10kΩ NTC Thermistor (Beta 3950)
The de facto standard temperature sensor in 3D printer hotends and beds, chosen for its steep beta-3950 resistance curve that gives precise, sensitive temperature readings over the 0-250°C range typical of those applications.
How this component works
A thermistor is a special type of resistor whose electrical resistance changes significantly with temperature. They come in two primary types: NTC (Negative Temperature Coefficient), where resistance drops as it gets hotter, and PTC (Positive Temperature Coefficient), where resistance rises as it gets hotter.
NTC thermistors are incredibly common as cheap, robust temperature sensors in 3D printers, digital thermostats, and battery charging circuits. They are highly non-linear, so converting their resistance into a precise Celsius temperature reading requires solving a complex logarithmic equation (the Steinhart-Hart equation) in software.
Deep Dive: 10kΩ NTC (B3950)
The beta value is what actually distinguishes this thermistor from the other 10kΩ NTC entry in this category, and the difference is large enough to matter in practice, not just on paper. Using the standard beta-equation approximation R(T) = R0 × exp(B × (1/T − 1/T0)), with R0 = 10,000Ω at T0 = 298.15K (25°C) and B = 3950: at 100°C (373.15K), the term B×(1/T − 1/T0) works out to roughly −2.66, and exp(−2.66) ≈ 0.070, giving R(100°C) ≈ 700Ω — a roughly 14-fold drop from the 25°C value. That steep, well-behaved curve over the 0-250°C range is exactly why beta-3950 10kΩ thermistors became the standard sensor bundled with 3D printer hotend and bed heaters, and why firmware like Marlin ships default thermistor tables built around this specific beta value.
The practical consequence is that beta value, not just nominal 25°C resistance, has to match between a thermistor and whatever lookup table or equation is reading it — a project calibrated assuming B3950 will misread temperature if a B3435 part (see the sibling entry) is substituted, because the two produce meaningfully different resistance at the same real temperature once you're away from the shared 25°C reference point.
Common Use Cases
- 3D printer hotend and heated-bed temperature feedback for firmware PID control loops
- General consumer appliance temperature sensing (e.g. a coffee maker or space heater cutoff)
- Battery pack or motor temperature monitoring where a simple analog voltage-divider reading is sufficient
Frequently Asked Questions
How much does this thermistor's resistance actually change between room temperature and a hot 3D-printer nozzle?
Using the beta equation R(T) = R0 × exp(B×(1/T − 1/T0)) with R0 = 10,000Ω, B = 3950, and T0 = 298.15K (25°C): at 100°C (373.15K) the resistance drops to roughly 700Ω — about a 14x drop from its 25°C value, illustrating how steep the NTC curve is.
Why is beta 3950 specifically the common choice for 3D printers rather than a lower beta value?
A higher beta value means a steeper resistance-vs-temperature curve, which gives more resolution (a bigger resistance change per degree) in the temperature range hotends actually operate in — useful for tight PID temperature control where small deviations matter.
Can I substitute a beta-3435 thermistor (see the other entry in this category) in a project calibrated for beta 3950?
Not without recalibration. The two curves diverge meaningfully at higher temperatures — at 100°C, a B3950 part reads roughly 700Ω while a B3435 part of the same 25°C nominal value reads roughly 990Ω — so firmware or lookup tables built for one will misread temperature with the other.
Comparison Notes
Both this part and the B3435 entry in this category are nominally "10kΩ NTC thermistors," but their beta values give them meaningfully different steepness: this B3950 part is the sharper, more sensitive curve favored in modern 3D-printer firmware, while B3435 is an older, more gradual industrial standard still common in legacy designs.