DEV Community: flarelab

PrusaSlicer 3.0 Preview: What the Rebuild Actually Changes for Your Printer

flarelab — Fri, 11 Sep 2026 17:30:06 +0000

Fifteen years to the day after the first line of Slic3r was committed, Prusa Research pushed PrusaSlicer 3.0 into public preview — and it is not a normal point release. Josef Přůša calls it the biggest set of changes in the software's history, and for once that is not marketing. The interface was rebuilt from scratch, the profile system was re-architected, and the slicer now accepts plugins.

The headline change for hobby printers is the new project system. Print beds are no longer just copies of one machine's settings. Each bed is an independent part of a single project with its own configuration, so you can combine different printers and profiles inside one file, move models from bed to bed, and slice several beds in parallel. The old nine-bed ceiling is gone, multiple projects can sit open in separate tabs, and the application quietly keeps local backups of whatever you have open. Multi-tool machines like the Prusa XL can finally assign a different nozzle size and print profile to each tool head.

The rest of the rebuild is about getting out of your way. Tools now run along the top of the screen, a collapsible scene browser sits on the left, and the right-hand panel changes depending on whether you have a bed, an object or a tool selected. There is a View Cube for camera control, a light theme at last, and navigation presets that match Tinkercad, Blender, SolidWorks or Fusion so your muscle memory carries over. You can star individual settings as favourites and they appear directly in that contextual panel. Underneath it all sits a sandboxed Lua plugin system: plugins run with no disk, project or network access by default, two calibration-tower generators ship bundled, and a reviewed community marketplace is on the way.

Installing it is low-risk, which is the whole point of a preview. PrusaSlicer 3.x installs alongside 2.x using a separate profile directory, so nothing you already rely on gets touched. Your existing 3MF projects open and have their settings mapped across automatically, though profiles themselves have moved from .ini to .yaml. Two caveats: third-party 3MF files from Bambu Studio or OrcaSlicer import as geometry only, with no settings attached, and the preview is not yet feature-complete against 2.9.6 — some functionality is not due back until 3.1.0. There is also a proper offline mode if you would rather the slicer never phoned home.

Try it on your printer. Download the preview, open a model you have already printed successfully, and slice the same job in both 2.x and 3.x. Compare the time and filament estimates, then skim the G-code preview layer by layer. It is the fastest way to build trust in a new slicer without burning a single gram of filament. Keep 2.9.6 as your production slicer for now, and send anything odd to Prusa's GitHub or community forum, because preview builds only improve when people report. Stocking up on filament to test with? Browse the range at flarelab.com.

Frequently asked questions

Is PrusaSlicer 3.0 safe to install over my current version?

It does not overwrite anything. PrusaSlicer 3.x installs alongside 2.x and uses its own separate profile directory, so your existing setup stays intact. Keep 2.9.6 as your production slicer while the preview matures.

Will my old projects and profiles still open?

Yes. 3MF projects created in PrusaSlicer 2.x load in 3.x and have their settings matched to the new system automatically. Profile files themselves have moved from the .ini format to .yaml.

Can I open Bambu Studio or OrcaSlicer files in it?

Partly. Third-party 3MF files import as geometry only, so the models come through but the print settings do not. You will need to reapply your own profile after importing.

What do the new plugins actually do?

The initial Lua plugins generate parametric objects, and two calibration towers for flow rate and temperature are included by default. Plugins run sandboxed with no disk, project or network access unless granted, and a reviewed community marketplace with ratings is planned.

Reporting based on coverage by 3D Printing Industry. Written and rewritten for beginners by the Flarelab team.

Originally published at flarelab.com.

Color-Changing 3D Prints: How Lenticular Layers Make an Object Shift Hue

flarelab — Thu, 10 Sep 2026 17:30:08 +0000

Most 3D prints have one look and stick with it. A team at MIT CSAIL just showed off prints that change color when you twist a knob, with no LEDs, no electronics and no paint anywhere. The technique is called ShiftLens, and the physics behind it is old enough that you have almost certainly held it in your hands before.

The idea borrows from lenticular printing, the effect behind those novelty cards that flip between two pictures as you tilt them. A ShiftLens object is printed as two stacked parts. The top layer is a sheet of tiny linear lenses printed in transparent material. Underneath it sits a patterned layer printed in alternating colors, lined up so each stripe of color hides behind its own lens.

What makes this different from a novelty card is that the object itself moves. Instead of you tilting your head, a built-in mechanism slides the lens layer sideways across the pattern layer by a fraction of a millimetre. Each lens suddenly magnifies a different color stripe, and the whole surface flips hue at once. That mechanism can be a knob, a switch or a roller, anything that shifts the two layers relative to each other. The demo people latched onto was a bottle: close the container properly and the band around it changes color, so the object tells you at a glance that it is sealed.

Want to experiment with the concept? The researchers built their design tool inside Rhino and have not released it publicly yet, so treat this as a build to explore rather than a file to download. The ingredients are all ordinary multi-material 3D printing, though: a clear PETG or clear PLA lens layer printed at a fine layer height, a two-color pattern layer beneath it, and a sliding fit loose enough to move but tight enough to stay flat. Start small. A 40 mm flat test tile with roughly 1 mm lens pitch will teach you more about lens alignment in one print than an entire afternoon in CAD.

