95A vs 85A TPU: Which Shore Hardness Your Flexible Part Actually Needs
The Shore A number is the first thing you see on a TPU spool, but it does not tell you how your part will behave once printed. This guide compares a 95A and an 85A using the Prusament and Bambu Lab datasheets, explains how geometry changes the result, and lists which extruders, nozzles and multi-spool systems actually accept each hardness.
The short answer
Shore A hardness describes indentation response; it does not by itself predict the flexibility of a printed part. Compare an 85A and 95A TPU with their formulation, geometry, walls and infill. The sheets considered here concern Prusament TPU 95A and Bambu TPU 85A. Bambu recommends 0.6 or 0.8 mm nozzles for its 85A and excludes the A Series, AMS, AMS lite and AMS 2 Pro; AMS HT is listed as compatible subject to feeding instructions. Manufacturer use examples can help select a prototype but do not guarantee cushioning or wear resistance for a particular part.
What Shore A hardness measures, and what it leaves out
Shore A hardness measures a specimen's response to an indenter under a defined method. Under comparable conditions, a lower number indicates softer material. An 85A or 95A rating therefore helps select a filament, but describes neither its full formulation nor the deformation of a printed part. Keep the product name and measurement method with the number: different polymer formulations or geometries cannot be reduced to their hardness gap.
That number describes the raw material, not your print. A thin 95A strap bends easily, while a solid 85A block still feels fairly stiff. Prusament's own page makes the point for its TPU 95A: soft and bendable at thin layers, strong and durable at thick layers. Treat hardness as a starting point for choosing a family of filament, not as a forecast of how the finished part will feel in your hand. The TPU filament guide covers the other properties these materials share, from wear resistance to moisture sensitivity.
95A TPU: the firm, wear-resistant option in Prusament's datasheet
Prusament TPU 95A is a good example of what a firm TPU is for. The datasheet lists mechanically loaded parts: O-rings, spacers, rugged housings and pieces of sports equipment. Prusament also calls out durable grips, sharp-edge covers for skis and skates, cable protectors and cleaning air blowers. What ties these uses together is wear and impact resistance rather than a squishy feel.
The page explains that this filament uses an ether diol chemistry, while most TPU filaments are ester diol types. Prusament credits that choice with very low moisture absorption, high hydrolysis resistance and good resistance to oil and grease. Elasticity and impact resistance are stated to hold down to -50 °C, and heat deflection temperature is listed at 78.6 °C under 1.80 MPa. For print settings, the datasheet gives a nozzle temperature of 230 ± 10 °C and a bed of 65 ± 10 °C, with no enclosure required. Drying is usually unnecessary, but if a spool has sat in humid conditions Prusament recommends at least 4 hours at 60 °C followed by storage in a dry box. The datasheet also notes that some extra tweaking may be needed depending on your printer, and that a smooth sheet needs a separation layer.
85A TPU: the soft, cushioning option on Bambu Lab's product page
Bambu Lab sells its TPU 85A alongside a TPU 90A, and the product page describes 85A as the softest TPU in the lineup. The listed uses are impact absorption and soft support: padding, protective cases, squishy fidget toys and cushioning elements. The slightly firmer 90A is aimed at shoe soles, RC tires and functional parts that need both strength and elasticity. Bambu Lab stresses wear resistance after repeated bending for both grades.
The same page carries a plain warning: because 85A and 90A are so soft, Bambu Lab asks you to read the wiki printing guide before buying. All parameters come pre-set in Bambu Studio, and the company tells you to turn off Flow Dynamics Calibration and keep the default maximum volumetric speed. For unloading, the page recommends heating the nozzle to 250 °C, retracting with the extruder and pulling the filament out by hand, since repeated automatic unloads can leave uncut TPU stuck in the extruder. If you print TPU often, Bambu Lab suggests keeping an eye on the filament cutter at the right inlet, which wears faster with this material. The page also states that the filament is not food grade.
Walls and infill: compare the part as well as hardness
Ten points separate 85A and 95A on the Shore A scale. On a printed part, wall count, infill and solid layers also affect mechanical response. Prusa recommends increasing infill and perimeters for strength, and reducing perimeters and solid layers to preserve flexibility. This does not quantify their effect against a ten-point hardness difference. To compare two spools, record geometry and profile, print a prototype suited to the intended load, then observe its deformation.
Distinguish a part that needs to bend from one that needs to compress. A cable protector and a cushion call for different motion; manufacturer use examples can suggest prototype candidates. They do not establish that one hardness suits every hinge or vibration foot. Prusa also flags bridging and overhang difficulties with flexible filaments. Its support suggestions, including a 0.3 mm Z gap and XY separation of at least 100 percent, remain starting points to check against the relevant profile and geometry.
Feed path: which extruders, nozzles and AMS units accept each hardness
Softness costs you at the loading stage. Prusa's guide puts it simply: the softer the filament, the harder it is to print, because it twists into the drive gears, clogs the nozzle and strings. Both manufacturers therefore publish specific compatibility lists. For TPU 85A, Bambu Lab lists the X Series, P Series and H2S, plus the right filament inlet only on the H2D and H2C. The A Series is not supported. The firmer 90A does work on the A Series. The AMS, AMS lite and AMS 2 Pro are listed as not compatible with either grade. Only the AMS HT is, and Bambu Lab points you to the wiki for feeding instructions.
On nozzles, the Bambu Lab page recommends 0.6 mm and 0.8 mm for 85A and rules out 0.2 mm, 0.4 mm and the high flow nozzle. The 90A adds 0.4 mm to the list. The guide to 3D printer nozzles helps you weigh detail against reliability. On the Prusa side, Prusament describes its TPU 95A as very close to failproof with the Nextruder, with profiles for the Core One, XL, MK4S, MK4 and HT90. The documentation still flags that the XL's long path needs special attention, that on the Core One you may need to unplug the PTFE tube and push the filament through by hand, and that the CORE One+ and CORE One L have a latch that loosens the side filament sensor. On the MK3S+ and older machines, loosening the idler is advised.
How to read a TPU datasheet before you buy a spool
A TPU datasheet reads in three blocks. The first describes the material: Shore A hardness, ether or ester chemistry, temperature resistance, moisture absorption. The second gives recommended settings: nozzle and bed temperatures, build surface, drying. The third, often the most useful, lists compatible printers, nozzle sizes and multi-spool systems. That third block decides whether the spool will actually run on your machine.
Prusament recommends the PA Nylon or Satin sheet for its 95A, accepts PP and textured sheets, and requires a separation layer on Smooth PEI. Bambu Lab lists Smooth PEI and Textured PEI for its 85A and 90A. The PEI build plate guide helps identify the exact surface. Speed advice also needs its source: Prusa's general flexible guide cites 20 mm/s as typical and 30–40 mm/s as a maximum, while Bambu asks users to keep its presets. Do not transfer those values automatically to another spool. Finally, humid storage can justify a filament dryer under the product instructions; moisture alone does not explain every print defect.
Sources and limits
- Prusament TPU 95A | Prusament
- TPU 85A / TPU 90A | Bambu Lab EU Store
- Flexible materials | Prusa Knowledge Base
We organize manufacturer and community documentation; we have not measured these products ourselves. The manual for your exact model takes precedence over general guidance.
A clearer shortlist starts with your projects.
Use the free buying checklist and budget worksheet before comparing offers.