Last updated: 2026-09-04 · 5 min read
Mold Trial Press Matching: Why Does the Trial Machine Have to Match Production?
The mold does not have a process — the mold-and-press pair does. Every trial number is a joint product: melt temperature and shear history come from the trial press’s screw, fill profile from its valves, packed state from its clamp. The part is the mold’s answer to that specific machine’s questions. Press matching is making the trial machine speak with production’s voice — and transferability is the only reason to run a trial at all.
Key Facts
<>
Shrinkage — The dimensional reduction from mold cavity to cooled part. Amorphous resins: 0.4–0.7% (uniform). Semi-crystalline: 1.0–2.5% (directional). Mold makers apply shrinkage factor when machining cavity. Actual shrinkage varies with wall thickness, gate location and packing pressure. The related guide covers adjacent topics.
Which Machine Dimensions Actually Shape the Process?
Five dimensions, ranked by how silently they change the part:
| Dimension | Why It Changes the Process | Risk When Mismatched |
|---|---|---|
| Shot-to-barrel ratio | Residence time and melt heat history | Degradation or unstable melt at production |
| Screw geometry and class | Plasticizing quality and rate | A melt condition production cannot repeat |
| Injection dynamics | Speed response and switchover accuracy | Profiles that transfer only on paper |
| Clamp and platen condition | Tonnage margin and parallelism | Flash or shorts appearing only in production |
| Interfaces and signals | Cores, ejector, robot sequences | Trial passes; production line stalls |
The shot-to-barrel ratio is the quiet one with the longest memory: a small shot from a large barrel spends its melt’s youth in long residence, and every extra minute brings oxidation, shear aging and color drift. The screw row is the melt’s biography: diameter, compression ratio and design determine how uniformly pellets become melt — two presses of the same tonnage can deliver melts that differ enough to move a process window visibly.
Injection dynamics decide whether the speed profile means anything: fast-response machines track setpoints closely, sluggish ones round the corners, and switchover accuracy differs by machine class. The clamp row is where money hides: tonnage margin projects differently on machines with different platen condition, and a tool that flashed imperceptibly at trial on a stiff machine can flash steadily in production on a tired one. The interface row is the least glamorous and the most schedule-destroying: core-pull sequences, ejector interlocks and robot handshakes differ across controllers.
How Do You Plan a Trial So Its Data Survives to Production?
Four steps, in order:
- Write the production press into the tool specification. The press is a design input, belonging beside the resin and cavity count: tonnage class, screw diameter and L/D, injection rate, controller generation, interface signals, nozzle radius. With the press named, every downstream decision inherits it: runner and gate sizes are checked against the machine’s real capability; shot-to-barrel ratio is verified before steel is cut. Programs that skip this step trial against whatever was free that week, and their report quietly describes a process that exists on exactly one machine.
- Match the melt-shaping dimensions first, and document the rest. Not all mismatches weigh the same. Priority one: shot-to-barrel ratio and screw class — the residence and plasticizing story — because a melt arriving in different condition invalidates every setting downstream. Priority two: injection dynamics — rate capability and switchover response. Priority three: clamp condition and tonnage margin. Everything else goes in the delta table: each mismatch recorded with its predicted direction of effect. The delta table is the discipline’s core artifact — its absence is how mismatches become mysteries.
- Separate tool verification from process qualification in the trial’s goals. Tool verification asks: do the actions work? — sequences, cores, ejector, leaks, gate freeze; this transfers across machines almost perfectly because the mechanics belong to the tool. Process qualification asks: is the window real? — dimensioned parts at documented settings; this transfers only as far as presses match. Labeling findings by which kind they are keeps the report honest: a T1 on a mismatched press can legitimately declare the tool mechanically sound and refuse to declare the process qualified.
- Close the gap with a production-press verification before release. When the trial ran on anything other than the production press, the plan includes a verification stage on the real machine — and the schedule reserves time for it. The verification re-runs the process-defining checks: window corners re-confirmed, cavity balance re-measured, weight and dimensions re-sampled, delta table predictions checked. Where deltas were benign, the process signs off quickly; where a delta bit, adjustments happen in a planned session with the tool’s history available, not during launch.
Gate vestige — The visible mark left on the part surface where the gate was trimmed. Edge gate: 0.5–2.0mm vestige. Pin-point gate: 0.2–0.5mm. Valve gate: 0.05–0.1mm (flush). Vestige size affects cosmetic appearance and may require secondary trimming. The related guide covers adjacent topics.
The mold does not have a process — the mold-and-press pair does. A trial report describes a joint product, and transferability is the only reason to run a trial at all. Undocumented deltas are the ones that surprise.
Frequently Asked Questions
Can we trial a mold on a smaller press than production uses?
Sometimes — the smaller press must clear three floors: clamp tonnage with real margin, shot capacity in the sane ratio band, and platen geometry that holds the mold square. If those clear, a smaller-press trial can verify tool mechanics, run short-shot studies, and map a preliminary window — all labeled provisional.
What it cannot promise is production behavior: the production press’s larger injection unit, different screw, and different response class will redraw the window’s edges. Do not size gates against the trial press’s fill capability alone, and do not let the trial window’s dimensional results enter the customer file as qualified numbers.
What is the shot-to-barrel ratio rule, and why does it matter?
The shot should occupy a sane fraction of the barrel’s capacity — classically 20 to 80 percent, with the comfortable middle nearer 30 to 70. The ratio is really a residence-time rule wearing a geometry label: melt sits in the barrel between cycles, and the fraction decides how long any molecule waits.
A tiny shot in a big barrel means long residence — oxidative aging, color drift, visible degradation on sensitive resins. The mismatch between trial and production moves this ratio silently: the same mold on a bigger machine drops the ratio and inherits a melt condition the trial never saw. Compute the ratio for every press the mold will run on; when a trial press forces it outside the band, mark melt-dependent findings as provisional.
The moldmaker’s press is all we have for T1 — how do we keep data honest?
Declare the trial press’s identity in the report: machine, screw, controller, and the delta table against the specified production press, filled in before the trial so predictions are made in advance. Direct the T1 at questions it can answer truthfully: mechanical actions, sequences, leaks, basic fill, gate freeze — these are the tool’s own properties and they transfer.
Harvest but quarantine the process data: record everything, but stamp it as press-relative rather than process-qualified. Write the production-press verification into the project plan at this moment — while the delta table is fresh — with budget and dates. A T1 that says “the tool works; the process awaits production verification” is a success, not a shortfall.
Key Takeaways
<>
Separate tool verification (transfers perfectly) from process qualification (transfers only as far as presses match), and close the gap with production-press verification before release.
Press matching is not ceremony — it is what makes trial data transferable. State the production press as part of the tool’s specification, match what can be matched, document what cannot, and verify on the real machine before release.
Undocumented deltas are the ones that surprise, and every one of them costs more per hour at launch than the verification session would have cost in total.
Sources
- Plastics Technology — Scientific Molding and Press Matching
- Society of Plastics Engineers (SPE) — Mold Trial Procedures
- RJG — Decoupled Molding Concepts
- Plastics Today — Process Transfer Between Presses
- ENGEL — Injection Molding Process Control
Sources: Plastics Technology · Manufacturing.net · Society of Plastics Engineers (SPE) · ThomasNet