Skip to content
Request a Quote

Engineering Plastics Injection Molding: How to Choose the Right Grade

Fixed bracket plastic molding

Engineering plastics are where industrial injection molding becomes a materials problem rather than a tooling problem. A consumer housing can be specified as “ABS” and the supplier will pick a grade that looks right on the bench; an industrial bracket has to survive a named chemical at a named temperature under a named load for a stated life, and the answer to that is a specific resin grade – not a polymer family. Get it right and the part performs like the engineering drawing says. Get it wrong and the failure does not appear at the bench, it appears eight months into a pump skid installed at a chemical plant, and it appears as a warranty claim rather than as a tooling problem. This guide covers how engineering plastics are actually selected and molded: which family fits which duty, the failure modes that decide the grade, what a usable specification contains, and how molding engineering plastics differs in the tool from molding commodity resins.

Injection molded load-bearing industrial bracket in glass-filled engineering plastic
An industrial part is decided by the resin grade and the duty, not by the mold – the tool is the easy part to get right.

What engineering plastics are, and how the grade is chosen

Engineering plastic selection is a specification exercise, not a shopping exercise. The polymer family is almost never the decision – the decision is the grade, and the grade is chosen by identifying which property is actually binding in service: stiffness under sustained load, continuous service temperature, chemical resistance against a named fluid, wear against a named counterpart, or dimensional stability across a temperature band. In practice that narrows the field quickly. POM for precision gearing and sliding parts where friction and dimensional stability matter and temperature stays moderate. PA66-GF for load-bearing brackets where creep is the concern. UHMW-PE for wear strips and rollers. PEEK only where the duty genuinely justifies a cost that is eight to fifteen times that of a polyamide.

Two things follow from that, and they are the reason this topic is worth a page of its own. First, the supplier cannot choose the grade for you. If the specification on the drawing says “PA66”, the supplier can legally quote any of dozens of PA66 grades – unfilled, 30% glass filled, impact modified, hydrolysis resistant – and they will choose the one that is in stock, which is usually the one that is cheapest. Second, engineering plastics are harder to mold than commodity resins: narrower processing windows, higher shrinkage variation, hot tooling requirements for crystalline grades, and wear that attacks the tool steel if you put POM or UHMW-PE through it for years. A supplier who has only ever run PP and ABS will produce acceptable brackets and then produce one batch of gears that does not hold size.

Our industrial injection molding page covers the part families we build and the volumes we build them at, and the materials page covers the resin families in more depth. This guide is about the decision between them, and about what has to be written down for a supplier to be able to make it correctly.

The engineering plastic families and what each is for

The families below are ordered roughly by cost, not by performance – because performance and cost are not correlated in the way people expect. A 30% glass-filled PA66 bracket costs more per kilo than unfilled POM but may be the only thing that survives the duty, and a PEEK part may be replaceable with a glass-filled PA66 if the real constraint turned out to be stiffness rather than temperature. Selecting by cost first and by performance second is how programmes end up paying for PEEK and still failing.

Engineering plastic families and where each one actually fits
Family Continuous service temp Key property Typical industrial use The trap to avoid
POM (acetal) 80-100 C High stiffness, low friction, excellent machinability Precision gears, cams, bushings, sliders, valve parts Creeps under sustained load above ~60 C; absorbs little moisture but still moves
PA66 / PA6 80-120 C Tough, fatigue resistant, good chemical resistance Structural brackets, hydraulic fittings, connectors, clips Moisture absorption changes dimensions and stiffness – condition the resin
PA66-GF / PA6-GF 100-140 C Stiffness far above unfilled, low creep Load-bearing brackets, housings, gear boxes, automotive carriers Fibre orientation is anisotropic; flow direction must be designed, not assumed
UHMW-PE 60-80 C Outstanding wear and impact resistance, chemically inert Wear strips, rollers, guides, chute liners, bushings Very high shrinkage and high creep – wall thickness and clamping need designing
PBT 90-110 C Good stiffness, easy to mold, good dimensional stability Sensor bodies, electrical housings, fluid fittings Moisture sensitivity is modest but hydrolysis matters in hot water
PPS 150-200 C High temperature, excellent chemical and creep resistance Automotive under-bonnet, chemical processing, high-temp electrical Expensive and demanding to mold; requires specific drying and temperature control
PEEK 200-250 C Highest mechanical and chemical performance of the family Semiconductor tooling, medical sterilisation, aerospace, high-temp bearings Cost is 8-15x PA and cycle times are long – it needs a commercial case
PC / PC-ABS 100-130 C Transparency, toughness, dimensional stability Machine guards, sight glasses, instrument housings, fluid sight components Hydrolysis and stress cracking with PC; PC-ABS is the safer general choice

