---
title: "How to eliminate visible sink marks on thick-wall injection molded plastic components?"
description: "Address persistent sink mark defects on injection molded parts during pre-mass production trials, with actionable process, tooling and material strategies to lower defect rates, meet dimensional tolerances and ensure part functional compliance."
url: "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html"
language: "en"
type: "Q&A"
category: "Injection Molding Q&A"
datePublished: "2026-09-28"
dateModified: "2026-09-28"
brand: "OK TOOL"
manufacturer: "OK TOOL"
answerCount: 7
---

# How to eliminate visible sink marks on thick-wall injection molded plastic components?

## Question

 I am currently running first article trials for a custom ABS tool housing that will go into mass production in 4 weeks, and we are facing a persistent sink mark issue right now. The marks are 0.2mm to 0.3mm deep, located directly above three 2.5mm thick support bosses on the non-cosmetic side of the part. Our current trial runs already pushed holding pressure up to 95 bar and extended holding time to 12 seconds, but the sink marks are still present, and 12% of the sampled parts are exceeding our 0.1mm depth tolerance requirement. If we push holding pressure any higher, we start getting part flash at the edge gates and 0.08% over the specified outer diameter tolerance. We cannot move the gate position at this stage because the mold steel has already been finished machining. We need to lock in stable parameters this week to meet the mass production launch timeline, and do not want to rework the mold if that can be avoided. What concrete steps can we take to reduce these sink marks properly without triggering new defects or delaying our project? 

## Answers
                            
### Answer 1 — Best Answer

The sink marks you are seeing above the support bosses stem from uneven volumetric shrinkage of the ABS material in the thick isolated section, where the material cools at a much slower rate than the adjacent thin nominal 1.8mm wall. The holding pressure you have already applied cannot transfer effectively to the inner core of the boss once the gate has frozen off, which is why increasing holding pressure beyond 95 bar only adds stress to the melt at the gate location, causes flash, and does not pack the thick boss area sufficiently.

The first set of actionable adjustments can be implemented within 2-3 trial runs without any mold modification. **Increase the back pressure from the current 3 bar to 7-9 bar, and raise the barrel rear zone temperature by 8 to 10 degrees Celsius**. This improves the melt homogeneity and reduces the material pre-shrinkage that happens in the barrel before injection, so the total shrinkage that occurs inside the cavity after fill is lower. Next, extend the second stage holding profile from a single 12 second 95 bar step to a 3-step ramp down: 90 bar for 5 seconds, 70 bar for 7 seconds, 45 bar for 6 seconds. This maintains packing pressure for longer before the gate fully solidifies, and prevents sudden pressure drop that pulls material away from the thick boss surface. You will also need to raise mold coolant temperature on the core side by 5 degrees for the first 8 seconds of the cooling cycle, then drop it 15 degrees to speed up overall cooling: this avoids early skin formation on the boss surface that traps internal shrinkage and creates visible dents.

Once you run these adjusted parameters, the sink mark depth will typically drop to under 0.08mm, which is well within your tolerance. If you still see 2-3% of parts that fall outside tolerance after these process tweaks, add a small 0.8mm deep vent insert at the back face of each boss on the mold core side, to allow trapped air to escape so the full packing pressure can reach every part of the thick section. Do not raise melt temperature higher than recommended for ABS, as this will increase total post-mold shrinkage and lead to warpage issues 72 hours after parts are demolded.

To prevent this issue from recurring in mass production, **add a specific sink mark check point at 1 hour intervals during IPQC with a 0.1mm feeler gauge** to avoid bad parts flowing to downstream assembly. Record the actual gate freeze time for this material and part design during your final trial run, and lock the holding time to 1.2x that measured value in your standard work instruction, to eliminate the risk of operator variation leading to defect spikes. For future similar parts, mark all thick boss to nominal wall ratio values in the DFM review stage to flag high sink mark risk areas before the mold is machined.

**status:** accepted
**Author:** Emily Chen
**Date:** 2026-09-28

### Answer 2

You can start by mapping the current defect rate across 5 consecutive 50-shot trial runs to isolate if the sink mark variation comes from parameter drift or inherent part design constraints. First, log every shot’s actual holding pressure curve, melt temperature reading, and cycle time deviation, then filter out runs where the hydraulic pressure fluctuates more than 3 bar to eliminate machine variability as the root cause.

