Venting is the part of a mold that is almost never on a drawing. It is a set of grooves a few hundredths of a millimetre deep, cut into a parting surface, and it decides whether a tool runs clean for a decade or burns the same corner of every part until someone blames the process. When a molder says a tool “runs difficult”, and the difficulty is a mark that will not move no matter what is changed at the press, the answer is usually not in the process window. It is in the gap that was never cut.
What is actually inside the cavity
A mold cavity is not empty when it closes. It is full of air at atmospheric pressure, and its volume is the volume of the part plus the runner. That air has to go somewhere, and the only place it can go is out through the parting line, which means through a gap measured in thousandths of a millimetre, in the same fraction of a second in which several grams of molten polymer are being forced in behind it.
That original air is the honest part of the problem, because it is the same on every shot and can be calculated. The rest of what ends up in the cavity is less predictable and is where most troubleshooting goes wrong: it varies with the resin, the drying, the melt temperature and the additive package, so two tools that look identical on the bench can behave completely differently on the same press.
| What is in the cavity | How much, and how often | What it does when it cannot get out |
|---|---|---|
| Air that was in the cavity when it closed | Roughly the swept volume of the cavity, every single shot, for the life of the tool | Burn marks at the last place to fill, in the same place every time |
| Air carried in among the granules | A small fraction of the shot; most of it escapes back up the feed throat | Adds to the total, and matters most on a short cycle where there is no time to spare |
| Moisture driven off the resin | Set by the drying you did or did not do | Silver streaks and splay, usually near the gate rather than at the end |
| Volatiles from additives, flame retardants and colorants | Small per shot, but it accumulates on the tool | A pale deposit on the vent land that looks like chalk and slowly closes the gap |
| Decomposition products from overheated melt | Grows quickly once the melt is above the recommended band | Dark streaks and a smell that is still there after the part has cooled |
Two conclusions follow from that table. The first is that cavity air is a constant and everything else is a variable, so a tool that burns on the first shot with correctly dried resin has a venting problem by definition — there is nothing else in the cavity to blame. The second is that three of the five rows leave a deposit on the tool, which means venting is not a feature you build once. It is a feature you maintain, and the maintenance is the part that gets forgotten.
The diesel effect, and why trapped air burns
This is the mechanism, and it is worth understanding properly because it explains why the damage is so localized and so severe. When the flow front reaches the end of the cavity and there is nowhere left to go, the melt keeps coming, driven by the injection pressure behind it. The gas in front of it is compressed. Compression is fast — a few hundredths of a second — and fast compression of a gas is close to adiabatic, which means the work done on it turns into heat rather than escaping into the steel.
The arithmetic is simple enough to do on the back of an envelope. For air, the ratio of the final to the initial absolute temperature is the pressure ratio raised to about 0.286. Compress air at 20 °C to a tenth of its volume and it reaches roughly 290 °C. Compress it to a hundredth — which is what a cavity pressure in the tens of megapascals produces — and the same expression gives a little over 800 °C.
Real compression is not adiabatic. The gas is in contact with steel at 60 to 90 °C and loses heat continuously, and some of it escapes through whatever gap exists. So the true peak is far below that figure. It does not need to reach 800 °C to do damage: polypropylene begins to degrade around 300 °C, POM and PVC well below that, and any of them will char on contact with a surface hot enough to brown a piece of paper.
What this explains, and what a process-only explanation cannot, is the localisation. The gas is hottest in the last few cubic millimetres to be compressed, which is the last few millimetres of the cavity to fill. That is why a burn mark sits in one corner and nowhere else, and why it comes back in the same corner after every attempt to fix it from the press.
It also explains a diagnostic detail that is genuinely useful. Lowering the injection speed often reduces the burn, which makes people conclude that the tool is fine and the process was wrong. What actually happened is that a slower fill gives the gas more time to leak out through the inadequate gap that is there. The tool was marginal before and is marginal now; the process has simply stopped pushing it over the edge. On the next run, with a slightly different batch of resin, the mark will be back.
