Different industries have different tolerance expectations. Here are typical ranges you will see in purchase specifications:
- Consumer electronics housings: ±0.1 to ±0.2 mm (MT2). Surface appearance matters, but internal snap-fits have some give.
- Automotive connectors: ±0.05 mm (MT1). Electrical contact geometry must be precise to ensure signal integrity and vibration resistance.
- Medical device components: ±0.02 to ±0.05 mm (MT1, sometimes tighter). Biocompatibility and sterilization repeatability require tight dimensional control.
- Industrial enclosures: ±0.2 to ±0.5 mm (MT3). Large parts with non-critical mating surfaces.
- Packaging caps and closures: ±0.1 mm (MT2). Thread engagement must be consistent for torque and sealing performance.
If you are sourcing from China and your spec was written for a European supplier, the tolerance class may not transfer directly. Always clarify the ISO 20457 class explicitly — do not assume that “±0.1 mm” means the same thing across all suppliers.
Avoiding Common Tolerance Mistakes
Even experienced engineers make tolerance specification errors. Here are the most frequent ones and how to avoid them:
- Over-specifying tolerances. Tightening every dimension to ±0.05 mm is the fastest way to blow your mold budget. Ask which dimensions are truly functional before tightening.
- Omitting reference conditions. A dimension measured hot after ejection is not comparable to one measured after 24-hour conditioning. Always state the measurement condition.
- Specifying tolerances on non-machinable features. Wire-cut EDM can hold ±0.01 mm; sandblasted texture cannot. Do not specify a tolerance on a surface feature that cannot be measured consistently.
- Ignoring stack-up analysis. A single dimension may look fine, but when five parts assemble, the tolerance stack may exceed your assembly clearance. Run a tolerance stack analysis on critical assemblies.
- Assuming all cavities are equal. In a multi-cavity mold, cavity 1 may produce a slightly different dimension than cavity 8 due to flow imbalance. Specify that tolerances apply across all cavities, not just the first one.
When to Request a Tolerance Audit
If you are new to injection molding or working with a new supplier, request a tolerance audit before committing to the tooling budget. A tolerance audit is a review of your drawings where the mold maker identifies every dimension that may be difficult to hold and suggests alternatives. The audit typically takes 1–2 days and can save thousands of dollars in mold rework or production scrap.
At RCH Plastic, we include a tolerance capability matrix in every DFM report. Each critical dimension is tagged with the expected process tolerance, the required tooling grade, and any special process controls needed. This transparency lets you make informed decisions before the mold is cut — not after the first shot is rejected.
Bottom Line
Injection molding tolerances are achievable — but only when they are specified clearly, matched to the right material and process, and backed by a mold maker who documents the tolerance capability before the steel is cut. The most expensive tolerance is the one you discover is impossible to hold after the mold is already built. Share your drawing with your supplier during the DFM phase, and let them flag every tolerance that needs special attention. That conversation will save you time, money, and disputes on every program you run.
For a detailed DFM review of your tolerance spec, or a tolerance capability quote for your part, contact RCH Plastic with your drawing and material selection. We will return a dimension-by-dimension tolerance analysis within 48 hours.
A well-written tolerance spec prevents disputes and reduces the cost of a mold program. Follow these rules:
- Specify only critical dimensions. Non-critical dimensions should use ISO 20457 MT2 or MT3 by default. Every tightened tolerance adds cost.
- Use ISO 20457 classes. Write “MT2” next to the dimension, not just ±0.1 mm. This removes ambiguity.
- Define the reference condition. State whether dimensions are measured at 23°C / 50% RH, after 24-hour conditioning, or as-ejected.
- Separate functional from cosmetic dimensions. A mating surface may need ±0.05 mm; a visible A-surface may only need ±0.2 mm for appearance.
- Include datum references. GD&T (Geometric Dimensioning and Tolerancing) with datums is the professional standard for critical assemblies.
When in doubt, ask your mold maker for a tolerance capability matrix before you finalize the drawing. A good supplier will show you what each dimension can realistically achieve with the selected material and process — and flag any that require special tooling or process controls.
