A sink mark is a shallow dimple in the surface over a thick section, a rib or a boss. The thick area stays hot after the surface around it has frozen. As it cools it shrinks and pulls the skin inward. Start with packing: hold pressure, hold time and cushion. Then look at the geometry. The resin sets how much shrinkage has to be made up.
Also described as: sinks, dimples, depressions over ribs, shrink marks.
Quick diagnosis
| What you see | Likely cause | Check first |
|---|---|---|
| A dimple on the show face, directly opposite a rib or boss. | The rib or boss is thick compared with the wall, so the junction cools last and shrinks most. | Rib thickness against wall thickness; hold pressure and hold time |
| Sinks all over the part. Parts are light. | Not enough packing: hold pressure too low, hold time too short, or no cushion. | Cushion; part weight |
| Sinks come and go from shot to shot. The cushion wanders. | A worn or leaking check ring is letting pack pressure escape back past the screw. | Cushion and part weight over 20 shots |
| Sink in a thick section far from the gate. Areas near the gate are fine. | The flow path between the gate and the thick section freezes first, so pack pressure never reaches it. | Hold pressure; melt and mold temperature; where the gate sits relative to the thick section |
| Sink near the gate, or part weight still rising when you add hold time. | The gate is not frozen when hold ends, so material flows back out. Or the gate is small and freezes before the part is packed. | A gate freeze study; gate size |
| The surface looks good as the mold opens and the sink appears as the part cools on the bench. | The part is ejected with a hot core and a skin too soft to resist it, or one area of the tool runs hot. | Cooling time; steel temperature next to the sink |
| Sinks started with a new lot or a new resin. | The new material shrinks more, or flows differently and packs differently. | COA of the good lot against the bad lot; shrinkage of the old and new grade |
Checks in order, cheapest first
- Weigh the parts. Part weight is the quickest measure of packing. Light parts with sinks need more material packed in. Parts at full weight with a sink over one rib point to geometry.
- Check the cushion. A cushion of zero means hold pressure is pushing on nothing. A cushion that drifts points to the check ring.
- Run a gate freeze study. Raise hold time in steps and weigh the parts at each step. When the weight stops rising, the gate is frozen. Set hold time just beyond that point.
- Raise hold pressure in steps. Watch for flash, sticking and over-packing near the gate. If a thick section far from the gate still sinks at the highest pressure the tool will take, pressure is not reaching it.
- Adjust temperatures and cooling time. A melt temperature toward the low end of the range leaves less shrinkage to make up, as long as the gate does not freeze before the part is packed. A longer cooling time stiffens the skin before ejection. A colder mold also stiffens the skin, but the shrinkage can then show up as a void inside the part.
- Compare lots. If the sinks arrived with a new lot, compare melt flow and any filler content on the COAs before you change the process.
- Look at the tool. Ribs and bosses should be noticeably thinner than the wall they join. Thick sections can often be cored out. The gate should feed the thick section first, and be large enough to stay open while it packs. Check cooling near the sink.
- Change the resin last. Sink is usually not the resin. If packing and geometry are as good as they can be and the sink still fails the part, a lower-shrink material is the next place to look.
The numbers
Typical mold shrinkage, in percent. The more a resin shrinks, the more volume a thick section has to make up, and the more it tends to sink. Where two figures are given, the first is along the flow and the second is across it.
| Resin | Unfilled | Filled |
|---|---|---|
| ABS | 0.4–0.7 | n/a |
| Polycarbonate | 0.5–0.7 (one producer gives 0.70 / 0.75) | 0.1–0.4 along the flow (10–30% glass) |
| General-purpose polystyrene | 0.3–0.8 | n/a |
| Nylon 6 | 0.90 / 0.90 | 0.30 / 0.75 (30% glass) |
| Nylon 66 | 1.50 / 1.80 | 0.50 / 1.10 (30–33% glass) |
| PBT | 1.6–2.0 | 0.30 / 1.10 (30% glass, one producer) |
| Acetal copolymer | 1.2–2.4 | See the grade sheet |
| Polypropylene homopolymer | 0.8–2.5 | 1.10–1.50 with 20% talc; 0.70–1.00 with 40% talc; 0.24 / 1.15 (30% glass) |
| HDPE | 1.70–2.90 | n/a |
Typical values from published processing guides. Shrinkage depends on the grade, wall thickness and packing, so confirm against the data sheet for your grade. The full list is in the melt temp, mold temp and shrinkage chart.
