Scientific Molding: The Five Studies and Why the Order Matters

Why five tests and not one

Scientific molding is often described as a philosophy. In practice it is a fixed sequence of five studies, run in a fixed order, each of which answers one question and depends on the answer to the one before it. Skip a step and the later numbers are meaningless.

The sequence exists because injection molding variables interact. You cannot find the gate freeze time before the cavity is balanced, because an unbalanced cavity never reaches a stable weight. You cannot balance cavities before you know the pressure drop, because the imbalance may be a fill-rate effect rather than a tool defect.

Table 1 — The five studies and their order. Each depends on the previous one being complete, which is why the sequence is not negotiable.

Study 1: the rheology curve

The rheology study fills the part at a series of different fill times — typically six to ten — holding everything else constant, and records the pressure required at transfer for each. The result is a U-shaped curve: too fast and shear heating thins the melt, too slow and the melt cools and thickens.

Table 2 — A typical rheology curve for an ABS part. The flat bottom is the operating window; the process should sit in the middle of it, not at the edge.

The point of the study is not to find the lowest pressure. It is to find the flat portion of the curve, because a process sitting on a slope has no tolerance for variation. A 5% change in fill time on the steep part of the curve moves pressure by 40 bar; on the flat part it moves it by 5. Only one of those is a process you can leave running.

Practical range: most studies bracket 0.3 s to 3 s, or roughly the natural fill time of the part ±60%. Below 0.3 s the machine cannot control the velocity profile accurately; above 3 s the melt front has cooled enough that the numbers stop meaning anything.

Study 2: the pressure drop study

With the fill time fixed from study 1, the pressure drop study measures how much of the available pressure is used getting the melt from the nozzle to the end of the cavity. It is done by filling to a series of short-shot positions and recording the pressure at each.

Table 3 — Where injection pressure goes. If the nozzle and runner together take more than half, the tool is fighting the machine and no process setting will fix it.

The study also establishes the pressure at transfer, which becomes the upper limit for the pack pressure in study 4. Pack pressure is normally set at 50–80% of the pressure at transfer; setting it above that just holds the screw against a frozen gate and wastes energy.

The most useful single number from this study is the pressure at the end of fill. If it is above 80% of the machine's rated injection pressure, the process has no headroom and will drift as the melt viscosity changes through the shift. That is a tool or machine problem, and it is better discovered during qualification than in production.

Study 3: the cavity balance study

For a multi-cavity tool, the balance study determines whether all cavities fill at the same time and reach the same pressure. It is done with a short-shot series: fill to 90%, 95%, 98% and 100%, and see which cavities are full at each stage.

Table 4 — Reading a cavity balance study. An imbalance that changes with fill rate is a rheological problem, not a tool geometry problem, and correcting the steel will not fix it.

The acceptance criterion is usually a weight spread of under 2% across cavities, measured on parts from a single shot after the process has stabilised. Some industries accept 1%; very few accept more than 3%. Beyond that, the parts are effectively coming from different processes and downstream assembly will show it.

Study 4: the gate freeze study

Gate freeze is the moment the gate solidifies and no more material can enter the cavity. After that point, additional hold time does nothing except keep the screw forward. The study finds that moment by running a series of shots at increasing hold time and weighing the parts.

Table 5 — A weight plateau. The plateau is the answer: hold time should be set at the first point where the weight stops rising, plus a small safety margin.

Two rules follow. First, hold time set below the plateau leaves the part under-packed and the weight varies with every small change in melt viscosity. Second, hold time set well above the plateau adds cycle time for nothing — and on a 25-second cycle, 3 seconds of wasted hold is 12% of capacity.

The plateau method needs a balance readable to 0.01 g and parts weighed after they have cooled to room temperature. Weighing warm parts gives a reading that drifts with the cooling, which flattens the plateau and makes it look as though gate freeze happened earlier than it did.

Study 5: the cooling study

The cooling study determines how long the part must stay in the tool to be rigid enough to eject without distortion. It is done by reducing cooling time in steps and inspecting the parts for ejection marks, warpage and dimensional change.

Table 6 — Cooling study results. The correct setting is the shortest time that gives stable dimensions, not the time that gives the best-looking part off the tool.

The cooling study also gives the heat load, which is what the machine's cooling capacity has to remove. Heat load is part weight × enthalpy change ÷ cycle time, and comparing it with the chiller capacity explains a class of problems that look like process instability but are actually a warm mould.

What the five give you together

Run in sequence, the five studies produce a process window rather than a process setting. That distinction is the whole point. A setting works until something changes; a window tells you how far you can move before quality is affected, and therefore which variations you can absorb and which you cannot.

The window also makes the process portable. A documented window — fill time range, pressure at transfer, pack pressure range, hold time, cooling time — can be transferred to another machine of the same tonnage and reproduced. A recipe of setpoints cannot, because the same numbers on a different machine produce a different cavity pressure.

Doing it without a consultant

All five studies need a machine, a balance, a stopwatch and a way to read pressure at transfer. They do not need a cavity pressure sensor, although one makes studies 2 and 4 faster and more precise. A plant that runs the sequence once per tool and records the results has most of the benefit of a formal scientific molding programme.

The time cost is real: half a day to a day per tool for the full sequence, plus the material consumed. Against that, the first production run after qualification typically saves more than that in reduced debugging, and the tool's process window is documented once rather than rediscovered every time the job returns to the press.