Filament Specs: Tolerance, Density, and Drying Temps
Most people buy filament by price and color. What actually kills a print is the six numbers on the box, and those six belong to three completely different owners: some are properties of the material, some are properties of your printer or your dryer, and one of them, net weight, is something vendors mostly refuse to pin down.
This piece puts four sources on the same table: the Prusament technical data sheet (version 1.1), the Bambu Lab official store page plus the official wiki's drying table, Polymaker's technical pages, and the title fields of four listings that are actually in stock. At the same nominal value, four vendors land on two tolerances, 0.02 and 0.03 mm. For the same material class, official PLA density is published as both 1.17 and 1.24. Where I could not get a number, the text says so.
Purchase links here are Amazon affiliate links. If you buy through them this site may earn a commission, and it does not change what you pay. Everything was fetched on 2026-11-12; from this machine's region Amazon returned an empty price container, so this article quotes no prices. Check the product page and your order page before you buy. If you are still picking a machine, start with the 2026 3D printer roundup for programmers; that one answers which printer, this one only answers how to read the spool.
TL;DR: Of the six numbers on a spool, only two can be compared across brands. Diameter tolerance is written as 0.02 mm by Prusament PLA and FlashForge PLA PRO, and 0.03 mm by everyone else in this group; through the cross-section that is a 2.3% versus 3.4% swing in extruded volume. One kilogram is not one length: official PLA density spans 1.24 and 1.17, a 5.98% difference, which is 335.3 m against 355.3 m. Drying temperature is not a material property: in Bambu's own table, PLA is 60-70 C on the heated bed, 50 C in a convection oven, and 45 C in both AMS 2 Pro and AMS HT. Your hardware picks the number. If you want the parameters preloaded, Bambu Lab PLA Basic writes them into RFID, but its tolerance is the looser 0.03 mm. If you need to dry PETG and own a 65 C dryer, Creality Space Pi SE covers it; ABS it does not.
Six numbers, three owners
Sort out ownership first, and every later comparison has a rule.
| Number on the box | Who decides it | Comparable across brands? |
|---|---|---|
| Diameter tolerance 0.02 or 0.03 mm | The production line; material only indirectly | Yes, but note Blend and pure stock sit in one table at two tolerances |
| Net weight 1 kg | The spooling step | No, convert with density first |
| Density g/cm3 | The formulation | Yes, if the test standard matches |
| Drying temperature and time | Your drying hardware | No, four official values for one spool |
| Nozzle temperature range | Material, plus nozzle and speed | Only the width; vendors do not define the endpoints |
| Heat resistance (HDT, Tg, Vicat) | The material | No, three methods, three numbers |
The table is the conclusion: of the six, only diameter tolerance and density are independent of your hardware. Change the machine and the other four change value. This is the same shape as the USB power meter piece, where volts times amps is not the power ceiling. The number is not wrong. Believing it has one meaning is.
Diameter tolerance: 0.02 versus 0.03 is 3.4% of flow
The Prusament data sheet is the bluntest about it. The first line of the Identification block reads "Diameter 1.75 +/- 0.02 mm (+/- 0.03 mm for PLA Blend)". Note the bracket: pure stock is 0.02, Blend is 0.03, one notch looser, in the same document. The Bambu Lab store page states "Diameter: 1.75mm +/- 0.03mm". eSUN's listing title says "Dimensional Accuracy +/- 0.03mm" and FlashForge's says "+/- 0.02mm". Four vendors, two tolerances.
Why does the second decimal matter? Because extrusion depends on cross-section, not diameter, and cross-section is the square of diameter.
| Tolerance | Diameter range | Cross-section vs 1.75 mm | Flow deviation |
|---|---|---|---|
| +/- 0.02 mm | 1.73-1.77 mm | 2.4053 mm2 nominal | +2.30% / -2.27% |
| +/- 0.03 mm | 1.72-1.78 mm | same | +3.46% / -3.40% |
| +/- 0.05 mm | 1.70-1.80 mm | same | +5.80% / -5.63% |
The nominal area is pi times 0.875 squared, or 2.4053 mm2; deviations are (d/1.75) squared. That arithmetic is mine, not the vendors'.
So a spool that is perfectly within a 0.05 mm spec can swing your flow by more than 11 percentage points end to end. Even 0.03 mm gives 6.8 points. A flow coefficient calibrated at mid-spool is wrong at the tail. This is the same trap as in the solder wire spec piece, where the millimetre column is nominal and the area error reaches 12%: the box gives you a diameter, the machine eats a volume.
