Molding Mistakes

Injection Molding Mistakes That Wreck Your Numbers and How to Catch Them

The specific errors that make injection molding estimates and processes fail, each with the symptom to watch for and a numeric fix.

The most expensive mistake in injection molding is quoting part weight from CAD volume without accounting for resin density and packing. Symptom: your Plastic Part Weight number reads 12.4 g but the press delivers 13.1 g parts, a 5.6 percent miss that silently eats margin across a 500,000 piece run. Root cause is using a generic 1.0 g/cm3 density instead of the resin sheet value, and forgetting the part gets packed 2 to 4 percent above nominal. Fix: pull the actual density (ABS 1.04, PC 1.20, glass-filled nylon 1.35 to 1.55 g/cm3) and weigh 30 shots to calibrate. A 0.5 g error at 8 cents per gram is 0.04 dollars per part.

Sizing shots without the runner is a classic underestimate. Symptom: the barrel runs dry mid-injection or the machine flags short shots even though your Shot Size math looked fine. Root cause: engineers plug in cavity weight and skip the Runner Weight, which on a cold runner eight-cavity tool can be 25 to 40 percent of total shot mass. If eight cavities hold 6 g each, that is 48 g, but a 16 g runner pushes the true shot to 64 g. Fix: always compute shot as (cavities times part weight) plus runner, then add 10 to 15 percent barrel cushion so you are not injecting from an empty screw.

Losing cushion is a process failure that shows as dimensional drift, not a math error. Symptom: parts creep undersize and sink marks appear intermittently across a shift. Root cause: the screw bottoms out because Cushion Percentage fell below the 5 to 10 percent safe band, often near 2 percent, so the machine cannot hold pack pressure. A screw check-ring wearing out drops cushion 0.1 mm per thousand cycles. Fix: log cushion every hour, hold it between 3 and 8 mm on a 40 mm screw, and pull the tool for a ring change once cushion variation exceeds plus or minus 1.5 mm shot to shot.

Under-specifying clamp tonnage cracks tools and flashes parts. Symptom: flash along the parting line and a press that alarms on clamp force. Root cause: engineers use 2 tons per square inch as a blanket rule when filled or thin-wall resins need 4 to 5. Projected area matters: a 30 square inch part at 4 tons per square inch needs 120 tons, but sizing it at 2 tons gives 60 and the mold blows open. Fix: run Clamp Tonnage with the resin-specific factor (unfilled polyolefins near 2, engineering resins 3 to 4, thin-wall packaging 5 plus) and keep the press at 80 percent of rated tonnage, never 100.

Ignoring regrind blows up both cost and quality assumptions. Symptom: your Resin Cost Per Part looks low but scrap climbs and parts get brittle after a few weeks. Root cause: runners and rejects get reground and blended back without tracking Regrind Percentage, so mechanical properties degrade past the 15 to 25 percent blend limit most engineering resins tolerate. At 30 percent regrind, impact strength on ABS can fall 20 percent. Fix: cap regrind at a documented ratio, weigh virgin against reground on a scale each shift, and if runner mass is 20 percent of shot, that alone sets your minimum regrind stream before any reject.

Cycle time estimates fail when cooling is guessed instead of derived from wall thickness. Symptom: your Injection Molding Cycle Time predicts 22 seconds but the floor runs 31, so quoted capacity is 40 percent optimistic. Root cause: cooling scales with the square of wall thickness, so a 3 mm wall does not cool in the same time as 2 mm; it takes roughly 2.25 times longer. Fix: use the thickest-wall cooling estimate, add mold-open and ejection (typically 3 to 6 seconds), and validate against a stopwatch on 25 cycles. A 9 second miss at 45 dollars per hour machine rate adds 0.11 dollars per part.

Overstating cavitation without checking machine limits wrecks capacity plans. Symptom: Mold Cavitation says 16 cavities fit the platen, but the press cannot fill or clamp them. Root cause: cavity count gets set by platen area alone, ignoring that 16 cavities multiply shot size, tonnage, and plasticizing demand beyond the machine. Sixteen 6 g cavities need 96 g plus runner, and if the barrel maxes at 110 g you have no cushion. Fix: reconcile Mold Cavitation against Shot Size, Clamp Tonnage, and barrel capacity together, and derate to the tightest constraint, usually landing at 8 to 12 cavities, not the platen maximum.

Confusing theoretical and actual parts per hour double-counts output. Symptom: Mold Parts Per Hour reads 1,600 on paper but the shift ships 1,150, a 28 percent gap. Root cause: the calc uses raw cycle time and ideal cavity count with zero downtime, ignoring the 85 to 92 percent uptime real cells hit and the scrap that pulls good parts below gross. Fix: multiply theoretical output by an availability factor (start at 0.88) and a quality factor (0.97 for a stable tool), so 1,600 gross becomes about 1,366, then reconcile weekly against counter data and adjust the factors from measured reality, not hope.

Published 2026-07-01.