Tolerance vs. Cost: The Real Trade-off
Tightening a tolerance from ±0.2 mm to ±0.05 mm can increase mold cost by 30–50 percent and part cost by 15–25 percent. The reasons are straightforward: harder tool steel, more machining time, tighter fit tolerances, longer trial cycles, and stricter process control. For most commercial parts, MT2 (±0.1–0.2 mm) is the sweet spot — tight enough for assembly, affordable for volume production.
If your design requires MT1 tolerances on multiple features, discuss this with your mold maker early. There may be a cheaper alternative: a different material with lower shrinkage, a revised part geometry that reduces the number of tight features, or a post-molding machining step on a single critical dimension instead of holding it in the mold.
When to Use Coordinate Measuring Machine (CMM) vs. Go/No-Go
CMM inspection is the standard for production validation — it measures X, Y, Z coordinates with micron-level accuracy and generates a full dimensional report. It is the right choice for first-article inspection, PPAP submissions, and ongoing process validation.
Go/No-Go gauges are faster and cheaper for high-volume production checking. They are best suited for single features (hole diameter, pin length, thickness) rather than complex geometries. For a part with 20 dimensions, CMM is more efficient; for a part with one critical hole diameter, a custom go/no-go gauge is faster and sufficient.
Real-World Tolerance Examples by Industry
Different industries have different tolerance expectations. Here are typical ranges you will see in purchase specifications:
- Consumer electronics housings: ±0.1 to ±0.2 mm (MT2). Surface appearance matters, but internal snap-fits have some give.
- Automotive connectors: ±0.05 mm (MT1). Electrical contact geometry must be precise to ensure signal integrity and vibration resistance.
- Medical device components: ±0.02 to ±0.05 mm (MT1, sometimes tighter). Biocompatibility and sterilization repeatability require tight dimensional control.
- Industrial enclosures: ±0.2 to ±0.5 mm (MT3). Large parts with non-critical mating surfaces.
- Packaging caps and closures: ±0.1 mm (MT2). Thread engagement must be consistent for torque and sealing performance.
If you are sourcing from China and your spec was written for a European supplier, the tolerance class may not transfer directly. Always clarify the ISO 20457 class explicitly — do not assume that “±0.1 mm” means the same thing across all suppliers.
Avoiding Common Tolerance Mistakes
Even experienced engineers make tolerance specification errors. Here are the most frequent ones and how to avoid them:
- Over-specifying tolerances. Tightening every dimension to ±0.05 mm is the fastest way to blow your mold budget. Ask which dimensions are truly functional before tightening.
- Omitting reference conditions. A dimension measured hot after ejection is not comparable to one measured after 24-hour conditioning. Always state the measurement condition.
- Specifying tolerances on non-machinable features. Wire-cut EDM can hold ±0.01 mm; sandblasted texture cannot. Do not specify a tolerance on a surface feature that cannot be measured consistently.
- Ignoring stack-up analysis. A single dimension may look fine, but when five parts assemble, the tolerance stack may exceed your assembly clearance. Run a tolerance stack analysis on critical assemblies.
- Assuming all cavities are equal. In a multi-cavity mold, cavity 1 may produce a slightly different dimension than cavity 8 due to flow imbalance. Specify that tolerances apply across all cavities, not just the first one.
When to Request a Tolerance Audit
If you are new to injection molding or working with a new supplier, request a tolerance audit before committing to the tooling budget. A tolerance audit is a review of your drawings where the mold maker identifies every dimension that may be difficult to hold and suggests alternatives. The audit typically takes 1–2 days and can save thousands of dollars in mold rework or production scrap.
At RCH Plastic, we include a tolerance capability matrix in every DFM report. Each critical dimension is tagged with the expected process tolerance, the required tooling grade, and any special process controls needed. This transparency lets you make informed decisions before the mold is cut — not after the first shot is rejected.
Bottom Line
Injection molding tolerances are achievable — but only when they are specified clearly, matched to the right material and process, and backed by a mold maker who documents the tolerance capability before the steel is cut. The most expensive tolerance is the one you discover is impossible to hold after the mold is already built. Share your drawing with your supplier during the DFM phase, and let them flag every tolerance that needs special attention. That conversation will save you time, money, and disputes on every program you run.
For a detailed DFM review of your tolerance spec, or a tolerance capability quote for your part, contact RCH Plastic with your drawing and material selection. We will return a dimension-by-dimension tolerance analysis within 48 hours.
