an alternative, several smaller, identical molds could be built which aresimpler and can be made faster by contracting out to more than one moldmaker if necessary. These (smaller) molds can be built faster and then be runon smaller machines, which are also usually easier to locate; if necessary,

Conclusion: Investing in the bestpossible mold is usually the key to asuccessful operation

Figure 3.7 High volume production moldfor a stadium cup (2×12 cavities, air eject,modular construction) (Courtesy: Husky)

8/13/2019 03086_03a


59

even at different custom molders. This will probably increase the piece cost,(a) because the smaller and less expensive molds are not as productive as alarger mold, and (b) because of the added cost of dealing with more than

one source. However, this approach will also ensure that any breakdown willbe less serious to the customer. In all these cases we assume that the moldsare complete, self-contained molds.

3.6.2.4 Short Runs, Small Production

For short runs or for very small total production, when products are relatively simple and small, and when it is known that the annual requirements are

also small, there are two alternatives:

Individual molds, with the least amount of “high productivity” features(especially good cooling, hot runners, etc.) or

Making inserts for so-called “universal mold shoes” which are listed inmost of the mold supply house catalogues.

Such mold inserts for universal mold shoes usually do not cost much more

than the mold stack for a regular mold. They can be mounted in the “shoe”and run by itself, in pairs, or in multiples, or even in combination with insertsfor another product. Mold changing is usually simple and fast.

There is also the alternative of using a regular mold shoe, such as would beused for a conventional (”designated”) mold, and solely changing the stack or the inserts. This is quite practical if the molding shop personnel are familiarwith mold work or where a mold shop is connected with the molder, otherwisethere is always the danger of damage to delicate stack parts while changing.

Larger products, which will not fit a standard universal mold shoe, or whichare too large to fit into a common mold shoe, will have to be built as designatedmolds, but can be using any shortcuts available, as mentioned above in Section3.6.1.1, to keep the mold cost as low as possible while still getting good quality products.

3.7 Forecasting the Cycle Time

After we are clear on the question of how many of the products will berequired, our next step is to arrive at an estimate of how fast the piece can bemolded. Some molders like to indicate the number of seconds (cycle time)to mold a piece (or shot); others prefer the number of shots per hour orpieces produced per hour. Either method is suitable.

The cycle time of any molding operation depends on a number of parameters,which will be discussed in detail in the following sections. It is important tounderstand these dependencies before settling on a reasonable figure for thecycle time.

Conclusion: Use standard moldshoes if possible, or use simple,dedicated molds


3,600 s (1 hour) divided by the cycletime (in seconds) equals the numberof shots per hour. Shots per hourtimes the number of cavities equalsthe number of pieces produced perhour

Figure 3.8 Standard mold base for a strippedclosure application (Courtesy: Husky)

Conclusion: Go small and use severalsources

8/13/2019 03086_03a



3.7.1 Type of Plastic Molded

There are several issues to consider:

Melt temperature required to be able to inject and to fill the cavity space.Higher melt temperatures require longer cooling times before the piecescan be ejected.

Thermal conductivity of the plastic. With lower heat conductivity it takeslonger for the heat within the melt to travel to the cooled mold wallsthan with higher conductivity. However, with very thin-walled products,without heavy sections, the difference in conductivity can be insignificantbecause of the very short distance the heat travels to the cooled walls.

Injection speed , especially through the gate(s). Some plastics are sensitiveto shear stresses caused by high injection speeds (especially in the gatearea) and exhibit degradation of the plastic flowing into the cavity space.In this case, lower injection speed (and pressure) will be required, whichwill affect the molding cycle. Some plastics, especially the largest groups(“commodity plastics” such as PS, PP, and PE) used for many products,

are little affected by shear stresses, but many heat sensitive plastics can bedamaged (degraded) by high injection speeds.

Crystallinity of the plastic . Crystalline plastics such as PE and PP requiremore heat input than amorphous plastics such as PS to reach the requiredmelt temperature. The resulting higher heat content in the plastic willthen require more time to cool before ejection is possible. Occasionally,a mold may be planned to produce both types of plastics: Not only willthe shrinkage be different but also the molding cycle.