Once you confirm the machine is running stable, implement a controlled DOE with 3 levels of back pressure and 3 levels of switch over position from injection to holding, to identify the optimal parameter window that keeps sink marks under 0.1mm while not generating flash. This lean DOE approach typically narrows down your stable operating window within 12 trial shots, and reduces long-term production defect rate to under 0.3% without adding extra cycle time that impacts your overall production efficiency. You can also document these parameter ranges for future similar ABS part projects to cut down trial validation time by 40%.

**status:** suggested
**Author:** Rachel Huang
**Date:** 2026-09-28

### Answer 3

There is a low effort, low cost mold tweak you can implement in less than 4 hours without full mold disassembly if process adjustments do not hit your target. You can polish the core side surface of each support boss by 0.05mm, then apply a high thermal conductivity boron nitride based coating only to those local boss surfaces, which improves heat transfer speed at the thick section by roughly 22%. This helps the material at the boss surface cool evenly with the surrounding wall, and prevents the surface from sinking inward as the internal core of the boss shrinks.

This modification does not change the part’s critical dimensional dimensions, and will not reduce the overall mold service life, as the coating has a proven wear resistance rating for over 250,000 injection cycles. You will also need to add those 3 local areas to your monthly mold maintenance cleaning checklist, to ensure no residual plastic buildup accumulates on the coated surface that could reduce heat transfer performance over long production runs.

**status:** suggested
**Author:** Olivia Chen
**Date:** 2026-09-28

### Answer 4

Even if full gate relocation is not feasible at this stage, you can make small adjustments to the existing gate land dimensions to improve packing efficiency. You can extend the gate land length from the current 1.2mm to 1.8mm, which slows down the gate freeze rate by roughly 30% and allows holding pressure to be transferred into the cavity for a longer total period.

This modification only requires a simple bench machining step on the mold insert edge, and can be completed in a few hours without altering any other mold structure. The extended gate land will also reduce the shear heat generated during injection, so the melt inside the cavity maintains a more uniform temperature profile, lowering uneven shrinkage at the thick boss sections. You do not need to adjust existing cycle time after this change, as the extra gate freeze time is offset by slightly reduced required holding pressure.

**status:** suggested
**Author:** Eric Zhao
**Date:** 2026-09-28

### Answer 5

Since these sink marks are located on the non-cosmetic side of the tool housing, you can also verify if minor sink mark depth below 0.15mm has no impact on the part’s actual end use performance. Assemble 20 parts with 0.08mm to 0.12mm sink mark depth onto your final product assembly, then run the full 72 hour thermal cycle test, drop test, and load bearing test for the support boss to confirm there is no functional performance degradation.

If test results show that sink marks up to 0.15mm do not affect the boss’s structural strength or assembly fit, you can adjust your tolerance specification accordingly to avoid over-tight process constraints that cause higher reject rates in mass production. This avoids unnecessary mold rework, and does not compromise the final product’s field performance for end users, as long as the final sink mark depth does not cause any interference with mating components during assembly.

**status:** suggested
**Author:** Kevin Liu
**Date:** 2026-09-28

### Answer 6

You can implement a layered defect classification system to separate acceptable parts from non-conforming units after trial parameters are locked, to prevent mixed defect parts from being shipped. First, define 3 distinct sink mark grades: grade 0 for depth under 0.05mm, grade 1 for 0.05mm to 0.1mm, grade 2 for over 0.1mm, then create a physical reference sample board with parts of each grade for all production line inspectors to reference.

Add a dedicated 2 second visual check step on the non-cosmetic side of the part right after the parts are demolded, before the deflashing process. Use a calibrated 0.1mm feeler gauge for random sampling every 20 parts, to confirm visual inspection judgment is consistent across different operators. You can also add a non-contact laser thickness measurement station on the end of the production conveyor line for 100% inspection of the 3 boss positions, if your mass production volume is over 50k units per month to eliminate human inspection error.

