Indirect Manufacturing Injection Molding with MJP Parts

It is often advantageous to use rapid prototyping in the fabrication process of another part.  The 3D printed part is typically destroyed or discarded after being used and can make one or more of the final desired part.  This type of process is termed “indirect manufacturing” or “indirect prototyping” and the 3D printed parts are commonly referred to as “indirect tooling.”  Many traditional manufacturing processes can leverage rapid prototyping in this way, the most common uses being plastic injection molding, thermoforming, sheet metal forming, sand/investment/flask casting, indirect and direct printed silicone/urethane molding and wax molding for investment casting.

Injection Molding.jpg

Injection Molding

Thermoforming .jpg

Thermoforming

Sheet Metal Forming.jpg

Sheet Metal Forming

Investment Casting.jpg

Investment Casting

Direct Printed Eggshell Mold.jpg

Direct Printed Eggshell Mold

2-Half Direct Printed Mold.jpg

2-Half Direct Printed Mold

Indirect Manufactured MJP Mold.jpg

Indirect Manufactured MJP Mold

(Using MJP as a Master Pattern)

Wax Molding.jpg

Wax Molding

(for casting)

 

Rapid Tooling / Indirect Manufacturing Examples using MultiJet Printing Technology

Injection molding is a manufacturing process that produces parts by injecting a molten material into a mold.  Injection molding can be accomplished with many materials including metals and glass, but is most commonly done with thermoplastics and is termed “Plastic Injection Molding.”  Plastic injection molding can produce highly complex parts inexpensively and quickly with relatively tight tolerances, excellent surface qualities and in an array of different material properties.  For these reasons, it is by far the most common way to produce large volumes of finished plastic parts for every kind of commercial and industrial need.  In this process, plastic in pellet form is fed into a hopper that feeds a heated barrel that melts and mixes the material using a helical shaped screw.

BOY 22D Plastic Injection Molding Machine.jpg Plastic Injection Molding Pellets.jpg
Left:  BOY 22D Plastic Injection Molding Machine, Right:  Plastic Injection Molding Pellets

 

The material is injected under pressure into a machined mold cavity that has the inverse shape of the required final part geometry.  The molten plastic conforms to the contour of the mold and quickly cools and hardens in the configuration of the cavity.  The mold is often made with water cooling lines to facilitate higher cycle rates.  After cooling, the parts are typically removed from the mold with the use of ejector pins that are integrated into the part and tool design.  Numerous different thermoplastic materials are prevalent in this process due to their ability to repeatedly soften and flow upon heating.  They are also commonly available in a wide variety of highly functional material properties and are easily recycled as part of the manufacturing process.  The mold is typically made of metal, usually either steel or aluminum and is precision-machined to form the features of the desired part.  The mold is composed of two halves that come together to form the correct cavity.  The side with male features is called the “core” and the side with the female features is called the “cavity.”  These are often referred to as the “A” and “B” sides or “top” and “bottom” halves of a mold.  The molds are made by a mold-maker (or toolmaker) and can be made to form a single part (single cavity) or numerous parts at once (multi-cavity).  Molding machines (or presses) are rated by tonnage, which expresses the amount of clamping force that the machine can exert.  This force keeps the mold halves closed during the injection process and can range from 5 tons to tens of thousands of tons.  The machines typically have standard mounting features that accommodate the installation of a standard metal piece on both the A and B sides that together are called a Master Unit Die (MUD).   The Master Unit Die (MUD) is a metal part designed to hold interchangeable mold halves. These frames are typically purchased as blanks and then modified with the molding features.  The MUD blank is standardized for use in a variety of injection molding machines and serves as a quick-change system that allows increased productivity, design simplification, and reduced tooling costs.

Blank Plastic Injection Molding Master Unit Die (MUD.jpg

Blank Plastic Injection Molding Master Unit Die (MUD) Core and Cavity Halves

The “mold” or “die” are both common terms used to describe the overall tool.  High volume molds are designed with hardened steel.  Aluminum molds are used for lower volume and are quicker to machine and are easier and less expensive to create.  The thermoplastic resin flows through the barrel of the machine and enters the mold through a long flow channel in that is part of the injection molding machine and is called the “sprue”.  The end of this channel is composed of a flat or rounded piece known as the “nozzle” of the injection barrel.  A “sprue bushing” is built into one halve of the mold and is used to seal tightly against the nozzle.  Past the sprue bushing within the mold, the molten plastic runs through channels that are machined into the faces of the A and B halves of the mold.  Plastic runs along these channels and they are thus referred to as “runners”.  The plastic flows through the runners and enters the actual part through a specially designed feature called the “gate”.  There are different types of gates based on their geometry.  A special feature knows as a “cold slug well” or “cold well” is added somewhere in the mold before the plastic enters the gate and functions to trap the colder plastic material at the flow front.  This cold well is needed because the tip of the nozzle is pressed against the cold water-cooled mold and the material at this tip is therefore not at the appropriate temperature for robust part formation.  The mold is clamped together with a very high pressure and is sealed tightly to avoid plastic leaking out of the seal of the mold halves (the leaked plastic is known as ”flashing”).  Small “vents” are often needed to allow the air within the mold to escape as the molten plastic fills the cavity.  The location along the edge of the part where the two halves of the MUD tool come together is called the “parting line” (P/L).  Angled sidewall draft of the part surface is used on both sides of the parting line to facilitate part release from the mold.

