Calibrating E-Steps for Different Filaments

This table shows when E-steps usually stay fixed and which filament-specific settings are more likely to need tuning.

Filament Family	Change E-Steps?	Typical Starting Temperature	Tune First	What Usually Changes the Result
PLA [e]	Usually keep fixed	210–215 °C nozzle, 60 °C bed	Flow, temperature, cooling	Low warp, easy feeding, visual overfill can look worse than it really measures
PETG [f]	Usually keep fixed	230–240 °C nozzle, 85–90 °C bed	Flow, retraction, temperature	Higher tack, stronger layer bonding, more stringing and glossy surfaces
TPU / TPE [g]	Keep fixed, verify only if feed slips	230–245 °C nozzle, 60–75 °C bed	Speed, retraction, idler pressure, drying	Soft filament deforms in the feed path and reacts strongly to speed
ASA [h]	Usually keep fixed	260 °C nozzle, 105–110 °C bed	Warm air around the print, flow, cooling	Warp and layer split can look like under-extrusion when the real issue is thermal control
Nylon / PA [i]	Usually keep fixed	285 °C nozzle, 110 °C bed	Drying, flow, warm enclosure	Moisture changes surface quality and feed behavior very quickly
Carbon / Glass Filled [k]	Usually keep fixed	240–285 °C nozzle, 70–110 °C bed	Nozzle choice, flow, layer height	Abrasion, clog risk, and nozzle wear often matter more than firmware feed values
LW PLA [l]	Keep fixed	210–250 °C nozzle, 55–60 °C bed	Temperature, speed, flow	Foaming changes line volume during printing

Most extrusion trouble that feels “filament-specific” is not a firmware E-step problem. In normal FDM printing, E-steps describe how far the extruder motor feeds filament for a commanded distance; once that hardware feed value is in the right zone, material changes are usually handled with flow tuning, temperature, speed, retraction, drying, or nozzle choice rather than by rewriting the printer’s E value every time you swap spools [a].

• Hardware Feed
• Flow Rate
• Temperature
• Retraction
• Moisture
• Nozzle Wear
• Filament Diameter

🔧 What E-Steps Actually Control

In Marlin, E-steps are the extruder’s steps-per-unit value. That means the firmware is deciding how many motor steps are used to move a commanded length of filament. Marlin also ties that setting to persistent storage through EEPROM, so once the value is set correctly and saved, it should stay stable until something mechanical changes in the extruder system [a].

Klipper uses a different name for the same job. Instead of E-steps, it asks for rotation distance, which is the amount of filament moved per full motor rotation. The useful part is the idea, not the label: both firmwares are calibrating the feed mechanism, not the chemistry of PLA, PETG, TPU, or Nylon [b].

There is one nuance worth keeping. Real feed behavior can drift a little from spool to spool because grip on the filament is not perfectly identical every time, especially when the material is soft or inconsistent. Still, that small drift is usually handled faster with a filament profile than with a new firmware feed value for every spool [b].

🧪 How to Calibrate Marlin and Klipper

For a trustworthy measurement, use dry filament, a heated hotend, and a slow feed rate. Klipper’s own procedure is explicit about this: heat the hotend, measure carefully with calipers, and extrude slowly enough that pressure in the nozzle does not distort the result. That detail matters. A fast push can make a healthy extruder look wrong [b].

Marlin Workflow

1. Heat the nozzle to a realistic printing temperature for the filament used in the test.
2. Use a caliper and a clear marker line so your start and finish points are easy to read.
3. Extrude a known amount slowly.
4. Measure how much filament actually moved.
5. Apply the proportional formula above.
6. Write the corrected value with M92 E..., then save it with M500.
7. Repeat once. If the second pass lands very close, stop there.

Cold-extrude shortcuts exist in Marlin through M302, but a warm-nozzle calibration usually matches real print conditions much better because the filament is moving under normal melt resistance, not under an artificial bench condition [b].

Klipper Workflow

Klipper documents a very clean method: mark the filament around 70 mm from the extruder intake, measure it with calipers, then extrude 50 mm at G1 E50 F60 after switching to relative movement. The new value is calculated as rotation_distance = previous_rotation_distance × actual_extrude_distance ÷ requested_extrude_distance. Klipper also warns against using the graphical “extrude” button for this because it usually runs too fast for a clean test [b].