Two properties deserve more attention than they usually get, because they are the ones that silently change a part’s behaviour over its service life rather than at the bench.

Creep. Most engineering thermoplastics are semi- crystalline, and semi-crystalline polymers creep under sustained load that is below their short-term strength. The modulus of unfilled POM falls from roughly 3 GPa at room temperature to about 1 GPa at 70 C, and under a continuous load the fall is steeper still. For a bushing or a gear tooth carrying a constant force, the relevant number is not the tensile strength on the datasheet but the creep modulus at the service temperature under the actual load. Filled grades are the usual answer: adding 30% glass roughly triples the creep modulus relative to the unfilled grade, which is why a filled gear holds size where an unfilled one drifts.

Moisture conditioning. Polyamides are hygroscopic, and PA66 absorbs roughly 3 to 4% water at 23 C / 50% RH, rising toward 8% in humid conditions. Absorbed moisture plasticises the resin: the part gets softer, and it swells. A PA66 part that is dimensionally correct when it is molded and measured dry may be measurably larger after a month in a humid environment. This is not a defect and it is not fixable at the bench – it is a consequence of the material being selected. It is, however, entirely predictable, which means the design can compensate for it or the specification can require a lower-absorption grade.

Our injection molding materials guide covers the commodity-to-engineering transition in more detail, and the mold design checklist covers the geometric decisions that interact with material choice.

The failure modes that decide the grade

In industrial work the resin is usually decided by a failure that has already happened somewhere – a returned part, a field complaint, a premature wear-out on a line. The table below maps the failures that recur most often to their causes, because naming the failure correctly is what makes the grade selection defensible rather than a matter of preference.

The failure modes that decide the resin, and what causes each one
Failure observed in service Usual root cause The correct response Where it shows up first
Part dims out of tolerance over months Creep under sustained load, or moisture conditioning in polyamide Move to a filled grade, or remove the sustained load from the design Bearing bores, seal grooves, threaded bosses
Browned or stained surface on a cosmetic part Degraded polymer carried forward from a hot idle barrel or long hot-runner dwell Control dwell time, purge correctly, log barrel temperature against throughput First parts after a shutdown or a short run
White or grey deposits on the part Additive migration from the resin grade, or contamination from another material Identify the deposit, change the purge discipline, review the grade High-temperature grades with high additive load
Brittle fracture after a few months Environmental stress cracking from a chemical that was never specified Write the chemical exposure into the requirement as a named list Seals, gaskets and fasteners in chemical service
Warpage that changes with the season Moisture conditioning in PA, with unbalanced packing on top Condition the resin, then rebalance the runner and cooling Large flat brackets and housings
Wear at a contact surface Wrong wear family chosen for the counterpart material and the lubricant Specify the wear pair, not just the plastic Bushings, rollers, cam followers, slides

The pattern worth noticing is that most of these failures were never molding problems at all. They were specification problems that surfaced in service. A gear that wears out in eight months is a material-pair problem. A seal that cracks in a chemical bath is a missing requirement. A bracket that is out of size after a year is a creep calculation that was never done. Once the team is looking at this table rather than at the tool, most of these programmes are solved by writing one line into the specification that was missing, and that line usually costs less than the tooling change it replaces.

What a usable specification actually contains

The table below is the practical core of this guide. It is a list of the things that must be explicit on the drawing or in the accompanying requirement document. Every one of them has been the missing item on a programme where the tool was built correctly and the part still did not work.