Side view of the mold showing the tip of the nozzle.jpg Top view of the mold showing the runner.jpg

Left:  Side view of the mold showing the tip of the nozzle, the cold slug within the nozzle and the cold slug well. 

Right:  Top view of the mold showing the runner, gate and vents as well as a cold well

(Note:  the nozzle is part of the molding machine and the interface between the nozzle and the MUD is called the sprue bushing which is part of the mold)

 

Often there are areas where the core and cavity sides of the mold come together to form a feature.  This area is called a “Shut-off” and if the plastic flows past the shutoff it will create flashing on the interface edge similar to that on the main parting line between the mold halves.

Shut-off feature between the core and cavity .jpgShut-off feature between the core and cavity that crates the underside of a snap fit
 

 

Rapid Prototyping vs. Traditional Machined Metal Injection Molding and RTV Silicone Molding

Traditional plastic injection molding that utilize a metal mold is a highly developed and extremely common technology.  Engineers are introduced to this type of design starting in college and there are vendors worldwide with the capability to mass-produce parts.  Parts to be injection molded must be very carefully designed to facilitate the molding process.  Common design rules for parts include the desire for relatively thin and constant wall thickness, use of draft angles to allow for part removal, and the use of rounds to allow for optimal filling and flow.   Senior design engineers are well versed in injection mold design rules and can spend many months or years working closely with suppliers, vendors and mold-makers on a specific design.  It is a well developed, but very complex and lengthy process involving many specialists.  There are also service bureaus that specialize in making prototype aluminum tooling faster, but that still requires weeks of time instead of the normal months long process required to make a steel mold.  Because of this long lead time and complex process, early in a design cycle, it is also possible and a common engineering prototyping workflow to make an RTV mold (Silicone mold) of a part using a 3D printed master pattern.  The RTV mold is able to numerous polyurethane parts.  This is typically done in a design phase where numerous parts are needed with reasonably functional properties, but before the final injection molding tooling is ready.  However, this typically takes specialized skills and equipment and can take numerous days or weeks to complete.  Even more important, the final RTV molded part does not have thermoplastic material properties as the process can only use two-part curable resins that can be poured into the RTV mold.  Rapid prototyping of printed metal is possible; however, 3D metal technology is expensive and post processing involves many of the traditional cutting, machining, sanding, and polishing operations that plague machined metal mold lead times.  Also, support structures can cause scarring defects and can be difficult to remove.  Three places where rapid prototyping with metal additive manufacturing can add value are

  1. Conformal cooling
  2. Geometrically complex inserts for traditional tools
  3. Custom inserts for traditional tools. 

Conformal cooling involves adding complex cooling channels into a part which can otherwise not be machined.  Also, many very small metal features can be difficult to machine with traditional methods, but are easy to print with metal rapid prototyping.  For example, relatively tall and slender rods and complex upward facing small features or surface textures are difficult to machine with traditional methods, but can be easy to 3D print.  Finally, metal printers can be used to make an insert that fits into a traditional machined metal mold.  In this way, a few parts can be made that have a unique marking compared to other parts (like a serial number or batch date).  Of course, rapid prototype plastic tools do not come close to the durability of metal tools needed for long life and do not have the thermal conductivity required to remove the heat for high cycle rates.  However, there are places where plastic rapid prototyping add tremendous value to injection molding. 

  1. Custom or rapid, low-volume prototyping or production parts using traditional thermoplastic
  2. Unique Inserts for traditional tooling
  3. Assist in the rapid development of a metal tool design

The two most important properties 3D plastic tools bring are 1) lead time and 2) mechanical properties of the final part.  In terms of prototyping with printed plastic molds, it is possible to design a part, print it overnight, and injection mold the part the next day using the product intent thermoplastic material.   Therefore, rapid prototyping with plastic materials is much faster in terms of lead-time and is much more cost competitive compared to machined metal and is able to inject the actual thermoplastic material.  Also, 3D plastic parts that can be inserted into a traditional metal mold can be very competitive to print vs. using a traditional machining process as each individual insert is typically unique and that type of character-based marking can be complex to machine.  Also, inserts are smaller (faster and cheaper to print) compared to the full part size and will only have to undergo fractions of cycles compared to the main tool usage.   Finally, the low lead time and low cost for plastic prototyping can add value in the design of a traditional metal tool.  Iterative testing is often required in the design of a metal tool.  The process requires a complex CNC machining operation, likely with numerous part and machine setups.  This requires an expert machinist and can take substantial time even after the job makes it out of the queue and to the shop floor.  High cycle time or high number of total shots is not important if the rapid prototyped plastic tool is only going to be used a few times for tool testing purposes or to make a few form/fit/function prototypes for design qualification.  Therefore, even if a machined steel tool will ultimately be needed, rapid prototyping can be used in the tool design process to quickly test and iterate on the final design.  This allows for rapid review, modification and redesign.