Every thermoplastic shrinks as it cools from melt to solid. The shrinkage rate is material-specific and direction-dependent — flow direction shrinks less than the transverse direction. Common shrinkage ranges:
- PP: 1.5–2.5%
- PE: 1.5–4.0%
- ABS: 0.4–0.9%
- PA66: 0.7–1.5% (moisture-dependent)
- POM: 1.5–2.5%
- PC: 0.5–0.7%
- PTFE: 3.0–4.0%
A 100 mm feature in PP with 2% shrinkage will target 98 mm in the mold cavity. If the shrinkage varies between lots — and it will, unless the material is pre-dried and processed consistently — the finished part will drift. This is why material certification and batch control are part of tolerance management.
2. Part Geometry
Thin walls cool faster and shrink less, but they also require higher injection pressure, which can increase flash risk. Long, slender features flex during ejection and may spring back unpredictably. Undercuts and ribs create local cooling variation that warps the part. Simple, symmetric geometries are easier to hold to tight tolerances than complex, asymmetrical ones.
3. Mold Design and Construction
Tight tolerances require tight tooling. This means hardened steel cores and cavities (HRC 50+), precision ground parting surfaces, matched ejector pins, and adequate venting to prevent gas traps that cause local dimensional variation. A mold designed to MT1 tolerances will cost 20–40 percent more than one designed to MT3 because of the extra machining time, tighter fit tolerances, and longer trial cycles.
4. Machine and Process Control
The injection molding machine must be stable: consistent melt temperature, repeatable screw position, and adequate clamp force. Process parameters — injection speed, packing pressure, cooling time, and mold temperature — must be documented and controlled. Statistical process control (SPC) with CpK ≥ 1.33 is the minimum expectation for production parts. Processes running without documented parameters will produce variable dimensions.
5. Post-Molding Conditioning
Parts continue to shrink and distort after ejection, especially hygroscopic materials like PA66 and PC. Post-molding conditioning — drying, annealing, or humidity conditioning — stabilizes dimensions. A PA66 part measured immediately after ejection may be 0.3 mm undersize; after 24 hours at 50% relative humidity, it may be 0.1 mm oversize. Always specify the conditioning state in your drawing.
How to Specify Tolerances in Your RFQ
A well-written tolerance spec prevents disputes and reduces the cost of a mold program. Follow these rules:
- Specify only critical dimensions. Non-critical dimensions should use ISO 20457 MT2 or MT3 by default. Every tightened tolerance adds cost.
- Use ISO 20457 classes. Write “MT2” next to the dimension, not just ±0.1 mm. This removes ambiguity.
- Define the reference condition. State whether dimensions are measured at 23°C / 50% RH, after 24-hour conditioning, or as-ejected.
- Separate functional from cosmetic dimensions. A mating surface may need ±0.05 mm; a visible A-surface may only need ±0.2 mm for appearance.
- Include datum references. GD&T (Geometric Dimensioning and Tolerancing) with datums is the professional standard for critical assemblies.
When in doubt, ask your mold maker for a tolerance capability matrix before you finalize the drawing. A good supplier will show you what each dimension can realistically achieve with the selected material and process — and flag any that require special tooling or process controls.
Tolerance vs. Cost: The Real Trade-off
Tightening a tolerance from ±0.2 mm to ±0.05 mm can increase mold cost by 30–50 percent and part cost by 15–25 percent. The reasons are straightforward: harder tool steel, more machining time, tighter fit tolerances, longer trial cycles, and stricter process control. For most commercial parts, MT2 (±0.1–0.2 mm) is the sweet spot — tight enough for assembly, affordable for volume production.
If your design requires MT1 tolerances on multiple features, discuss this with your mold maker early. There may be a cheaper alternative: a different material with lower shrinkage, a revised part geometry that reduces the number of tight features, or a post-molding machining step on a single critical dimension instead of holding it in the mold.
When to Use Coordinate Measuring Machine (CMM) vs. Go/No-Go
CMM inspection is the standard for production validation — it measures X, Y, Z coordinates with micron-level accuracy and generates a full dimensional report. It is the right choice for first-article inspection, PPAP submissions, and ongoing process validation.
Go/No-Go gauges are faster and cheaper for high-volume production checking. They are best suited for single features (hole diameter, pin length, thickness) rather than complex geometries. For a part with 20 dimensions, CMM is more efficient; for a part with one critical hole diameter, a custom go/no-go gauge is faster and sufficient.