There is a second, quieter version of the same mechanism. Gas that does not burn still has to go somewhere, and if the path is too restrictive it becomes a hydraulic lock: the flow front stalls a millimetre short of the end of the cavity, the press reaches its pressure limit, and the part comes out short with a smooth, rounded edge. That is a venting failure that gets diagnosed as insufficient injection pressure and answered with a bigger machine, which does not fix it. How much force a given part actually needs, and why a press at its pressure limit is usually a sign of restricted flow rather than a machine that is too small, is set out in clamp force and tonnage.
What a vent actually is
A vent is not a channel. It is a controlled leak. The distinction matters because almost every venting mistake comes from treating it as a channel and making it big enough to work.
The geometry has three parts. The land is the shallow section at the cavity edge, typically 1 to 3 mm long and a few millimetres wide, and its depth is the whole design decision. Behind it comes the relief, a groove 0.3 mm deep or more that carries the gas to the outside of the mold block without resisting it. Then the exit, which must actually reach atmosphere — a vent that ends against another plate, a shim or a clamp is not a vent.
Why the land is shallow
The land is shallow for a reason. Polymer melt at normal processing viscosity cannot enter a gap of 0.02 mm under the pressures in a cavity, because the resistance to flow through such a gap over a 2 mm length is enormous; gas at the same pressure passes through it easily because its viscosity is roughly a thousand times lower. The land is therefore a filter that passes gas and blocks melt, and it works only as long as the depth stays inside a narrow band.
Too shallow and the filter blocks everything: the gas cannot get through in the time available, and the tool behaves as if it had no vents at all. Too deep and the filter passes melt as well, which produces a thin fin of flash along the vent, and that flash then sits in the land and closes it permanently. A vent that has flashed once has usually stopped venting.
Two practical consequences follow, and both are routinely ignored. The first is that the depth tolerance on a vent is tighter than the tolerance on most cavity dimensions. A nominal 0.02 mm land with a plus or minus 0.01 mm tolerance is a feature that is either half size or double size, and both are failures. The second is that the land has to be produced by a method that can hold that tolerance — surface grinding, or wire or sinker EDM followed by hand stoning to depth. A milling cutter leaves a radius at the bottom of the groove and chatter along its length, and the depth that results is nominal at best. What a drawing can realistically hold on a feature this small is part of the same argument as mold tolerances, and the reason a vent is routinely specified at a tolerance tighter than the cavity it vents.
How deep, and why the grade decides
The correct depth is set by the viscosity of the melt at the moment it arrives at the vent, which follows the resin family far more than it follows the part. Low viscosity grades need a tighter land because they will find any gap; high viscosity grades need a more open one because otherwise they cannot be vented at all.
| Resin family | Vent land depth | Why it sits where it does |
|---|---|---|
| PP, PE, TPO | 0.02–0.03 mm | High viscosity melt that resists entering a thin gap, so the generous end of the range is safe |
| PS, ABS, SAN | 0.02 mm | Moderate viscosity; flash risk climbs quickly above 0.03 mm and the gain in venting is small |
| PA6, PA66, POM | 0.01–0.015 mm | Very low viscosity when hot; these grades will find any gap you leave and fill it permanently |
| PC, PC/ABS | 0.03–0.04 mm | Stiff, high viscosity melt that needs a freer path, and a low tendency to flash |
| PET, PBT | 0.015–0.02 mm | Low viscosity and shear sensitive, so the vent also has to be clean to avoid drag |
| Rigid PVC | 0.02–0.03 mm | Depth is the secondary problem; the vent must be cleaned often because decomposition products seal it |
| Any glass-filled grade | 0.03 mm, cleaned on a schedule | Fibers pack into the land and bridge it, so the effective depth falls faster than on an unfilled grade |
These are starting points rather than rules, and the adjustment that matters is nearly always in one direction. If a tool burns at the last place to fill and the land is already at the recommended depth, the answer is almost never to go deeper — it is to cut more vents, or to widen the existing ones.