Key_ injection_ molding properties for traditional machine metal, 3D Rapid Prototype Metal or Plastic.png

Key injection molding properties for traditional machine metal, 3D Rapid Prototype Metal or Plastic

 

Designing and Leveraging the Properties of MJP Technology in Injection Molding

There are a number of important printer and material requirements that enable a safe and functional use of rapid tooling with the injection molding process.  Of course, the printer needs to reproduce geometry sufficiently to create the basic features and the build size is important to allow the most ease-of-use and flexibility.  There are numerous other important capabilities of MJP technology that are difficult to achieve and bring substantial value to injection molding application.

  1. The material’s ability to handle a high temperature
  2. The material’s ability to withstand high and repeated compressive forces
  3. The material’s sufficient tensile elongation for durability and fixture needs
  4. The technology’s smooth and blemish free molding surface quality
  5. The overall bulk geometric accuracy of the parts
  6. The ability to form extremely small features for things like text marking and/or to create high fidelity and complex surface texture

 

In terms of the high temperature requirements, the material itself is most important.  One of the great strengths of 3D Systems MultiJet Printing (MJP) technology is in its broad and individually optimized material set focused on different customer applications.  For example, all the Rigid and Engineering materials can be used for drilling/taping/machining/pressing and achieve the good surface finish, sharp corners, fine features, and create high fidelity true-to-CAD parts that is well known for MJP technology.   The Rigid white, clear, and gray materials (M2R-CL/WT/GRY) are optimized for general-purpose properties.  They are fairly rigid with good heat deflection temperatures, and yet maintain adequate elongation.  They are good for concept modeling and light end-use prototype part needs.  The enGineering materials M2G-CL (Armor) and M2G-DUR (ProFlex) were designed for the most aggressive engineering applications as they can be substantially twisted, flexed and deformed without cracking or breaking and are excellent for things like prototyping snap fits and for room temperature forming operations like sheet metal forming.   However, the Specialty material M2S-HT90 was designed and optimized for applications like plastic injection molding and is the material of choice.  The operational melting temperatures of most thermoplastics are in the range of 140 to 200C.

Typical thermoforming plastics with respective forming temperatures.png

Typical thermoforming plastics with respective forming temperatures

 

HT90 heat deflection temperature of 90C provides very good resistance to these hot plastic contact temperatures during the injection process.  This high heat deflection temperature also enables good cycle times before overheating and protects small features from distortion during the forming process.  In addition, HT90 also has an elongation before break of 6-7%.  Foremost, this elastic property allows the molds to be tightly bolted into place for the molding operation without cracking or breaking.

Brittle failure of mold insert .jpgBrittle failure of mold insert that occurred as the 3D plastic insert was clamped into the MUD

 

The elastic property of the material also allows sufficient flexibility that is required to remove final formed parts without damaging the tool and for small features to be formed at elevated temperatures without cracking or breaking.  While the HT90 material address the elevated temperature and elongation requirements, the MJP technology itself address the remaining surface, feature, and fidelity needs.  3D Systems MultiJet Printing (MJP) uses a high-resolution inkjet technology combined with a melt-away support material system.  The MJP printing process generates a part with over 1.1 billion drops per cubic inch of part volume allowing the creation of even the smallest details.  The support wax technology instantly solidifies in place during part formation and greatly contributes to sharp features and smooth surfaces.  Injection molded plastics are able to reproduce even the smallest of surface imperfections.  MJP technology creates surfaces that are smooth enough that it creates an aesthetically pleasing and functionally smooth surface desirable to designers and engineers.  Also, because of its inherent smooth surface quality and its high fidelity capability, MJP is able to create very small and highly detailed surface textures which are often a desirable attribute of an injection molding tool.

M2R-GRY part showing detailed surface textures.jpg Surface features on part created with MJP .jpg
Left:  M2R-GRY part showing detailed surface textures, Right:  Surface features on part created with MJP printed mold

 

Additionally, the wax support also allows for simple and hands-free post processing removal of support from even the most complex geometry and entrapped cavity without scarring of the part surfaces.  There is nothing that typically needs to be done to accurately match your part design dimensions with an MJP printer.  For most geometries, the part should come out true-to-CAD the very first time printed and every time printed.  If needed, the printer was designed with a scale factors for each material that is built into the 3D Sprint software.  Also, part accuracy is independent of orientation for even the smallest and sharpest of features so you don’t have to worry about numerous special part design rules, part orientation and complex placement issues or special printer setups.  The high-resolution capability of MJP also allows for things like the formation of printed threads for assembly and fixture needs. 