Fillers. Filling of the plastic with inert materials (fibers, talcum, etc.) canalso adversely affect the cycle time. In addition, the shrinkage will besmaller than when using the same but unfilled plastic. We must alsoconsider that many fillers tend to erode the mold materials, especially the gate; when selecting the materials for the gates and the mold in general,this must be taken into account because of its effect on the mold cost.For example, a high strength, wear resistant steel for a gate insert is goodto resist abrasion but because it is also a poor heat conductor, the gate

area is more difficult to cool. This will result in a slower molding cycle.This is especially important with fast cycling molds.

3.7.2 Wall Thickness of Product

The wall thickness plays a significant role in the cooling process and thusinfluences the cycle time. Ideally, the walls should be uniform throughout

the whole surface of the product; however, this is rarely achievable or practical,except in some lids (covers), some containers, and some flat products. Mostproducts have thicker sections in design features such as hubs for fasteners,but even with otherwise uniform walls, there will be heavier sections at the

Figure 3.9 Resin to be molded cansignificantly affect the cycle time(such as this PS)

8/13/2019 03086_03a


61

junctions of walls and at the base of ribs. It is not the uniform wall thicknessthat governs the cycle time but these heavier sections. This becomes evenmore complicated, because these heavier sections are usually more remote

from the areas that are easily provided with good cooling and thereforedepend on longer paths for the heat to travel to the cooling channels. Betterheat conducting materials are occasionally used to help to remove the heatfaster from these “hot spots”. There are charts, nomograms, and computerprograms prepared to relate the wall thickness to the cooling time; they aremostly based on the simplest but often unlikely cases, namely perfectly uniform wall thickness throughout, and equally well cooled surfaces through-out the mold (see Fig. 3.10).

3.7.3 Mold Materials

The selection of mold materials for the stack parts also has an effect on themolding cycle. There is some, but relatively little, difference in (the ratherpoor) heat conductivity between the various hardened alloy steels (“moldsteels”) commonly used. The conductivity of pre-hardened machinery steelsthat are often used for larger stack parts is somewhat better but still poor.So-called “mild” steels have a still better conductivity, but are rarely used forstack parts because of their low physical strength, poor polishing quality,and the frequent dirt enclosures.

Metals with much higher conductivity, such as aluminum and copper, arenot used because of their softness; certain aluminum alloys are easily machinable and relatively inexpensive and are used in blow molds wherepressures are much lower and occasionally in injection molds in areas of low

stresses, and even in prototype molds.Beryllium copper (BeCu) alloys have aheat conductivity about four to seventimes better than steel. Their use in mold stacks (usually as inserts in cavitiesand cores) is often of advantage in areas that require the highest heat removalrate possible. In fast running molds, the difference between steel and BeCucan be a few seconds, or even just a fraction of a second. It may not appear tobe much, but a saving of 0.5 s can translate into a large increase in production.

For example, a mold with steel cavities runs at a 4 s cycle, it will produce3600 ÷ 4 = 900 shots per hour. By using BeCu for the cavities it may runat a 3.5 s cycle; 3600 ÷ 3.5 = 1,028 shots per hour, an increase of 14% inproductivity!

The reason that BeCu is not used more in molds is that it is much moreexpensive than steel and not as strong. Because BeCu is softer than hardenedsteels, it is usually inserted in steel (cavities or cores) and it should never beused on the parting line or on alignments or shut-off tapers. With largerpieces, there is the danger of porosity. Smaller parts can usually be machinedfrom forged or drawn rods and bars, but for larger parts, pressure casting of the blanks for mold parts is required.

Figure 3.10 Schematic relationshipbetween material, uniform wall thickness,and cooling time. These graphs are onlyshown to demonstrate that there is adefinite advantage in designing withthinner walls and with uniform thickness

Note: BeCu requires specialprecautions in the machiningoperations because of hazardousgases created when working withmachine tools


Figure 3.11 This 2-cavity lid mold usesBeCu inserts (copper color) on the cavityring and the gate inserts to significantly cutcooling time (Courtesy: Topgrade Molds)

8/13/2019 03086_03a



3.7.4 Efficiency of Cooling

The purpose of mold cooling is to remove, in the shortest possible time, the

heat energy that entered into the cavity space during the injection of the hotplastic melt. The higher the efficiency of removing the heat, the higher theproductivity of the mold.

Molds for Small Production (Fewer than Approx. 1,000 Pieces)

With any injection mold, over time, the heat will dissipate through the moldinto the surrounding air and into the machine platens. This could beconsidered as sufficient in cases where only a few pieces are required andwhere the cycle time is not important. In this case, there would be no need toprovide any mold cooling at all.

