Powder Quality

Setting and Holding Powder Refresh Rate Targets

Refresh rate is the dial between powder cost and part properties. Here is how to set the target by evidence, blend to it every build, and catch drift before parts fail.

Refresh rate is the single dial that trades powder cost against part properties, and both directions cost real money. Run too much virgin and you burn cash: on PA12 at 70 dollars per kilogram, every 10 points of unnecessary refresh on a 15-kilogram charge is 105 dollars per build, roughly 21,000 dollars a year at 4 builds a week. Run too little and heat-aged polymer produces orange peel surfaces, porosity, and elongation at break falling from 20 percent to under 10, which shows up as customer rejects worth 5 to 20 times the powder you saved. The refresh number is where finance and metallurgy argue, and it deserves data instead of folklore.

The blend math is simple; the control is not. Refresh rate equals virgin powder mass divided by total blend mass, times 100. Work it: 6 kilograms of virgin blended into 24 kilograms of recycled powder makes a 30-kilogram charge at 20 percent refresh. The Powder Refresh Rate calculator compares your actual blend against target in one step, which matters because operators blending by scoop routinely land 5 to 10 points off. Note the compounding: at 20 percent refresh, a given powder particle can survive 5 or more thermal cycles, so the blend's effective age keeps climbing for the first 10 to 15 builds before stabilizing. Your target must account for that steady state, not the first build.

Anchor targets to known ranges, then validate on your own machine. SLS PA12 typically runs 30 to 50 percent refresh; MJF PA12 is engineered around 20 percent; PA11 tolerates aggressive reuse; TPU commonly needs 25 to 35. Metal powders in laser powder bed are different physics: with sieving and interstitial control, titanium and nickel alloys commonly sustain 10 to 30-plus reuse cycles with refresh driven by losses rather than aging, though oxygen pickup limits apply, such as roughly 0.13 to 0.20 percent oxygen ceilings on Ti-6Al-4V depending on spec. Treat every one of these as a starting hypothesis and confirm with mechanical coupons before you cut a single point of virgin.

Manage the levers around the ratio, not just the ratio. Thermal exposure is the aging engine, so shorter builds, lower chamber soak, and part placement away from repeated hot zones let the same blend survive more cycles. Sieving at 100 to 150 microns for polymers removes agglomerates that seed surface defects. Storage matters: PA12 above 50 percent relative humidity degrades flow within days, so sealed containers with desiccant protect blend quality for free. And melt flow rate testing, at about 30 minutes per sample, gives polymers an aging gauge: many shops set an MFR floor and refresh to hold it rather than blending to a fixed percentage forever.

The failure modes are predictable and expensive. Blending by volume instead of mass misses 5 to 8 points because virgin and recycled powders pack differently. Skipping the blend log means that when elongation drops on a customer part, you cannot say which of 30 builds contaminated the lot, and the conservative response is scrapping the entire 200-kilogram working inventory. Chasing cost by cutting refresh 5 points per quarter without coupon data works right up until it does not, and the failure arrives as a field return. And treating overflow, cake, and part-bed powder as identical ignores that they aged at different temperatures and should be tracked separately.

Run refresh as a per-build control. Every build: weigh virgin and recycled additions, log the computed refresh percentage against target, record powder lot genealogy; ninety seconds of work. Weekly: review refresh compliance and any builds beyond plus or minus 3 points of target. Monthly: print tensile coupons from the working blend and test elongation, and run MFR on polymers; plot both on a control chart with limits set from your qualification study. Quarterly: revisit the target itself using the coupon trend, powder pricing, and reject data. A target that was right at 70 dollars per kilogram may be wrong at 90.

World-class refresh management hits four marks: actual blend within 2 points of target on 95 percent of builds, mechanical properties inside control limits for 12 straight months, powder-related rejects under 1 percent, and virgin consumption 20 to 40 percent below what an always-safe blend policy would spend. On a two-machine SLS cell consuming 2,000 kilograms a year, evidence-based refresh versus fear-based refresh is commonly worth 15,000 to 30,000 dollars annually with zero property risk, because the coupons prove it every month. The dial only pays when someone owns it, logs it, and defends it with data.

Published 2026-07-02.