Widen rather than deepen
That is the single most useful idea on this page. Widen rather than deepen. The cross-sectional area of the land is its width multiplied by its depth, so a land taken from 3 mm to 8 mm wide gains two and a half times the flow area at no extra flash risk, because flash is governed by depth alone. Doubling the depth gains the same area and doubles the risk of a fin. Yet in practice, when a tool burns, the first thing that happens is that someone stones the vent deeper, and the first thing that happens after that is a flash line along the parting surface that nobody connects to the repair.
Where a land has to be long — around a large cavity, for example — the length works against you, because flow resistance rises with length. That is the argument for relieving the land as soon as possible rather than carrying the shallow depth further than necessary, and for placing several short lands at the places gas actually collects instead of one continuous shallow groove around everything.
Where to put them, and how to find the place
Vent locations are usually chosen from the part drawing, on the assumption that gas ends up in the corners. It does not. Gas ends up where the flow front arrives last, and that is a function of wall thickness, gate position and flow balance, not of the outline of the part. On a part with a thick section at one end and a thin rim at the other, the thin rim fills last even if the thick end is further from the gate in millimetres.
The cheap and reliable way to find the real location is a short-shot trial. Set the shot size down until the cavity is 90 to 95 percent full, mold a few parts, and look at where the melt stopped. The last place to fill is the place the gas is being pushed to, and it is often somewhere nobody would have chosen from the drawing — the middle of a long edge, the base of a rib, the far side of a boss. It is the same trial that answers the rest of the questions in the DFM checklist, run on steel instead of on a screen.
| Location | Why gas collects there | What to cut |
|---|---|---|
| The last place to fill | By definition, that is where the gas has been pushed | One land, 3–6 mm wide, on the parting surface |
| Every weld line | Two flow fronts meet and close the gap behind them | A short land at the meeting point, on the parting line |
| Around the whole cavity edge | Cheap insurance: gas escapes wherever it happens to arrive | A continuous land, same depth, along the parting line |
| Deep ribs and bosses | Gas is trapped at the base of a blind feature and cannot rise through the melt | A land at the base, or a vented ejector pin in the feature |
| End of each runner | The runner fills before the cavity, so gas is compressed there first | A land at the sprue well and at each runner end |
| Under a slide that closes against the cavity | The slide face seals the pocket as the tool closes | A land across the slide face, relieved behind it |
There is one more location that is worth cutting on every tool regardless of the analysis: a continuous land around the whole cavity edge, at the same depth as the local vents. It costs almost nothing at the machining stage, it cannot hurt, and it means that gas which arrives somewhere unexpected has a path out instead of pressurising. Tools that vent only at the calculated last-to-fill point are the ones that burn the first time the resin batch changes.
Weld lines: where venting is worth the most
A weld line is where two flow fronts meet and stop. The gas between them has been pushed from both sides into a moving point, and when the fronts touch, that gas has nowhere to go unless there is a vent exactly at the meeting point — not near it, at it.
The mechanical consequence is larger than most people expect. A weld line in a well-vented tool typically reaches 60 to 80 percent of the strength of the surrounding material, which is already a design allowance that has to be made. In a poorly vented tool the same weld line can fall to 30 to 50 percent, because the trapped gas is hot enough to degrade the polymer at the interface before the two fronts can diffuse into each other. The part looks acceptable and fails in service, which is the worst combination.
Two details make the difference. The included angle between the two fronts matters: a head-on meeting produces a stronger weld than a shallow-angle one, where the fronts run alongside each other and the gas is smeared along the seam. And the vent has to be at the seam, on the surface where the seam reaches the parting line; a vent 5 mm to one side does nothing for the gas trapped at the meeting point itself.
For any part that will carry load, it is worth specifying venting at the weld line as a drawing requirement and then checking it on the first samples by breaking one. A dark, porous fracture surface at the weld is gas damage. A clean fracture that simply follows the seam is the normal, expected weakness of a weld line, and it can be designed around.