Try it on your printer. If your machine can handle two materials, or even a manual filament swap at a set layer, you already have everything you need to attempt a first lenticular tile tonight. If you would rather have someone dial in the tolerances for you, the Flarelab team prints and tests multi-material parts every week and can help you get from sketch to working prototype. Browse the shop and the workshop builds at flarelab.com and tell us what you would make change color.

Frequently asked questions

What is a lenticular 3D print?

It is a print made of two layers: a transparent sheet of narrow linear lenses on top, and a striped multi-color pattern underneath. Each lens shows one stripe at a time, so shifting the layers changes which color you see.

Do I need a multi-material 3D printer?

It helps, because the pattern layer needs at least two colors. You can approximate it on a single-extruder machine with a scripted filament change at a specific layer, though alignment is harder and results vary.

Which filament works for the lens layer?

Clear PETG is the usual starting point because it stays glossy and prints reliably. Clear PLA also works. Fine layer heights and a smooth top surface matter far more than the brand of filament you choose.

Can I download the ShiftLens files right now?

Not yet. The MIT CSAIL team built their design tool for internal use in Rhino and has not published a public version, so makers recreating the effect are currently modelling the lens and pattern layers themselves.

Source: 3D Printable Lenticular Indicators on Hackaday, covering the ShiftLens research from MIT CSAIL. Rewritten for beginners by the Flarelab team.

Originally published at flarelab.com.

80 printers, one factory: what a drone maker's print farm teaches your single machine

flarelab — Tue, 08 Sep 2026 17:30:06 +0000

Eighty 3D printers just came online in a single building in Bridgeport, Connecticut — and none of them are there for show. Autonomous defense company Quantum Cyber has finished installing all 80 units of its drone production farm inside a roughly 50,000-square-foot facility, where the machines will turn out structural airframe parts and enclosures as the site's primary manufacturing line. It is a useful reminder that 3D printing stopped being only a prototyping toy a while ago. What is more useful, though, is what a setup like that would demand of your printer.

A print farm is simply many identical machines running the same job. There is no clever trick that makes one printer eighty times faster; instead you scale sideways, adding machines until throughput matches demand. That design choice quietly changes what counts as a good part. On a farm, nobody is standing over the bed rescuing a first layer or babysitting a support removal. A part either comes off correctly, unattended, on every machine in the building — or it is not a production part at all.

That standard is where hobbyists can learn something. Most of us judge a print by whether it eventually worked after two failed attempts, a nozzle wipe and a bit of luck. Factories judge by yield: out of a hundred starts, how many are usable without rework? Chasing yield instead of one heroic print pushes you toward boring, repeatable habits — the same habits that make a shelf full of parts look like they came from the same machine, because they did.

Getting there is mostly bookkeeping. Lock one slicer profile per filament type and stop editing it mid-project. Print the same part three times and measure the same feature on each with calipers — variation across those three tells you far more than any single print. Design supports out of the model rather than adding them in the slicer: chamfer overhangs to about 45 degrees, split awkward geometry into two parts and glue or pin them. Orient for strength so layer lines run across the load, not along it. Keep a short log of filament brand, spool, nozzle temperature and the date, then batch several copies on one plate and walk away. If the plate comes off clean, you have a process, not a fluke.

Try it on your printer. Pick one small functional part you actually use — a bracket, a hook, a cable clip — and run five copies on a single plate with no supports and no mid-print adjustments. Count how many are usable. That number is your yield, and improving it is the whole game. If you need reliable filament and printer parts to keep that number climbing, browse the workshop gear at Flarelab.

Frequently asked questions

What is a 3D printing farm?

A print farm is a group of identical 3D printers running the same job in parallel. Output scales by adding machines rather than speeding one up, which is why factories use farms for low-to-medium volume parts.

Do I need more than one printer to benefit from this?

No. The valuable part is the mindset: design parts that print unattended and first-try. Batching five copies on one plate on a single machine tests the same thing a farm tests.

What is print yield and how do I measure it?

Yield is the share of prints usable without rework. Run a batch of identical parts, count how many pass inspection, and divide. Track it over time as you change one variable at a time.

Can a normal FDM printer make real functional parts?

Yes, within limits. Choose the right material, orient the part so layer lines run across the load rather than along it, use enough walls, and test the part in its actual application before trusting it.

Reported by 3D Printing Industry. Rewritten and expanded for beginners by Flarelab.

Originally published at flarelab.com.

Print your own handheld PC: what a Raspberry Pi 5 cyberdeck teaches about enclosure design

flarelab — Mon, 07 Sep 2026 17:30:06 +0000

A cyberdeck is a homebuilt handheld computer, and it has quietly become one of the best teachers in the maker world. Maker MadHias recently shared a new one: a portable deck built around a full-size Raspberry Pi 5, a high-resolution display and a swappable power bank, with the printable files posted free on MakerWorld. It is a fun build on its own, but the more useful part is what the design choices tell you about printing enclosures for any electronics project.