Real-World Tolerance Examples by Industry
Different industries have different tolerance expectations. Here are typical ranges you will see in purchase specifications:
- Consumer electronics housings: ±0.1 to ±0.2 mm (MT2). Surface appearance matters, but internal snap-fits have some give.
- Automotive connectors: ±0.05 mm (MT1). Electrical contact geometry must be precise to ensure signal integrity and vibration resistance.
- Medical device components: ±0.02 to ±0.05 mm (MT1, sometimes tighter). Biocompatibility and sterilization repeatability require tight dimensional control.
- Industrial enclosures: ±0.2 to ±0.5 mm (MT3). Large parts with non-critical mating surfaces.
- Packaging caps and closures: ±0.1 mm (MT2). Thread engagement must be consistent for torque and sealing performance.
If you are sourcing from China and your spec was written for a European supplier, the tolerance class may not transfer directly. Always clarify the ISO 20457 class explicitly — do not assume that “±0.1 mm” means the same thing across all suppliers.
Avoiding Common Tolerance Mistakes
Even experienced engineers make tolerance specification errors. Here are the most frequent ones and how to avoid them:
- Over-specifying tolerances. Tightening every dimension to ±0.05 mm is the fastest way to blow your mold budget. Ask which dimensions are truly functional before tightening.
- Omitting reference conditions. A dimension measured hot after ejection is not comparable to one measured after 24-hour conditioning. Always state the measurement condition.
- Specifying tolerances on non-machinable features. Wire-cut EDM can hold ±0.01 mm; sandblasted texture cannot. Do not specify a tolerance on a surface feature that cannot be measured consistently.
- Ignoring stack-up analysis. A single dimension may look fine, but when five parts assemble, the tolerance stack may exceed your assembly clearance. Run a tolerance stack analysis on critical assemblies.
- Assuming all cavities are equal. In a multi-cavity mold, cavity 1 may produce a slightly different dimension than cavity 8 due to flow imbalance. Specify that tolerances apply across all cavities, not just the first one.
When to Request a Tolerance Audit
If you are new to injection molding or working with a new supplier, request a tolerance audit before committing to the tooling budget. A tolerance audit is a review of your drawings where the mold maker identifies every dimension that may be difficult to hold and suggests alternatives. The audit typically takes 1–2 days and can save thousands of dollars in mold rework or production scrap.
At RCH Plastic, we include a tolerance capability matrix in every DFM report. Each critical dimension is tagged with the expected process tolerance, the required tooling grade, and any special process controls needed. This transparency lets you make informed decisions before the mold is cut — not after the first shot is rejected.
Bottom Line
Injection molding tolerances are achievable — but only when they are specified clearly, matched to the right material and process, and backed by a mold maker who documents the tolerance capability before the steel is cut. The most expensive tolerance is the one you discover is impossible to hold after the mold is already built. Share your drawing with your supplier during the DFM phase, and let them flag every tolerance that needs special attention. That conversation will save you time, money, and disputes on every program you run.
For a detailed DFM review of your tolerance spec, or a tolerance capability quote for your part, contact RCH Plastic with your drawing and material selection. We will return a dimension-by-dimension tolerance analysis within 48 hours.
The international standard for injection molding tolerances is ISO 20457:2018. It defines three tolerance classes:
- MT1 — Highest precision. Typically ±0.05 to ±0.1 mm for small features. Used for optical, medical, and connector housings.
- MT2 — Standard precision. Typically ±0.1 to ±0.2 mm. The default for most automotive and industrial components.
- MT3 — General purpose. Typically ±0.2 to ±0.5 mm. Used for large structural parts, enclosures, and non-mating surfaces.
ISO 20457 does not replace supplier judgment. It provides a common language so that a tolerance written by a German engineer and a tolerance held by a Chinese mold maker describe the same requirement. When you write your RFQ, reference the ISO class and the specific dimension — not just “±0.1 mm” without context.