**status:** suggested
**Author:** Sophia Wang
**Date:** 2026-09-28

### Answer 7

You can make a minor change to the part’s internal boss geometry that does not impact its assembly or functional performance, to reduce the volumetric shrinkage difference between the boss and the adjacent wall. Machine a 1mm diameter through hole on the back side of each support boss, so the total material volume of the thick section is reduced by roughly 35%, which eliminates the isolated heavy mass of material that causes extreme uneven shrinkage.

This hole can be located on the inside of the boss that will be covered by the mated screw after assembly, so it is completely invisible to end users and does not affect the boss’s screw pull out strength. This small design adjustment cuts down sink mark tendency drastically, and can be implemented by adding a small ejector pin insert on the mold core side for each boss, with minimal machining work and no impact on other part dimensions.

**status:** suggested
**Author:** Amy Li
**Date:** 2026-09-28

## Related Resources

- [Injection Molding Q&A](https://www.ok-tool.com/qa/injection-molding/)
- [Plastic Injection Molding](https://www.ok-tool.com/capabilities/plastic-injection-molding/)
- [Injection Molding Guide](https://www.ok-tool.com/manufacturing/injection-molding/)
- [Injection Molding Buying Guides](https://www.ok-tool.com/buying/injection-molding/)
- [Products](https://www.ok-tool.com/products/)
- [Custom Manufacturing](https://www.ok-tool.com/custom-manufacturing/)
- [Manufacturing Knowledge Base](https://www.ok-tool.com/knowledge/)
- [Injection Molding](https://www.ok-tool.com/knowledge/injection-molding/)

