is part of the Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Critical Considerations for Selecting a Prototyping Method

Image: Michael Paloian 3D-printed medical model
Material properties, quantities, part size, and cost are among the key considerations when choosing a prototyping technology, from 3D printing to CNC machining.

In the previous article on this topic, I stated the importance of prototyping during product design and development to verify your design. I reviewed the many plastic prototyping options available to us, including the three most popular rapid prototyping technologies. Part 1 discussed the first half of the 12 considerations that should be evaluated before selecting the best process. This second installment will review the other six considerations for the same prototyping technologies to help you decide which method is best suited for your needs.

Below is a list of currently available plastic prototyping methods:

  • Rapid prototyping — 3D part creation directly from CAD files. The top three technologies are FDM (fused deposition modeling), SLA (stereolithography), and SLS (selective laser sintering).
  • Handmade/fabricated
  • CNC machining
  • Casting — polyurethane, epoxy, or silicone
  • Injection molding — temporary tooling
  • Vacuum forming
  • Composites

The remaining six considerations for selecting the best prototyping method are provided below:

  • Material properties
  • Quantity of prototypes required
  • Size of parts
  • Cost
  • Degree of clarity/color
  • Parameters to be tested or verified

I will review each of these considerations in this article.

Material properties

All the processes previously listed can be specified for prototyping parts to verify overall appearance, interference checks, ergonomics, and overall concept. However, part-design verification based on specific material properties is much more difficult to complete. Prototype injection molding is the only prototyping option that will produce parts that are almost identical to production parts. The reason I’ve described it as almost identical is because tool design, quality, construction, and materials, as well as processing conditions, could introduce some differences between a prototype injection molded part and a production part.

The two second-place prototyping options are CNC machining and FDM. CNC machined parts cut from a single slab of a specific plastic material will provide you with parts that closely represent the production pieces. It should be noted that slab stock is either extruded or injection molded in thick cross sections. Exact matches to production materials are rarely achieved. Instead, general matches to a specific material are highly probable. Examples include materials like GP polystyrene, polyethylene, polycarbonate, acetal, ABS, and so forth. It’s important to realize that machined slab stock is limited to operations that can be performed on a CNC machine and the availability of stock material sizes. Also, machined parts should be annealed to relieve machined-in stresses. One should also know that the molecular structure of extruded or injection molded slab stock is different from that of an injection molded part. This is especially true for glass-reinforced materials.

Unlike CNC machining, FDM prototypes can be fabricated in virtually any thermoplastic material. Mini extruders allow you to extrude most thermoplastics into filaments for your FDM printer. The printed part is geometrically identical to your 3D CAD file except for tolerances. The drawback of FDM is the anisotropic behavior of the printed part. Material properties in the X-Y plane can be significantly different from those in the Z axis. The degree of difference is dependent on material and, most importantly, the printer. Some printers claim the Z axis material strength is 90% that of the X-Y plane. Most printers, however, are nowhere near this level of correlation. One of the most significant property differences is in the material’s flexural strength. A simple work-around is to print features like snap locks as separate parts, with the build grain running in the direction of flex. The snap can later be bonded to the main part with solvents or adhesives.

Handmade prototypes, which comprise several smaller bits bonded together into a single piece, are limited to commercially available stock materials like CNC machined pieces. Handmade prototypes can more closely resemble an injection molded part because of the process; however, completed prototypes are much more delicate than a CNC machined piece since the main part is typically fabricated by bonding several smaller pieces together with solvents or adhesives. Handmade prototypes, therefore, cannot be drop tested or impact tested with any reliability.

Cast polyurethanes, SLS, and SLA are all limited to a small family of resins: Cast polyurethanes are limited to polyurethane, SLS is limited to nylon, and SLA is limited to epoxy or acrylic resins.

If a product is to be subjected to harsh chemicals, creep tests, or impact testing, you are advised to test the actual material sample under your anticipated use conditions. Testing a prototype may be too expensive or impractical. 

Quantity of prototypes required

You are sometimes required to create more than one prototype. Multiple prototypes are required when you’re performing destructive testing or distributing several prototypes to users for feedback or to multiple sites for evaluation and testing. Selecting the ideal prototyping option is not as simple you think. You must consider cost, investment, lead time, quality, and what you are evaluating. The table below compares the various prototyping options based on these considerations. It has been prepared as a guide. There are overlaps between the options, and your experiences may vary to some degree depending on the vendors with whom you are working.

Process Investment Lead Time Quality Similarity to Production
Rapid Prototyping — FDM 1 5 4 4
Rapid Prototyping — SLA 1 5 5 3 – 4
Rapid Prototyping — SLS 1 4 4 3 – 4
Handmade/fabricated 1 3 3 2 – 4
CNC Machining 1 3 3 3 – 4
Cast 3 4 3 2
Injection Molding 5 5 5 5
Vacuum Forming 3 3 3 3
Composites 3 3 3 3

Rating: 1 = low, slow; 5 = high, fast

Parameters to be tested or verified

Product designers and engineers must maintain a balance of creativity, innovation, and confidence that their proposed designs will function as intended when the project is completed. Prototyping and testing throughout the development process enable developers to efficiently verify their ideas well before the product design cycle is completed. This methodology reduces the risk of investing hundreds or thousands of man-hours on a design that ultimately may require a complete redesign. Designers must always analyze the trade-offs between cost and time of prototyping versus the validity of the results derived from testing.