The first choice is the board itself. MadHias skipped the Compute Module route, which needs a custom carrier board and costs more, and designed around a standard Pi 5 instead. That is the same trade-off you face every time you print a case: designing around common, cheap, easily replaced hardware means your printed shell stays useful when a part dies. The second choice is power. The deck runs from a 5V 3A power bank for portability, but the official Pi power supply can be swapped in for heavy loads without unscrewing anything. Serviceability designed in from the start, rather than bolted on later.

The third choice is the one most beginners miss. The Mini HDMI ports are left fully accessible so an external monitor can be plugged in at any time. A sealed, beautiful case that buries a port you need in six months is a case you will end up cutting open with a knife. Before you slice anything, list every port, button, LED and vent you might want to reach, then design the shell around that list.

To print a case like this well, a few habits do most of the work. Use PETG rather than PLA for anything enclosing a Pi 5 or a battery, because a small sealed shell traps heat and PLA softens early. Set 4 perimeters and 0.2 mm layers so the walls have real strength. Leave about 0.3 mm of clearance per side on every port opening, and print a small test coupon of just one USB cutout before committing to a six-hour shell. Orient each part so its largest flat face is on the bed, which kills most support requirements. And use brass heat-set inserts at the screw bosses so the case survives being opened repeatedly.

Try it on your printer. You do not need to build a whole cyberdeck to practise this. Print a simple two-part box for a spare Pi or microcontroller, dial in your port clearances, and press in your first heat-set insert. Once those three things work, every enclosure after it gets easier. If you need PETG that behaves predictably or hardware to finish the build, browse the filament and maker gear at Flarelab.

Frequently asked questions

Do I need a big printer to make a cyberdeck case?

No. Most handheld shells are split into two to four parts that each fit inside a 220 x 220 mm bed. Splitting the model also lets you orient every piece so its flat face sits on the bed, which is what actually gets you clean port cutouts.

Should I print an electronics enclosure in PLA or PETG?

PETG is the safer pick for anything holding a Raspberry Pi 5 or a battery. PLA starts softening around 55-60 degrees Celsius, and a small sealed shell with a hot board inside can get there on a warm day. PETG holds its shape roughly 20 degrees higher.

How much clearance should I leave around USB and HDMI cutouts?

Start at 0.3 mm per side and adjust from there. Elephant's foot on the first layers and slight over-extrusion both eat into openings, so a cutout modelled at exactly nominal size almost always prints too tight.

What is the easiest way to get strong screw holes in a printed case?

Brass heat-set inserts. You press them in with a soldering iron at about 200 degrees Celsius and get a real metal thread you can open and close dozens of times, instead of a printed hole that strips on the third disassembly.

Originally spotted via Adafruit. Project by MadHias, files on MakerWorld. Rewritten and expanded by the Flarelab team.

Originally published at flarelab.com.

Snap-Fit Cases, Ball Joints and Magnets: 3D Printing an Enclosure for Your ESP32-CAM

flarelab — Sun, 06 Sep 2026 17:30:06 +0000

An ESP32-CAM board costs about the price of a coffee, but out of the bag it is a naked green rectangle with a camera lens hanging off one edge. Point it at your front door and you have a security camera; leave it loose on a shelf and you have a board that falls behind the shelf. A new case kit shared on MakerWorld by the maker i-BoxIt closes that gap, and it is a small clinic in three techniques every 3D printing beginner should learn: snap-fit closures, printed ball joints, and magnet pockets.

Start with the snap fit. Instead of screws, the lid has a thin lip that bends outward as it slides past a ridge on the case, then springs back to lock. The whole trick lives in the clearance, usually 0.15 to 0.25 mm of deliberate gap between the two parts. Too tight and the lid will not go on. Too loose and it drops off when you pick the camera up. The kit covers both micro-USB and USB-C board variants, because the connector cutout is the one dimension you cannot fudge.

The mount is the more interesting half. A printed ball joint gives you a camera that aims anywhere, and the kit fixes the ball to the case using a short piece of filament as a pin, which is about as elegant as scrap-bin engineering gets. Add an optional 10x2 mm magnet in the base and the camera sticks to any steel surface: a radiator, a metal door frame, the side of a filing cabinet. A separate wall mount takes a screw or an 8x2 mm magnet. If you are running the board from a cable rather than a battery, the cable threads through the ball mount so nothing dangles.

Printing it is undramatic, which is the point. A 0.4 mm nozzle at 0.2 mm layer height is plenty. Use three perimeters so the snap arms have material to flex without delaminating, and 15 to 20% infill. Orient the case open-face down and you will need no supports at all. PLA works for an indoor build; choose PETG if the camera will sit in a window or anywhere warm, since PLA softens around 60 C. Print one lid first and test the fit before committing to the full set.

Try it on your printer. Enclosures are the ideal second or third project after the calibration cube, because they teach tolerance in a way no test print can: either the lid clicks or it does not. Grab a spool, print a single test lid, dial in your clearance, then run the whole kit. Filament, magnets and printer supplies are all at flarelab.com, and Flick is always happy to see another board get a proper home.