Molds for Large Production of Thin-Walled Products

On the other end of the scale for cooling efficiency is the cooling for a moldfor fast running, thin-walled products. The mass of each product may berelatively small, but because the cycle times are short, large amounts of heat

(in the hot plastic) per unit of time enter the mold, which must be well cooledto ensure that its temperature is kept stable and at an optimal (low) tempera-ture. This will require the best possible cooling methods, which are more costly to design and to manufacture. These molds use the most suitable (andsometimes expensive) mold materials to facilitate the rapid removing of theheat. The higher costs incurred will usually be worthwhile, because they resultin a mold with higher productivity and in lower costs per molded piece.

Molds for Large Production of Heavy-Walled Products

As the plastic cools during molding, it shrinks onto the core, away from thecavity walls. After losing contact with the cavity walls, even the best cavity cooling will not do much in removing heat from the product. But the relatively short time the plastic is in contact with the cold cavity walls is enough tocreate a rather thin, rigid but still warm surface, while the core coolingcontinues to remove heat from the inside of the plastic walls. While the rigidouter skin allows early opening of the mold and to pull the product (whilestill on the core) out of the cavity without risk of damage, it would not bepossible to eject it at this time, because the outside of the product could bedamaged despite the rigid skin.

It is usually fairly easy to provide adequate cooling to the cavity but oftennot easy to cool the core, because

The volume of the core is usually much smaller than the volume of the cavity,

There may be ejector pins going through the cores, and

There are sometimes air channels in the core.

Figure 3.12 Mold cooling schematic

8/13/2019 03086_03a


63

The latter two conditions reduce the space to accommodate effective coolinglines.

But there are a number of possibilities to increase the productivity:

The first method going back to the earliest manufacture of injectionmolded product is simple and still being used occasionally for very heavy,simple, and not particularly closely dimensioned products. The moldopens and the outwardly cool but inwardly still hot and soft (and thereforeeasily damaged) products drop into a container with cooled, circulatingwater, from which they are then removed and dried either by hand or aconveyor carries them through an air cooling tunnel for drying. This is arather crude method, but can be quite efficient in some cases. For nylonsand other materials that require high water content to reach their physicalproperties, this water immersion may even be an advantage. For othermaterials, water absorption could be damaging and therefore this methodshould not be used.

Another typical method is the handling of flat products such as trays,but also other – usually larger – shapes that tend to warp after beingremoved early from the mold. The still hot (and easily damaged) productis placed into a cooling fixture (some are simple, others quite elaborate),where it is held by weights, clamps, or in a mechanically locked frame orany other suitable method, until it is cool enough and keeps its proper(as molded) shape without warping. There could be a small number of such fixtures beside the machine where pieces are successively placed asthey are ejected and then packed once they are cold. This method too israther crude and labor-intensive, but can shave quite a few seconds fromthe cycle time. The alternative is to keep the mold closed until the molded

product is cold enough and will not warp after ejection.

Hot products can be held in actively cooled fixtures (post-mold-cooling).

Typical examples for post-mold cooling are the heavy walled preformsused for the manufacture of blown PET bottles. These products arerequired daily by the millions world wide, and every fraction of a secondsaved amounts to huge savings over the years. The biggest problem is theintense cooling and the long time required to cool the very thick walls

which, in the next step of operation, will be reheated and blown up tothe final bottle shape in special machines. The usual wall thickness is inthe order of 3–4 mm (1/8–5/32 in.). Cooling must be very intense andefficient to prevent crystallization of the plastic as it cools, requiring very cold water with large flow. But while it is relatively easy to cool both thecavity and the core, the cycle could be still in the range of 30–35 s or evenmore. The problem of efficiently cooling the preforms has been, and stillis, the subject of much research and many improvements are developedin this narrow field of injection molding. The simplest approach toincrease productivity is to increase the number of cavities; while thisobviously has improved the productivity, the main target of research ishow to reduce the cooling time (note: from the early beginnings of these


Figure 3.13 A part is cooled very efficientlywhen dropped into cold water

8/13/2019 03086_03a



molds, the number of cavities has increased from 4 to 192 in many production molds).