🧵 Why Filaments Vary Even When E-Steps Stay Fixed

Filament Diameter and Roundness

Diameter tolerance changes volume. That part is simple geometry. Prusament states that its standard tolerance is usually ±0.02 mm for most materials and ±0.03 mm for blended materials, while also noting that ±0.05 mm is a common industry number. On 1.75 mm filament, a swing from 1.75 to 1.77 mm changes cross-sectional area by about 2.3%; a swing to 1.80 mm is closer to 5.8%. So two spools can behave differently even when the motor feed is identical [d].

Moisture Changes the Print Before Firmware Changes the Feed

Moisture is another reason a spool can look “miscalibrated.” Prusa notes that some materials shrug off humidity better than others, while highly hygroscopic materials need much tighter storage habits. Polyamide is heavily affected, and even materials that often seem easy can start stringing or laying down inconsistent lines once they pick up enough moisture [j].

Nozzle Condition and Abrasion

Filled materials can wear hardware faster than many users expect. Carbon-, glass-, and kevlar-filled filaments need a hardened nozzle, and Prusa also notes a higher risk of clogging with these materials. If line width drifts after a run of abrasive filament, inspect the nozzle before touching E-steps. A worn or partially restricted nozzle can mimic under-extrusion very convincingly [k].

🧩 Tuning PLA and PETG

PLA

PLA is usually the easiest place to set a baseline. Prusa’s material page puts standard PLA around 215 °C for the first layer, 210 °C after that, with a 60 °C bed. Because PLA is easy to feed and not very prone to warp, it is a good material for the initial feed check. After that, small print-to-print differences are better handled with flow adjustment and temperature tuning than with repeated firmware edits [e].

PETG

PETG often fools people into chasing the wrong setting. Prusa lists PETG around 230 °C for the first layer, 240 °C after that, with an 85–90 °C bed. It bonds strongly, has a glossy surface, and is more tenacious than PLA, so its over-extrusion signs can look different. In practice, PETG usually wants a different material profile from PLA, not a different E-step value from the same extruder [f].

PETG profiles often end up with small flow and retraction changes because the material stretches more and sticks to itself differently. When PETG strings or leaves thick seams, start with temperature, flow, and retraction. Leave the firmware feed number alone unless the slow measured extrusion test says otherwise.

⚙️ Tuning TPU, ASA, and Nylon

TPU and Other Flexible Filaments

Flexible materials are where people most often assume they need “special E-steps.” Sometimes the printer really is slipping a little. More often, the material is simply soft enough that speed, retraction, idler pressure, and moisture dominate the behavior. Prusa recommends printing flex materials slowly, often around 20 mm/s, with 30–40 mm/s as an upper practical range, plus gentle gear pressure and lower retraction. That is a feed-path control problem far more than a firmware-number problem [g].

If TPU bunches near the gears or shows erratic feed, run one careful low-speed measurement to confirm the extruder is still close. After that, move back to speed, retraction, and drying. Soft materials can make the extruder look inconsistent even when the underlying calibration is still fine.

ASA

ASA prints hot and wants stable thermal conditions. Prusa places it around 260 °C nozzle temperature and 105–110 °C bed temperature, with higher ambient warmth helping a lot. If ASA corners lift or layers separate, that can look like under-extrusion from a distance, but the first suspect is usually thermal control around the part, not a fresh E-step number [h].

Warm, steady air matters here. Use flow changes for measured wall-thickness errors. Use enclosure warmth, bed adhesion strategy, and cooling restraint for warp-related defects. Those are different problems, and mixing them together wastes time.

Nylon / Polyamide

Nylon is where moisture can wreck a calibration session. Prusa lists polyamide around 285 °C at the nozzle and 110 °C on the bed, while also noting that it can absorb water up to 10% of filament weight if stored badly. Their Nylon page also recommends drying before printing, for at least 4 hours below 90 °C. So if one spool suddenly gives rough walls, bubbles, or uneven feed, dry it first. Do not rush to rewrite E-steps [i].