A few real-world test and evaluation examples are listed below. Hopefully these examples may help you understand the process that is best suited to a specific situation.

Michael Paloianprototyping aesthetic models

All of the prototyping options are appropriate for producing aesthetic models, except for prototype injection molding.

Aesthetic presentation

Aesthetic models can be very simple or quite complicated. In all cases, product function is not important unless the function is somehow directly related to appearance. For example, doors, buttons, and handles — which are functional — affect appearance. This requires a level of detail that clearly communicates design intent. Interior spaces may be revealed, requiring them to comply with the overall appearance standards. All of the prototyping options, therefore, are appropriate except for prototype injection molding. A large, complex electronic device with hundreds of internal components can be clearly represented by a solid or hollow shell with fine details for all the external parts visible to the viewer. The cost of these prototypes is invested in the artistry and craftsmanship of the model maker — mixing paints, masking, painting, sanding, polishing, and painstakingly resolving fine details are extensive and costly.

Ergonomic evaluation

Prototypes are essential when you are trying to optimize human factors considerations. Parameters such as weight, balance, finger clearance, comfort, handle designs, and so forth all require extensive research and evaluation well before the design is finalized. Ergonomic prototypes are not required to be 100% functional. Their level of functionality is determined by its effect on ergonomic considerations. For example, if the prototype is being evaluated based on its acceptable level of comfort when held in one’s hand, the handle contours must be precisely defined. If the prototype is being evaluated for trigger, button, or latch tactile feedback, the force profile for these features should be accurately represented. Other features such as locations for handles, clearances for legs, visibility of displays, and so forth must all be accurately represented in the prototype. In most cases, a simplified non-functional model can be used for these studies. The models may not require finishing or painting. A simple monochromatic raw model may be enough for most ergonomic studies; however, finished prototypes may be required if color, surface texture, or material hardness are critical.

Depending on the type of ergonomic study, handmade prototypes are highly effective, low cost, and quick. They can be fabricated in a variety of materials ranging from clay, plaster, or wood to plastic and metal. If more accurate models are required, all the other processes are suitable except prototype injection molding. The determining factors narrow down to cost, availability, and final requirements.

Michael Paloiansmall, intricate prototyped parts

Small quantities of microscopic parts are best suited for rapid prototyping.

Size of parts

Part size is a significant factor when deciding on the best prototyping option. Parts can range in size from microscopic to piano-like. Small quantities of microscopic parts are ideally suited for rapid prototyping. There are a few companies offering specialized 3D printers, which are specifically designed to print parts as small as a few microns. The material choices for this process, of course, are limited. Although most injection molded plastic parts fit within the standard 12-in. cube volume, larger parts can be printed as a single piece on larger machines. FDM machines are currently surpassing 1000 x 1000 x 1000 mm in volume. These very large machines are commercially available, but most 3D print fabricators don’t have them and print large parts in smaller sections that are later bonded together into one larger piece. Since all the processes can yield large — greater than 2 x 2 ft — parts, one must consider part geometry, complexity, and tolerances when deciding on a process. Prototype injection molding is certainly not a cost-effective option. However, cast polyurethanes, vacuum forming, and composites are ideally suited for medium to large parts. As a matter of fact, vacuum formed and composite prototypes can be identical to production parts if the materials and processing parameters are similar.


Cost is always a significant consideration when selecting a prototyping option. Some processes require a substantial upfront investment while others require no investment at all. Some processes are labor intensive, incurring high per-unit costs, while others are more efficient and much more cost effective. The table below compares processes based on investment and unit cost.

Process Investment Cost/Unit
Rapid Prototyping — FDM 1 4
Rapid Prototyping — SLA 1 5
Rapid Prototyping — SLS 1 5
Handmade/fabricated 1 1 – 4
CNC Machining 2 3 – 4
Cast 3 2
Injection Molding 5 1
Vacuum Forming 3 3
Composites 4 3

Rating: 1 = low; 5 = high

Per-unit cost for all the prototyping processes listed above, except for injection molding, can vary significantly based on exterior finish. Functional prototypes requiring little to no finishing work will be much less expensive than those that are finished with a Class 1 paint finish, graphics, and other details requiring many man-hours of hand work. The cost of an FDM part can double or triple, depending upon the quality of finish required. Typically, FDM parts require more hand work and finishing time than SLA parts, which have much higher resolution. CNC machined parts are most cost effective for functional prototypes that must perform when subjected to stresses or other environmental conditions.

Clarity and color

Prototypes often require clear parts. All the prototyping processes that have been discussed in this article can produce clear parts except for SLS, which is primarily limited to nylon and composites that are reinforced with glass or carbon fibers. However, if we add the requirement of optical clarity, the options narrow.

Michael Paloiancast polyurethane parts

Polyurethane materials can be cast in colors or formed into optically clear parts. They may require minimal buffing to achieve a clear finish.