The key to reducing the cycle time is to increase the time of the heattransfer from the plastic to the cooling media (water or air), either inspecial cooling fixtures or by removing the products earlier from thecavities but leaving them longer on the cores, as explained below:

– The still very hot products can be ejected from the core into water-cooled, tightly fitting sleeves, advanced by a robot into the moldingarea. The whole array of still hot preforms is then transported out of the molding area and a new injection cycle can begin. There are several

(mostly patented) executions of this method and the cycle times havebeen reduced by 1/2, 2/3, or even better. This significant increasein productivity is not inexpensive, but the cost of the necessary equipment can be written off in a very short time by the savingsachieved.

Figure 3.14 shows a 48-cavity mold for PET preforms from the rearof the machine. It shows (right) the array of cores (A) on the movingplaten. The side-entry robot (B) carries 3 × 48 cooling receptacles

(C), their position timed so that every time after the machine ejectsan array of still hot preforms into the cooling receptacles, the robotplate shifts so that by the next cycle, another (empty) set of coolingreceptacles faces the cores. Before the third cycle of unloading themold, the now cold preforms of the first unloading cycle are ejected.For every ejection, the plate swings 90° so that the now cold preformsdrop onto a conveyor (D) for removal to a shipping crate. At a 12–14 s cycle time, this system yields between 14,400 and 12,340 preforms

per hour.– A more recent development is the use of identical sets of cooled cores

mounted on a rotary moving half of the machine, either in sets of 2or of 4 arrays. This requires a special machine; however, it does notrequire the above-described robot. The plastic is injected and themold opens as soon as the cool skin has formed on the outside of thepreform and permits the products still on the cores to be pulled outof the cavities. As soon as the first array of cores is in the open position,

the core carrier rotates and a new array of cold cores enters the cavity for the next shot. The preforms stay on the cores until they are ready to be safely ejected. A core side with two arrays of cores will rotate180° every cycle and eject as it rotates, while the products are pointingdownwards, before reaching the position where they are again in linewith the cavity for re-closing the mold. A core side with four arrayswill rotate 90° at every cycle, and will eject when the products arepointing downwards before the rotation that brings the empty cores

into line with the cavity and closing the mold. These are very sophisticated systems, but they can be easily justified, especially because they require much less floor space than the systems usingrobots.

A B

C D

Figure 3.14 48-cavity mold for PETperforms with cooling receptacles(Courtesy: Husky)

8/13/2019 03086_03a


65

Figure 3.15 shows a portion of the special indexing machine, with a96-cavity mold for PET preforms. The mold consists of one array of 96 cavities, shown on the right (A). There are 4 arrays of 96 cores (B)

on the indexing section of the machine. The machine is timed sothat after the first injection, as soon as the preforms are cool enoughthat they can be pulled out of the cavities, the mold opens, theindexing clamp (C) rotates 90°, and the mold closes again for thenext injection cycle. The injection repeats. When the second shot isready to be removed from the cavity, the indexing repeats for another90° turn. The first molded preforms arrive now in a position oppositethe injection unit. The same step is repeated for the third injection

cycle. After the clamp rotates another 90°, the first array of preforms(from the first injection), which has been on the core through morethan 3 complete injection cycles, has arrived in the position facingdownwards. By this time, the preforms are cold enough to be ejectedsafely onto a conveyor (D) located below the clamp. The now un-loaded (empty) array is ready to enter the cavities again for the nextcycle. At a 11 s cycle time, this system yields more than 31,400preforms per hour.

These indexing machines can be used for any very-heavy wall product,wherever large quantities are required, such as cosmetic products jars.

For more details on the economics of PET preforms, see Section 3.4.10.

AB

C D

Figure 3.15 A special indexing machinewith a 96-cavity mold for PET preforms(Courtesy: Husky)

Figure 3.16 Schematicof the indexing clamp(Courtesy: Husky)

Drive pinion

Tension wheel

Movin platen assemblyIndex clamp unit (Operator side)

Shuttercylinder

Stationaryplaten

Mold strokecylinder (4)

Tiebar (4)Clamp base

Guide rail

Runner block

Tension plate

Tiebar nut

Turret gear

Timingdrive belt


8/13/2019 03086_03a



Molds for Most Other Products

Molds for most other products are equipped with a cooling system some-

where between these extremes. Often, in molds for intricate shapes requiringmany stack inserts (usually technical products), the cavities and cores arealmost impossible to cool close to the cavity space. Cooling can only beachieved by conducting the heat through the stack walls and the inserts tothe cooled mold plates located immediately behind them and/or surroundingthem. This results in slower heat removal than cooling the cavity or corewalls directly, but is often the only way to keep the mold at a stable tempe-rature. Similarly, in some products, which have unavoidable hot spots (thick sections), there is not much point in providing excellent cooling for the areas

that could be easily cooled just because there is enough room for such cooling,unless it is possible to provide better cooling to these hot spots.