Frequently asked questions

How much clearance should a snap-fit lid have?

Start at 0.2 mm between the lid and the case wall, then adjust. If the lid rattles, drop to 0.15 mm; if you have to force it, go to 0.25 mm. Every printer and every filament shrinks slightly differently, so one test lid saves a whole failed enclosure.

Which filament is best for an electronics case?

PETG is the safer pick for anything near heat or sunlight, because PLA starts softening around 60 C. PLA is fine for an indoor desk camera and prints more easily. For snap arms that flex repeatedly, PETG or ABS holds up far better than PLA, which fatigues and snaps.

Do I need supports for a case like this?

Usually no. Well-designed enclosures orient the open face down so walls print as vertical perimeters. Lens holes and USB cutouts print as bridges, which most printers handle up to about 15 mm. If your slicer wants supports everywhere, try rotating the part instead.

How do I stop magnets from popping out of their pockets?

Model the pocket about 0.1 mm undersized and pause the print one layer before the pocket closes, drop the magnet in, then resume so the print seals it permanently. If you prefer glue, a drop of cyanoacrylate works, but check magnet polarity before it cures.

Why use a piece of filament as a hinge pin?

A 1.75 mm filament offcut is a free, perfectly round pin that is already the same material as the print, so it expands and contracts at the same rate. Push it through the ball joint, snip it flush, and touch the ends with a hot soldering iron tip to mushroom them into place.

Source: the ESP32-CAM Case Kit by i-BoxIt, spotted via Adafruit's #3DThursday roundup. Rewritten and expanded by Flarelab.

Originally published at flarelab.com.

3D Printed Microneedle Patches: Belfast Researchers Take Aim at Skin Cancer

flarelab — Sat, 05 Sep 2026 17:30:06 +0000

Imagine a patch the size of a postage stamp, studded with needles so fine you would not feel them go in. That is what a research team at Queen's University Belfast has been printing — and the goal is to treat localised skin cancer without repeated injections or messy daily creams.

The patch works on a simple idea. Your skin's outer layer is a very good barrier, which is exactly why creams struggle to get drugs deep enough to do their job. Microneedles solve that by physically crossing the barrier, but they stop short of the nerves and blood vessels that make an injection hurt. The Belfast patch goes one step further: the needles are made from a material that dissolves once it is in place, so the medicine is left behind and there is no sharp waste to throw away. The team, led by Professor Dimitrios A. Lamprou, loaded the array with two different anti-cancer drugs at once.

So where does 3D printing come in? Precision, and the speed of changing your mind. A microneedle array lives or dies on its geometry — how tall each spike is, how sharp the tip is, how tightly they are packed. Traditionally you would cut a mould for every variation, which is slow and expensive. Printing turns that into a software problem: tweak the model, print a new array, test it, repeat. That is the same iteration loop desktop makers already know, just scaled down to microns.

Making one is a high-resolution job. Practically, teams either print the needle array directly on a resin-based machine — SLA, DLP, or two-photon polymerisation for the finest features — or print a master and cast a dissolvable polymer into it. The drug is blended into that polymer before casting or printing, so the dose is baked into the structure itself. Everything happens under laboratory conditions with biocompatible, medical-grade materials, which is the part that separates a research prototype from a desk ornament.

Try it on your printer

You cannot — and should not — print a medical patch at home. But you can chase the same skill it demands: resolution. Try printing a test array of tiny pins at descending heights and spacings on your resin machine and see where the tips stop forming cleanly. It is one of the fastest ways to learn your printer's real limits, and it makes the leap to microneedle research feel a lot less abstract. If you need resin, fine-nozzle hardware or filament to experiment with, browse the range at flarelab.com and start small.

Frequently asked questions

What is a 3D printed microneedle patch?

It is a small square of dissolvable material covered in hundreds of tiny spikes, each far thinner than a hair. The spikes sit in the outer layers of skin and melt away, releasing whatever medicine was mixed into them.

Why does 3D printing matter for something this small?

Needle height, tip sharpness and spacing all change how much drug gets delivered. Printing lets researchers redesign the array in software and produce a new version the same day, instead of commissioning a new mould each time.

What kind of printer makes these?

Not a desktop FDM machine. Microneedle work relies on high-resolution resin processes such as SLA, DLP or two-photon polymerisation, where features are measured in microns rather than tenths of a millimetre.

Can I print medical patches at home?

No. These are research prototypes made with medical-grade, biocompatible materials under laboratory conditions and regulatory oversight. Hobby resins are not safe for skin penetration. Treat this as inspiration for precision printing, not a project.

This article is a Flarelab summary written for beginners. Reporting and original details: 3D Printing Industry. Nothing here is medical advice — speak to a qualified clinician about any treatment.

Originally published at flarelab.com.

