Short answer
The biggest single marker of filament quality is how constant and round the diameter stays along the spool: the gap between a ±0.02 mm and a ±0.05 mm tolerance is the gap between roughly 2% and 6% error in extruded volume. Beyond that, a sealed pack with desiccant, neat winding, a lot number and an accessible TDS and SDS are the signs of a well-made filament. With digital calipers and a kitchen scale you can check most of this yourself in a few minutes.
Key takeaways
- Measure diameter on two axes at the same spot; ovality is invisible to a single reading.
- Cross-section area scales with diameter squared, so a small diameter error roughly doubles in the flow.
- A sealed pack, desiccant and tidy winding show how seriously the maker takes the process.
- Without a lot number, TDS and SDS, a problem spool is close to impossible to trace.
- Calipers, a scale and one short test print expose most quality problems on day one.
Why filament quality shows up before you print
However good your printer is, it cannot print a better part than the material you feed it. When the slicer works out each layer, it assumes the filament is a given diameter, dry and clean. As long as that holds, your settings behave predictably.
When it doesn't, the symptoms look like something else: gaps in walls, holes in top surfaces, sudden stringing, a print that stops halfway. Most people start adjusting settings at that point. The problem may not be the settings at all, but the spool.
On a single printer that costs you an evening. On a print farm the effect multiplies: hand out ten spools from one batch to ten printers and a single quality problem can stop many jobs at once. As you scale, checking material when it arrives is far cheaper than hunting for the problem mid-print.
This post covers what to look at when a new spool arrives and what you can confirm with simple tools. The general route from symptom to cause is in our troubleshooting guide; here the point is to catch the problem before you print.
Diameter tolerance and ovality: ±0.02 versus ±0.05 mm
The “1.75 mm” on the label is a target; the tolerance tells you how far from it the filament may stray. A ±0.05 mm tolerance means the filament can be anywhere between 1.70 and 1.80 mm. 3D-Fuel describes that as the generally accepted industry minimum, while makers such as Prusament aim for ±0.02 mm and publish measured data for each spool.
Geometry explains why the small difference matters. The extruder pushes a set length of filament; the volume of plastic entering the nozzle depends on the cross-section area. Area scales with the square of the diameter, so a 1% diameter error becomes about a 2% volume error.
| Actual diameter | Cross-section difference | Likely effect on the print |
|---|---|---|
| 1.70 mm | about 5.6% less | Thin walls, gaps in top surfaces |
| 1.73 mm | about 2.3% less | Often invisible, can show in fine detail |
| 1.77 mm | about 2.3% more | Slight over-extrusion, parts grow a little |
| 1.80 mm | about 5.8% more | Blobs, bulging, friction in tight paths |
So a ±0.05 mm tolerance allows a volume range of about 11% from end to end within one spool; at ±0.02 mm that range narrows to about 4.6%. Flow calibration corrects the average, but it cannot correct the swings within a spool. A study at Indiana University reports the same thing: an irregular diameter disturbs the flow rate, causing poor surface quality, gaps between lines and extruder jams.
Ovality: why one reading misleads
Ovality is the difference between the largest and smallest diameter of the same cross-section. On an oval filament you might read 1.80 with the calipers one way and 1.70 after turning it ninety degrees. Even a filament whose average looks right will be gripped unevenly by the extruder gear and can rub in a narrow feed path.
Moisture and packaging: how a spool should arrive
Most filaments draw moisture from the air. A spool that wasn't dried after production and sealed in an airtight pack collects moisture over weeks on a shelf, and you meet it on the first print as crackling, stringing and a rough surface.
- Is the pack sealed: expect a vacuum or heat-sealed bag. A hole, a tear or an open zip is a bad sign.
- Is there desiccant: a silica gel or similar sachet should be inside. With colour-changing types, check whether it already shows saturation.
- Can you reuse the pack: a resealable bag helps you protect the spool between jobs.
- Is the box clean: dust and grit ride into the extruder with the filament.
Even with a good pack, checking moisture once more is cheap insurance. We cover the signs and the principles of safe drying in the storage and drying guide. From a quality point of view, the question is simple: was this spool dry when it left the factory, and did the pack keep it that way?
Winding quality, tangles and the spool itself
On a well-wound spool the filament sits in neat, side-by-side layers. If one turn slips under another during winding, a knot forms: the spool locks at that point, the extruder can no longer pull, and a long print fails hours in. The worst part is that the fault can hide deep inside the spool.
- Look at the side: are the turns in orderly rows, or crossed over in places?
- The spool shouldn't be filled to the rim; winding that rises above the flange shifts in transit.
- The loose end should be fixed in a hole in the flange. A free end works loose, slips under and creates a knot.
- The hub should fit your printer's or dry box's holder and turn freely.
- The flange should be free of cracks, burrs and warping, which all add drag.
Colour, lot consistency and purity
For a single spool, colour is a matter of taste. If you print the same product for months, it becomes a quality issue: if the “same” grey on a new spool comes out slightly warmer, parts side by side on a shelf show it at once. We cover colour consistency and how it is measured in a separate post.