## Structured Data

```json
[
    {
      "@context": "https://schema.org",
      "@type": "QAPage",
      "mainEntity": {
        "@type": "Question",
        "name": "How to eliminate visible sink marks on thick-wall injection molded plastic components?",
        "text": "I am currently running first article trials for a custom ABS tool housing that will go into mass production in 4 weeks, and we are facing a persistent sink mark issue right now. The marks are 0.2mm to 0.3mm deep, located directly above three 2.5mm thick support bosses on the non-cosmetic side of the part. Our current trial runs already pushed holding pressure up to 95 bar and extended holding time to 12 seconds, but the sink marks are still present, and 12% of the sampled parts are exceeding our 0.1mm depth tolerance requirement. If we push holding pressure any higher, we start getting part flash at the edge gates and 0.08% over the specified outer diameter tolerance. We cannot move the gate position at this stage because the mold steel has already been finished machining. We need to lock in stable parameters this week to meet the mass production launch timeline, and do not want to rework the mold if that can be avoided. What concrete steps can we take to reduce these sink marks properly without triggering new defects or delaying our project?",
        "answerCount": 7,
        "upvoteCount": 12,
        "datePublished": "2026-09-28T13:42:07Z",
        "dateModified": "2026-09-28T13:42:42Z",
        "author": {
          "@type": "Person",
          "name": "Anonymous",
          "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html"
        }
                ,"acceptedAnswer": {
            "@type": "Answer",
            "text": "The sink marks you are seeing above the support bosses stem from uneven volumetric shrinkage of the ABS material in the thick isolated section, where the material cools at a much slower rate than the adjacent thin nominal 1.8mm wall. The holding pressure you have already applied cannot transfer effectively to the inner core of the boss once the gate has frozen off, which is why increasing holding pressure beyond 95 bar only adds stress to the melt at the gate location, causes flash, and does not pack the thick boss area sufficiently. The first set of actionable adjustments can be implemented within 2-3 trial runs without any mold modification. Increase the back pressure from the current 3 bar to 7-9 bar, and raise the barrel rear zone temperature by 8 to 10 degrees Celsius . This improves the melt homogeneity and reduces the material pre-shrinkage that happens in the barrel before injection, so the total shrinkage that occurs inside the cavity after fill is lower. Next, extend the second stage holding profile from a single 12 second 95 bar step to a 3-step ramp down: 90 bar for 5 seconds, 70 bar for 7 seconds, 45 bar for 6 seconds. This maintains packing pressure for longer before the gate fully solidifies, and prevents sudden pressure drop that pulls material away from the thick boss surface. You will also need to raise mold coolant temperature on the core side by 5 degrees for the first 8 seconds of the cooling cycle, then drop it 15 degrees to speed up overall cooling: this avoids early skin formation on the boss surface that traps internal shrinkage and creates visible dents. Once you run these adjusted parameters, the sink mark depth will typically drop to under 0.08mm, which is well within your tolerance. If you still see 2-3% of parts that fall outside tolerance after these process tweaks, add a small 0.8mm deep vent insert at the back face of each boss on the mold core side, to allow trapped air to escape so the full packing pressure can reach every part of the thick section. Do not raise melt temperature higher than recommended for ABS, as this will increase total post-mold shrinkage and lead to warpage issues 72 hours after parts are demolded. To prevent this issue from recurring in mass production, add a specific sink mark check point at 1 hour intervals during IPQC with a 0.1mm feeler gauge to avoid bad parts flowing to downstream assembly. Record the actual gate freeze time for this material and part design during your final trial run, and lock the holding time to 1.2x that measured value in your standard work instruction, to eliminate the risk of operator variation leading to defect spikes. For future similar parts, mark all thick boss to nominal wall ratio values in the DFM review stage to flag high sink mark risk areas before the mold is machined.",
            "upvoteCount": 12,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#acceptedAnswer",
            "datePublished": "2026-09-28T14:31:55Z",
            "author": {"@type": "Person","name": "Emily Chen","url": "https://www.ok-tool.com/team/emily.html"}        }
                ,"suggestedAnswer": [
                  {
            "@type": "Answer",
            "text": "You can start by mapping the current defect rate across 5 consecutive 50-shot trial runs to isolate if the sink mark variation comes from parameter drift or inherent part design constraints. First, log every shot’s actual holding pressure curve, melt temperature reading, and cycle time deviation, then filter out runs where the hydraulic pressure fluctuates more than 3 bar to eliminate machine variability as the root cause. Once you confirm the machine is running stable, implement a controlled DOE with 3 levels of back pressure and 3 levels of switch over position from injection to holding, to identify the optimal parameter window that keeps sink marks under 0.1mm while not generating flash. This lean DOE approach typically narrows down your stable operating window within 12 trial shots, and reduces long-term production defect rate to under 0.3% without adding extra cycle time that impacts your overall production efficiency. You can also document these parameter ranges for future similar ABS part projects to cut down trial validation time by 40%.",
            "upvoteCount": 0,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#suggestedAnswer-2",
            "datePublished": "2026-09-28T14:29:51Z",
            "author": {"@type": "Person","name": "Rachel Huang","url": "https://www.ok-tool.com/team/rachel.html"}          }
          ,          {
            "@type": "Answer",