What a usable engineering plastic specification contains
Item Why it must be explicit Acceptable form Common failure
Resin grade Performance is a property of the grade, not the family Manufacturer grade number plus manufacturer name “Use PA66” – dozens of grades behave differently
Filler content and type Changes stiffness, creep, shrinkage and finish e.g. 30% glass fibre, mineral filled, carbon filled Filled substituted for unfilled on price
Continuous service temperature Sets the temperature ceiling for design life A stated value with the load condition noted Quoting a short-term deflection temperature instead
Chemical exposure list The only way to prevent stress cracking Named fluids, concentrations and temperatures “Chemical resistant” with nothing behind it
Critical dimensions and method Determines whether the part is inspectable Dimension, tolerance, measurement method, sampling plan Tolerances on features that need post-processing
Wear partner and lubricant Wear behaviour depends on the pair Counterpart material, surface finish, lubricant type and load Specifying the plastic alone
Annual volume and scrap allowance Determines cavity count and grade economics Stated volume and the acceptable scrap rate Volume unknown at quote, so the tool is over- or under-built
Documentation required This is usually the real cost driver CoC, material lot traceability, dimensional report, capability data Discovered at the audit rather than the quote

Read that table as a cost list as well as a quality list. Item eight – the documentation required – is what distinguishes an industrial part from a consumer part in price, and it is also the item most often left implicit. A supplier who knows from the outset that they will have to produce a certificate of conformance with resin lot traceability, a dimensional report with a measurement method, and capability data on the critical features, can plan the tool and the trial for it. A supplier who discovers the requirement after the first production order has a much worse conversation, and the cost usually lands on the buyer.

Our guide to choosing an injection mold manufacturer covers the general supplier evaluation, and the quality page covers the systems behind the documentation.

When plastic replaces metal, and when it does not

Replacing metal with plastic is one of the few genuine cost improvements in industrial manufacturing, and it is also one of the most frequently mis-specified. The comparison has to include the machining cost that disappears, not just compare material prices. A stainless bracket that needs fifteen operations, a deburring stage and a passivation process will usually cost more in total than a molded glass-filled PA66 equivalent – and the molded one arrives finished and repeatable, which is worth more than the unit price in high-mix programmes.

The cases where metal still wins are worth stating plainly, because they are real and they recur:

  • Sustained service well above 150 C. The polymer families top out where engineering plastics give way to thermosets, composites and metals. A PPS part at 180 C continuous is near the ceiling of what a molded thermoplastic will do.
  • Fatigue rather than creep. Thermoplastics perform badly in cyclic load at elevated temperature. A metal part that survives millions of cycles is not being asked for something polymers do well.
  • Mandated material. Pressure equipment, aerospace qualification and some food-contact regulations specify the material, and the discussion is over before it starts.
  • Impact without deformation. Where a part must not bend or crack on impact, a filled thermoplastic may simply not be adequate, regardless of how strong the datasheet number looks.

The honest engineering caveat is that stiffness is not interchangeable between the two. GF30 PA66 has a flexural modulus around 8 to 11 GPa against aluminium’s 70 GPa and mild steel’s 200 GPa. At a given stiffness, the plastic needs several times the wall thickness or the section has to be redesigned as a ribbed structure rather than a solid plate – and because the modulus falls with temperature and with sustained load, the calculation has to include both. Plastic does not replace metal at equal dimensions; it replaces metal at equal function, and the translation between them is a design task.

The die casting page covers the alternative route when the part is genuinely metal, and the die casting versus injection molding guide covers how to choose between the two processes.

Molding engineering plastics: where the tool changes

Molding an engineering grade is a different job from molding PP, and the differences show up in the tool, in the process window and in the quality report. The table below is the short version.

