How and Why to Read Data Sheets for 3D Printing Filaments
A filament datasheet is a test report, not a spec sheet. Read it wrong and you pick a material that prints well but fails in service. This guide shows engineers which lines decide the part, and which lines are noise.
In this article
- 1
- 2
- 3
- 4
- 5
- 6
Datasheet Warning Signs and What to Do
Match your symptom in the left column, confirm the cause, then apply the fix.
| Symptom | Likely cause | What to do |
|---|---|---|
| Part splits along layer lines | Z tensile far below XY tensile | Reorient the load; treat Z values as the design limit |
| Ductility looks fine on paper | Elongation quoted at 5 mm/min, not 50 mm/min | Check test speed; fast pulls flatter brittle parts |
| Stiffness drops in service | Modulus quoted dry; nylon absorbs moisture | Dry to the datasheet spec before printing |
| Warping on large flat parts | High mold shrinkage, low fill content | Use a heated chamber or a lower-shrink grade |
| Part softens near a heat source | HDT quoted at 0.45 MPa, no load stated | Compare at the load your part actually sees |
What Data Sheets for 3D Printing Filaments Actually Report
A filament datasheet is built from tests on printed specimens, not on the raw pellet. That single fact explains most of the confusion. Two vendors selling the same generic PLA can publish tensile strengths 20% apart, because one printed a flat bar and the other printed it upright. Neither is lying. They measured different things.
Start with the test standard column, not the value column. ISO 527 and ASTM D638 are both tensile tests, but specimen geometry and pull speed differ. If one sheet says ISO 527-2 and another says ASTM D638, the numbers are not directly comparable. Ask the vendor which standard and which build orientation were used before you compare anything.
Then find the moisture and drying line. Most engineering filaments, especially PA and PC, absorb water from the air. A datasheet listing tensile strength of 70 MPa usually means the specimen was dried first. Print the same spool after two weeks on an open shelf and you will not reach that number.
Only after those three checks should you look at the headline strength figure. It is the least useful number on the page for a real part, because it is measured in the strongest direction of the print.
Which Values Predict Real Part Behavior
Z-axis tensile strength is the number most datasheets bury or omit. In FDM, the bond between layers is weaker than the extruded strand itself. A filament rated at 50 MPa in XY may reach only 20 to 30 MPa in Z. If your load pulls layers apart, the XY number is irrelevant.
Heat deflection temperature (HDT) comes with a load condition attached. A value quoted at 0.45 MPa is not the same as one quoted at 1.8 MPa. Many sheet summaries list only the higher number because it looks better. A part that sits near a motor or in a hot enclosure needs the value at the load it actually carries.
Glass transition temperature (Tg) tells you when the material starts to soften, not when it melts. For amorphous plastics like ABS and PC, service temperature should stay well below Tg. For semi-crystalline materials like PA and PEEK, HDT sits closer to the melting range, which is why they hold up better in warm environments.
Mold shrinkage predicts warping before you print. A shrinkage of 1.5% on a 200 mm part means roughly 3 mm of movement during cooling. Low-shrink grades with fillers cut that number, but they also cut elongation. There is always a trade.
Why Test Speed and Specimen Shape Change the Verdict
Elongation at break is the most abused value on a filament datasheet. Pull a specimen slowly and a ductile polymer stretches far before it snaps. Pull it fast and the same material breaks short. ISO 527-2 and ASTM D638 allow several speeds, commonly 5 mm/min and 50 mm/min. A sheet quoting 50 mm/min will show lower elongation than one quoting 5 mm/min for the same polymer.
Specimen type matters too. ASTM D638 Type I and Type V differ in gauge length and width, so strain values shift. If you are comparing two vendors and one uses Type I while the other uses Type V, you cannot rank them by elongation alone.
Flexural modulus is often quoted where tensile modulus is missing. They are not the same test. Flexural values usually run higher because the outer fibers carry the load in bending. Use flexural modulus for stiffness estimates on a part in bending, and tensile modulus when the part is in tension.