FDM can produce clear parts with PET and acrylic materials, but the parts will not be optically clear. Light will pass through, but visibility will be obscured because of the layering. Hand-fabricated and CNC-machined parts made from acrylic can be buffed to optical clarity. Polyurethanes can be cast in colors or clear materials, forming optically clear parts. Typically, cast parts require some minimal buffing to attain a clear finish. One should also be aware that cast parts can sometimes entrap air bubbles if not processed properly. Vacuum-formed parts can be molded in clear polycarbonate or acrylic sheet. Of course, injection molding can produce optically clear parts in virtually any thermoplastic if the molds are highly polished. The best rapid prototyping methods for producing clear parts is SLA. Prototyping lenses or optical-quality parts is much more complicated than prototyping a clear window. A clear window does not require the same level of perfection required of an optical-quality window that must transmit light with virtually no distortion. Optical lenses require not only distortion-free geometry, but the refractive index of the material must match that of the production parts.

Custom color matching is often required for production-like prototypes. FDM, casting, CNC machining, vacuum forming, and injection molding can produce parts with molded-in color. However, it is not practical to prototype any part with a molded-in custom color. Ordering custom pigmented resins for FDM filament, injection-molded pellets, or vacuum-formed sheet is highly impractical, costly, and time consuming. SLA and SLS parts are typically restricted to natural material colors — white, translucent beige, or water clear for SLA, and white for SLS. Custom color-matched polyurethane will never be identical to that of an injected-molded thermoplastic. Therefore, virtually all prototypes are painted with precisely matched pigments to comply with stringent color standards. When paints are applied to prototypes, you must be careful to define close-fitting features that must be masked from paint. Paint thicknesses will sometimes build up to more than five-thousandths of an inch, causing interferences. 

Properties to be tested or verified

Prototypes are extremely helpful to test and verify your design well before it has been finished or released for production tooling. A few of the more common tests and evaluations that are performed on prototypes are listed below:

  • Thermal testing
  • Friction and wear resistance
  • Drop impact test
  • Ease of assembly
  • Safety
  • Balance and feel

The first three are exclusively dependent on the materials and method of fabrication. As previously mentioned, material-based testing and evaluation should ideally be determined using the same material. However, in most situations this is not practical or cost effective, and you must find alternative methods of performing these tests. A viable alternative would be to isolate the property you are interested in testing and replacing the material with a suitable substitute. An example might be testing the feel of a snap fit. If your production part was specified in polycarbonate, for example, and you require an SLA prototype for fine details, you could substitute the polycarbonate with an epoxy resin with the same modulus as polycarbonate.

Michael Paloianprototype material properties

While ideal, material-based testing and evaluation using the same material that will be used in production is not practical or cost effective in most cases.

Evaluating ease of assembly, especially for compact electronic products with numerous connectors, small screws, and snap fits, is a critical step in the design process. The ideal process for this type of evaluation is SLA, which can replicate very fine details accurately and cost effectively. Larger assembly studies can be evaluated using FDM or even hand-fabricated models, depending on the stage of development. 

Evaluating product safety is essential for most products, which are becoming more complex and sophisticated through each generation. All of the processes that have been discussed in this article can be specified for testing product safety. If safe use is directly linked to material properties, the options are limited, as previously discussed. However, if you are interested in evaluating the ideal location of an emergency button or shield, sometimes a handmade foam model may be enough. The ideal process is highly dependent on the level of detail required to represent the feature being evaluated.

It is essential for handheld products, including power tools, sporting goods, medical devices, and so forth, to be well balanced and comfortable during use. The feel of a product can have a big impact on its acceptance or rejection in the marketplace. It is therefore imperative for you as a designer or engineer to validate these parameters early in the design process. These parameters are a function of mass distribution within an assembly as well as materials specified for the external structure. Non-slip over-molded elastomers often provide a level of extra tactile comfort that must be felt by an individual for proper evaluation. A simple handmade wood or cardboard model may be adequate for evaluating balance during the early stages of a design. However, a more refined, printed FDM model may be more suitable during the middle of a project, followed by a very refined, finished SLA model toward the end. The major factors determining the optimum process are cost, personnel, skill, and equipment.

Hopefully these two articles have provided you with enough information to help you decide which prototyping options will best satisfy your requirements. The boundaries between each of them can get a bit fuzzy, but there are definite pros and cons to all. If you have any comments or questions, please feel free to contact me at and we can talk. Best of luck on your next project.

About the author

Michael Paloian is President of Integrated Design Systems Inc. (IDS), located in Oyster Bay, New York. He has an undergraduate degree in plastics engineering from UMass Lowell and a master's of industrial design from Rhode Island School of Design. Paloian has an in-depth knowledge of designing parts in numerous processes and materials, including plastics, metals, and composites. Paloian holds more than 40 patents and was past chair of SPE RMD and PD3. He frequently speaks at SPE, SPI, ARM, MD&M, and IDSA conferences. He has also written hundreds of design-related articles for many publications. He can be reached by phone at 516/482-2181 or via e-mail,

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.