Molding differences between engineering plastics and commodity resins
Factor Commodity resin (PP, ABS) Engineering plastic (POM, PA, PPS) Why it matters to the tool
Melt temperature 190-230 C 250-330 C depending on family Higher melt means more heat into the tool and a hotter, longer cycle
Shrinkage 1.5-2.5% 1.5-2.5% unfilled, but 0.3-0.8% with 30% glass fibre Filler shrinks an order of magnitude less – the cavity size cannot be copied
Mold temperature 20-60 C 80-160 C for crystalline grades Hot tooling is essential for crystallinity and for warpage control
Fill behaviour Forgiving Narrow window, sensitive to velocity and shear Gate and velocity profile have to be simulated, not guessed
Surface finish Easy to achieve Demanding; marks, weld lines and jetting show Cosmetic parts need more gate work and slower fill
Moisture Low for PP/ABS PA6 and PA66 absorb 3-9% and must be dried Wet resin means splay, bubbles and dimensional drift
Wear on the tool Low POM and UHMW-PE are abrasive to steel Tool steel and surface finish need to be specified for these grades
Cycle time Faster Longer, because of the higher melt and hot mold Cycle time changes the cost per part at the same volume

The row that surprises people most is the wear one. POM and UHMW-PE are abrasive to tool steel – a POM gear mold running at volume will eventually wear the cavity surface, and the wear shows up as a dimensional drift on the gear teeth long before anything visibly fails. This is a reason to specify tool steel with higher hardness and a more corrosion- resistant finish for those grades, and it is also a reason to keep the process parameters stable: a grade that runs well outside its window degrades and the degradation products then act as abrasives themselves.

The shrinkage row deserves emphasis for anyone copying a cavity from a previous part. Unfilled PA66 shrinks around 1.5 to 2% depending on grade and direction; PA66 with 30% glass fibre shrinks around 0.3 to 0.8%, and it shrinks differently along the flow and across the flow. A cavity dimensioned for unfilled resin will produce a part that is badly oversized in one direction and marginally undersized in the other – and the error is often mistaken for a molding fault rather than a material fact.

Our injection molding process guide covers the five phases of the cycle, and the mold flow analysis guide covers how the simulation is used to settle the filling and balance questions that engineering grades raise.

Warpage and dimensional drift: two different problems

Buyers and suppliers use the word warpage for at least two unrelated things, and separating them is the fastest way to a correct answer. The first is warpage at the bench: the part comes out of the tool visibly distorted. The second is dimensional drift in service: the part was correctly molded and has changed shape over months or years. They have different causes and they need different responses.

Bench warpage is a design and process problem. It comes from uneven cooling, from fibre orientation being inconsistent shot to shot, from unbalanced packing, and from the fact that a filled grade’s fibres align along the flow path and shrink almost nothing in that direction while shrinking across it. The responses are tool-side: symmetric cooling about the part’s centre of mass, a runner designed for balanced filling, and a velocity profile that keeps the fibre orientation consistent. This is where mold flow analysis earns its cost, because the imbalance is measurable in simulation and invisible in a drawing.

Service drift is a material fact. Moisture conditioning in polyamides, creep under sustained load and stress relaxation all change dimensions over time, and no process adjustment stops them. The responses are on the design side: select a lower-absorption or higher-modulus grade, remove the sustained load from the interface, or design the joint so that it has enough clearance to allow for the movement that will occur. The most expensive version of this mistake is treating a creep problem as a process problem and re-tuning the tool for two years.

The injection molding defects guide covers the full diagnostic order, and the mold maintenance page covers the tool-side measures that reduce bench variation.

Deposits, scale and the cleaning sequence that works

Deposits on engineering plastic parts are a common complaint with a small number of distinct causes, and they need diagnosing before cleaning. Cleaning the wrong cause wastes time and, in the case of a degraded polymer, makes it worse by carrying degraded material further into the machine.

Three causes cover most cases. The first is degraded polymer: a barrel held at temperature with no throughput, or a long dwell in a hot runner held at temperature, breaks the polymer down and the residue then carries forward to the part. The tell is that the deposit appears on the part after a shutdown or a short run. The second is additive migration: some engineering grades contain stabilisers or processing aids that deposit under a specific thermal history, and the deposit behaviour can change completely with a different grade of the same family. The third is contamination from another resin on a shared machine – the residue is that other material, not the one in the datasheet.