There are numerous different 3D rapid prototyping plastic technologies that one could consider for this application.  However, most typically fall short of that needed for plastic injection molding.  For example, only a few technologies are able to create the required smooth surfaces, feature fidelity for complex shapes, controlled surface texture, good mold release, and adequate sealing at parting line interfaces.  Also, many products and/or materials simply do not create the geometric accuracy needed for mold insert set-up and mounting or for final part geometry.   Many materials are too flexible to allow proper feature creation or are too rigid and will crack and fail during installation or use.  It takes a very special material design to achieve the temperature, strength, and durability that is required for both assembling the tooling with hardware and for functional operation in the injection molding process.  For applications that require low volume of parts with reduced cycle times and real thermoplastic properties, there is no better technology than 3D Systems MultiJet Printing with M2S-HT90.

Key injection molding properties for competitive technologies vs. 3D Systems MultiJet Printing.png

Key injection molding properties for competitive technologies vs. 3D Systems MultiJet Printing

 

 

Mold Design and Injection Molding with MJP Technology

Part and Mold Design

There are numerous basic design principles that are documented and well know which help direct tooling design to create highly functional and error free products using plastic injection molding.  However, as previously discussed, the process is complex and there are many possible defects for different geometries and plastic properties that ultimately make it impossible to predict design based on 3D CAD.  However, the use of MJP rapid tooling does not substantially change the complexity or basic design rules.  Therefore, any literature, experience, or testing a person may have related to their design most likely transfers.  For example, always keep the part walls a consistent thickness to reduce shrinkage.  Also, generous use of drafts, rounds and fillets allow for better material flow and reduce stress concentrations for easier part removal and longer tool life.  However, some key adjustments may need to be considered when using MJP plastic tooling.   Foremost, compared to a metal tool, the plastic tool has greatly reduced mechanical properties and reduced thermal conductivity and both of these differences should be considered in the part and mold design. 

Tool Wall Thickness:  Often thicker tool walls 0.04-0.08” (1-2mm) need to be used to survive the molding process.    

Part Size:  Larger parts are possible, but applications with smaller parts that requiring less total material are recommended.  Larger parts bring in more material and more material carries in more heat which is difficult to dissipate quickly and results in a larger mold temperature for a given part generation cycle.  Larger part designs might have reduced tool life and suffer from inconsistent quality. 

Tall and/or Thin Features:  Taller and thinner mold features or larger parts will reduce the mold life.  Deep draws are also not recommended.  Thin walls with deeper (longer) draws can benefit from increased draft angle.

Draw Depth and Draft Angles:  Draws should be less than about 1” (25.4mm).  Always consider using the highest draft angles possible when using 3D tools if possible.  Increased draft angles will allow the part to release from the mold more easily and decrease the likelihood for part or mold damage. 

Sprue Bushing:  Avoid using the 3D material itself as the sprue bushing.  Machine the sprue bushing into the MUD half or use a machined sprue bushing that is inserted into the plastic mold.  A nice press fit can be created with MJP by printing the hole in the 3D part the exact size as the diameter of the sprue bushing. 

Vents:  Design vents into your mold at strategic locations that allow air to escape which could otherwise lead to part defects or mold damage.  Vents can be created in plastic tools on-the-fly with the tip of a sharp object against the face of the tool starting from the point of defect location at the parting line and extending outward to the side of the tool.

Vent Geometry.jpg Vent Geometry Details.jpg
Vent Geometry Details

Cooling Channels:  Cooling channels are not recommended.  The plastic mold material is an effective thermal insulator and this reduces any possible advantage of the use of cooling channels.  They will also weaken the mold due to higher stresses caused by the channel voids. 

Printed Inserts vs. Total Tool:  The entire MUD can be printed, however, it often makes sense to use a 3D plastic insert within a cavity of the normal steel MUD. 

 

Core and cavity mold halves that are inserts into pockets.jpgCore and cavity mold halves that are inserts into pockets in a traditional MUD

 

This does not defeat the rapid turnaround time as the metal MUD only has to be created once.  Numerous custom plastic inserts can be printed using the single metal MUD design.  Use a mold size that is suitable for various size inserts and design your core and cavity inserts in CAD to fit those standard mold bases.  The use of inserts dramatically reduce cost and print time.  The insert can be held into place with a clamping pressure created with standard hardware.  M2S-HT90 has sufficient elongation such that it can be clamped and flattened.  However, when using an insert that is held into the MUD with metal hardware, use large and wide washers to clamp the insert as this distributes the load.  Clamp to a relatively thick section of material and avoid sharp corners.   Another design option is to hold the insert lightly in the MUD using a setscrew with an angle that press against a feature that is printed in the mold halves. 