Ejector pins, slides and inserts
An ejector pin runs in a hole, and the clearance between pin and hole — typically 0.02 to 0.03 mm on the diameter, per side — is a ready-made vent. Many tools run acceptably on nothing else, which is why venting is often underestimated: the ejector pins are doing the work invisibly.
It is a fragile arrangement. A pin 150 mm long has enough length that the clearance is not uniform, and once the pin wears, the same gap that used to pass air starts to pass melt, producing a string of flash on the pin boss every shot. The fix is a vented pin: two or three flats ground along the side of the pin, 0.02 to 0.03 mm deep, which give a defined path for gas without relying on a worn clearance. A vented pin in a deep boss is one of the most effective single changes that can be made to a difficult tool.
Slides are the most commonly missed location. A slide closes against the cavity block, and when the tool is fully closed that interface is sealed by the closing force, so the pocket behind the slide has no vent at all. The gas has to be taken out across the slide face itself, with a land cut into the face and a relief behind it, arranged so that the land is only exposed to cavity pressure when the slide is home. Inserting the land into the block is a twenty-minute job at the machining stage and a two-day job once the tool is finished, which is the ordinary arithmetic of every detail decided during mold design and mold building. On a part with four slides, that is four separate pockets that can each burn independently.
Inserts are different and worth a caution. The joint line between a cavity insert and its pocket is a leak, not a vent: it is uncontrolled, it changes as the insert heats, and it usually leaks plastic rather than air once it has run for a while. Relying on an insert joint for venting is not a design, and the flash that appears along the joint line is the price of having done so.
Vents on the runner side
The runner fills before the cavity, which means gas in the runner is compressed first and has the least time of all to escape. On a cold runner tool, the places that need a vent are the end of the sprue well and the end of every secondary runner. A burn mark at the gate, on the runner side rather than on the part, is nearly always a runner vent that was never cut.
Hot runner tools have a different problem, and it runs against the usual assumption that a hot runner is automatically cleaner. The melt arriving through a hot drop is hotter and therefore less viscous than the same resin arriving through a cold runner, so it enters the vent land more easily. The result is that a hot runner tool with the same vent depth as its cold runner equivalent will flash earlier. The practical consequence is that hot runner tools often need to be vented slightly more conservatively, and vented in more places, not fewer. The trade-off between the two runner types is covered in hot runner versus cold runner.
The manifold itself needs attention too: the gap between the manifold and its pocket, the valve pin guides and the drop tips all collect gas and degraded material, and a manifold that has been running hot for a season will have a film of decomposed resin in exactly those places. It is one of the reasons a tool that was clean at T1 can burn six months later with nothing changed on the press.
When to evacuate instead of vent
Vents work by letting gas out through a very small gap in the time available. When the amount of gas is large, or the time is very short, or the geometry gives the gas nowhere to go, no arrangement of lands will keep up and the correct answer is to remove the gas before the shot rather than during it.
| Condition | Why vents alone are not enough | What evacuation changes |
|---|---|---|
| Deep, narrow cavity, depth greater than three times the width | The flow front closes over the top of the pocket and seals the gas underneath | The gas is already gone before the front arrives |
| Large flat part with a long flow path | The volume of gas is proportional to the part, while the vent area is a line | Removes the bulk of the gas in the seconds before injection |
| Optical or high-gloss surface | A burn mark anywhere on the part is a reject, and one vent is one failure point | Removes the whole class of defect rather than moving it |
| Thin wall with a fill time under half a second | There is not enough time for gas to travel along a 0.02 mm land | Starts the shot with almost nothing to expel |
| Many cavities in one tool | Every cavity needs its own vent, and the ones at the back of the tool are hard to reach | One seal around the whole cavity block covers all of them |
Evacuation is a system, not a feature, and it has to be treated that way. A sealed cavity block with an O-ring, a vacuum pump, a reservoir and a control that pulls the cavity down to a fraction of an atmosphere before the screw starts — each of those is a maintenance item, and the seal is the one that fails first because it is squeezed on every cycle.