The sequence that works: identify the deposit by colour and by whether it is on the part or only in the machine; check the resin supplier’s recommended purge material and maximum dwell time for that grade; verify that the barrel is not being held hot with no production running; and only then reach for mechanical cleaning of the screw and barrel. For deposits on the tool itself, the causes are different – thermal degradation of the resin trapped in the tool, or the breakdown of the mold release – and the injection molding defects guide covers that diagnosis separately.

Why industrial plastic parts cost more than consumer parts

When an industrial part comes in above the consumer equivalent, it is almost never the mold. It is the specification, and the verification behind it. The industrial part carries obligations that the consumer part does not – a named grade rather than a generic resin, a continuous service temperature, a chemical exposure list, a wear partner, dimensional reporting with a measurement method, and traceability on the resin lot. Each of those is a documented commitment that a supplier has to be able to discharge, and it is where the cost difference lives.

Geometry contributes too. Industrial parts frequently need tight tolerances on functional interfaces – a bearing bore, a seal groove, a threaded boss, a pilot diameter – and holding those on a molded part requires a different tool and a different measurement regime from a housing. A 20-micron bearing bore is not a drawing tolerance; it is a statement about the process capability that has to be demonstrated, not asserted.

It is also worth keeping the cost structure in proportion. For a typical industrial programme running long, the tooling amortises and the recurring cost is material, cycle time and scrap. That means the two levers that matter most for unit cost are not the tool price but the cycle time and the scrap rate – and cycle time for engineering grades is longer, because the melt temperature is higher and the mold runs hot. The injection mold cost breakdown covers how the tooling number is built, and the lead time guide covers the schedule implication of the same decisions.

Screening a supplier on engineering grades

Most of the difference between a supplier who runs engineering grades well and one who is running an industrial programme as a general molding job shows up in the first technical conversation, before any price is discussed. That makes it possible to screen without references or a site visit, which is what the table below is for.

Questions that separate a supplier who can run engineering grades from one who cannot
Question What a good answer sounds like What a weak answer sounds like
Which grade did you quote, and why that one? Names a grade and ties it to a specific property in the duty “A standard engineering plastic” or “whatever is suitable”
Will the tool run hot-molded, and what mold temperature? States a temperature for the quoted grade and explains its effect “We mold it normally”
How do you control shrinkage with fibre-filled grades? Describes either a validated cavity dimension or a post-mold plan “We adjust the tool after the first shots”
What is the drying specification for this resin? Gives a dew point, a temperature and a dwell time “We dry it”
How do you handle dimensional reports on critical features? Describes a measurement method and a reporting format “We check the parts”
Which grades have you run in this tool? Names specific families with volume or lifetime context “We mold engineering plastics”
Can you show per-cavity shot weight from the trial? Produces per-cavity data on request Offers only a sample part
What happens if the grade changes later? Explains the qualification path and the risk “We can substitute anything”

The first row is the one that matters most. A supplier who asks for the service temperature, the chemical exposure and the load in the first conversation is a supplier who will recommend a grade for a reason. A supplier who quotes from the drawing alone is quoting a part, and the grade will be chosen later by whoever is easiest to source. Our guide to choosing an injection mold manufacturer covers the general evaluation, and the engineering-specific addition is that the material question belongs in the first conversation rather than the last.

The last row is the one most buyers never ask. If the grade changes later – because the original is unavailable, or because a cost reduction programme is initiated – there is a qualification path to follow and a risk to understand. A supplier who treats a grade change as a like-for-like substitution is telling you that the part was never really specified.

Our industrial injection molding page describes the part families we build, the turnkey solutions page covers running the programme through to assembled and packed sub-assemblies, and the mold design checklist covers the design decisions that are cheap to fix on a drawing and expensive to fix on a tool.

Three industrial programs and what actually decided them

Industrial plastic selection goes wrong in a small number of ways, and the pattern is consistent across the three cases below: the molding was rarely the problem, and what decided the outcome was whether the duty was written down before the material was chosen. These are drawn from our own toolroom.