🧠 Filled Filaments and LW PLA Need a Different Mindset

Carbon-, Glass-, and Kevlar-Filled Filaments

Filled blends are not just “the same material, but stronger.” Prusa notes that carbon-, glass-, and kevlar-filled filaments can improve dimensional stability and sometimes heat resistance, but they also need a hardened nozzle and have a higher clog risk. Their material note also points out a lower risk of clogging with a 0.4 mm or larger nozzle and a 0.2 mm layer height or higher. In other words, the printer often needs hardware readiness more than a new E-step profile [k].

LW PLA

LW PLA makes the separation between feed calibration and material behavior very obvious. Prusa says LW PLA can expand roughly 2.3 to 3 times its original size during printing, and that the expansion depends on temperature and speed. That means wall thickness and volume shift because the filament is foaming inside the melt zone. Changing E-steps to chase that behavior usually creates more confusion than clarity. Temperature, speed, and flow are the right levers here [l].

🛠️ Problems That Look Like Wrong E-Steps

This table separates common “wrong E-step” symptoms from the checks that usually solve them faster.

What You See	More Likely Cause	Change First
PLA prints fine, TPU does not	Soft filament path issues, speed too high, retraction too aggressive	Slow down, reduce retraction, ease idler pressure
One Nylon spool looks foamy or rough	Moisture in the filament	Dry the spool and re-test
PETG seams look fat and shiny	Flow or temperature a little high	Trim flow or nozzle temperature
Carbon-filled filament starts under-extruding mid-roll	Nozzle wear or partial clog	Inspect nozzle condition before firmware edits
LW PLA walls come out thick	Foaming expansion at current temperature and speed	Lower temperature or adjust flow profile
Only one brand runs lean or rich	Diameter tolerance, moisture, or profile mismatch	Measure, dry, and tune flow before E-steps

📏 Calibration Order That Saves Time

1. Check the nozzle, drive gear, idler pressure, and filament path.
2. Use dry filament at a realistic print temperature.
3. Calibrate the extruder feed once with a slow measured test.
4. Save that value in firmware or configuration.
5. Create separate material profiles for flow, temperature, speed, cooling, and retraction.
6. Return to E-steps only after a hardware change or a repeatable measured feed error.

That order keeps hardware calibration on one side and filament behavior on the other. It is cleaner. It is also much easier to troubleshoot later when a spool acts strange.

❓ FAQ

Do PLA and PETG Need Different E-Steps?

Usually no. On the same extruder, PLA and PETG normally share the same hardware feed calibration. The better place to separate them is the filament profile: flow, temperature, retraction, cooling, and sometimes speed.

Why Does TPU Sometimes Feel Like It Needs Its Own E-Step Value?

Because soft filament can compress, drag, or tangle more easily in the feed path. A slow measured test can confirm whether the extruder is still close, but day-to-day TPU tuning is usually handled with speed, retraction, idler pressure, and drying.

Is Flow Rate the Same as E-Steps?

No. E-steps or rotation distance calibrate the printer’s feed mechanism. Flow rate or extrusion multiplier tunes how a filament profile behaves during printing. They work together, but they are not the same control.

Can Filament Diameter Really Change Extrusion That Much?

Yes. Small diameter changes alter cross-sectional area, which changes how much plastic is pushed for the same feed length. That is why one spool can run slightly rich or lean even when the extruder motor is feeding the same commanded distance.

Should I Calibrate Through the Hotend or With Cold Extrusion?

A heated, slow test is usually closer to real print conditions because the filament is moving under normal melt resistance. Cold-extrusion tricks can be useful for narrow bench checks, but they are not the most faithful way to judge real printing behavior.

What Should I Change First if Only One Spool Prints Badly?

Check moisture, diameter consistency, temperature, flow, and nozzle condition before touching firmware feed values. A single bad spool or a single specialty filament profile is often a material problem, not an extruder-calibration problem.