The slowest cooling area of the mold always governs the molding cycle.

It is amazing how many mold designers and mold makers overlook this pointand then wonder why “despite the massive cooling provided” in the moldthey cannot achieve a better cycle time. As already pointed out, especially inmolds for containers or other cup-shaped products, some mold parts, such

as cavity blocks, have ample space for cooling circuits, while the core, whichreally should have more cooling, does not provide enough space, because it ismuch smaller than the cavity block and has to accommodate the ejectormechanism as well, which is encroaching on the available space for coolingand air channels. Unfortunately, many mold designers don’t understand thisand see the large available space in the cavity, provide more cooling than isnecessary for the job in non-critical areas, and thereby waste money.

3.7.5 Venting

Venting is another feature that can affect the molding cycle; it is importantto evacuate the air in the mold in front of the inrushing plastic duringinjection. Venting is important for any mold to ensure that the plastic canenter freely into all areas of the cavity space and must be properly specified.Venting is especially important with fast running molds.

A typical example is an experiment with a 4-cavity mold for a smalldisposable container. With standard, yet ample venting, the mold ran at17 shots per minute. It had continuous vent gaps along the rim; by simply providing more vent channels (8 instead of 4) to allow the air to escapeeasier into the open, the production could be increased to 20 shots perminute, an increase in productivity of more than 17%!

Figure 3.17 shows a 4-cavity mold for a 4 lb (PP) margarine tub.

The slowest cooling area of the moldalways governs the molding cycle

Although some ribs and bosses mayfill without venting, all points wherethe plastic finishes filling should bevented. Regarding vent sizes, it isbest to consult the materialsuppliers for their recommendation

67

8/13/2019 03086_03a


67

3.7.6 Effect of Molding Machine on Cycle Time

Several features of the molding machine affect the mold productivity andwill be discussed in the following. It is important to be familiar with themachine for which the mold is to be built in order to arrive at a more accurateestimate of the probable cycle time. In case of similar or even identicalproducts and molds, the cycle time can vary considerably when run ondifferent make and size machines. Machine factors affecting the mold pro-ductivity are dry cycle, injection speed/pressure, tonnage, and recovery time.

The molding cycle time is the dry cycle time plus the time required to injectand cool the molded piece(s) sufficiently for ejection, plus any added moldopen (MO) time.

3.7.6.1 Dry Cycle

The dry cycle is probably the most significant variable from machine to

machine and the feature that can influence the cycle time more than otherfactors. Dry cycle is defined as the time (in seconds) it takes the moving platento move over the length of the stroke, from the mold open (MO) position toclose, clamp up, unclamp, and then return to the MO position. Obviously,the larger the masses (platen and moving mold half) to be moved, the morepower will be required to accelerate, to move, and to decelerate them for asoft stopping in both the Mold Closed (MC) position and in the MO position.Therefore, smaller machines can have shorter dry cycles than larger machines,but also, better machines have shorter dry cycles than lower performancemachines. Machines in the 500 kN to 10,000 kN (50–1,000 ton) range canhave dry cycles from 1.5 s up to about 10 s. But there are also other, mainly older machines with dry cycles up to 20 s! It obvious that for large production

A

B

C E

D

D

Figure 3.17 4-cavity mold for a 4 lb (PP)margarine tub (Courtesy: Husky).

The mold exhibits modular construction,floating cores, and cavity lock. Air ejection(absence of an ejector box) allows veryshort shut height of the mold. Air jets (A)(4 per cavity) and the 2 blow-down air jets

(B) on top assist fast removal of productsfrom the molding area. Note the intricatesystem of continuous vent gaps (C), ventinggrooves (D), and channels (E) to permit fastfiling of the cavities. The productivity of themold at 6.0 s cycles yields 2,400 tubs perhour.


Molding cycle (s) =Dry cycle + Injection + Cooling+ Ejection + Mold open

68 3 C F Aff i P d i i

8/13/2019 03086_03a



and a mold with a short molding cycle, the length of the dry cycle is muchmore important than with a mold that requires a long molding cycle.