Case 1: The pump housing where the grade was the whole business

A chemical dosing pump housing for a North American water treatment customer had been running in unfilled PA66 for two years. The tool was sound, the cycle time was competitive, and the parts were dimensionally consistent. What was not sound was the material: the housing sat in a metering chamber where the wetted parts saw a 60 C caustic solution continuously, and the customer had started seeing stress cracks around the mounting bosses in the field. The replacement was not a better tool – it was a better grade, in PA66 with a higher glass loading and a hydrolysis-resistant variant of the base resin, at a material cost increase that was material but trivial against the field failure rate. The general lesson is that when an industrial part fails chemically, the answer is almost never in the mold. Our materials guide covers the resin families involved, and the mold flow analysis guide covers why a chemical-resistant filled grade needs a different filling profile from the unfilled one it replaces.

Case 2: The gearbox where the gear was fine and the shaft was not

A small gear reducer for an agricultural equipment customer came to us after a supplier had delivered what was, technically, a conforming part. The POM gears were within tolerance, the material certificate matched, and the inspection report was complete. The assembly still failed in service, and the failure was at the gear shaft interface: the shaft was steel, the gear was unfilled POM, the lubricant was a heavy gear oil, and the two surfaces were running with a combination of sliding and rolling contact under a continuous load. The wear pattern on the first samples showed polishing rather than galling, which is the signature of a correct material pair running under too much load for too long. The fix was not a better POM – it was a filled POM grade with a deliberately different wear behaviour against steel, plus a shaft diameter adjustment that reduced the contact pressure. The general lesson is that wear behaviour is a property of the pair and the lubricant, not of the plastic alone, and a wear part cannot be specified without all three. The design checklist covers the geometric decisions that interact with it.

Case 3: The bracket where the drawing was right and the resin was wrong

A load-bearing mounting bracket for a piece of instrumentation was specified as “PA66” with no filler content, because that is how it appeared on the customer’s original drawing. The duty was a cantilever carrying a static load at 70 C in an enclosure that was not temperature controlled. The unfilled PA66 part was dimensionally correct at delivery and had crept enough within four months that the mount was visibly out of alignment, with the fastener holes no longer aligned to their datum. The fix was a 30% glass-filled PA66, which roughly tripled the creep modulus, together with a change to the fixing detail that removed the continuous load from the boss entirely. The molding of the bracket was never in question – the material and the joint were. The general lesson is that a creep calculation, or at least a creep-aware design decision, belongs in the specification for any sustained load at temperature, and “PA66” without filler content does not convey it. The mold flow analysis page covers the tool-side half of the answer, and the design checklist covers the other half.

Engineering plastics terms quick reference

Term What it means Why it matters to the part
Engineering plastic A polymer selected for a defined mechanical, thermal or chemical performance rather than general utility The family is rarely the decision – the grade within it usually is
Grade A specific formulation from a specific manufacturer, with fixed filler content and additives Two grades of the same family can differ in stiffness, shrinkage and price by a factor of three
Continuous service temperature The temperature a material can hold under sustained load over its design life Not the same as the short-term deflection temperature on a datasheet
Creep Time-dependent deformation under a sustained load below the material’s short-term strength The reason a correct part goes out of size in service
Creep modulus The effective stiffness of a material under a given load, temperature and duration The number to design a sustained-load part against, not tensile strength
Moisture absorption The mass of water a polymer takes up from the environment PA6 and PA66 absorb several percent, which swells and softens the part
Filler content The proportion and type of filler – glass, mineral or carbon fibre – in a grade Raises stiffness and creep resistance while cutting shrinkage sharply
Fibre orientation The alignment of reinforcing fibres along the flow path Makes the part anisotropic – the same dimension differs with and across flow
Hot mold A tool run above room temperature, typically 80-160 C for crystalline grades Needed for crystallinity, surface quality and warpage control
Mold scale A deposit on the tool cavity, usually degraded resin trapped in the tool Prints onto the part and is a different problem from machine deposits
Purge The material run through a machine to clear a previous resin The correct purge material and dwell time are grade-specific
Environmental stress cracking Failure under combined tensile stress and chemical exposure The most common long-term failure in industrial service, and entirely preventable in the specification
Wear pair The combination of the plastic, the counterpart material and the lubricant Wear behaviour cannot be specified for the plastic alone
Process capability The statistical ability of the process to hold a tolerance, expressed as Cpk The evidence that a tolerance is achievable, rather than an assertion that it is