Practically: calibrate flow at mid-spool rather than at the outer end, and when you switch brands, print a single-wall square and measure the wall before trusting any printed tolerance.
How long is 1 kg: density decides a 5.98% spread
One kilogram is a weight, not a length. Length is mass divided by mass-per-metre, and mass-per-metre is cross-section times density. The cross-section is fixed at 2.4053 mm2 for 1.75 mm, so length depends only on density.
Densities I could source: Prusament PLA lists "Density [g/cm3] 1.24, ISO 1183"; Prusament PETG lists 1.27 on the same standard; Polymaker's PolyLite PLA page lists "Density 1.17 g/cm3 at 23 C, ISO1183, GB/T1033". Two official PLA values, 5.6% apart. ABS is taken at 1.04 as an order-of-magnitude reference only, because I did not find a matching vendor data sheet.
| Material | Official density g/cm3 | Mass per metre | Length of 1 kg |
|---|---|---|---|
| PolyLite PLA | 1.17 | 2.81 g/m | 355.3 m |
| Prusament PLA | 1.24 | 2.98 g/m | 335.3 m |
| Prusament PETG | 1.27 | 3.05 g/m | 327.4 m |
| ABS (1.04, reference) | 1.04 | 2.50 g/m | 399.8 m |
Between the two PLA values the difference is 5.98%; between PLA and ABS it is 19.2%. "The cheaper spool is better value at 1 kg" therefore fails across materials, and is off by 6% across formulations. To cost by the metre, find the density line first; search the technical page for ISO 1183, it usually sits next to MFR.
There is a nastier wording problem: net or gross. Official pages say "1kg" or "Size 1kg" and never say whether the spool is included. Bambu lists "Refill" and "with spool" as two variants, both at 1kg, which implies the kilogram is filament; but third-party spools run from roughly 150 g to over 250 g, so the same "1kg" can arrive as 1.15 kg or 1.25 kg. I could not obtain an official tare weight from any vendor, so this column stays marked as not published, with no conversion attempted.
Drying temperature belongs to the hardware, not the filament
This is the single most useful item here. Bambu Lab's official wiki splits its drying table by the device doing the drying, and one row for PLA Basic carries four temperatures:
| Method | PLA Basic / Matte | PETG | ABS / ASA |
|---|---|---|---|
| Printer heated bed | 60-70 C / 12 h, flip every 6 h | 75-85 C / 12 h | 90-100 C / 12 h |
| Convection oven | 50 C / 8 h | 60-65 C / 8 h | 75-85 C / 8 h |
| AMS 2 Pro | 45 C / 12 h | 65 C / 12 h | Not compatible |
| AMS HT | 45 C / 12 h | 65 C / 12 h | 80 C / 8 h |
One spool of PLA, four official temperatures spanning 45 to 70 C, a 25 K spread. ABS is worse: in the AMS 2 Pro column it simply reads "Not compatible". So the answer to "what temperature do I dry PLA at" depends on your hardware, not on the table you found online.
Why the spread? The same wiki gives two clues. First, the spool: Bambu's reusable spool comes in "Low Temp (up to 70 C)" and "High Temp (up to 90 C)" versions, meaning the spool itself is a temperature-rated part. The same page warns that "Some third-party spools may not be heat-resistant enough and could deform during drying". Second, softening: the wiki states that static drying mode is not suitable for PLA, PVA, Support for PLA/PETG, and TPU, because these soften at low temperature and the layers stick or deform when dried without rotation. That is why AMS uses 45 C with rotation.
Together those explain the table. A heated bed can reach 60-70 C, but the spool has to survive it and you have to flip the roll. AMS uses 45 C because it rotates rather than heats hard. And 65 C is the AMS 2 Pro ceiling, which is why PETG sits exactly there and ABS is excluded.
One listing note: the Creality Space Pi SE title reads "45~65 C". That ceiling covers PLA and PETG on Bambu's numbers, but not ABS at the official 80-100 C. Decide which materials you will dry before you shop on capacity or looks.
Nozzle range: the width differs by a factor of two
Prusament PLA says "Nozzle Temperature [C] 210 +/- 10", which is 200-220, a 20 K window. Polymaker's PolyLite PLA says "190 - 230", a 40 K window. Same material family, double the width, from two official pages.
Prusament PETG is "250 +/- 10". The two vendors are expressing different things: a centre value with a tolerance, versus a floor and a ceiling. Both work, but they are not the same claim. A wide window usually means the formulation is insensitive to temperature, and also that the vendor promises nothing about what a specific temperature buys you. A narrow window means the operating point is clear and the cost of leaving it is higher.