Purity, on the other hand, you can check by eye. Hold the filament to the light and run a few metres through your fingers:
- Bubbles and voids: small bubbles on or inside the strand can point to moisture during or after production.
- Black specks and foreign matter: can trace back to contaminated raw material or a line that wasn't cleaned, and can partly clog a nozzle.
- Colour streaks: can mean the pigment wasn't dispersed well.
- Surface texture: good filament has the same sheen along its length; patches that switch between matte and glossy suggest the process drifted.
Bubbles are easy to spot in translucent filament. With opaque colours, you can also bend a short length to test brittleness: PLA that snaps easily often points to material that has absorbed moisture and aged.
Documentation: TDS, SDS and the lot number
A quality filament is more than a good spool: there is an open record behind it. Look for three things before you buy.
| Document | What it tells you | What to check |
|---|---|---|
| TDS (technical data sheet) | Recommended print temperatures, drying conditions, mechanical values | Does it state test method, sample conditions and print orientation, with a version and date |
| SDS (safety data sheet) | Composition, safe handling and storage, fire and disposal information | Is it specific to the product, or a generic polymer form |
| Lot number | Which production batch the spool came from | Is it readable on the label, and can the maker look up that lot's records |
A lot number on its own is not a quality claim; its value comes from the records behind it. When something goes wrong, it lets you find other spools from the same batch and speak the same language as the maker. What lot data records is covered in our lot data guide.
The quality system AMENT is building lines up with this list: lot tracking, diameter and ovality monitored throughout production, moisture management from the line to the sealed pack, colour consistency and versioned documents. Today these are goals; our filaments are not on sale yet and we have not published measured data. You can see the planned products on the materials page.
Simple tests with calipers and a scale
You don't need a laboratory. Digital calipers with 0.01 mm resolution and a kitchen scale that reads to the gram are enough to expose most problems.
- 1
Zero the calipers
Clean the jaws, close them and zero. Close them on the filament just until they touch; squeeze and the reading drops.
- 2
Measure ten points on two axes
Unwind a few metres and measure roughly every metre. At each point, measure once, rotate the filament 90 degrees and measure again. Write the two values side by side.
- 3
Work out average, range and ovality
Is the average of all readings close to target? Are the smallest and largest within tolerance? The difference between the two readings at one point is the ovality.
- 4
Weigh the spool
Weigh the full spool and subtract the empty spool weight the maker gives. The net weight should be close to the label. If the empty weight isn't published, it is worth asking for.
- 5
Run a short test print
Print a model you know, with settings you know. Sound, stringing and surface usually tell you within half an hour whether the spool is dry and consistent.
The scale tells you something else too: how much filament is left. At 1.75 mm and a density of about 1.24 g/cm³ (a figure common on PLA data sheets), a kilogram of filament is roughly 335 metres. Take your material's density from its data sheet; the number changes noticeably between materials.
The one-page checklist
| Check | Good sign | Bad sign |
|---|---|---|
| Diameter | Readings close to target, in a narrow range | Clear swings from metre to metre |
| Ovality | The two axes very close | Reading changes clearly when rotated |
| Packaging | Sealed bag, desiccant, resealable | Open or punctured bag, no desiccant |
| Winding | Neat rows, end fixed to the flange | Crossed turns, loose end |
| Surface | No bubbles, even sheen | Bubbles, black specks, colour streaks |
| Documents | Lot number, versioned TDS and SDS | No lot, no documents, or generic ones |
If a spool passes this list and still gives you trouble, you now have solid ground for looking at settings. The order for dialling in a new material is in the print settings guide.
Frequently asked questions
What diameter tolerance should good filament have?
For 1.75 mm filament, ±0.05 mm is a common industry minimum; more careful makers aim for ±0.02 mm. What matters is that the value is given as an upper limit that holds along the whole spool, not an average, and that ovality is stated separately.
How do you measure filament ovality?
Measure once at a point with calipers, rotate the filament 90 degrees and measure again. The difference between the two readings is the ovality at that point. Repeating this roughly every metre over a few metres gives you the overall picture of the spool.
Can cheap filament damage a 3D printer?
Not directly, but contaminated material can partly clog a nozzle, and a tangled spool can make the extruder grind the filament. The real cost is usually failed prints and lost time. Rather than price, check whether a spool passes the checks in this post.
Do I need to dry a brand-new spool of filament?
A spool from a sealed pack with desiccant is usually ready to use. If the first print crackles, strings or comes out rough, dry it and reprint the same test part. If the difference is clear, the pack or storage did not keep moisture out.
Why does a lot number on a filament spool matter?
The lot number shows which production batch the spool came from. If a problem appears, it lets you find other spools from the same batch, lets the maker check that batch's records, and helps you see what changed on a repeat order.
Sources
- 1Prusament PLA product page: ±0.02 mm manufacturing precision and per-spool data (Prusa Research)
- 2Does filament quality really matter? (3D-Fuel)
- 3Cardona, Curdes, Isaacs: Effects of Filament Diameter Tolerances in Fused Filament Fabrication (IUPUI Journal of Undergraduate Research, 2016)
- 4Prusament PLA technical summary, density 1.24 g/cm³ (Filament2Print)