Clamped insert.jpg Set screw holding 3D mold insert .jpg
Left:  Clamped insert, Right:  Set screw holding 3D mold insert

 

Design the thickness of the insert mold halves such that they are taller than the pockets in the MUD by about 0.2mm (0.008”).  Alternatively, it can be better to design the insert thickness equal to or slightly less than the pocket depth and use shims during the molding process under the insert to allow control over the interference pressure between the insert mold halves.  This interference is needed to create pressure at the parting line and seal the mold halves.   IMPORTANT:  If the metal carries the clamping load, the plastic inserts will not seal properly.

Shut-offs:  Shut-offs in the design should interfere to create sufficient sealing pressure and avoid flash.  Typically, interference of about 0.1mm allows sealing between mold halves. 

Long features:  Consider using metal rods that are pressed into 3D printed holes in the plastic insert for very long core features that may lack strength.  Reaming of holes is typically not needed for MJP technology.  Again, set the 3D printed hole size equal to the pin size to create a robust fit for the pin installation.

Runners:  Use a generous and direct material feed system if possible.  Round runner geometries are recommended over square runners.  Square runners can lead to shorter tool life.

Runner design can improve tool life.jpg

Runner design can improve tool life

 

Gates:  Avoid tapered gates that lead to localized areas of heat and pressure.  Enlarge the gates compared to that used in a metal design to reduce stress and achieve longer tool life.  Take advantage of the 3D printing capability to crate generous rounds on the gates.  The gate should be located so that the melt entering the cavity will not impinge on small/thin features in the mold. 

Injector Pins:  Use normal ejector system if possible.  Add printed injector holes as needed.  Set the printed sizes to exactly what one would need if it were machine drilled.  Reaming is not typically needed with MJP technology.  Possibly, test your fit clearances with a separate part as part of the mold design and stick to that tolerance. 

Printing and Post Processing the Mold

Print the molds in UHD mode.  This allows for the best part geometry, feature fidelity and surface quality.  The best practice is to orient the mold so that the mold cavities face up.  This will provide the best flatness and surface quality.

Non-Optimal Orientation.jpg

Non-Optimal Orientation

This orientation places the functional molding features against the base support.

This will likely result in the downward face warping up slightly.

Optimal Orientation.jpg

Optimal Orientation

This orientation places the functional molding surface up which will eliminate warp.

 

If any warp occurs on large flat surface, the mold can be positioned on its side for printing.  This will reduce the surface quality and create a rougher surface, but the mating surfaces will be flatter.  The increased roughness can be mitigated somewhat with the use of a dry or water-based bead-blasting (Refer to example #3 at the end of this paper).  Typically, the mold will be printed face up and no additional processing of the part will be needed (no sanding or polishing, etc…).  Post process the part normally in the EasyClean system or a convection oven using mineral oil as the fine wax removal solvent.   If needed, the sides or bottom surface can be post machined for the purpose of proper fit.

Materials and Injection Molding Equipment

Thermoplastic Recommendations:  Most all injectable thermoplastics have the potential to be used with MJP printed molds.  Some materials will be more difficult compared to others depending on the geometry and the user must iteratively test their material and geometry to determine its capability, cycle time and tool life.  Materials with barrel temperatures up to 350-400C have been successfully used and are dependent on the geometry.  Thermoplastic with relatively low melting points and good flow work best, like polypropylene and thermoplastic elastomers, and will create the longest tool life.  Higher temperature plastics, clear plastics, and fillers can be more difficult. 

Release Agents:  A release agent typically helps prevent adhesion and acts as a cooling agent between cycles.  While often not desirable in an injection molding shop floor, use of silicone release spray is recommended if possible.   Depending on the material and release agent type, the use of a mold release agent might help or hinder the part formation and release.  It is not uncommon for release agent to be required between each shot or every few shots.

Use of Shims:  It is often advantageous to have a set of shims readily available to make minor interference corrections between the 3D inserts as needed if there if flash is present at the parting line during molding.

Tool Startup Process:  When starting a molding run begin with lower shot size and pressure that creates a short shot.  Slowly increase shot size, temperature, and pressures, etc. until a good shot is achieved.  Try to achieve a “steady state” and then lock down process and continue to shoot.  Reuse these set points for similar materials in the future. 

Air Cooling:  Use air cooling on the outside of the mold for cooling between shots.  The mold can reach 100-200C immediately after the plastic is injected.  However, it will cool quickly with air cooling.  Cool the mold down to about 40-50C before shooting the next part.  This can be done manually, or with fixture that blows air on the mold faces.  Have a thermal measurement device like a thermal camera or single point non-contact measurement available to track the process.  Pay special attention to locations of the mold that heat the most and contain fragile features, such as locations where the runners connect to the gate or a large material area of the part.  

Hold Time:  Often you will need the longest practical holding time.  The part and tool need sufficient time to cool to avoid defects or part deformation as the tool opens and/or when the part is removed.