Figure 3.18 depicts the approximate motion of the closing of the movingplaten. It is typically an S-curve. The speed at the start of closing is ratherslow; then, a fast speed is reached for much of the closing stroke, until theplaten slows down again and finally creeps the last few millimeters beforereaching the closed position, when the stretching of the tie bars begins as theclamping force is created. Similarly, but in reversed order, the mold is first“unclamped”; then starts opening slowly, accelerates and moves fast until itis slowed down again for a gentle stopping in the MO position.

Note that the following illustrations (Fig. 3.19 and following) show the speed(velocity) of the moving platen as a straight line that really represents theaverage speed from start to stop (see Fig. 3.18). The opening speed is notnecessarily the same as the closing speed. In most machines, segments of thespeeds in either direction can be adjusted to best suit the molding conditions,but molds are often run at the maximum available speed.

Note also that the straight line representing the average speed terminatessooner than at the end of the tie bar stretch time, i.e., at the moment when

the mold halves meet and the tie bars begin to stretch. This fact allows theinjection to start earlier; as soon as the mold closes, just at the moment themold halves “kiss off” and the tie bars begin to stretch. Any significant forcesinside the mold will commence only when the cavity spaces are almostcompletely filled. The filling takes usually longer than the time required forthe final clamp-up.

Starting the injection sooner, even by only a second or even a fraction of asecond, will result in a significant gain of cycle time and productivity,

especially with molds running at short cycles.

The tie bars’ stretch provides the necessary clamping force (preload). Thisforce must be greater than the force created by the injection pressure insidethe cavity space, which tends to crack the mold open at the parting line.

The difference between the time required for the shorter and the longer dry cycle is wasted time.

For example, if the estimated combined injection and cooling time is 3 sand the dry cycle is 3.5 s; because MO = 0, the molding cycle is 6.5 s; thiscorresponds to 554 shots per hour. If the dry cycle were 5 s, the moldingcycle would be 8 s, and the mold would yield only 450 shots per hour, aconsiderable loss of production. This is certainly enough to seriously consider the choice of a faster machine, especially when the expectedproduction is high.

Production with zero mold open time (MO = 0) has been achieved on many smaller molds, even with short dry cycles (in the order of 2 s), but can alsobe achieved with molds equipped with automatic unloading equipment wherethe action of the “take-out”, i.e., the mechanism that reaches into the mold

Mold Open time of zero (no moldopen time) is an ideal condition

Figure 3.18 Schematic showing the closingmotion of the clamp

Figure 3.19 Schematic showing a shorterand a longer dry cycle

693 7 Forecasting the Cycle Time

Next Page

http://03086_03b.pdf/

http://03086_03b.pdf/

http://03086_03b.pdf/

http://03086_03b.pdf/

8/13/2019 03086_03a


69

to remove the products while the mold is opening and closing, is mechanically linked with the mold open and close motion.

Figure 3.20 shows a 2 × 8-cavity stack mold for dairy tubs with modularconstruction, built for a machine equipped for stack molds. Air ejection,cam (A) operated swing arms (B) with suction cups (C) to pick the productsfrom the cores and deliver them into the chutes (D) – shown in blue – on theside of the cavity plate. This mold operates with zero mold open time

(MO = 0) and has a productivity at 4.5 s cycle of 12,800 tubs per hour.

In the early 1990s, I observed several 8,000 kN machines producing largeproducts requiring an injection and cooling cycle in the order of 10 s.The dry cycle of these (then new) machines was 18 s (!), resulting in amolding cycle of 28 s and yielding 128 shots/h. On a comparable sizemachine, but with a reasonable 6 s dry cycle, the total cycle would havebeen 16 s, or 225 shots/h, or an increase of 76% in production. At the

time, I asked the factory mechanic who installed these machines, if therewas any possibility to decrease the dry cycle time and was told that themachines were designed to run so slow to save on the expensive hydraulicand electrical components needed for higher speeds. The owner hadbought these eight new machines mainly because their price was muchlower than comparable size good machines. Did he really save money?After 3 years of running these machines, this molder got rid of them allby selling them as scrap iron. Nobody else wanted the machines eventhough they were in “good” running condition.

This is probably an extreme case but it highlights how important it is toconsider the dry cycle when buying a molding machine.

A

B

C

DFigure 3.20 Modular 2 × 8-cavity stack mold for dairy tubs, built for a machineequipped for stack molds (Courtesy: Husky)


03086_03a

Documents