Engineering plastics questions, answered

What are engineering plastics and how do they differ from commodity plastics?
Engineering plastics are polymers selected for a specific mechanical, thermal or chemical performance rather than for price and general utility. The practical dividing line is not marketing language but a usable property range: continuous service temperature, tensile and flexural modulus, chemical resistance against a named fluid, and whether the grade is filled or unfilled. Commodity ABS and PP are adequate for housings and packaging but creep under sustained load well before they reach their stated temperature limit, and they cannot hold a dimensional tolerance across a wide temperature band. Engineering grades cover the range from POM (high stiffness and low friction at low cost) through PA66 and PA66-GF (structural, moisture sensitive), to PPS, PEEK and liquid crystal polymers for genuinely demanding service. In an industrial part, the decision usually comes down to which property is actually binding in service, and that is worth writing down before anyone quotes a tool.
Which engineering plastic should I use for precision gears and cams?
For most precision gearing and cam duty in the automotive, appliance and industrial equipment segments, the answer is POM (acetal) unfilled or lightly filled, and the reason is a combination of stiffness, dimensional stability and wear behaviour rather than any single headline number. POM holds a tight tolerance across a temperature band, machines to a good finish, and has a low coefficient of friction against steel – which is what makes it the default for sliding and rolling contact. Where the gear carries a continuous high load or runs above roughly 80 C continuously, POM is the wrong answer, because it creeps: it is a semi-crystalline polymer whose stiffness falls steadily with sustained load and temperature, and a gear that is dimensionally correct on day one can be out of specification after a year in service. In that regime the usual alternatives are a glass- or carbon-filled PA66, or a PPS if temperature and chemical exposure are both severe. For a gear that has to run against a shaft with no lubricant, or in a food or medical environment, the choice shifts again towards the filled POM grades or PTFE-filled variants. The point to take away is that the answer depends on the load, the temperature and the lubricant, not on the plastic.
Is it cheaper to mold engineering plastic parts in metal?
Usually yes, and considerably so, but the comparison has to be made correctly or it produces the wrong decision. The cost that disappears is not just the material cost of the metal – it is the machining cost, because a molded plastic part arrives finished. A machined aluminium or stainless bracket that requires fifteen separate operations and a deburring stage will usually cost more in total than an injection-molded equivalent in glass-filled PA66, and it will take longer. The case where metal still wins is specific and worth naming: sustained temperatures well above 150 C, loads that create fatigue rather than creep, environments where the part must not deform at all under impact, and applications with a mandated material – aerospace and pressure equipment being the obvious ones. It is also worth remembering that glass-filled nylon is not a universal replacement for aluminium. The modulus of GF30 PA66 is around 8 to 11 GPa against aluminium’s 70 GPa, so for a load-bearing bracket at a given stiffness you need considerably more material, and the creep rate of the plastic must be included in the deflection calculation. The engineering answer is that plastics replace metal where the constraint is corrosion, weight, cost or wear rather than stiffness or temperature.
How do you prevent warpage and dimensional drift in engineering plastic parts?
Start by identifying which of the two mechanisms is actually at work, because they have different fixes. Processing-induced warpage – the part pulling out of the tool because of uneven cooling, fibre orientation or packing – is a design and process problem and is usually fixable in the tool. Service-induced drift over months or years is a material problem: moisture absorption in polyamides, creep under sustained load, or stress relaxation, and no amount of process adjustment will stop it. For the first, the measures that matter are balanced filling through a properly designed runner, cooling that is symmetric about the part’s centre of mass, and controlling shear rate so the fibre orientation is consistent from shot to shot. For the second, the answers are material selection and specification: a lower-absorption grade, a filled grade where stiffness is required, or designing the interface so that it is not load-bearing in the first place. The common expensive mistake is treating a moisture problem as a process problem and re-tuning the process for two years.
Why do engineering plastic parts form mold deposits and how do you clean them?
