If you’ve ever worked in injection molding, you know that even the tiniest miscalculation can turn a perfect mold into a pile of scrap. As someone who’s run an injection molding supplier for 12 years, I’ve seen it all: parts that are too big to fit into an assembly, parts that warp so bad they get sent back, and even a customer who lost a $50k contract because their gears were 0.2mm off spec. Most of those issues trace back to one thing you have to get right right away: shrinkage rate. Today I’m breaking down exactly how to calculate it, why it’s not just a number, and how we make sure it works for every job that leaves our shop. Injection Molding
First, let’s get one thing straight: shrinkage in injection molding isn’t a flaw—it’s a predictable part of the process. When molten plastic gets injected into a cool mold, it cools, hardens, and pulls away from the mold walls as it solidifies. That’s where the shrink happens. Every plastic material has its own baseline shrinkage rate, but that number almost never works straight off the shelf. I learned that early in my career when I took a job making simple plastic containers for a local bakery. I used the generic 2% shrinkage listed on the polypropylene (PP) resin data sheet, ran the first 1000 parts, and none fit the customer’s lid mold. Turned out, their part was thin-walled, cooled fast, and shrank a little less than the thick, dense test samples the resin maker tested. That mistake cost me a week of rework and a few late nights, so I’ve never trusted generic numbers since.
So how do you actually calculate shrinkage rate for your specific part and mold? It starts with the two measurements you need, and these have to be as accurate as possible—no rough guesses with a tape measure. First, the mold cavity dimension. That’s the number you get from the final, properly calibrated mold tool, the one you used to make the master mold that sits in your press. Second, the molded part dimension, the one you measure after it’s fully cooled and removed from the mold, no less than 24 hours after ejection. Wait—24 hours? That’s not a random rule. Plastic parts keep shrinking after they come out of the mold, especially amorphous plastics like polystyrene (PS) or acrylic (ABS). Crystalline plastics like PP, polyethylene (PE), or nylon (PA) shrink more, and they finish shrinking in 12 hours, but I always wait 24 to be safe, even if it adds a day to our timeline. It’s better to wait than to have a part that’s too small a week later.
The basic formula for linear shrinkage rate is simple, but I’ve messed up the order of subtraction more times than I want to admit. Let me write it like we do it in our shop, no fancy calculator tricks: Shrinkage Rate = [(Mold Cavity Dimension – Final Part Dimension) / Mold Cavity Dimension] × 100. Let me use an example to make this real. Suppose I’m making a 100mm long PP gear, and the mold cavity for that gear is exactly 102mm long. I run the part, cool it properly, wait 24 hours, and measure the final part—it’s 99.9mm. Plug those numbers in: (102 – 99.9) / 102 = 2.1 / 102 = 0.02058. Multiply by 100, and my actual shrinkage rate for this exact part is ~2.06%. That’s almost spot on with the resin’s baseline, but if the part was a thick, solid block instead of a thin gear, shrinkage would probably be 1.8% instead. The formula doesn’t change, but the inputs sure do.
Wait a second—why do different parts made from the same plastic have different shrinkage? That’s the million-dollar question, and it’s why I don’t let new operators run jobs without me checking their math. Let’s break down the factors that throw shrinkage off, because ignoring these is how scrap piles up. First, part wall thickness. Thick sections take longer to cool, so the plastic molecules have more time to pack tightly as they shrink. Thin walls cool fast, so there’s less time for packing, meaning less shrinkage. I learned that when I made a custom tray for a medical client: the thin edges shrank 1.2%, but the thick handle shrank 1.8%, so I had to adjust the mold to account for both. Second, mold temperature. If your mold is too cold, the plastic solidifies too fast, so it doesn’t have time to pack—less shrinkage. If the mold is too hot, the plastic stays molten longer, packs more, and shrinks more as it cools. Third, injection pressure and hold time. Hold pressure is the extra pressure you apply after injecting the plastic to push more material into the cavity, filling gaps and compacting the plastic. More hold pressure means less shrinkage, because you’re essentially forcing the material to stay close to the mold walls. Hold time is how long you keep that pressure on—too short, and plastic can’t stay packed, so shrinkage goes up. Fourth, plastic type. Crystalline plastics (PP, PE, PA) have a tight molecular structure, so they shrink way more than amorphous plastics (ABS, PC, PS), which have loose, random structures. A generic PP has a shrink rate between 1.5% and 2.5%, while ABS is usually between 0.4% and 0.7%—that’s a big difference.
Now, here’s the part that separates a good injection molding supplier from a great one: you can’t just calculate shrinkage and adjust the mold once. You have to validate it. I’ve had customers give me a design, a resin, and say “just give me the mold adjusted for 2% shrink.” That’s a setup for failure. Our process is a four-step validation that we’ve refined over a decade. First, we run a small test mold (often a 4-cavity mold) with the baseline shrink rate from the resin’s data sheet. We make 10 test parts over a few press runs, measure each one with a digital caliper accurate to 0.01mm, and average the measurements. Second, we calculate the actual shrink rate from those parts, not the baseline. Third, we adjust the mold cavity dimensions based on that actual shrink rate. The formula for adjusting the mold is: Mold Cavity Dimension = Part Dimension / (1 – (Shrinkage Rate / 100)). Let’s use that same gear example: our part needs to be 99.9mm, actual shrink rate is 2.06%, so mold dimension is 99.9 / (1 – 0.0206) = ~102mm, which matches the example. If we’d used the generic 2% shrink rate, mold would be 99.9 / 0.98 = ~101.94mm, which would make the part 99.84mm, a tiny difference but enough for a tight-fitting gear. Fourth, we run another trial, measure again, and lock in the mold once the parts are within the customer’s spec. We never skip the trial runs—even for repeat customers, because resin batches change a little, press temperatures shift, and parts can have subtle geometry changes.