Reference Notes

• [a] Marlin Firmware — M92: Set Axis Steps-per-unit — Used for the definition of E-steps and the note that Marlin stores this setting with M500. (Trustworthy because it is the official Marlin documentation.)
• [b] Klipper Documentation — Rotation Distance — Used for the heated slow-feed measurement method, the 50 mm calibration move, the warning about high-speed extrusion during testing, and the rotation-distance formula. (Trustworthy because it is the official Klipper documentation.)
• [c] Prusa Knowledge Base — Extrusion Multiplier Calibration — Used for the wall-thickness method and the idea that material-specific print behavior should be tuned with extrusion multiplier rather than by constantly rewriting hardware feed values. (Trustworthy because it is an official Prusa support document.)
• [d] Prusament — How It’s Made — Used for diameter tolerance details, two-axis laser measurement, and the comparison between tighter tolerance and the more common ±0.05 mm range. (Trustworthy because it is the manufacturer’s own production-process page.)
• [e] Prusa Knowledge Base — PLA — Used for PLA temperature ranges, easy-print behavior, low warp tendency, and practical print characteristics relevant to baseline feed tests. (Trustworthy because it is an official material reference from a major printer manufacturer.)
• [f] Prusa Knowledge Base — PETG — Used for PETG temperature ranges, stronger layer adhesion, and general print behavior that often calls for profile tuning rather than E-step changes. (Trustworthy because it is an official material reference from Prusa.)
• [g] Prusa Knowledge Base — Flexible Materials — Used for TPU/TPE temperature ranges, slow print-speed guidance, gentle idler pressure, retraction advice, and moisture sensitivity. (Trustworthy because it is an official vendor support article focused on flexible filaments.)
• [h] Prusa Knowledge Base — ASA — Used for ASA temperature ranges and the need for warmer, more stable print conditions. (Trustworthy because it is an official material page from Prusa.)
• [i] Prusa Knowledge Base — Polyamide (Nylon) — Used for Nylon temperature ranges, moisture behavior, and drying recommendations before printing. (Trustworthy because it is an official material page from Prusa.)
• [j] Prusa Knowledge Base — Drying Filament — Used for storage and drying context, including the difference between materials that tolerate humidity better and those that do not. (Trustworthy because it is an official maintenance and materials reference.)
• [k] Prusa Knowledge Base — Composite Materials Filled With Carbon, Kevlar, or Glass — Used for hardened-nozzle requirements, abrasion, clog risk, and practical nozzle and layer-height notes for filled filaments. (Trustworthy because it is an official material reference page.)
• [l] Prusa Knowledge Base — LW PLA — Used for the foaming-expansion behavior of LW PLA and the way temperature and speed affect output volume. (Trustworthy because it is an official material page from Prusa.)

{ “@context”: “https://schema.org”, “@type”: “FAQPage”, “mainEntity”: [ { “@type”: “Question”, “name”: “Do PLA and PETG need different E-steps?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “Usually no. On the same extruder, PLA and PETG normally share the same hardware feed calibration. Separate them with filament-profile settings such as flow, temperature, retraction, cooling, and speed.” } }, { “@type”: “Question”, “name”: “Why does TPU sometimes feel like it needs its own E-step value?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “Because soft filament can compress, drag, or tangle more easily in the feed path. A slow measured test can confirm whether the extruder is still close, but daily TPU tuning is usually handled with speed, retraction, idler pressure, and drying.” } }, { “@type”: “Question”, “name”: “Is flow rate the same as E-steps?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “No. E-steps or rotation distance calibrate the printer’s feed mechanism. Flow rate or extrusion multiplier tunes how a filament profile behaves during printing.” } }, { “@type”: “Question”, “name”: “Can filament diameter really change extrusion that much?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “Yes. Small diameter changes alter filament cross-sectional area, which changes how much plastic is pushed for the same feed length. One spool can run slightly rich or lean even when the extruder motor is feeding the same commanded distance.” } }, { “@type”: “Question”, “name”: “Should I calibrate through the hotend or with cold extrusion?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “A heated, slow test is usually closer to real print conditions because the filament moves under normal melt resistance. Cold-extrusion checks can be useful for narrow bench tests, but they are not the most faithful way to judge real printing behavior.” } }, { “@type”: “Question”, “name”: “What should I change first if only one spool prints badly?”, “acceptedAnswer”: { “@type”: “Answer”, “text”: “Check moisture, diameter consistency, temperature, flow, and nozzle condition before touching firmware feed values. A single bad spool or a specialty filament profile is often a material problem, not an extruder-calibration problem.” } } ] }