Five Factors That Control Your Tolerances
1. Material Shrinkage
Every thermoplastic shrinks as it cools from melt to solid. The shrinkage rate is material-specific and direction-dependent — flow direction shrinks less than the transverse direction. Common shrinkage ranges:
- PP: 1.5–2.5%
- PE: 1.5–4.0%
- ABS: 0.4–0.9%
- PA66: 0.7–1.5% (moisture-dependent)
- POM: 1.5–2.5%
- PC: 0.5–0.7%
- PTFE: 3.0–4.0%
A 100 mm feature in PP with 2% shrinkage will target 98 mm in the mold cavity. If the shrinkage varies between lots — and it will, unless the material is pre-dried and processed consistently — the finished part will drift. This is why material certification and batch control are part of tolerance management.
2. Part Geometry
Thin walls cool faster and shrink less, but they also require higher injection pressure, which can increase flash risk. Long, slender features flex during ejection and may spring back unpredictably. Undercuts and ribs create local cooling variation that warps the part. Simple, symmetric geometries are easier to hold to tight tolerances than complex, asymmetrical ones.
3. Mold Design and Construction
Tight tolerances require tight tooling. This means hardened steel cores and cavities (HRC 50+), precision ground parting surfaces, matched ejector pins, and adequate venting to prevent gas traps that cause local dimensional variation. A mold designed to MT1 tolerances will cost 20–40 percent more than one designed to MT3 because of the extra machining time, tighter fit tolerances, and longer trial cycles.
4. Machine and Process Control
The injection molding machine must be stable: consistent melt temperature, repeatable screw position, and adequate clamp force. Process parameters — injection speed, packing pressure, cooling time, and mold temperature — must be documented and controlled. Statistical process control (SPC) with CpK ≥ 1.33 is the minimum expectation for production parts. Processes running without documented parameters will produce variable dimensions.
5. Post-Molding Conditioning
Parts continue to shrink and distort after ejection, especially hygroscopic materials like PA66 and PC. Post-molding conditioning — drying, annealing, or humidity conditioning — stabilizes dimensions. A PA66 part measured immediately after ejection may be 0.3 mm undersize; after 24 hours at 50% relative humidity, it may be 0.1 mm oversize. Always specify the conditioning state in your drawing.
How to Specify Tolerances in Your RFQ
A well-written tolerance spec prevents disputes and reduces the cost of a mold program. Follow these rules:
- Specify only critical dimensions. Non-critical dimensions should use ISO 20457 MT2 or MT3 by default. Every tightened tolerance adds cost.
- Use ISO 20457 classes. Write “MT2” next to the dimension, not just ±0.1 mm. This removes ambiguity.
- Define the reference condition. State whether dimensions are measured at 23°C / 50% RH, after 24-hour conditioning, or as-ejected.
- Separate functional from cosmetic dimensions. A mating surface may need ±0.05 mm; a visible A-surface may only need ±0.2 mm for appearance.
- Include datum references. GD&T (Geometric Dimensioning and Tolerancing) with datums is the professional standard for critical assemblies.
When in doubt, ask your mold maker for a tolerance capability matrix before you finalize the drawing. A good supplier will show you what each dimension can realistically achieve with the selected material and process — and flag any that require special tooling or process controls.
Tolerance vs. Cost: The Real Trade-off
Tightening a tolerance from ±0.2 mm to ±0.05 mm can increase mold cost by 30–50 percent and part cost by 15–25 percent. The reasons are straightforward: harder tool steel, more machining time, tighter fit tolerances, longer trial cycles, and stricter process control. For most commercial parts, MT2 (±0.1–0.2 mm) is the sweet spot — tight enough for assembly, affordable for volume production.
If your design requires MT1 tolerances on multiple features, discuss this with your mold maker early. There may be a cheaper alternative: a different material with lower shrinkage, a revised part geometry that reduces the number of tight features, or a post-molding machining step on a single critical dimension instead of holding it in the mold.
When to Use Coordinate Measuring Machine (CMM) vs. Go/No-Go
CMM inspection is the standard for production validation — it measures X, Y, Z coordinates with micron-level accuracy and generates a full dimensional report. It is the right choice for first-article inspection, PPAP submissions, and ongoing process validation.
Go/No-Go gauges are faster and cheaper for high-volume production checking. They are best suited for single features (hole diameter, pin length, thickness) rather than complex geometries. For a part with 20 dimensions, CMM is more efficient; for a part with one critical hole diameter, a custom go/no-go gauge is faster and sufficient.
