Cavity Pressure Monitoring — How Do You Use In-Cavity Sensors to Validate the Process, Catch Defects and Hold Part Quality Across Every Cavity?
Last updated: 2026-09-03 · Reading time: about 6 min
Cavity pressure monitoring puts a piezoelectric sensor in the cavity wall and reads the pressure the melt actually sees — not the hydraulic pressure back at the screw. Every shot’s curve is compared to a known-good reference, and a curve outside the tolerance band flags the part before it reaches the customer.
See also related article.
Key facts
<>
Process parameter — A controllable variable in the injection molding process: melt temperature, injection speed, holding pressure, cooling time, mold temperature. Each parameter affects part quality — deviations cause defects. Parameters are set during T1 and locked for production. See also related guide.
What sensor types are there — and how do you choose the right one?
Four sensor types cover most molds. The choice turns on accuracy, wear and whether you can machine the cavity surface.
| Sensor type | Range | Temp | When to use |
|---|---|---|---|
| Piezoelectric (flush mount) | 0–2,000 bar | 400 °C | Direct at the cavity surface; most accurate (±1%) |
| Piezoelectric (subsurface) | 0–2,000 bar | 400 °C | 1–2 mm below the surface; protected from wear on abrasive resins |
| Strain gauge (ejector pin) | 0–1,000 bar | 200 °C | Least expensive; indirect force on the pin (±5%) |
| Piezoelectric (nozzle) | 0–2,000 bar | 400 °C | At the nozzle before the runner; no mold modification |
Flush-mount is the reference. The sensor face must sit within 0.01 mm of the cavity surface — proud and it marks the part, recessed and the reading lags. For glass- or mineral-filled resins that would wear a flush face, drop to a subsurface sensor and let the steel transmit the pressure.
How do you read the cavity pressure curve — and what does each phase tell you?
One shot traces four phases. Each has a signature, and each deviation names a defect.
- Fill (0 → peak). Pressure climbs as the cavity fills. A linear rise is a balanced fill; a sudden spike is a flow restriction — a weld line, a thickness change or a blocked vent. A fill time longer than the reference means melt too cold or speed too slow.
- Peak and compression. Peak pressure (typically 500–1,500 bar) must stay under the mold design limit (1,200–1,800 bar) or the parting line opens and the mold flashes. The drop after peak marks the screw switching from injection to holding.
- Packing (plateau). Packing pressure runs 30–70% of peak and holds as the melt cools and shrinks. If it collapses early, the gate has frozen off — the time from injection start to that drop is your gate freeze time.
- Cooling (→ 0). After gate freeze the pressure falls as the part shrinks away from the wall. A faster fall means more effective cooling; pressure reaches 0 when the part is rigid enough to eject. Cooling is 50–70% of the cycle.
“We put a cavity pressure curve next to every process we qualify. It shows the fill balance across all cavities and the exact gate freeze time — so the settings we hand you are the settings that actually made a good part, not a guess that looked right on the machine.”
— Project Engineer, aCatMould
Frequently asked questions
How is it used for quality control — and what defects does it catch?
Every curve is compared to a known-good reference.
A lower peak is a short shot; a higher peak is flash and over-packing; a packing plateau that drops too fast is a sink mark; a spike during fill is a weld line; a shifted pressure at gate freeze is dimensional drift.
Set the window at ±5% of the reference and the system accepts or rejects each shot automatically, diverting bad parts to the reject bin. That typically cuts scrap 30–60% because defects are caught before they ship.
How many sensors do you need — and where do they go?
Single cavity: 1–2 sensors, one near the gate and one at the far end of the flow path. Multi-cavity: one per cavity minimum to prove fill balance, two per cavity for a fuller picture.
Near the gate reads peak and packing; the far end reads fill balance and venting. Keep sensors away from thickness changes, weld lines and vents, and mount them in a replaceable cavity insert — $500–2,000 per sensor plus $200–500 for the amplifier.
What does a system cost — and when is it justified?
Add it up: sensors $500–2,000 each, amplifiers $200–500 per channel, a data-acquisition unit $5,000–20,000 and software $2,000–10,000. A 4-cavity, 4-sensor system lands at $10,000–30,000.
It earns out above 1M shots/year — a 30–60% scrap cut saves $0.01–0.05 per part — or on medical, automotive and aerospace parts held below 10 ppm. Payback is usually 1–3 years.
Key takeaways
- Cavity pressure reads the melt directly at the part — the fill, packing and cooling behavior hydraulic pressure cannot see.
- Pick the sensor by accuracy and wear: flush-mount (±1%) is the reference, subsurface survives abrasive resins, ejector-pin is the least expensive.
- Compare every shot to a reference curve, reject outside ±5%, and justify the $10,000–30,000 system above 1M shots/year or on critical parts.
Cavity pressure monitoring is the most direct way to validate a process — the curve shows fill, packing and cooling in real time and catches defects you cannot see, from internal voids to dimensional drift. If your program needs process validation, or your scrap runs above 2%, it is the next step.
Sources
Kistler — cavity pressure sensing fundamentals
Priamus — cavity pressure and process monitoring technology
RJG Inc — scientific (decoupled) molding and process control
Plastics Technology — cavity pressure monitoring practice
Society of Plastics Engineers — injection molding process control
Sources: Plastics Technology · Plastics Technology Online · Society of Plastics Engineers (SPE) · ThomasNet