# Ohmforge Transient Simulation

> Simulate a circuit over time: waveforms, Vpp/RMS/frequency, CSV and a chart. Runs a transient (time-domain) simulation of a SPICE netlist with trapezoidal or backward-Euler integration, adaptive time steps and breakpoints at source corners. Returns resampled waveforms of the chosen nodes, oscilloscope-style measurements (min, max, Vpp, mean, RMS, frequency, duty cycle, rise time), rating warnings, a chart and CSV.

Part of usefulweb.site (https://usefulweb.site/). Results are deterministic calculations from the inputs and the listed assumptions; the API returns exactly what the web page shows.

## Use it
- Human interface: https://ohmforgetransient.usefulweb.site/ (interactive tool at https://ohmforgetransient.usefulweb.site/tool)
- JSON API: POST https://ohmforgetransient.usefulweb.site/api/v1/calculate with a JSON object of inputs (schema: https://ohmforgetransient.usefulweb.site/api/v1/tool)
- OpenAPI: https://ohmforgetransient.usefulweb.site/openapi.json
- MCP (Streamable HTTP): https://ohmforgetransient.usefulweb.site/mcp - tool `ohmforge_transient`

## Inputs
- netlist (textarea): SPICE netlist - Subset: R C L V I D Q M E G K, .model, SIN/PULSE/PWL sources, engineering suffixes (k, m, u, n, p, meg). Node 0 is ground. Leave empty when you pass an Ohmforge project instead.
- project (textarea): Ohmforge project JSON (optional) - Alternative to the netlist: a .ohmforge.json project as saved by the app (format "ohmforge", version 1).
- tstop (number): Stop time - Simulated duration. Empty = use the .tran line of the netlist (or the project's transient settings).
- tstep (number): Output step - Spacing of the returned samples (also the largest internal step). Empty = .tran value or stop time / 1000.
- method (select): Integration
- uic (boolean): Start from initial conditions (skip the operating point) - Capacitors start at their IC (default 0 V). Needed for oscillators.
- probes (list): Nodes to return - Node names; empty = all nodes (max 8 in the chart).

## Method
- Starts from the DC operating point (or from initial conditions with "uic"), then integrates capacitors and inductors with companion models: trapezoidal (2C/h) or backward Euler (C/h); inductors and coupled inductors use branch currents.
- Each time step is solved with Newton–Raphson. The step size adapts: the difference between the corrector and a linear predictor estimates the local truncation error; failed steps are retried with h/8.
- Source corners (PULSE edges, PWL points) are breakpoints: the integrator lands on them and restarts with backward Euler to avoid trapezoidal ringing. Behavioural events (555 latch, relay pull-in, fuse, current limit) are localised by step bisection.
- Returned samples are linearly interpolated onto a uniform grid (output step). Measurements: time-weighted mean and RMS, Vpp, frequency from mid-level crossings with 10 % hysteresis, 10–90 % rise time.

## Limits
- Teaching-grade simulation: device models are simplified (Ebers–Moll BJT, level-1 MOSFET, behavioural op-amps and 555) and part ratings are typical values. Verify real designs on the bench and against manufacturer data before building anything connected to mains voltage.