Adidas 3D Printed an Entire Sneaker — What Futurecool Means for Your Printer

flarelab — Fri, 04 Sep 2026 17:30:07 +0000

Adidas just printed a shoe. Not a midsole, not an insert, not a little printed logo glued onto a normal sneaker — the whole thing. The new Futurecool comes off the machine as one continuous piece of upper and sole, and the first 400 individually numbered pairs were raffled through the CONFIRMED app in late August, with a wider global release announced for 2027 at around €180.

To see why that is a big deal, count the parts in the sneaker you are wearing. A normal running shoe is somewhere between twenty and forty separate pieces — mesh panels, foam, a rubber outsole, a heel counter, laces, glue — cut in one factory, assembled in another, held together by adhesive that eventually gives up. Futurecool has no assembly step at all. It is grown as a single lattice, and the ventilation that gives the shoe its Climacool name is not holes punched in fabric. It is just the gaps between the printed struts.

The magic ingredient is that lattice. A printed mesh lets one single material behave like several: fatten the struts under the heel and it feels firm, thin them out under the forefoot and it feels springy, open them up along the sides and air walks straight through. Adidas is doing this with resin printing — liquid photopolymer cured layer by layer with light, the same basic idea as the desktop LCD printers people use for miniatures, just scaled up and repeated a few hundred thousand times. That last part is the actual news. Printed footwear has existed since 2018; printed footwear at volume has not.

You cannot print a running shoe on your Ender or your P1S, but you can absolutely steal the technique. Load a spool of TPU, slow the printer down to roughly 15–25 mm/s, drop retraction close to zero, and set your infill to gyroid at 8–15%. Gyroid is the closest thing an FDM slicer has to a proper lattice: it is continuous in all three directions, so it compresses and springs back instead of collapsing. Print a small cube at 5%, 10% and 20% and squeeze each one. That five-minute test tells you more about cushioning than any spec sheet will.

Try it on your printer. Start with something with a real job: a gyroid-infill insole, a pair of TPU sandal straps, a springy phone bumper, or a set of shoe trees in cheap PLA. If you want filament that actually behaves at those speeds — flexible TPU, quality PLA, and the odd spool of PETG for the stiff bits — grab it from Flarelab and print something today.

Frequently asked questions

Is the Adidas Futurecool really 100% 3D printed?

The upper and the sole are printed together as one continuous structure rather than cut, glued and stitched from separate panels. Adidas describes it as a fully printed shoe, which is a big step past earlier models that only printed the midsole.

What kind of 3D printing does Adidas use for sneakers?

Resin-based photopolymer printing, in the same family as DLP and LCD machines. Liquid resin is cured layer by layer with light, which is what allows the fine lattice struts that give the shoe its cushioning and airflow.

Can I print shoes on a normal desktop FDM printer?

Not a wearable running shoe, no. But you can print the same ideas: flexible TPU parts, gyroid lattices, and print-in-place hinges. Sandals, insoles, clog-style slip-ons and shoe trees are all realistic desktop projects.

Why do 3D printed shoes use lattices instead of solid foam?

A lattice lets one single material behave differently in different places. Thicker struts under the heel feel firm, thinner struts under the forefoot feel springy, and the gaps between them do the job of ventilation holes for free.

Reported from coverage by 3D Printing Industry. Rewritten and expanded for beginner makers by Flarelab.

Originally published at flarelab.com.

Why Your PLA Snaps: What Really Makes Old Filament Go Brittle

flarelab — Thu, 03 Sep 2026 17:30:06 +0000

You pull a spool off the shelf, feed it in, walk away for an hour, and come back to filament snapped clean off at the extruder. If that has happened to you, the usual advice is to blame moisture and throw the spool in a dryer. That advice is not wrong, exactly. It is just missing half the story, and the missing half explains why drying sometimes fixes nothing at all.

PLA is not one material. The plastic on your spool is almost always poly(L-lactide), the cheap and common chiral form of polylactic acid, blended with additives the manufacturer will not name. Check the safety data sheet from any of the big filament brands and you will usually find the ingredients listed as PLA plus additives, full stop. Those additives are plasticizers and nucleating agents, and they are the reason fresh PLA bends a little before it breaks. Base PLA on its own is genuinely brittle stuff.

Two things then happen to that spool over time. The first is hydrolysis: water molecules work their way into the plastic and chop the long polymer chains into shorter ones, and short chains break easily. The second is slower and sneakier. PLA drifts from its amorphous state toward a crystalline one, and crystalline PLA is the stable, rigid, brittle version. Plasticizers can also migrate out of the plastic entirely over a few years. A three-year storage study published in Polymers found that samples kept in vacuum-sealed bags came out essentially unchanged, while identical samples in ordinary zipper bags had measurably degraded and crystallized. Same room, same three years, very different filament.

So what can you actually do? Drying helps only with the moisture half. To push crystallinity back down you need to hold the filament above its glass transition temperature, roughly 65 degrees Celsius for a typical PLA blend, for several hours rather than the twenty minutes a quick dry cycle gives you. Most filament dryers with a manual temperature and time setting can manage this. What no amount of heat will fix is hydrolysis, because once the polymer chains are cut short they stay short. If a spool still snaps after a long high-temperature soak, that is your answer.