I get questions all the time from small business owners who do their own injection molding and say “why can’t I just use the generic shrink rate?” Let’s be honest, for simple, low-tolerance parts that don’t need to fit with anything else, a generic number works 80% of the time. But if you’re making parts for medical devices, automotive components, or consumer electronics that have to line up with other parts? Generic shrinkage is a recipe for waste. I had a customer last year who came to us after 3 months of trying to make their own phone case clips, using generic PP shrinkage. They had 70% scrap, because each clip was either too big to snap into place or too small to stay. When we ran our trial, calculated the actual shrink rate, and adjusted the mold, they had zero scrap on the first production run. That’s the value of getting shrinkage right—it’s not just math, it’s saving you time and money.
Another common mistake I see is measuring parts too soon. Some operators measure parts right after ejection, when they’re still warm, and think that’s the final dimension. Plastic is flexible when it’s warm, so a part that’s 100mm right out of the mold can shrink another 0.5mm by the next day. I can’t tell you how many times I’ve caught that: an operator brings me a part, says “it’s 100mm, perfect,” I tell them to put it on the shelf and measure it tomorrow. 9 times out of 10, it’s 99.5mm, so I know we have to adjust the mold. We also account for post-mold shrinkage, which is the shrinkage that happens after the part has cooled fully. For crystalline plastics, post-mold shrinkage can be up to 0.2%, so we add that into our shrinkage calculation. It’s a small number, but for parts with tight tolerances (under 0.1mm), that difference is huge.
Now, let’s talk about tolerances, because shrinkage doesn’t exist in a vacuum. If a part has a tolerance of ±0.1mm, our shrinkage calculation has to hit that within 0.05mm. If it’s a low-tolerance part, like a toy that doesn’t need to fit anything, we can work with ±0.3mm. Our process is to first get the customer’s tolerance spec, then calculate shrinkage to fit that spec. For example, if a customer needs a 50mm long PE part with a tolerance of ±0.1mm, and our trial parts came in at 48.9mm with a shrink rate of 2.2%, we adjust the mold to make it 50 / (1 – 0.022) = ~51.12mm. That way, when the part shrinks, it will be exactly 50mm, within the tolerance.
I also want to mention that different axes of a part can have different shrinkage rates. Wait—what? You thought shrinkage was the same across all sides? For parts with long, thin geometries, like ribs or gears, the flow of the plastic during injection makes the molecules align in the direction of flow. That means shrinkage along the flow direction (longitudinal) is different than shrinkage across the flow direction (transverse). For example, a long thin rod of PP might have 2.0% shrink along its length and 1.8% shrink across its width. That’s why we always measure both directions when we run trials, especially for parts with large aspect ratios. We adjust the mold cavity for each direction separately, which is a detail most small shops miss.
Over the years, we’ve built a library of shrinkage rates for common materials and part geometries, but we never rely on it blindly. For every new job, we run at least 10 trial parts, measure each one, and recalculate shrinkage. We also use calibrated digital micrometers and coordinate measuring machines (CMMs) for high-tolerance parts, because a caliper can’t measure to 0.001mm, which is what we need for parts that go into automotive engines or medical devices.
At the end of the day, getting shrinkage right is all about attention to detail. It’s not just plugging numbers into a formula—it’s understanding how your specific material, part design, mold, and press all interact. We’ve been refining this process since we started our injection molding supplier, and that’s why we’ve built a customer base that comes back for every new project, from small prototype runs to full-scale production of 1 million parts a year.
If you’re working on an injection molding project, whether it’s a prototype, a custom part, or a full production run, don’t leave shrinkage to guesswork. Let our team walk you through the process, from material selection to mold calibration and final part validation. We don’t cut corners on calculations, we don’t skip trial runs, and we make sure every part that leaves our shop meets your exact specs. Reach out to us to discuss your project, and we’ll help you avoid the scrap and delays that come with bad shrinkage calculation.
References
SLS Printing Rosato, D. V., Rosato, M. G., & Rosato, D. V. (2000). Injection Molding Handbook. Kluwer Academic Publishers.
Harper, C. A. (Ed.). (2002). Modern Plastics Handbook. McGraw-Hill.
Kazmer, D. O. (2007). Injection Mold Design Engineering. Hanser Gardner Publications.
Dongguan Xinchuan Precision Manufacturing Co., Ltd.
As one of the most professional injection molding manufacturers and suppliers in China, we offer a quick lead time for customized orders. Please feel free to buy precision injection molding made in China here from our factory. Contact us for quotation.
Address: #16 Gangjian Road, Changping Town, Dongguan City, Guangdong, China. 523000
E-mail: FreeQuote@xcprecision.com
WebSite: https://www.xcprecisions.com/