Wait time:  Once the mold is open, apply more cooling directly to the part and the mold.  Measure the mold surface temperature during each shot and adjust the wait time to prevent heat build-up.  Larger parts or more complex part geometry can require longer holding and separation times.

 

Conclusion

In conclusion, metal tooling is expensive and has a long lead time.  RTV Silicone molds are also relatively slow and do not create a part with the desired thermoplastic material properties.  3D printed metal is expensive and has complex post process requirements.  3D Rapid prototyping using plastic technology is very quick (1 day) and relatively inexpensive and is able to use standard thermoplastic materials.  It is a great option for specific applications for prototyping needs and even for some low volume production workflows.  3D Systems MJP technology using M2S-HT90 material fulfills these needs more than any other technology available.

A few injection molding examples will be presented next to facilitate best practices and key information related to using MultiJejt Printing for injection molding.  However, ultimately each design is unique and one must iteratively test to determine the appropriate geometry, process parameters and load/pressure requirements for specific needs.

 

Example 1:  Basic Insert Mold Design and Operation, Cycle Time, and Total Number of Cycles

An injection molding insert tool was designed and 3D printed with MJP 2500 acrylate in four MJP materials (M2S-HT90, M2R-CL, M2G-CL Armor, M2G-DUR ProFlex). The 3600 material “X” was also tested as a benchmark. The goal was to test the materials for basic design rules, special capability, and best practices including tool life. Two materials were tested, polypropylene (PP) and thermoplastic polyurethane. The design intent was to create a two-cavity mold that could shoot both sides at once, or one side individually by blocking off the material flow. A MUD blank was chosen that fit the BOY 60D injection-molding machine that was used for testing (08/09 ALU210 from www.masterunitdie.com). The inserts measured 3.5” by 6.5” (165.1mm) and were 0.75” (19mm) thick. Pockets were milled into the A and B MUD plate halves. The pocket was designed to be about 0.001” to 0.002” (0.25 to 0.5 mm) larger than the printed insert geometry. Many inserts were produced and tested and it was not uncommon to use a light touch of sandpaper on the edges of the mold to facilitate smooth insertion, but the inserts typically were used as-printed. The corners of the inserts serve no functional purpose and so generous rounds were be added to facilitate easier assembly.

Injection mold core.jpg Core and cavity mold halves that are inserts into pockets.jpg

Injection mold core and cavity with machined pockets to hold printed inserts.

 

 

  The pockets were design to be identical in thickness as the inserts.  Shims were used during injection to create a small interference for parting line sealing purposes.  A ball-type sprue bushing was machined into the MUD.   The sprue was also machined into the MUD within a round metal cylinder that protruded through the mold insert up to the parting line of the inserts.   Two short runners were made in this metal feature to allow material flow into either one or both halves of the mold (driven by the 3D insert geometry that either is closed off, or has an open runner).   To install the inserts into the MUD, thru holes were used on all four corners of the insert and large washers were used between the screw and the mold to reduce pressure/stress.   To allow for maximum material thickness at these locations, four low profile socket head cap screws were used on both sides (which allowed the 3D printed mold to be thicker). 

A and B halves of the MUD showing insert cutout.jpg

A and B halves of the MUD showing insert cutout, sprue feature, and attachment hardware

 

A and B mold halves installed .jpg BOY injection molding machine.jpg
A and B mold halves installed in the BOY injection molding machine

 

A few different geometries were tested. A simple rectangular part with a flat surface was chosen as it allowed measurement of surface roughness and gloss as well as some simple dimensional measurements. The same part, but with some surface texture added was also tested to explore the feature size capability of the technology and the molding process.

Simple flat and textures surfaces.jpgSimple flat and textures surfaces to allow roughness, gloss and fine surface feature testing

 

To allow ease of part removal, a draft angle of 5 degrees was used.  Two runner geometries were tested: trapezoidal and round. 

Runner design can improve tool life_2.jpg

Runner design can improve tool life

 

Round runners are the most commonly used geometry in injection molding because they uniformly distribute the high heats and pressures encountered during shooting, naturally create draft and are easy to machine. Testing confirmed that the round runner is the best option for 3D printed molds. The trapezoidal runner produced greater stress concentrations at the gate that caused premature mold failure. A sprue puller was built into the plastic insert. The sprue puller was located on the half of the mold opposite side of injection. The sprue puller was a simple round feature with a slight undercut. Material shot into the mold fills all the cavities (including the undercut of the sprue puller. Upon opening, the undercut of the sprue puller captures the cured part and pulls the part away from the sprue and first half of the mold.

Side view of sprue puller geometry.jpg

3D view of runner and sprue puller geometry.jpg

Top:  Side view of sprue puller geometry

Bottom:  3D view of runner and sprue puller geometry

 

Vents were initially not used in the printed mold halves.  This lead to cavitation defects on the surface of the part and/or entrained air defects.

incomplete shot (short shot) during startup of the machine.jpg

Parts 1 and 5 respectively. (1) incomplete shot (short shot) during startup of the machine, (2) successfully printed part, (3 & 4) cavitation central to the causing part rupture (5) deformation due to trapped air in mold

 

Later designs included a simple 3D printed air vent at the back side of the part. This eliminated cavitation and air entrainment defects and led to the creation of numerous molds each producing numerous successful parts.

