This is a 2D companion to the 3D orbit-camera version of this simulator. It computes the exact same electrothermal actuator model — the coiled carbon-nanotube (CNT) yarn muscle heats resistively, untwists, and contracts axially, pulling a tendon that curls a three-phalanx prosthetic finger — but renders it as a fixed side-view mechanism plus a live strip-chart of the underlying signals, instead of an orbitable 3D scene.
Joule heating: P = V² / R
Thermal balance: C·dT/dt = P − h·(T − T_amb)
Actuation strain: s(T) = s_max·(1 − e^(−(T−T_amb)/Tc))
Load line: s_eff = s·max(0, 1 − F_load / F_stall)
Finger curl: θ = θ_max·(s_eff / s_max)
- Drive voltage — sets Joule heating power P = V²/R in the yarn.
- Grip load — an opposing force (an object being grasped) that reduces achievable contraction via the actuator's force–stroke trade-off, exactly like a real linear-actuator load line.
- Cooling airflow — scales the convective heat-loss coefficient h, controlling how fast the yarn (and the finger) relax back open once voltage drops. It also sets the thermal time constant τ = C/h, visible directly on the strip chart.
- Pulse mode — square-waves the voltage on/off every second instead of holding it, so you can compare the 1 s electrical period against the yarn's own (much slower) thermal response time on the chart.
Verified numerically (standalone Node script, not shipped): integrating the thermal ODE to steady state reproduces the closed form T_ss = T_amb + P/h exactly at every tested (V, cooling) pair, and the load-line/closure/force formulas stay internally consistent (e.g. force = closure·F_stall by construction) — the 3D source's own model is self-consistent, so this 2D port reuses its constants unchanged.
Real-world relevance: Haines et al. (2014, Science) demonstrated twisted-and-coiled nylon and CNT-yarn muscles generating over 100× the specific work of natural muscle — a candidate nanomaterial actuator for tendon-driven prosthetic fingers that need to be light, silent and compliant.