Try it on your printer. Pick your oldest spool and run a simple test: pull off a metre, bend it into a tight loop, and see whether it flexes or cracks. If it cracks, try a long dry at the top of your dryer's PLA range and test again. And whatever survives, store it properly from now on: vacuum bag with fresh desiccant, sealed the moment the print finishes, not just tossed back on the shelf. Storage does more for filament life than any recovery trick. Browse our filament range at flarelab.com.

Frequently asked questions

Can I always save brittle PLA by drying it?

No. Drying removes moisture and can partly reverse crystallisation if you hold the filament above about 65 degrees Celsius for several hours. It cannot repair hydrolysis, which permanently shortens the polymer chains. Filament that still snaps after a long hot dry cycle is chemically damaged rather than damp.

How long does PLA filament actually last?

It depends almost entirely on storage. Vacuum-sealed PLA has been shown to survive three years with no meaningful degradation, while the same filament left in a loosely sealed bag degraded noticeably over the same period. Assume months in open air and years in a proper vacuum bag with desiccant.

Is PLA+ more resistant to going brittle?

Not reliably. PLA+ is a marketing label rather than a defined formulation, and the additives behind it vary by brand, sometimes including simple fillers like calcium carbonate. Some PLA+ blends are tougher out of the box, but none of them are immune to moisture or long-term crystallisation.

What is the best way to store PLA between prints?

Vacuum-seal the spool with fresh desiccant as soon as the print finishes, rather than leaving it loaded in the printer. Room air is the main culprit, so the goal is minimising exposure time. A sealed dry box with active desiccant is a good alternative if you print from the same spool often.

Reported by Flarelab. Based on reporting by Maya Posch at Hackaday, drawing on published research in the journal Polymers.

Originally published at flarelab.com.

Bambu Lab Patents Hint at Printers That Swap Their Own Nozzles

flarelab — Wed, 02 Sep 2026 17:30:08 +0000

Your printer already watches its first layer. The next question the industry is asking is whether it can watch its own nozzle — and swap that nozzle out on its own when the job calls for something different.

Three new Chinese patent applications, filed by Shenzhen Tuozhu Technology (the company behind Bambu Lab), sketch out exactly that. One group of designs builds illumination directly into the part-cooling duct, so an onboard camera gets a clean, evenly lit view of plastic leaving the nozzle instead of squinting into the shadow the hotend casts on itself. Another describes a storage rack that holds several replaceable printheads, letting the machine park one and pick up another mid-job without a human touching anything.

Why would a printer want more than one head? Because the way multi-color printing works today is wasteful. A single nozzle has to purge the old color out before the new one goes down, and that purge becomes a pile of scrap that can outweigh the actual model. Dedicated printheads sidestep the purge entirely — one head per material. The same trick unlocks mixed nozzle sizes in a single print: a 0.8 mm head blasting through infill and interior walls, then a 0.2 mm head handling the visible detail. The camera work matters for a plainer reason. Clogs, under-extrusion and heat creep all announce themselves at the nozzle tip seconds before they ruin a print, and a well-lit tip is one a machine can actually diagnose. Worth saying clearly, though: a patent filing is an idea with a lawyer attached, not a product. Plenty of them never ship.

You do not have to wait for hardware to get most of this benefit. Start by matching the nozzle to the filament — brass is fine for PLA and PETG, but carbon-fiber, glow-in-the-dark and glitter filaments are abrasive and will round out a brass tip in a spool or two, so fit hardened steel before you print them. Keep a spare hotend assembly pre-built so a clog costs you ten minutes rather than an evening. If your printer has a camera, point it at the nozzle rather than the bed and add a cheap LED so the footage is usable. And run a flow-rate calibration after every nozzle change; a new tip is a new flow profile, and skipping that step is where most post-swap first-layer problems come from.

Try it on your printer this week: swap in a fresh nozzle, run a flow calibration cube, and print the same benchmark model before and after so you can see what a worn tip was costing you. If you need hardened nozzles, spare hotends or abrasive-ready filament to test with, we stock the lot at Flarelab — and our team can tell you which tip your machine actually takes.

Frequently asked questions

Can my current 3D printer swap printheads automatically?

Almost certainly not. Automatic printhead changing is still mostly patent filings and a handful of tool-changer machines. Most desktop printers use one hotend and purge between colors.

How often should I replace my 3D printer nozzle?

A brass nozzle printing plain PLA can last several hundred hours. Printing abrasive filament such as carbon-fiber or glow-in-the-dark can wear one out in under 50 hours. Replace it when hole diameter looks uneven or first layers stop sticking predictably.

Is a hardened steel nozzle better than brass?

Only for abrasive filament. Brass conducts heat better, so it stays the better choice for PLA, PETG and ABS. Hardened steel resists wear but may need a slightly higher printing temperature.

What does purge waste mean in multi-color 3D printing?

When a single nozzle switches colors it must extrude the old filament out before the new color prints cleanly. That discarded plastic is purge waste, and on a four-color print it can weigh more than the model itself.

Reporting on the patent filings via 3D Printing Industry. Written and expanded for beginners by the Flarelab team.

