Injection Mold Venting: Where the Air Goes and Why Parts Burn

The diesel effect: a burn mark is compression ignition

A brown or black scorch mark on an injection moulded part looks like the melt got too hot. It did, but not because the barrel was too hot. The air trapped in front of the melt was compressed into a fraction of its volume in a few tenths of a second, and compressing a gas that fast is close to adiabatic: the work of compression has nowhere to go, so it becomes heat.

The arithmetic is worth doing once because it explains everything else in this article. For air, the adiabatic relation gives a temperature ratio of (V₁/V₂)^0.4. Compress the cavity air to a tenth of its volume and the ratio is 2.5, which turns 20 °C room air into roughly 460 °C. Compress it to a twentieth and it reaches about 700 °C. That second number should be familiar: it is the compression ratio and the temperature at which a diesel engine ignites its fuel.

That is exactly what is happening. The trapped gas is not merely hot, it is hot enough to pyrolyse the polymer in contact with it. The residue is carbon, and carbon is brown-black. The smell is acrid and distinct, and an operator who has worked with it can identify a venting problem from across the shop before seeing the part.

Where the air actually comes from

Operators tend to think of the cavity as empty. It is not. Three separate sources put gas in front of the melt, and they need different fixes.

Table 1 — Gas sources in a moulding cavity. Cavity air is the largest by volume; volatiles cause the most inconsistent defects.

Cavity air is the big one by volume, and it is unavoidable: a 200 g polypropylene part displaces roughly 220 cm³ of air, and that air has to be out of the way in the 0.4 to 0.8 seconds the cavity takes to fill. That is a flow of several hundred cubic centimetres per second through gaps measured in hundredths of a millimetre. It is not a trivial amount of gas to move.

Volatiles are smaller in volume but far more damaging, because they arrive late and they condense. Moisture in an under-dried polyamide flashes to steam at the melt front and expands by a factor of over a thousand; that steam is at the melt front, it is at the end of fill, and it carries the heat with it. This is why a part can be perfectly dry in summer and burn in the monsoon, on the same tool and the same settings.

Leaking water is the one that gets misdiagnosed most often. A hairline crack in a cooling channel or a bad O-ring lets water into the cavity, and the result is a continuous stream of steam that looks exactly like a venting problem but does not move when you cut the vent deeper. If the defect is present on every shot including the first one after a long stop, pressure-test the tool before touching the vent.

Cutting a vent that actually works

A vent is not a groove. It is a shallow land followed by a deep relief, and both dimensions matter for different reasons.

The land is the shallow section, typically 0.8 to 1.5 mm long, and its depth is set by the material's viscosity at the moment it reaches the vent. Too deep and the melt flashes into the vent and blocks it on the first shot; too shallow and the air cannot escape fast enough. The relief behind the land is 0.3 to 0.5 mm deep and leads to atmosphere, and its only job is to stop the shallow land from becoming a long restriction.

Table 2 — Vent land depths by material. These are starting points; the only real test is whether the part flashes or burns.

Depth control is where most venting jobs fail. A 0.02 mm land is 20 microns, which is finer than most shop measurement can verify and well inside the tolerance of a hand-held grinder. Vents should be cut by EDM, surface ground, or milled and then stoned flat against a surface plate. A vent cut freehand with a die grinder will be 0.05 mm in places and 0 in others, and the deep spots will flash while the shallow spots still burn.

Width matters less than depth but is not irrelevant. Three to six millimetres per vent is typical, and the total vent length around a perimeter should be sized on the gas flow, not on convenience. A large flat part with a single 5 mm vent at the end of fill will burn no matter how deep the vent is, because the air has to travel the full length of the part to reach it.

The four ways to vent a mould

Table 3 — Venting methods. Most tools use the first two; the last two are for parts that cannot be vented conventionally.

The parting line vent is the workhorse because it is cheap and it is adjustable. Its limitation is geometric: it only helps air that reaches the split face, and a deep rib, a boss, or a blind pocket traps air in a place the split face cannot reach. That is when the other three methods come in.