Air vent geometry on 3D printed mold.jpg

Air vent geometry on 3D printed mold

 

Examples of successful parts printed in polyurethane and thermoplastic elastomer.jpg

Examples of successful parts printed in polyurethane and thermoplastic elastomer

 

Each time a part is injection molded, it needs to cool and solidify prior to the release of the part.  Natural convection cooling worked as long as the wait time between shots was sufficiently long.  However, the disadvantage of this method is that it prolongs the time required between each shot, both in the time prior to halve separation (curing time) and in the time before part removal (wait time).  Some thermoplastic materials will degrade within minutes at temperature within the barrel of the injection molding machine.  For this reason and for general productivity improvement we also tested and typically used convective air cooling with a simple air spray using house air available on the shop floor.  If the part is removed too soon from the mold it can be deformed substantially as it is still very soft and is still adhered to the mold.  

Plastic deformation.jpgPlastic deformation due to no cooling treatment and aggressive part removal before full solidification.

 

Appropriate cooling was also important to the life of the tool.  The first failure was often observed at the location that was the hottest:  for example, in the runner at the entrance to the gate.  Flashing occurs when plastic seeps between the edges of the mold. This phenomenon is common and does not result in an unusable part if it can be removed with post processes.  Excessive flashing may cause a part to be unusable if the flashing is too thick or difficult to remove.  Flashing was controlled by tuning the clamping pressure and by controlling the interference created between the two 3D inserts with the use of shims located between the inserts and the MUD.

Example of excessive flashing .jpgExample of excessive flashing caused from insufficient clamping pressure or insufficient mold insert interference size

 

3D_ Printed_ inserts _installed into the MUD.jpg3D Printed inserts installed into the MUD showing machined sprue feature feed system, 3D printed runners and gates, cold well geometry and metal shims between the insert and the MUD

 

For this test a nozzle temperature of 370F (188C) was used with a clamp pressure of 1700 psi and an injection pressure of 200-300 psi.  A 40 second cure time was used before the mold was open.   After opening the mold it was cooled with air pressure to about 50-70C before reinjection.  The total process would take a number of minutes between cycles.  All the MJP materials tested were able to make at least a few successful shots.  M2S-HT90 material was able to make 40 shots, where the testing was suspended.  Failure was difficult to quantify, as it would be highly dependent on the customer requirements.  The main difference we saw between the first and 40th shot was an increase in flash near the gate which could be removed.  There was no visible part degradation, and the M2S-HT90 mold was visually inspected for warping and no obvious changes were detected.

left to right showed an increase in flashing.jpgM2S-HT90 Parts 1, 25, and 40, left to right showed an increase in flashing, but otherwise good quality parts

 

The molds with surface texture features worked well and could make numerous parts that had good aesthetic properties.

CAD geometry.jpg smooth_ surface _and fine surface texture.jpg
Left:  CAD geometry, Right:  smooth surface and fine surface texture capability

 

Example 2:  Printed Part Orientation:  Surface Quality and Dimensional Accuracy

            A test was performed on the surface roughness, gloss and part dimensional accuracy for upward, downward, and side printed molds.  A simple rectangular part was used to allow measurement of gloss and surface roughness.  Also, flatness of the mold was measured with a Coordinate Measurement Machine.   The part measured 1.5mm thick and was 31.75 mm long by 19.05 mm wide by 4.57 mm deep that formed a shallow rectangular prism.  A 7 deg draft was used. The design incorporates the use of vents to prevent air bubbles and cavitation.  Two vents were used, one at the end of the part and the other as a relief for the cold well feature.  The runner has a circular shape with a diameter of 4.6 mm and two 90 degree turns smoothed out by using an interior radius of 5.3 mm on both turns.  The runner fed into a fan gate that fed the part.

Injection molding insert design .jpg

Injection molding insert design with runners, gate and vents

 

Part print orientatin for MJP printer.jpg

Part print orientatin for MJP printer

 

A steel MUD was used to contain the 3D plastic inserts.  An ejector plate was used with a single ejector located at the sprue. 

MUD core and cavity with inserts installed.jpg

MUD core and cavity with inserts installed

 

A BOY 22D injection molding machine was used with polyproplyene material at the following approximate operating setpoints. 

BOY 22D injection molding machine.jpg

 

Two MJP materials were tested M2G-CL and M2S-HT90. A competitive SLA high temperature resin was also tested for comparison. However, the competitive SLA mold could not be printed with sufficient accuracy to allow proper insertion in the metal MUD inserts and was therefore abandoned.