Real-World Tolerance Examples by Industry
Different industries have different tolerance expectations. Here are typical ranges you will see in purchase specifications:
- Consumer electronics housings: ±0.1 to ±0.2 mm (MT2). Surface appearance matters, but internal snap-fits have some give.
- Automotive connectors: ±0.05 mm (MT1). Electrical contact geometry must be precise to ensure signal integrity and vibration resistance.
- Medical device components: ±0.02 to ±0.05 mm (MT1, sometimes tighter). Biocompatibility and sterilization repeatability require tight dimensional control.
- Industrial enclosures: ±0.2 to ±0.5 mm (MT3). Large parts with non-critical mating surfaces.
- Packaging caps and closures: ±0.1 mm (MT2). Thread engagement must be consistent for torque and sealing performance.
If you are sourcing from China and your spec was written for a European supplier, the tolerance class may not transfer directly. Always clarify the ISO 20457 class explicitly — do not assume that “±0.1 mm” means the same thing across all suppliers.
Avoiding Common Tolerance Mistakes
Even experienced engineers make tolerance specification errors. Here are the most frequent ones and how to avoid them:
- Over-specifying tolerances. Tightening every dimension to ±0.05 mm is the fastest way to blow your mold budget. Ask which dimensions are truly functional before tightening.
- Omitting reference conditions. A dimension measured hot after ejection is not comparable to one measured after 24-hour conditioning. Always state the measurement condition.
- Specifying tolerances on non-machinable features. Wire-cut EDM can hold ±0.01 mm; sandblasted texture cannot. Do not specify a tolerance on a surface feature that cannot be measured consistently.
- Ignoring stack-up analysis. A single dimension may look fine, but when five parts assemble, the tolerance stack may exceed your assembly clearance. Run a tolerance stack analysis on critical assemblies.
- Assuming all cavities are equal. In a multi-cavity mold, cavity 1 may produce a slightly different dimension than cavity 8 due to flow imbalance. Specify that tolerances apply across all cavities, not just the first one.
When to Request a Tolerance Audit
If you are new to injection molding or working with a new supplier, request a tolerance audit before committing to the tooling budget. A tolerance audit is a review of your drawings where the mold maker identifies every dimension that may be difficult to hold and suggests alternatives. The audit typically takes 1–2 days and can save thousands of dollars in mold rework or production scrap.
At RCH Plastic, we include a tolerance capability matrix in every DFM report. Each critical dimension is tagged with the expected process tolerance, the required tooling grade, and any special process controls needed. This transparency lets you make informed decisions before the mold is cut — not after the first shot is rejected.
Bottom Line
Injection molding tolerances are achievable — but only when they are specified clearly, matched to the right material and process, and backed by a mold maker who documents the tolerance capability before the steel is cut. The most expensive tolerance is the one you discover is impossible to hold after the mold is already built. Share your drawing with your supplier during the DFM phase, and let them flag every tolerance that needs special attention. That conversation will save you time, money, and disputes on every program you run.
For a detailed DFM review of your tolerance spec, or a tolerance capability quote for your part, contact RCH Plastic with your drawing and material selection. We will return a dimension-by-dimension tolerance analysis within 48 hours.
Injection Molding Tolerances Guide: How Tight Can You Go?
When you submit an RFQ for injection molded parts, the first line in your spec sheet should be the tolerance requirement — but “±0.1 mm” means something very different to a mold designer than it does to a procurement manager. The gap between what looks good on paper and what your supplier can reliably hold in production is where most tooling disputes start. This guide explains what injection molding tolerances actually are, how they are specified, which factors control them, and how to write a tolerance spec that gets you good parts without inflating your mold cost by 30 percent.
What Are Injection Molding Tolerances?
Injection molding tolerance is the allowable deviation from a nominal dimension. It is usually expressed as a bilateral value such as ±0.1 mm or as a unilateral stack such as +0.15 / −0.05 mm. Tolerances apply to linear dimensions, angular dimensions, wall thickness, hole position, and surface features. They are not a single number for an entire part — each critical dimension gets its own tolerance, and non-critical dimensions are left to the process window.
Standard injection molding tolerances for most thermoplastics fall in the range of ±0.1 to ±0.2 mm for dimensions up to 100 mm. Beyond 100 mm, tolerances typically widen by about 0.05 mm per additional 100 mm of travel. These are process capabilities, not guarantees — a reliable mold maker will confirm each tolerance against the material shrinkage table, the expected part geometry, and the tool steel selected.
