On the floor, everyone talks about injection speed first, but speed alone means nothing without melt temperature behind it. If the melt is running cold, you crank up speed and you get shear marks, jetting, or flash at the parting line; run it too hot and you fight stringing, sink, and longer cooling. In practice, I set the barrel profile so the melt reaches the cavity at the low end of the material supplier's window, then tune injection speed in stages — fast fill through the runner, slower through the gate, and a controlled profile to 95–98% cavity fill before switching to packing. That switchover point matters more than most people admit; short shots and overpacking usually trace back to a sloppy velocity-to-pressure transfer.
Packing and cooling are where the mold steel really shows its character. Packing pressure should be set as a percentage of injection pressure — typically 60–80% — and held until the gate freezes, which you can confirm with a gate freeze study rather than guessing. Cooling time is governed by wall thickness and the mold's cooling circuit layout: ribbed cores, bubblers, and baffles need to be positioned so the heat comes out evenly, or you get warpage that no amount of packing pressure will fix. I always check mold surface temperature at several points with a contact probe before touching process parameters, because a 5–10°C imbalance across the cavity will throw dimensional control off all shift.
Ejection is the last checkpoint, not an afterthought. Ejector pin marks, drag marks, and part sticking usually mean insufficient cooling at the core or too much packing near the gate. Keep ejection velocity low and stroke consistent, and verify ejector plate return before the next clamp cycle. Get the sequence right — clamp, fill, pack, cool, open, eject — and the process becomes repeatable instead of a fight every shift. For mold sourcing and supplier comparisons, visit MoldSo at www.moldso.com.