Part Weight

Getting Part Weight Right Before You Cut Steel

Part weight sets resin cost, shot size, and press selection before the mold exists. Here is how to estimate it within 2 percent and manage it as a controlled number.

Part weight is the first number on every molding quote, and it compounds. Resin runs 50 to 70 percent of a molded part's cost, so a 1 gram error on a part running 1 million units a year is 1,000 kilograms of resin, about $2,400 at $2.40 per kilogram, straight out of margin or straight into a lost quote. Get the weight wrong high and you lose the job to a sharper estimator. Get it wrong low and you eat the difference for the life of the program, often 3 to 7 years. Weight also drives press selection, cycle time, and shot size, so one bad gram ripples through the whole cost model.

The math is volume times density. Pull the solid volume from CAD, in cubic centimeters, and multiply by the resin's specific gravity. A housing at 24.5 cm3 in ABS at 1.04 g/cm3 estimates at 25.5 g. The same geometry in polypropylene at 0.905 lands at 22.2 g, and in 30 percent glass-filled nylon at 1.36 it jumps to 33.3 g. The Plastic Part Weight calculator runs this in seconds, but the discipline is in the inputs: use the exact grade's datasheet density, not the family average, because polycarbonate grades alone span 1.15 to 1.22.

CAD volume is not molded volume. Expect molded parts to land 2 to 5 percent off the raw estimate. Pack and hold pressure adds 1 to 3 percent as the machine compresses melt into the cavity. Sink and voids in sections over 4 mm pull weight down. Shrinkage is already accounted for in the steel, so do not double-count it. My rule: quote at CAD volume times density plus 3 percent for parts with walls under 3 mm, plus 1 percent for thicker parts, then correct to the actual scale weight at first shots and update the router the same day.

Weight sets shot size, and shot size sets which press can run the tool. Total shot equals part weight times cavitation plus runner: a 25.5 g part in 4 cavities with a 14 g runner is a 116 g shot. Keep that between 20 and 80 percent of barrel capacity; below 20 percent, residence time cooks the resin, and above 80 percent you starve the cushion. A 116 g shot wants roughly a 150 to 500 g barrel. Undersize the press and you get short shots; oversize it and residence time beyond 6 to 8 minutes degrades ABS and polycarbonate visibly.

The levers that pull weight out are design levers, and they pay every cycle. Coring out a solid boss can cut 10 to 20 percent of part mass. Taking nominal wall from 3.0 mm to 2.5 mm removes about 15 percent of volume and usually shortens cooling 20 to 25 percent as a bonus, since cooling time scales with wall thickness squared. Chemical foaming agents shave 5 to 10 percent on non-cosmetic parts. On a million-unit program, every gram removed at $2.40 per kilogram resin is $2,400 a year, plus the cycle time gift that comes with thinner walls.

The common failures are all input failures. Quoting off a similar part's weight instead of this part's CAD volume misses by 10 to 20 percent. Using the generic density for a filled grade when the customer specified 30 percent glass understates weight by 20 percent or more. Forgetting that the runner is bought resin too, which on small parts can equal the part weight. And trusting a datasheet's typical density on a regrind-blended job, where 25 percent regrind shifts effective density and weight enough to move a tight cost model by 1 to 2 percent.

Run weight as a controlled number. Every shift, weigh 5 parts per cavity on a 0.01 g scale and log the result against the standard; drift over 0.5 percent means pack, cushion, or check ring trouble before it becomes scrap. Weekly, reconcile resin issued against theoretical consumption, parts times standard weight plus runner; a gap over 2 percent is purge, startup scrap, or a wrong standard. Monthly, re-verify standards on the top 20 tools by resin spend and push corrections back into quoting so the next estimate starts from actuals.

World class looks boring. First-shot weights land within 2 percent of the pre-steel estimate because the estimating loop is closed. Part weight variation runs under 0.3 percent shift to shift. Material yield variance sits inside 1.5 percent monthly, and the quote database carries actual weights for every active tool. Shops at that level win the tight quotes without losing money on them, because their number is real, and they spot a worn check ring on the weight chart two weeks before quality sees a short shot.

Published 2026-07-02.