ISO 20457 and the Three Tolerance Classes
The international standard for injection molding tolerances is ISO 20457:2018. It defines three tolerance classes:
- MT1 — Highest precision. Typically ±0.05 to ±0.1 mm for small features. Used for optical, medical, and connector housings.
- MT2 — Standard precision. Typically ±0.1 to ±0.2 mm. The default for most automotive and industrial components.
- MT3 — General purpose. Typically ±0.2 to ±0.5 mm. Used for large structural parts, enclosures, and non-mating surfaces.
ISO 20457 does not replace supplier judgment. It provides a common language so that a tolerance written by a German engineer and a tolerance held by a Chinese mold maker describe the same requirement. When you write your RFQ, reference the ISO class and the specific dimension — not just “±0.1 mm” without context.
Five Factors That Control Your Tolerances
1. Material Shrinkage
Every thermoplastic shrinks as it cools from melt to solid. The shrinkage rate is material-specific and direction-dependent — flow direction shrinks less than the transverse direction. Common shrinkage ranges:
- PP: 1.5–2.5%
- PE: 1.5–4.0%
- ABS: 0.4–0.9%
- PA66: 0.7–1.5% (moisture-dependent)
- POM: 1.5–2.5%
- PC: 0.5–0.7%
- PTFE: 3.0–4.0%
A 100 mm feature in PP with 2% shrinkage will target 98 mm in the mold cavity. If the shrinkage varies between lots — and it will, unless the material is pre-dried and processed consistently — the finished part will drift. This is why material certification and batch control are part of tolerance management.
2. Part Geometry
Thin walls cool faster and shrink less, but they also require higher injection pressure, which can increase flash risk. Long, slender features flex during ejection and may spring back unpredictably. Undercuts and ribs create local cooling variation that warps the part. Simple, symmetric geometries are easier to hold to tight tolerances than complex, asymmetrical ones.
3. Mold Design and Construction
Tight tolerances require tight tooling. This means hardened steel cores and cavities (HRC 50+), precision ground parting surfaces, matched ejector pins, and adequate venting to prevent gas traps that cause local dimensional variation. A mold designed to MT1 tolerances will cost 20–40 percent more than one designed to MT3 because of the extra machining time, tighter fit tolerances, and longer trial cycles.
4. Machine and Process Control
The injection molding machine must be stable: consistent melt temperature, repeatable screw position, and adequate clamp force. Process parameters — injection speed, packing pressure, cooling time, and mold temperature — must be documented and controlled. Statistical process control (SPC) with CpK ≥ 1.33 is the minimum expectation for production parts. Processes running without documented parameters will produce variable dimensions.
5. Post-Molding Conditioning
Parts continue to shrink and distort after ejection, especially hygroscopic materials like PA66 and PC. Post-molding conditioning — drying, annealing, or humidity conditioning — stabilizes dimensions. A PA66 part measured immediately after ejection may be 0.3 mm undersize; after 24 hours at 50% relative humidity, it may be 0.1 mm oversize. Always specify the conditioning state in your drawing.
How to Specify Tolerances in Your RFQ
A well-written tolerance spec prevents disputes and reduces the cost of a mold program. Follow these rules:
- Specify only critical dimensions. Non-critical dimensions should use ISO 20457 MT2 or MT3 by default. Every tightened tolerance adds cost.
- Use ISO 20457 classes. Write “MT2” next to the dimension, not just ±0.1 mm. This removes ambiguity.
- Define the reference condition. State whether dimensions are measured at 23°C / 50% RH, after 24-hour conditioning, or as-ejected.
- Separate functional from cosmetic dimensions. A mating surface may need ±0.05 mm; a visible A-surface may only need ±0.2 mm for appearance.
- Include datum references. GD&T (Geometric Dimensioning and Tolerancing) with datums is the professional standard for critical assemblies.
When in doubt, ask your mold maker for a tolerance capability matrix before you finalize the drawing. A good supplier will show you what each dimension can realistically achieve with the selected material and process — and flag any that require special tooling or process controls.