The failure mode is what makes it worth being careful. A tool designed around evacuation that loses its seal does not degrade gracefully; it becomes worse than an unvented tool, because the design has no lands to fall back on. If evacuation is specified, the seal needs a replacement interval written into the tool documentation, and the pressure switch needs to interlock the press so that a lost vacuum stops the machine rather than producing scrap quietly.
Porous steel inserts
A porous or sintered steel insert is a block of tool steel with a controlled network of interconnected pores, typically 7 to 25 microns, through which gas passes and through which melt at normal viscosity does not. It is used where a vent cannot be cut: the bottom of a deep blind boss, the base of a tall rib, a small core surrounded by a closed pocket.
It works well in the right application and fails in a predictable set of wrong ones. Glass-filled grades are unsuitable because the fibers pack into the pore mouths and close them, and once closed they cannot be cleared. PVC and most flame-retardant grades are unsuitable because the decomposition products are corrosive and attack the sintered structure itself. And the material has lower strength and lower thermal conductivity than solid steel, so it cannot be used on a load-bearing face or where the cooling of that face matters — which is a real constraint, because the deep bosses and tall ribs that need venting most are also the places where the cooling circuit is already the hardest thing to place.
The maintenance regime is not optional. Porous inserts need periodic cleaning with an ultrasonic bath and the supplier’s recommended solvent, and they degrade irreversibly once the pores are blocked; there is no field repair. In practice, a porous insert that is not on a cleaning schedule becomes a solid insert within a year, and the burn mark that it was installed to fix comes back with the insert still in place — which then gets diagnosed as a process problem, because nobody suspects the feature that was supposed to solve it.
How vents fail over time
A vent is the only feature of a mold that is designed to be at the edge of working, and it degrades from the first shot. This is the reason two identical tools can behave differently after two years, and it is the reason a tool that ran clean in one plant burns in another where the maintenance routine is different.
| What happens | How it shows up | What to do |
|---|---|---|
| Release agent and oil film build up on the land | The vent looks clean to the eye, and the part starts burning again | Solvent wash and air blast, then stop spraying release agent into the cavity |
| A thin film of plastic bridges the land | Effective depth is halved or worse, and nothing looks wrong | Scrape with a copper or brass shim; if it will not lift, stone it back and re-measure |
| Polishing after a repair | The land is rounded and deepened, so the tool flashes instead of venting | Re-cut the land and measure it; polishing is the most common way a good vent is destroyed |
| Plating | The vent is filled completely and no longer functions at all | Mask the vents before plating, or re-cut them afterwards |
| Erosion from glass-filled resin | The land widens and shallows, so it leaks plastic instead of air | Re-cut, and put vent maintenance on a fixed schedule for abrasive grades |
| Oxidation and scale on a hot tool | The gap closes gradually over months, and the defect appears without any change to the process | Clean the vents on the same schedule as the cooling circuits |
Polishing is how a good vent dies
The single most destructive habit in that table is polishing. A vent that has flashed, or that has a mark on the parting line next to it, invites a toolmaker to run a stone over the area to clean it up, and that both deepens and rounds the land. The tool then flashes instead of venting, and the flash is stoned off in turn. Two or three cycles of that and the vent is gone, and nobody can say when it went.
Which is why the last two items on the checklist earlier in this page are administrative rather than technical. A vent has to be measured when the tool is built, the measurement has to be written on the tool card, and the cleaning has to be on a schedule with the same seriousness as the cooling circuits. A tool with a documented 0.02 mm land and a quarterly clean is a better tool than an identical one with no record, even if the second one was cut more carefully.