Why thick sections sink
- Plastic shrinks as it cools, and thick sections cool last. The skin against the steel freezes in the first moments. The core of a thick section is still molten and still shrinking long after.
- Hold pressure makes up the shrinkage, but only while the gate is open. Once the gate or the wall leading to the thick section freezes, no more material can get in. Whatever shrinkage is left has to come from somewhere.
- A soft skin gives a sink. A stiff skin gives a void. If the surface can still move, the core pulls it in and you see a dimple. If the surface is already rigid, the core pulls apart inside and leaves a hole. They are the same problem. See voids and bubbles.
- Semi-crystalline resins shrink more. Polypropylene, polyethylene, nylon, acetal and PBT shrink more than amorphous resins such as ABS, polycarbonate and polystyrene, so the same rib sinks more.
- Finish decides whether it gets rejected. A sink that fails a glossy, dark part can pass unnoticed on a textured or matte surface. The depth is the same.
Material changes, and what each one costs you
| Change | Why it helps | What it costs you |
|---|---|---|
| A glass-filled or mineral-filled grade | Fillers do not shrink, so the compound shrinks less, and the stiffer skin resists being pulled in. | A different surface finish, lower impact strength in most cases, more tool wear with glass, a heavier part, and with glass fiber uneven shrinkage that can cause warp. |
| An amorphous resin in place of a semi-crystalline one | Lower shrinkage, so less volume to make up. | Weaker chemical resistance and a higher risk of stress cracking near oils and solvents. A full part approval. |
| A chemical foaming agent, or a structural foam grade | Gas in the core pushes the skin against the mold from the inside, which packs out thick sections that hold pressure cannot reach. | A swirled surface that usually needs texture or paint, lower strength, a lighter part, and a process change your customer has to approve. |
| A melt flow range on the specification | Sinks that come and go by lot often trace to melt flow moving between lots. | Nothing in the part. It takes an incoming check, and it may narrow your choice of sources. |
Any resin change moves your dimensions. The tool was cut for the shrinkage of the resin you run now. A resin that shrinks less makes a bigger part from the same steel, so check critical dimensions before you approve a swap.
Often confused with
- Voids. The same shrinkage, hidden inside the part because the skin was too stiff to pull in. Fixing a sink with a colder mold can create one. See voids and bubbles.
- Warp. A sink is a local dimple and the part keeps its overall shape. Warp changes the shape of the whole part. See warped parts.
- Short shots. A short shot is missing material at an edge or at the end of fill, with a rounded, unfinished look. A sunk part is complete. See short shots.
Quick answers
What causes sink marks over ribs and bosses?
A rib or boss that is thick compared with the wall it joins cools last and shrinks most. The surface around it has already frozen, so the hot core pulls the skin inward as it cools. Check rib thickness against wall thickness, then hold pressure and hold time.
How do I get rid of sink marks?
Start with packing: hold pressure, hold time and cushion. Weigh the parts, confirm there is a cushion, and run a gate freeze study to set hold time. Then raise hold pressure in steps. If the sink remains, look at rib thickness, gate position and cooling in the tool.
Why does my part have dimples on the surface?
Dimples over thick sections are sink marks. The thick area stays hot after the surface around it has frozen, and as it cools it shrinks and pulls the skin inward. Sinks all over a light part point to not enough packing: low hold pressure, short hold time or no cushion.
Are sink marks caused by the resin or the process?
Sink is usually not the resin. Packing and part geometry come first, and the resin sets how much shrinkage has to be made up. If sinks started with a new lot, compare the COAs. Consider a lower-shrink material only after packing and geometry are as good as they can be.
Why do sink marks come and go from shot to shot?
Sinks that come and go while the cushion wanders usually point to a worn or leaking check ring. Pack pressure escapes back past the screw, so some shots are packed less than others. Track cushion and part weight over 20 shots to confirm it.
Send us a photo
A polymer chemist will look at it and reply by email. There is no charge and no obligation.