Ejector pin clearance is the free option that shops underuse. A pin sitting in a hole with 0.02 to 0.03 mm of diametral clearance is already a vent, and if a trap happens to be near a pin the problem is solved without touching the tool. The catch is that the clearance is a ring, not a slot, and the gas has to get to it.

Porous sintered steel inserts are the clean answer for large flat areas and for optical parts where a vent land would leave a witness line. They are pressed into the cavity wall and vent the whole area evenly through the porosity. They cost more than a machined vent, and they clog with mold release and degraded material over time, so they need a periodic cleaning regime — but for a part that has to be cosmetically perfect and cannot have a vent mark, they are often the only option that works.

Reading the burn to find the trap

The position of the burn tells you where the air is trapped, and the shape tells you how bad it is. This table is the fastest diagnostic in the shop.

Table 4 — Burn location against cause. The bottom row is the one that gets blamed on venting and never is.

The weld line case deserves a note because it is the one that requires a tooling change rather than a vent. When two flow fronts meet, they trap whatever gas is between them at the meeting point, and if that point is in the middle of a part there is nowhere for the gas to go. Adding a vent at the split face will not help, because the gas is not at the split face. The fix is to move the gate so that the last point to fill is somewhere ventable, or to add a flow leader that carries the weld line to an edge.

Vacuum venting, and when it earns its cost

Vacuum venting evacuates the cavity before the shot, typically to −0.3 to −0.6 bar, and holds it until the melt seals the gates. It removes the cavity air entirely rather than trying to push it out through gaps, which means it works on geometries that cannot be vented at all: deep blind pockets, parts with no split face at the trap, and optical parts that cannot carry a vent witness line.

The cost is real. A vacuum system needs a sealed cavity, O-rings on the parting line, a valve that opens and closes each cycle, and a pump. Cycle time goes up by a second or two for the evacuation and the valve actuation, and the sealing adds maintenance. On a high-cavity tool running millions of parts, that is a large investment for a problem that a properly cut vent would have solved.

Where it does earn its cost is when the part cannot be vented conventionally and the scrap rate is high. A thick-walled polycarbonate lens with a deep pocket, or a part with a decorative surface on every face, can run at 20% scrap on trapped gas; vacuum venting pays back in weeks at that rate. The rule of thumb is that vacuum venting is worth considering when the part geometry has a trap that no conventional vent can reach and the part value is high enough that scrap hurts.

What process settings can and cannot fix

Process settings can reduce a venting problem. They cannot solve one, and treating them as a solution is how a tool runs for years at 3% scrap without anyone investigating.

Table 5 — Process levers against trapped gas. Only the last one addresses a root cause.

Lowering injection speed is the standard field fix and it works by reducing the compression rate, which reduces the temperature rise. It is also a confession that the tool cannot vent, and it costs cycle time on every shot forever. On a part where the fill is already at the limit of short-shot, it is not even available.

Drying is the one process change that addresses a root cause, and it is the one most often skipped. Polyamide, polycarbonate, PET and PBT all absorb enough moisture to produce steam at the melt front, and the amount is measured in hundredths of a percent. A dryer that is working but not monitored is a common cause of intermittent burning that comes and goes with the weather.

Keeping a vent working

A vent that worked at tool acceptance and burns two years later has usually not changed depth. It has filled up. Mold release spray, degraded material, plasticiser bleed and rust all deposit in a 20 micron gap and close it.

The maintenance regime is simple and almost universally skipped: clean the vents on a schedule, not on a defect. For a tool running a volatile material like PVC, POM or a flame-retardant grade, that can mean every shift. For a clean polyolefin tool it might be monthly. The clean is a soft brass scraper and a solvent wipe, followed by a check that the land is still open with a feeler or a light behind the parting line.

Vents also wear open. A tool that has run a few million shots will have parting line vents that started at 0.03 mm and are now 0.06 mm, and the symptom is flash rather than a burn. That is the other half of the same maintenance job: measure the land depth periodically, and re-cut or re-stone when it has doubled.