Originally published at flarelab.com.

Why Your Filament Keeps Failing: Storage, Not Drying, Is the Fix

flarelab — Tue, 01 Sep 2026 17:30:07 +0000

If your prints have been coming out stringy, brittle, or peppered with tiny surface pops, the problem may not be your printer at all. It may be the air in your workshop.

Filament is hygroscopic — it pulls water vapour out of the surrounding air and holds onto it. PLA does this slowly. PETG, nylon, TPU and PVA do it fast enough to ruin a spool over a humid weekend. When damp filament reaches a 200°C nozzle, the trapped moisture flashes into steam inside the melt zone. You hear it as a faint crackling, and you see it as stringing, blobbing, rough surfaces, and layers that peel apart under almost no force. Wet filament also swells slightly, which is one of the quieter causes of mid-print clogs.

The standard answer has been the heated dryer box: bake a spool at 45–70°C for a few hours, then print while it is still warm. That works, but it treats moisture as an event rather than a constant. The moment a dried spool goes back on an open shelf the clock restarts, and in a tropical climate it can reabsorb most of what you drove out within a day. Repeated heat cycling is not free either, since warming temperature-sensitive materials over and over can soften or deform a spool. That tension is exactly what the new FilaDC i10, a collaboration between SUNLU and INSLOGIC, is designed around: the idea that filament should live in dry conditions permanently instead of visiting them occasionally.

You can apply the same principle today without buying anything new. Aim to keep spools below roughly 20% relative humidity, and under 15% for nylon and PVA. An airtight bin, a handful of rechargeable silica gel packs and a cheap hygrometer will get most makers there for the price of two spools. Recharge the desiccant when the indicator beads change colour, because saturated gel is just decoration. Dry a spool first if it is already wet, then store it dry: drying and storage solve two different halves of the same problem. If you print a lot of engineering materials, feeding straight out of a sealed dry box through a PTFE tube closes the exposure window entirely.

Try it on your printer. Take a spool that has been sitting out for a month and print a 20 mm single-wall cylinder. Dry that same spool for four hours, print the cylinder again, and put the two side by side under a lamp. The difference in surface finish is usually obvious at arm's length, and it costs you about ten grams of filament to prove. When you are ready to upgrade your storage or restock, our filament and printer gear is at flarelab.com.

Frequently asked questions

How do I know if my filament is wet?

Listen and look. A wet spool makes a faint popping or crackling sound at the nozzle, sometimes with visible wisps of steam. On the part you will see stringing between islands, rough or bubbled surfaces, and layers that snap apart with very little force. Brittle filament that breaks in your hands as you feed it is another strong hint.

What temperature should I dry filament at?

Stay below each material's glass transition point. As a rough guide: PLA around 45–55°C, TPU around 50–55°C, PETG around 60–65°C, and nylon around 70–80°C. Four to eight hours is typical, and badly soaked nylon can need longer. Always check the manufacturer's figure for your specific spool first.

Can I dry filament in a kitchen oven?

It is risky. Most home ovens overshoot their set point by a wide margin and their lowest setting often sits above PLA's glass transition temperature, which can fuse a spool into a solid block. A dedicated filament dryer or a food dehydrator with a real adjustable thermostat gives you far better control for similar money.

Does PLA really need dry storage?

Yes, just less urgently than nylon or PVA. PLA absorbs moisture more slowly, but a spool left open for months still turns brittle, strings more, and loses layer strength. If you print only occasionally, dry storage matters more for you than for someone who burns through a spool in a week.

Reported by 3D Printing Industry. Rewritten for beginners by Flarelab. Original source.

Originally published at flarelab.com.

Turn a Dead Build Plate Into a Magnetic PCB Holder

flarelab — Mon, 31 Aug 2026 17:30:07 +0000

Every 3D printing hobbyist eventually ends up with a warped, scratched, or delaminated build plate sitting in a drawer. It is too damaged to print on and too useful-looking to throw away. A design shared by Sanja 3D on MakerWorld gives that dead plate a second career: it becomes the base of an adjustable magnetic PCB holder for soldering and repair work.

The idea is simple and it is why it works so well. Instead of building a clamp with hinges, screws and springs, the design uses small printed jaws with magnets embedded inside them. Drop the jaws onto a steel surface and they stay exactly where you put them, but slide freely when you push them. Any PCB, from a tiny sensor breakout to a full Raspberry Pi HAT, gets pinned between two jaws in about three seconds. There is no fixed maximum size, because the working area is whatever metal sheet you place underneath.

It is also a good example of remix culture in 3D printing. Sanja 3D built on an earlier PCB holder by squinn rather than starting over, and credits the original openly. That is how most of the best functional prints evolve: someone solves the hard geometry, someone else notices it could be magnetic, and the community gets a better tool than either would have made alone.