Competitive SLA 3D printed inserts.jpg

Competitive SLA 3D printed inserts were catastrophically out of dimension for testing and was abandoned.

 

Approximately 0.002” clearance was allocated between the MJP part and the mold and the parts printed fit the mold well in all materials and trials.  Metal shims were used to control the interference between the core and cavity.  The dimensional accuracy and flatness was measured with a TESA Micro-Hite 3D Coordinate Measuring Machine.  The roughness was measured with an SRT-6210a Surface Roughness tester.  The gloss was measured with a BYK Gardner micro-tri-gloss meter with and angle of incidence of 60 deg.  The upward and downward facing surfaces had the lowest roughness (<1um) and highest gloss (40-50 GU).  The mold printed on the side had an increased roughness (~20um) and reduced gloss (<10 GU).

Roughness and gloss of molds.jpg

Roughness and gloss of molds by material and orientation

 

The dimensional accuracy and flatness was measured on 7 surfaces labeled A-G. All of the functional surfaces were extremely accurate with very good flatness. 

Flatness for each face.jpg

Flatness for each face

 

Extremely _flat functional mold halves.jpg

Extremely flat functional mold halves in M2S-HT90

 

Printing the insert on its side improved flatness, but increased roughness.  A water honing process was used on the parts and can be used to equalize the surface roughness and gloss, making upward or downward surfaces more rough and sidewalls less rough.  It is possible and could be advantageous for some applications to use a dry glass blasting or a water honing process to equalize the mold surfaces to create a more uniform part surface from the molds. 

Example 3:  CAD Workflow and High Volume Printing of 3D Systems Logo Part

            Of course, cavity and core mold design and creation is not typically needed to be done by an engineer as they will typically only create the desired part and will work with vendors and toolmakers who will design the actual mold based on the engineer’s part.  However, for the case of rapid prototyping with injection molding, it will likely be up to the designer or engineer to design the molds to be printed.  Due to the high usage of plastic injection molding, there are custom workflows in many engineering CAD packages that semi-automate the creation of the core and cavity based on a part design.  The workflows typically start with the part and have processes to define parting lines, parting surfaces, shut-off surfaces, etc. and will automatically make the core and cavity geometry based on the part geometry. 

Plastic injection mold design automation.jpg

Plastic injection mold design automation is often built into CAD packages

 

Visual tools such as draft analysis is also built into the CAD package.  These workflows are documented well in the CAD package documentation and there are numerous videos and instructions on-line that a user can reference.  Another method to make molds is to use a simple “cavity” or “Boolean subtraction” operation in CAD to create the mold halves.  Start by making the two mold halves based on the desired part dimensions.  It is best to leave about  ¼” (6-7mm) spacing between any edge of the part and the edge of the insert to allow for proper sealing of the mold halves.  In CAD, assemble the core and cavity together with the mold faces touching each other and aligned on the other two surfaces.  Design the part with integrated gate, runners and cold slug well.

3D Systems logo design .jpg 3D Systems logo design with integrated gate.jpg 3D Systems logo design with integrated gate, runner and cold slug.jpg
3D Systems logo design with integrated gate, runner and cold slug well

 

Assemble the part between the core and cavity halves in CAD. Locate the parting line of the part exactly on the face of the mold halves. Of course, the end of the runner will need to align with the side of the mold halves that will interface with the runner system on the MUD. Typically the engineer will also have the MUD design in CAD and so the entire assembly can be made with fewer chances of error.

3D insert core and cavity assembly .jpg

3D insert core and cavity assembly with desired part

(with runners and gate) assembly

 

Finally, perform a “cavity” or “Boolean subtraction” operation between each mold halve and the part to create the inverse of the geometry in the mold halves.

Part geometry created in core and cavity.jpg

Part geometry created in core and cavity with “Boolean subtraction” or “cavity” operation

 

After a successful Boolean/cavity operation the inverse of the part geometry will be cut into the core and cavity.

Cavity tool in SolidWorks .jpgCavity tool in SolidWorks creates the negative of the part/runner/gate geometry

 

Add the vents manually based on best guess, experience, and/or feedback from a known failure. Vents can be very small and erring on the side of too many typically does not create an issue as any flash is easily removed. 

Final core and cavity 3D plastic inserts with example of final part.jpg

Final core and cavity 3D plastic inserts with example of final part before removal of the sprue, runner, cold slug well and gate

 

The molds were printed using 3D Sprint software. Four sets could be printed at a time. They were printed with the mold face up and in HT90 material.

4x mold sets .jpg

4x mold sets each were printed per job

 

In all, about 25 mold sets were made and were delivered along with the MUD to a local injection molding shop. Each mold was able to make about 10-15 successful parts. A total of about 400 parts were made.

Total of 400 parts were manufactured from multiple 3D printed inserts.jpg

Total of 400 parts were manufactured from multiple 3D printed inserts.

Each insert could produce approximately 10-15 pieces each