Tolerance vs. Cost: The Real Trade-off
Tightening a tolerance from ±0.2 mm to ±0.05 mm can increase mold cost by 30–50 percent and part cost by 15–25 percent. The reasons are straightforward: harder tool steel, more machining time, tighter fit tolerances, longer trial cycles, and stricter process control. For most commercial parts, MT2 (±0.1–0.2 mm) is the sweet spot — tight enough for assembly, affordable for volume production.
If your design requires MT1 tolerances on multiple features, discuss this with your mold maker early. There may be a cheaper alternative: a different material with lower shrinkage, a revised part geometry that reduces the number of tight features, or a post-molding machining step on a single critical dimension instead of holding it in the mold.
When to Use Coordinate Measuring Machine (CMM) vs. Go/No-Go
CMM inspection is the standard for production validation — it measures X, Y, Z coordinates with micron-level accuracy and generates a full dimensional report. It is the right choice for first-article inspection, PPAP submissions, and ongoing process validation.
Go/No-Go gauges are faster and cheaper for high-volume production checking. They are best suited for single features (hole diameter, pin length, thickness) rather than complex geometries. For a part with 20 dimensions, CMM is more efficient; for a part with one critical hole diameter, a custom go/no-go gauge is faster and sufficient.
Real-World Tolerance Examples by Industry
Different industries have different tolerance expectations. Here are typical ranges you will see in purchase specifications:
- Consumer electronics housings: ±0.1 to ±0.2 mm (MT2). Surface appearance matters, but internal snap-fits have some give.
- Automotive connectors: ±0.05 mm (MT1). Electrical contact geometry must be precise to ensure signal integrity and vibration resistance.
- Medical device components: ±0.02 to ±0.05 mm (MT1, sometimes tighter). Biocompatibility and sterilization repeatability require tight dimensional control.
- Industrial enclosures: ±0.2 to ±0.5 mm (MT3). Large parts with non-critical mating surfaces.
- Packaging caps and closures: ±0.1 mm (MT2). Thread engagement must be consistent for torque and sealing performance.
If you are sourcing from China and your spec was written for a European supplier, the tolerance class may not transfer directly. Always clarify the ISO 20457 class explicitly — do not assume that “±0.1 mm” means the same thing across all suppliers.
Avoiding Common Tolerance Mistakes
Even experienced engineers make tolerance specification errors. Here are the most frequent ones and how to avoid them:
- Over-specifying tolerances. Tightening every dimension to ±0.05 mm is the fastest way to blow your mold budget. Ask which dimensions are truly functional before tightening.
- Omitting reference conditions. A dimension measured hot after ejection is not comparable to one measured after 24-hour conditioning. Always state the measurement condition.
- Specifying tolerances on non-machinable features. Wire-cut EDM can hold ±0.01 mm; sandblasted texture cannot. Do not specify a tolerance on a surface feature that cannot be measured consistently.
- Ignoring stack-up analysis. A single dimension may look fine, but when five parts assemble, the tolerance stack may exceed your assembly clearance. Run a tolerance stack analysis on critical assemblies.
- Assuming all cavities are equal. In a multi-cavity mold, cavity 1 may produce a slightly different dimension than cavity 8 due to flow imbalance. Specify that tolerances apply across all cavities, not just the first one.
When to Request a Tolerance Audit
If you are new to injection molding or working with a new supplier, request a tolerance audit before committing to the tooling budget. A tolerance audit is a review of your drawings where the mold maker identifies every dimension that may be difficult to hold and suggests alternatives. The audit typically takes 1–2 days and can save thousands of dollars in mold rework or production scrap.
At RCH Plastic, we include a tolerance capability matrix in every DFM report. Each critical dimension is tagged with the expected process tolerance, the required tooling grade, and any special process controls needed. This transparency lets you make informed decisions before the mold is cut — not after the first shot is rejected.
Bottom Line
Injection molding tolerances are achievable — but only when they are specified clearly, matched to the right material and process, and backed by a mold maker who documents the tolerance capability before the steel is cut. The most expensive tolerance is the one you discover is impossible to hold after the mold is already built. Share your drawing with your supplier during the DFM phase, and let them flag every tolerance that needs special attention. That conversation will save you time, money, and disputes on every program you run.
For a detailed DFM review of your tolerance spec, or a tolerance capability quote for your part, contact RCH Plastic with your drawing and material selection. We will return a dimension-by-dimension tolerance analysis within 48 hours.