Five symptoms and what they tell you
Venting problems and process problems present with the same list of defects, which is why the diagnosis is so often wrong and why so much production time is spent adjusting parameters that were never the cause. The distinguishing feature is repeatability: a venting fault is fixed in space, and a process fault moves.
| Symptom | Where it appears | Venting cause | How to tell it from a process fault |
|---|---|---|---|
| Burn marks, dark streaks | Always the same place, usually the last to fill | Gas cannot escape fast enough | A process fault moves when you change speed; a vent fault stays exactly where it is |
| Weak weld line | At the meeting point of two fronts | Gas trapped in the included angle | Break the part: a venting failure fractures with a dark, porous surface, not a clean one |
| Short shot with a rounded end | The far end of the flow path | Compressed gas is blocking the last few millimetres | A capacity short shot has a torn, ragged edge; a gas-blocked one is smooth and rounded |
| Silver streaks and splay | Often near the gate rather than at the end | Moisture or volatiles, only partly a venting issue | Drying the resin fixes it; cutting a bigger vent does not |
| Local gloss difference | The area around the vent, on the last part to fill | Gas leaves slowly enough that the surface cools at a different rate | Polishing the tool does not fix it, and it returns in the same place on the next run |
The one to internalise is the first row. If a defect appears in exactly the same place on every part over an entire shift, at every speed, at every hold pressure, it is not a process problem, because a process problem would have responded to at least one of those changes. Fixed position means fixed geometry, and in a cavity the fixed geometry that controls gas is the vent.
The fifteen-minute test
The corollary is a test that costs fifteen minutes. Take the tool apart, clean the vents properly with solvent and a copper shim, and run it again on the same settings. If the defect is reduced or gone, the vent was the problem, and the only remaining question is why it closed. If nothing changes, look at the vent locations against a short-shot trial before touching the process again.
Ten checks before the tool is signed off
The list below is written to be used at the design review and again at the trial, because most of the items can only be confirmed once there are parts to look at. Venting is also the cheapest thing on a tool to get right and one of the most expensive to discover late, for reasons that have nothing to do with the cost of the grooves themselves: the whole mold cost structure is built around a tool that runs first time, and a burn mark found at T1 usually moves the schedule as well as the budget.
| # | Check | What “wrong” looks like |
|---|---|---|
| 1 | Locations taken from a short-shot trial, not from the drawing | Vents placed at geometric corners where the flow never arrives last |
| 2 | Depth matched to the resin grade, with a tolerance the shop can actually hold | A single depth for every tool, or a tolerance of plus or minus 0.05 mm on a 0.02 mm feature |
| 3 | Land 1–3 mm long, then relieved to 0.3 mm or more | A shallow groove running all the way to the tool edge, which restricts flow for its whole length |
| 4 | Wide rather than deep | A narrow deep slot, which flashes and then blocks itself with the flash |
| 5 | A vent at every weld line | Weld lines marked on the part drawing and left unvented on the tool |
| 6 | Ejector pins able to act as vents, with flats ground where the pin is long | Reliance on a 150 mm pin as the only vent, which leaks plastic once it wears |
| 7 | Slides vented on the face that closes against the cavity | A slide with no vent, giving a burn mark at the edge of the slide every shot |
| 8 | Runner ends vented | Gas compressed in the runner, showing up as a burn at the gate |
| 9 | A written cleaning interval, and an instruction never to polish the land | No maintenance instruction at all, which is the same as planning for the defect |
| 10 | Measured depths recorded against the tool serial number | No record, so the next repair has no baseline and no one knows what changed |
The two items that carry the most weight are the first and the ninth. Locations taken from a short-shot trial rather than from the drawing will be right the first time, and a written cleaning interval is the only thing that keeps a vent working after the tool has left the shop that built it. Everything else on the list is a detail that follows from those two being taken seriously.
It is also worth saying what venting is not. It is not a substitute for a balanced runner, because a cavity that fills last will always be the one that burns; it is not a substitute for drying resin, because moisture produces gas faster than any vent can remove it; and it is not a way to fix a fill that is genuinely beyond the machine, because a short shot caused by insufficient injection capacity produces exactly the same rounded edge as one caused by trapped gas. The related subjects are covered elsewhere on this site: where gas collects follows from how many cavities the runner has to feed, how much gas the melt carries in follows from the process itself, and the tool’s ability to hold the vent depth over its life follows from the steel it is cut into. What a trial should prove, including whether the vents are working, is set out in mold trial approval, and the cleaning that keeps them open belongs with mold maintenance.