Deposits on engineering plastic parts are usually one of three things, and they need to be diagnosed rather than cleaned indiscriminately because each responds to a different remedy. First, degraded polymer: a barrel held at temperature with no throughput, or a long dwell in a hot runner, breaks the polymer down into a brown residue that forms on the screw and then carries forward to the part. Second, additives migrating out of the resin: some engineering grades contain processing aids or stabilisers that can deposit under a specific thermal history, and switching grade can change the deposit behaviour entirely. Third, contamination from another material – a shared machine that last ran a different resin will deposit that resin’s residue. The practical sequence is: identify the deposit by its colour and whether it is on the part or only in the machine, check the resin’s recommended purge material and dwell time, verify that the barrel is not being held hot with no production, and only then reach for mechanical cleaning of the screw and barrel. For mold deposits on the tool itself, which is a different problem with a different cause, see our injection molding defects guide.
Can engineering plastics be used in food or medical contact applications?
Yes, but only with grade-level documentation, and the distinction between the general engineering grade and the regulated grade is the whole point. A general POM or PA66 grade has no food contact or medical declaration attached to it, and using one in a food-contact or medical application is not merely improvised – it is usually non-compliant, because the compliance is attached to a specific resin grade rather than to the polymer family. The correct approach is to specify the application and require the supplier to quote a grade that carries the appropriate declaration: FDA or EU food contact compliance for food-contact parts, ISO 13485 quality system plus biocompatibility data for medical components, and full traceability on the resin lot. There are genuine technical constraints that follow from these applications too: detergents and hot water attack some grades, radiation sterilisation degrades others, and additives such as plasticisers or glass fibre are not appropriate for food-contact or implantable applications. Our medical device injection molding guide covers the regulatory side of this in more detail.
What makes industrial plastic parts expensive compared to consumer parts?
Usually the specification rather than the plastic. An industrial part carries requirements that a consumer part does not: a named resin grade rather than a generic resin family, a defined continuous service temperature, a chemical exposure list, a wear or fatigue life requirement, and often a dimensional report on critical features with process capability attached. Each of those is a documented obligation that a supplier has to be able to discharge, and the cost of an industrial mold is largely the cost of the documentation and the verification behind the part. The second reason is geometric: industrial parts frequently need tight tolerances on functional interfaces – a shaft bore, a seal groove, a bearing seat, a threaded boss – and holding a bearing bore to 20 microns in a molded part requires a different tool from a housing. It also pays to note that the expensive part is rarely the mold. For a typical industrial programme the tooling is amortised over a long run, and the recurring cost is the material, the cycle time and the scrap rate. Our injection mold cost breakdown covers how the tooling number is actually built.
How do you select a supplier for engineering plastic molding?
On whether they will discuss the material before the price. A supplier who quotes a tool without asking about service temperature, chemical exposure, load and annual volume is quoting a part, not a programme, and the material will be chosen later by whoever is easiest to source – which is how a glass-filled PA66 bracket becomes an unfilled one and nobody notices until the part is in service. The second criterion is whether they can demonstrate the relevant experience in engineering grades specifically: engineering plastics are harder to mold than commodity resins because of narrow processing windows, high shrinkage and sensitive surface finish, and a shop that has only run PP and ABS will need to learn the tolerance and finish discipline that engineering grades demand. The third is whether they will run mold flow analysis on the specific grade before cutting steel, because engineering grades fill differently from commodity ones. The fourth is whether they can give you per-cavity shot weight data from the trial, which is how you find out whether the tool is balanced before the first production order is placed. Our guide to choosing an injection mold manufacturer covers the general evaluation.

Send us the duty, not just the drawing

Tell us the service temperature, the chemicals the part will touch, the load it carries and the annual volume. We will return a material recommendation with the trade-offs in plain language, a DFM review, and an itemised quotation within 24 hours.

Request a Quote

No obligation. NDA available.

  • 23+ Years of experience
  • 2,000+ Molds per year
  • 24h Quote turnaround

Recent Blog Posts