What matters is what happens outside the window. Below 190 C, PLA layer adhesion drops; above 230 C, stringing and charring risk rises. Those are experience statements, not vendor claims, and this article ran no temperature ladder, so it quotes no single recommended temperature.
Heat resistance is HDT, Tg and Vicat, not nozzle temperature
"PLA prints at 210 C" and "PLA survives at 60 C" are unrelated numbers. Prusament PLA publishes "Heat Deflection Temperature (0.45 MPa) [C] 55, ISO 75" and "(1.80 MPa) [C] 55"; both loads give 55. Polymaker PolyLite PLA publishes Tg 61 C by DSC, HDT 58 C at 1.8 MPa and 60 C at 0.45 MPa, and Vicat 63 C. One material class, five numbers: 55, 58, 60, 61, 63.
Nobody is lying. Three methods measure three things: glass transition, deflection under a fixed load, and penetration by a standard needle, and the load itself differs between 0.45 and 1.80 MPa. Prusament PETG publishes HDT 68, thirteen kelvin above its own PLA, which is the real size of PETG being more heat resistant, rather than the 40 K higher nozzle temperature.
The engineering rule is simple. Do not leave PLA parts in a closed car in summer, where cabin temperatures routinely pass 60 C, and do not park them behind a laptop exhaust. If you need heat resistance, change material; raising nozzle temperature does nothing. Structurally this is the same as the digital caliper piece, where one tolerance figure splits into E and S columns: one word, two definitions.
Moisture: 0.19% water is 4.18 L of gas at 210 C
Prusament measures absorption at two points: "Moisture Absorption in 24 hours [%] 0.13" and "in 7 days [%] 0.19", at "24 C; humidity 22 %". PETG reads 0.07 and 0.10.
The counterintuitive part: PETG absorbs more slowly than PLA, yet its official drying temperatures (60-85 C) are higher than PLA's (45-70 C). It is more sensitive to water, not less. Absorption rate and moisture sensitivity are two different things.
How bad is 0.19%? On a 1 kg spool that is 1.9 g of water. Treating it as an ideal gas at 210 C and one atmosphere gives about 4.18 L: 1.9 g divided by 18.015 g/mol is 0.1055 mol, and V equals nRT over P with R at 8.314, T at 483.15 K and P at 101325 Pa. The 0.13% figure works out to 1.30 g and about 2.86 L. All of that has to pass a 0.4 mm nozzle. That is where popping, bubbles and stringing come from.
Vacuum sealing with desiccant only guarantees the spool was dry on arrival. Storage after opening is the variable, and no vendor publishes a number for it.
Six spools in stock, and who each one suits
All six returned a product title and an add-to-cart control when parsed from their /dp/ pages on 2026-11-12. Prices are omitted because the local price container came back empty.
Prusament PLA (1.75 mm, 1 kg)
The most complete official data here: tolerance 0.02 mm, density 1.24, HDT 55, nozzle 210 +/- 10, moisture at two time points, every figure with a test standard. The downside is that the data sheet is version 1.1, last updated 2022-02-16, and has not moved in four years and nine months; colour availability on Amazon is also much thinner than the vendor's own store. It suits people who want to trace every number to a source, and you can confirm the current colours on the product page.
Bambu Lab PLA Basic (1.75 mm, 1 kg)
The store page states "Diameter: 1.75mm +/- 0.03mm" and "All AMS Series Compatible", with print parameters stored in RFID for the AMS to read. The downside is the loosest tolerance in this group, about 1.1 percentage points more flow swing than 0.02 mm; the RFID parameters are only readable by its own printers, so third-party machines need a fresh calibration; and Refill versions require a reusable spool, which comes in low-temp (up to 70 C) and high-temp (up to 90 C) ratings, so the wrong one can deform while drying PLA. It suits people already inside the Bambu ecosystem, and the spec sheet can be checked on the product page.
Polymaker PolyLite PLA Pro (1.75 mm, 1 kg)
Official numbers: density 1.17 g/cm3 at 23 C to ISO 1183, Tg 61, HDT 58 and 60, Vicat 63, drying 55 C for 6 h, nozzle 190-230. The downside is that the product page carries less than the technical wiki, so a key figure like density requires digging; and "Pro" has no standard definition, with the same brand's CosPLA listed at 1.24, a 5.98% difference, so do not extrapolate from the brand name. It suits people costing by the metre who want a longer single spool, and the listing is here.
eSUN PLA+ (1.75 mm, 1 kg)
The listing states "Dimensional Accuracy +/- 0.03mm, 1kg Spool, 1.75mm", which is admirably direct. The downside is that the "Plastic Spool" carries no temperature rating, while Bambu's official wiki explicitly warns that third-party spools may deform during drying, so check the spool before using a heated bed or an oven. "PLA+" is also not a standard name, and Prusament's own sheet puts its Blend class one tolerance looser than pure stock, so similarly named products are not interchangeable. It suits budget-first buyers who do not plan on hot drying, and the listing is here.