Frequently asked questions
How deep should a mold vent be?
Between 0.01 and 0.04 mm depending on the resin, measured at the land. Polyamides and POM need 0.01 to 0.015 mm because they are thin enough when hot to enter any larger gap; polypropylene and polyethylene can take 0.02 to 0.03 mm; polycarbonate needs 0.03 to 0.04 mm because the melt is stiff and will not flash easily. Whatever the figure, the tolerance on it has to be achievable: plus or minus 0.005 mm on a 0.02 mm land is realistic with grinding, and a nominal depth with a loose tolerance is not a specification at all.
Do I need vents on both halves of the tool?
Almost always, yes, because the gas arrives wherever the flow front last touches steel and that can be on either half. On a simple part that fills from one edge, the gas collects on the parting surface at the far end, which may be on the cavity side or the core side depending on the geometry. The safe default is a land around the full parting line on both halves, at the same depth, which costs very little at the machining stage.
Can vents be added to an existing tool that already burns?
Usually yes, and it is often a small job: the land can be cut with a grinder or EDM in a few hours, sometimes without removing the cavity block. Two conditions apply. The location has to come from a short-shot trial rather than from an assumption, otherwise the new vent misses the gas. And the depth has to be measured afterwards, because a hand-cut vent is the one most likely to be too deep and to start flashing. Where the burn is at the base of a deep boss or the bottom of a rib, there may be no accessible surface and a porous insert or a vented ejector pin becomes the practical answer.
Why does my part burn in only one cavity of a multi-cavity tool?
Because that cavity is filling last, and the gas from the whole shot is arriving there. The cause is runner imbalance rather than the vent in that cavity, and adding vents on the slow cavity will reduce the symptom without fixing the imbalance. The test is to weigh parts from each cavity: if the spread between cavities is larger than the spread within one cavity over thirty shots, the runner is the problem. Vent the last-filling cavity as well, because it will still be the last one even after the runner is corrected, but fix the balance first.
Is vacuum venting worth the extra cost and maintenance?
It is worth it where the geometry or the part specification leaves no alternative: deep narrow cavities, large flat parts with long flow paths, optical surfaces, and sub-second fills. For an ordinary part at ordinary wall thickness, a properly cut set of lands will do the job for a fraction of the complexity. The test is whether a short-shot trial shows gas blocking the fill before the cavity is full. If it does not, evacuation is solving a problem the tool does not have, and it adds a seal, a pump and an interlock that all have to be maintained.
Do vents need maintenance during production?
Yes, and this is the most commonly skipped item in tool handover. Vents collect release agent, oil mist, a thin film of plastic and the deposit left by additives, and the effective depth falls long before anything looks wrong. A practical interval is every few hundred thousand shots for an unfilled commodity resin, and much more often for glass-filled, flame-retardant or PVC grades. Cleaning means solvent and an air blast, plus a copper or brass shim to lift any plastic film; it never means polishing, which removes the land geometry permanently.
Can I rely on ejector pin clearance to vent the tool?
You can rely on it as a contribution, not as the design. Pin clearance does pass gas, and many tools work on that basis, but the clearance is uncontrolled, it changes as the pin and hole wear, and the same gap that vents a new tool will leak melt once it has opened up. Where a pin sits in a deep boss and is the only route out, grind two or three flats along the side of the pin to a depth of 0.02 to 0.03 mm. That converts an accident into a feature.
Will a bigger vent stop the part from flashing?
The opposite: a deeper vent is one of the more common causes of flash along a parting line. Flash is controlled by the gap the melt can enter, which is the vent depth, while venting performance is controlled by the area, which is depth multiplied by width. So the way to improve venting without inviting flash is to make the land wider, not deeper, and to add more lands at the places gas actually collects. If a tool is flashing at the vent, the land is too deep or has been stoned, and the fix is to re-cut it to depth rather than to chase the clamp force.