            "text": "There is a low effort, low cost mold tweak you can implement in less than 4 hours without full mold disassembly if process adjustments do not hit your target. You can polish the core side surface of each support boss by 0.05mm, then apply a high thermal conductivity boron nitride based coating only to those local boss surfaces, which improves heat transfer speed at the thick section by roughly 22%. This helps the material at the boss surface cool evenly with the surrounding wall, and prevents the surface from sinking inward as the internal core of the boss shrinks. This modification does not change the part’s critical dimensional dimensions, and will not reduce the overall mold service life, as the coating has a proven wear resistance rating for over 250,000 injection cycles. You will also need to add those 3 local areas to your monthly mold maintenance cleaning checklist, to ensure no residual plastic buildup accumulates on the coated surface that could reduce heat transfer performance over long production runs.",
            "upvoteCount": 0,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#suggestedAnswer-3",
            "datePublished": "2026-09-28T14:22:25Z",
            "author": {"@type": "Person","name": "Olivia Chen","url": "https://www.ok-tool.com/team/olivia.html"}          }
          ,          {
            "@type": "Answer",
            "text": "Even if full gate relocation is not feasible at this stage, you can make small adjustments to the existing gate land dimensions to improve packing efficiency. You can extend the gate land length from the current 1.2mm to 1.8mm, which slows down the gate freeze rate by roughly 30% and allows holding pressure to be transferred into the cavity for a longer total period. This modification only requires a simple bench machining step on the mold insert edge, and can be completed in a few hours without altering any other mold structure. The extended gate land will also reduce the shear heat generated during injection, so the melt inside the cavity maintains a more uniform temperature profile, lowering uneven shrinkage at the thick boss sections. You do not need to adjust existing cycle time after this change, as the extra gate freeze time is offset by slightly reduced required holding pressure.",
            "upvoteCount": 0,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#suggestedAnswer-4",
            "datePublished": "2026-09-28T14:22:15Z",
            "author": {"@type": "Person","name": "Eric Zhao","url": "https://www.ok-tool.com/team/eric.html"}          }
          ,          {
            "@type": "Answer",
            "text": "Since these sink marks are located on the non-cosmetic side of the tool housing, you can also verify if minor sink mark depth below 0.15mm has no impact on the part’s actual end use performance. Assemble 20 parts with 0.08mm to 0.12mm sink mark depth onto your final product assembly, then run the full 72 hour thermal cycle test, drop test, and load bearing test for the support boss to confirm there is no functional performance degradation. If test results show that sink marks up to 0.15mm do not affect the boss’s structural strength or assembly fit, you can adjust your tolerance specification accordingly to avoid over-tight process constraints that cause higher reject rates in mass production. This avoids unnecessary mold rework, and does not compromise the final product’s field performance for end users, as long as the final sink mark depth does not cause any interference with mating components during assembly.",
            "upvoteCount": 0,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#suggestedAnswer-5",
            "datePublished": "2026-09-28T14:09:32Z",
            "author": {"@type": "Person","name": "Kevin Liu","url": "https://www.ok-tool.com/team/kevin.html"}          }
          ,          {
            "@type": "Answer",
            "text": "You can implement a layered defect classification system to separate acceptable parts from non-conforming units after trial parameters are locked, to prevent mixed defect parts from being shipped. First, define 3 distinct sink mark grades: grade 0 for depth under 0.05mm, grade 1 for 0.05mm to 0.1mm, grade 2 for over 0.1mm, then create a physical reference sample board with parts of each grade for all production line inspectors to reference. Add a dedicated 2 second visual check step on the non-cosmetic side of the part right after the parts are demolded, before the deflashing process. Use a calibrated 0.1mm feeler gauge for random sampling every 20 parts, to confirm visual inspection judgment is consistent across different operators. You can also add a non-contact laser thickness measurement station on the end of the production conveyor line for 100% inspection of the 3 boss positions, if your mass production volume is over 50k units per month to eliminate human inspection error.",
            "upvoteCount": 0,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#suggestedAnswer-6",
            "datePublished": "2026-09-28T14:02:41Z",
            "author": {"@type": "Person","name": "Sophia Wang","url": "https://www.ok-tool.com/team/sophia.html"}          }
          ,          {
            "@type": "Answer",
            "text": "You can make a minor change to the part’s internal boss geometry that does not impact its assembly or functional performance, to reduce the volumetric shrinkage difference between the boss and the adjacent wall. Machine a 1mm diameter through hole on the back side of each support boss, so the total material volume of the thick section is reduced by roughly 35%, which eliminates the isolated heavy mass of material that causes extreme uneven shrinkage. This hole can be located on the inside of the boss that will be covered by the mated screw after assembly, so it is completely invisible to end users and does not affect the boss’s screw pull out strength. This small design adjustment cuts down sink mark tendency drastically, and can be implemented by adding a small ejector pin insert on the mold core side for each boss, with minimal machining work and no impact on other part dimensions.",
            "upvoteCount": 0,
            "url": "https://www.ok-tool.com/qa/eliminate-sink-marks-thick-wall-injection-molded-parts.html#suggestedAnswer-7",
            "datePublished": "2026-09-28T13:42:42Z",
            "author": {"@type": "Person","name": "Amy Li","url": "https://www.ok-tool.com/team/amy.html"}          }
                  ]
              }
    },
    {
      "@context": "https://schema.org",
      "@type": "BreadcrumbList",
      "itemListElement": [
          {"@type": "ListItem", "position": 1, "name": "Home", "item": "https://www.ok-tool.com/"},{"@type": "ListItem", "position": 2, "name": "Q&A", "item": "https://www.ok-tool.com/qa/"},{"@type": "ListItem", "position": 3, "name": "Injection Molding Q&A", "item": "https://www.ok-tool.com/qa/injection-molding/"}          ,{"@type": "ListItem", "position": 4, "name": "How to eliminate visible sink marks on thick-wall injection molded plastic components?"}
      ]
    }
]
```