Process Development
Scientific molding is process development done in the right order. First you measure how the resin behaves. Then you set fill speed from a viscosity curve and weigh parts to find gate seal. Lock the window that results. A scientific process survives resin lot changes and shift changes because it controls the plastic, not the machine's buttons.
Scientific Molding Starts by Separating Fill from Pack
Scientific molding, sometimes called decoupled molding, develops fill first, then pack and hold, as separate stages.
Most shops develop a process by adjusting everything at once until the parts look right.
The separation matters because fill speed and pack pressure push the plastic in opposite directions. Fill too fast and shear heating burns the resin. Pack too hard and the gate freezes stress into the part.
Scientific molding switches from filling to holding on screw position or cavity pressure — never on a timer. Time-based transfer drifts with the machine and the operator. Position-based transfer follows the plastic itself.
Ask your supplier how the press switches from fill to hold. If the answer is "five seconds," that process is not scientific. Our injection molding process guide walks through the cycle stages this separation relies on.
The Viscosity Curve Finds a Fill Speed That Forgives
Buyers almost never ask what fill speed a supplier runs. That number tells you more about the process than the quote does.
The first study in scientific molding is the viscosity curve. You mold parts across a range of injection speeds and record the peak injection pressure at each speed. Plotted, the curve levels off across a range. Inside that flat zone, small speed changes barely move cavity pressure.
Production speed belongs in the middle of the flat zone, not at its edge. Running at the edge leaves no room when the resin lot changes. Inside the flat zone of the viscosity curve, small speed changes barely move cavity pressure. A scientific process absorbs normal operator and machine variation instead of passing it into the parts.
The curve belongs to one resin lot and one regrind ratio, and it shifts when either changes. Record fill time and part weight for a 95–99% full part at validation. If a new lot gives a different weight at the same fill time, the lot changed, not the machine.
Gate Seal Tells You Exactly When Holding Stops Working
Hold time gets set by habit. Five seconds, because the last job ran five seconds.
The gate seal study replaces habit with a scale. You mold parts at decreasing hold times and weigh each one. Part weight climbs, then plateaus. The gate has frozen; no more plastic can enter the cavity.
Once the gate freezes, extra hold time adds no part weight and only costs cycle time. Hold too short and thick sections sink. Hold past the seal point and you pay for seconds on every shot, thousands of times a day, for zero gain.
There is a second cost buyers never see. Holding hard after seal packs extra material into the gate region. The extra pack freezes stress into the part's thinnest zone. That stress waits for a cold morning or a dropped assembly, then cracks. A gate seal study draws the line between packed and overpacked.
The Cosmetic Window Is the Real Process Window
A process window is not a feeling. It is the tested range of settings that still makes good parts.
After fill and pack are set, scientific molding probes the edges. Melt temperature, injection speed, pack pressure, and mold temperature each get tested at both extremes. The combinations that still make acceptable parts define the window.
A process with only one workable setting leaves no room for the next shift. A wide window lets a night-shift operator miss a setting by a click and still ship good parts.
The margin also erodes on its own. Resin viscosity can drift roughly 20% over a program's life [VERIFY]. That quiet drift consumes pressure the process was counting on. A wide window today is an average job in two years.
When the window starts producing cosmetic rejects, our flash and weld line fixes separate mold causes from process causes.
What Happens When the Process Drifts at 2 A.M.?
Check the monitoring data before changing anything.
Parts go bad on night shift, and the instinct is to change the process. Scientific shops do the opposite.
They check fill time, peak pressure, melt temperature, and thermolator actuals against setpoints first. A plugged water channel or a worn check ring moves the plastic before any setting changes. Changing the process on top of a mechanical change masks the real cause.
Before changing the process, find what changed. Adjust settings only after machine and mold check out.
I've seen a night shift raise melt temperature across a whole job because one thermocouple read low. That baked degradation into a week of parts. When adjustment is unavoidable, return fill time and peak pressure to their validated values.
Try the Injection Molding Process Simulator
Move the sliders below and watch how melt temperature, injection speed, packing pressure, and cooling time move the part quality score. It is the fastest way to feel what a process window means.
Scientific Molding FAQ
What is scientific molding?
Scientific molding, also called decoupled molding, develops the injection process in measured stages instead of trial and error. A viscosity curve sets fill speed; a gate seal study sets hold time. Fill, pack, and hold are developed separately and documented as plastic variables like fill time and melt temperature. Any trained operator can then repeat the process.
How long does scientific process development take?
For a straightforward part, an experienced process engineer completes the core studies within a single shift at the press. Those studies cover the viscosity curve, gate seal, pressure drop, and a first window check. Complex multi-cavity tools take longer. The time pays back fast. A documented window cuts the trial-and-error loops that usually cost far more press time.
What should a supplier document at process validation?
Ask for a setup card built on plastic variables. It lists fill time, part weight with a photo at 95–99% full, transfer position, and gate seal time. Add hold pressure, shot size in volume, melt temperature from a hot probe, and coolant flow per channel. If the sheet only lists barrel setpoints and speeds, the process cannot transfer to another machine or survive a material lot change.
Can the same scientific process run on a different machine?
Yes, when it was documented in plastic variables rather than machine settings. Fill time, part weight, melt temperature, and cavity pressure describe what the plastic did, and those targets hold on any capable press. The new machine will use different barrel zones and hydraulic settings to reach them. Transfer fails only when the documentation recorded button positions instead of plastic behavior.
How do I know my supplier really runs a scientific process?
Ask for the viscosity curve from process development, the gate seal study with part weights, and the setup card. A shop that runs scientific molding produces them without hesitation because the studies are part of its routine. A shop that tweaks until parts look right will stall, and that hesitation is your answer.
The next time your supplier sends T0 samples, ask for the gate seal weights and the fill-time record alongside them. A shop running a scientific process sends them without blinking. A shop that does not will send an excuse. That tells you exactly what your parts depend on.