FlashForge PLA PRO (1.75 mm, 1KG)
The title carries both "1KG Spool" and "(2.2lbs)" and states "+/- 0.02mm", making it one of the tighter spools here. The downside is that 2.2 lbs is an approximation, since 1 kg is exactly 2.2046 lbs, and printing both units side by side reads like two specifications. PLA PRO shares PLA+'s lack of a standard. It suits people who want 0.02 mm without a premium price, and the listing is here.
Creality Space Pi SE dryer (45-65 C)
The title reads "45~65 C" and "for 1KG Spools", which covers the official PLA and PETG drying ranges. The downside is the 65 C ceiling: it cannot dry ABS or ASA at the official 80-100 C, and it misses the AMS HT 80 C step. Check which materials you actually dry before checking how many spools it holds, and the listing is here.
Troubleshooting: three symptoms, three numbers
Stringing, popping, bubbles on the surface. Suspect moisture before retraction. The evidence is Prusament's two absorption figures: if the spool has been open for a week in humid air, dry it within your hardware's limits, 45-70 C for PLA over 8-12 hours, and do not exceed 100 C because the spool may not survive it.
The same spool prints fine at the start and under- or over-extrudes at the end. Read the tolerance line. A 0.03 mm spool is enough for a 6.8 percentage point area swing, so a mid-spool flow coefficient will not hold at either end. Do not chase it with flow multipliers: print a single-wall part, measure the wall, set the coefficient from that, or move to a 0.02 mm spool.
Parts warp after a few days, or deform in a car. This is not a slicer problem, it is HDT. Official PLA HDT lands between 55 and 60 C, and beyond that the part creeps. Switch to PETG at HDT 68 or add ribs; raising nozzle temperature will not help.
FAQ
Q: How many metres are in a 1 kg spool?
It depends on density. One metre of 1.75 mm filament is about 2.405 cm3, so mass per metre is that volume times density: PLA at 1.24 is about 2.98 g/m, or 335.3 m per kilogram; PLA at 1.17 is about 2.81 g/m, or 355.3 m; PETG at 1.27 is about 327.4 m. The spread is 5.98%, and vendors publish the kilogram without the metres.
Q: Can the same spool be dried at different temperatures on different machines?
Yes, within each machine's ceiling. Bambu's table gives PLA 60-70 C on a heated bed, 50 C in a convection oven, and 45 C in both AMS units, because the heating method (contact, convection, rotation) and the spool's temperature rating differ, not the material. ABS is "Not compatible" in AMS 2 Pro, which is the line most often missed.
Q: How much does 0.02 versus 0.03 mm actually change?
It is 2.3% versus 3.4% of extruded volume through the cross-section. The numbers differ by 0.01 mm; the flow differs by 1.1 percentage points. If your parts are dimensionally sensitive, that is worth choosing a brand over.
Q: Are PLA+, PLA Pro and PLA Blend the same thing?
No, they are marketing names without a common definition. The checkable evidence is tolerance inside one data sheet: Prusament states 0.02 mm for pure stock and 0.03 mm for Blend, so the suffixed version is officially one notch looser. Read the sheet, not the suffix.
What this piece did and did not do
It did: transcribe and cross-check four official pages and three data sheets; convert tolerance and weight into flow and length using cross-section and density; unpack the four columns of Bambu's official drying table; and parse six /dp/ pages to confirm availability.
It did not: there is no 3D printer on this machine, so every print-behaviour statement is marked as experience rather than measurement; no spool was measured for diameter or density, only vendor declarations are used; no spool was weighed, so the net-versus-gross column stays unpublished; and the ABS density is an industry figure, not a vendor data sheet. Discount accordingly.
If you keep one line: before comparing two spools, ask whether the number belongs to the material or to the machine. Get that wrong and every comparison after it drifts.
👉 Join Xiaomi MiMo Platform: Leading AI model platform with cost-effective inference
👉 Join Aliyun AI: Top AI products with exclusive coupons for business innovation
📌 This article was AI-assisted generated and human-reviewed | TechPassive — An AI-driven content testing site focused on real tool reviews
🔗 Recommended Tools
These are carefully selected tools. Using our affiliate links supports us to keep producing quality content:
As an Amazon Associate, I earn from qualifying purchases at no extra cost to you.