Printing it is beginner territory. Use PLA or PETG at 0.2 mm layer height with four walls and 25 to 30 percent infill, since the jaws take side loads rather than crushing forces. The important part is the magnet pockets. Most designs use 6 x 3 mm or 8 x 3 mm neodymium discs, so measure yours before slicing and check the model's stated size. Print a single jaw first as a test. If the magnet is loose, a drop of superglue fixes it; if it will not go in, scale the pocket up by 1 to 2 percent rather than forcing it and cracking the wall. Insert magnets with the same pole facing down on every jaw, or two jaws placed side by side will shove each other apart.

Try it on your printer. If your old plate is aluminium rather than steel it will not work, so grab any thin steel sheet, a magnetic knife strip, or a spare PEI-on-spring-steel plate instead. Print two jaws, live with them for a week, then print four more once you know the height you actually want. For more beginner-friendly 3D printing guides, filament tips and printable project ideas, visit Flarelab.

Frequently asked questions

What size magnets should I use for a 3D printed PCB holder?

Six by three millimetre or eight by three millimetre neodymium discs are the most common sizes in these designs. Always check the model page first, measure your magnets with calipers, and print one test jaw before committing to a full set.

Will this work on any old build plate?

Only if the plate is ferromagnetic. Spring steel PEI sheets and standard steel plates work perfectly. Aluminium, glass and most flexible plastic plates are not magnetic, so use a steel sheet or magnetic knife strip as the base instead.

Should I print PCB holder jaws in PLA or PETG?

PLA is fine for occasional bench use and prints more easily. Choose PETG if the holder will sit near a hot soldering iron or in a warm workshop, because PLA softens at roughly 60 degrees Celsius and can deform.

Why do my magnetic jaws push each other away?

The magnets were inserted with opposite poles facing down. Mark one face of every magnet before gluing and keep that face consistent across all jaws so they attract the base plate and stay neutral to each other.

Design by Sanja 3D on MakerWorld, remixed from an original by squinn. Spotted via Adafruit's #3DThursday. Rewritten and expanded for beginners by Flarelab.

Originally published at flarelab.com.

Why Remixing Beats Redesigning: A Soldering Helper That Got Better Three Times

flarelab — Sun, 30 Aug 2026 17:30:06 +0000

Here is a small piece of 3D printing magic that beginners almost always miss: the best model on your printer probably was not designed by one person. It was designed by one person and then quietly improved by five more. A soldering "third hand" making the rounds this week is a perfect example — a maker named CoatOver took a well-loved clothespin-style holder and shipped a version that fixes three annoyances the original never solved.

The changes sound tiny written down. The slot the clothespin sits in was pulled back a couple of millimetres so the clip grips instead of rattling. A curved brace was added behind the jaw so it stops drifting sideways mid-solder. And the footprint was stretched out so the whole thing stops tipping over when you open it with one hand — which, if you have ever tried to hold a wire, an iron and a board at the same time, is the exact moment you need it to stay put. A little tray was tucked onto the front for the screws and offcuts that otherwise vanish into the carpet.

That is what a remix actually is. Nobody redrew the model from scratch. Somebody used the original enough times to learn where it was irritating, then fixed those three spots and put the result back out for everyone. This is the loop that makes 3D printing different from buying a finished product: a design is never really done, it just gets to the next person. Filament choice matters here too — this one is recommended in PETG rather than PLA, because a soldering iron lives at a few hundred degrees and PLA starts going soft somewhere around 60°C. A holder that droops next to a hot tip is worse than no holder at all.

To try it yourself: download the STL, slice it at roughly 0.2 mm layers, and use 3 or 4 perimeters with 20–25% infill so the base has enough mass to resist a tug. Print in PETG if you have it, and orient the base flat on the plate so the layer lines run across the direction of stress instead of along it. Add a brim if your first layer is temperamental — a wide, well-stuck base is the entire point of this remix. Then use it for a week and notice what still annoys you. That annoyance is your remix.

Try it on your printer. If you are still dialling in PETG, or you want filament that behaves the same way on the tenth spool as it did on the first, have a look at what we stock at Flarelab — and tell us what you would change about this design. We love a good remix.

Frequently asked questions

Why PETG instead of PLA for a soldering helper?

PLA softens at roughly 60°C, and anything sitting next to a soldering iron gets warm fast. PETG holds its shape closer to 80°C and is far less likely to sag or deform. ABS and ASA work too, but PETG is much easier to print for beginners.

What actually is a 'remix' in 3D printing?

A remix is a modified version of someone else's model, re-shared publicly. Most open model licences allow it as long as you credit the original designer and share under the same terms. Check the licence on the model page before you republish.

Do I need supports to print this?

Usually not, if you orient the base flat on the build plate. Clothespin-style holders are designed to print without overhangs. Preview the sliced file first — if your slicer shows a large unsupported bridge, rotate the part before adding supports.

My print keeps tipping over when I use it. What do I do?

That is exactly the problem this remix solves, so start with the extended-base version. Beyond that, raise your infill to 25% or more so the base carries weight, or add a coin or nut into a pocket in the base as ballast.

Spotted via Adafruit's #3DThursday. Original design by Wolfchild; extended-base remix by CoatOver, shared on MakerWorld. Written and expanded by the Flarelab team.

Originally published at flarelab.com.