Impact strength, when present, is usually Charpy or Izod. These are notched tests, so the notch geometry dominates the result. A high impact number does not mean the part survives a drop. It means a notched bar survived a swing.
Where Data Sheets Stop Being Useful
Datasheets describe a material, not your geometry. Layer height, nozzle temperature, chamber temperature, and print speed all change the outcome. A 0.1 mm layer bonds differently from a 0.3 mm layer, even with identical filament. No filament datasheet can capture that.
Chemical resistance tables are the weakest section on most sheets. They usually list a handful of common fluids at room temperature and 24-hour exposure. Real service involves temperature, stress, and time together. Stress cracking under a solvent at 60 °C will not show up in a 24-hour room-temperature soak.
Creep is rarely reported at all. If a part holds a constant load for months, the material will slowly deform even below its yield point. Without creep curves you are guessing. For long-term loaded parts, plan a test coupon instead of trusting a single tensile number.
UV and weathering data are also thin. Outdoor parts need accelerated weathering results, and most filament datasheets simply do not include them. Treat an indoor-only rating as indoor-only.
Step by Step: From Datasheet to Print Decision
Run these in order. Each step filters the material list further.
- 1List your real loads and temperaturesWrite down the direction of each load, the peak temperature, and the duration. A bracket in tension at 60 °C for years needs different data than a cover in compression at 25 °C.
- 2Confirm the test standardsCheck whether tensile data is ISO 527-2 or ASTM D638, and note the pull speed. Reject any comparison between sheets that use different standards unless you can convert.
- 3Find the Z-axis valuesIf the sheet lists only XY tensile, ask the vendor for Z data. If neither exists, plan a test print before committing to a production run.
- 4Check HDT at your loadMatch the HDT load condition to your application. A 0.45 MPa value is fine for light covers, not for structural brackets near heat.
- 5Read drying requirementsNote the drying temperature and time, for example 80 °C for 4 to 6 hours for PA. Print only after the filament is dry; wet filament shows up as popping and weak layers.
- 6Verify shrinkage against part sizeMultiply the shrinkage percentage by your longest dimension. If the movement exceeds your tolerance, add ribs, adjust orientation, or pick a lower-shrink grade.
- 7Build a small test couponPrint the actual load case and test it. This catches orientation effects, moisture, and creep that the datasheet cannot describe.
Frequently Asked Questions
Are datasheet values guaranteed for my printed part?
No. They describe a specimen printed under controlled conditions. Your printer, nozzle, layer height, and chamber temperature will shift the result. Treat the sheet as a starting filter and confirm with a test coupon.
Why do two PLA sheets show different tensile strength?
Different test standards, specimen types, pull speeds, or build orientations. Also, some values come from injection-molded specimens rather than printed ones. Injection-molded bars test noticeably stronger, so check the specimen description.
Is higher tensile strength always better?
No. Stiffer, stronger filaments usually break with less elongation. A part that needs to absorb impact or snap into place may perform better with a lower-strength, higher-elongation material.
How much does moisture change the numbers?
For PA and PC it can be significant. Wet filament often prints with visible popping and poor layer bonding, which lowers measured strength well below the dry datasheet value. Dry to the vendor spec and store the spool sealed.
What if the datasheet omits Z-axis data?
Ask the vendor directly. If they cannot provide it, print a test bar in the upright orientation and test it yourself. For any part loaded across layers, this step is not optional.
Should I trust chemical resistance tables?
Use them as a first screen only. They usually cover short exposures at room temperature. Combine temperature, stress, and long duration in your own test if the part contacts chemicals in service.
Need a Second Opinion on Your Filament Choice?
Send us the datasheet and the load case. Our engineers will tell you whether the material fits, or whether CNC machining or another process suits the part better.
12-hour quote100% inspectionNo minimum order quantity
More engineering notes
We publish setup notes, tooling trials and inspection data from the factory floor.
