pulp::signal::FeedbackClipperT< SampleType > Class Template Reference

The op-amp gain stage with the diode network either at its output or across its feedback resistor. More...

#include <distortion.hpp>

Public Types
using	Model = DiodeModel
using	Topology = ClipperTopology

Public Member Functions
	FeedbackClipperT ()
void	prepare (double sample_rate)
void	set_topology (Topology topology)
Topology	topology () const
void	set_diode_model (Model model)
void	set_symmetry (double symmetry)
void	set_feedback_resistance (double ohms)
void	set_input_resistance (double ohms)
void	set_knee_corner_hz (double hz)
double	linear_gain () const
	The stage's linear (small-signal) gain, Rf/Rin — the bound the loop can never exceed at frequencies below the feedback capacitor's corner.
void	reset ()
int	last_iteration_count () const
SampleType	process (SampleType input)

Static Public Attributes
static constexpr double	kFeedbackResistanceDefault = 51000.0
	Feedback resistor: sets the linear gain Rf/Rin and, with Cf, the in-loop knee corner.
static constexpr double	kInputResistanceDefault = 4700.0
	Input resistor.
static constexpr double	kKneeCornerHzDefault = 720.0
	In-loop knee corner, fc = 1/(2π·Rf·Cf).

Detailed Description

template<typename SampleType = float>
class pulp::signal::FeedbackClipperT< SampleType >

The op-amp gain stage with the diode network either at its output or across its feedback resistor.

The in-loop KCL at the virtual ground reduces to the SAME functional form as the to-ground ODE with v ≡ −vout, R ≡ Rf, C ≡ Cf, plus a vin/Rin forcing term. So this reuses the identical TR-BDF2 + Newton solver rather than introducing a second numerical method.

Small-signal gain and its bound (series law 1). At v → 0 the diodes' incremental conductance is Is/(n·Vt) — for silicon about 3.9e-13 S, utterly negligible against 1/Rf — so at zero signal the diodes are effectively an open circuit and the small-signal gain is exactly the linear Rf/Rin. As |v| grows the diodes' conductance grows exponentially, shunting more feedback current and REDUCING loop gain monotonically. The loop is therefore self-limiting, never self-amplifying, and Rf/Rin bounds it above at every amplitude. That is the tested invariant the Forge registry cites.

Member Typedef Documentation

Model

template<typename SampleType = float>

using pulp::signal::FeedbackClipperT< SampleType >::Model = DiodeModel

Topology

template<typename SampleType = float>

using pulp::signal::FeedbackClipperT< SampleType >::Topology = ClipperTopology

Constructor & Destructor Documentation

FeedbackClipperT()

template<typename SampleType = float>

pulp::signal::FeedbackClipperT< SampleType >::FeedbackClipperT ( )

inline

Member Function Documentation

last_iteration_count()

template<typename SampleType = float>

int pulp::signal::FeedbackClipperT< SampleType >::last_iteration_count ( ) const

inline

linear_gain()

template<typename SampleType = float>

double pulp::signal::FeedbackClipperT< SampleType >::linear_gain ( ) const

inline

The stage's linear (small-signal) gain, Rf/Rin — the bound the loop can never exceed at frequencies below the feedback capacitor's corner.

Above that corner the capacitor shunts feedback current and the gain is lower, which is the topology's defining behaviour rather than a caveat.

prepare()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::prepare ( double sample_rate )

inline

process()

template<typename SampleType = float>

SampleType pulp::signal::FeedbackClipperT< SampleType >::process ( SampleType input )

inline

reset()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::reset ( )

inline

set_diode_model()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::set_diode_model ( Model model )

inline

set_feedback_resistance()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::set_feedback_resistance ( double ohms )

inline

set_input_resistance()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::set_input_resistance ( double ohms )

inline

set_knee_corner_hz()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::set_knee_corner_hz ( double hz )

inline

set_symmetry()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::set_symmetry ( double symmetry )

inline

set_topology()

template<typename SampleType = float>

void pulp::signal::FeedbackClipperT< SampleType >::set_topology ( Topology topology )

inline

topology()

template<typename SampleType = float>

Topology pulp::signal::FeedbackClipperT< SampleType >::topology ( ) const

inline

Member Data Documentation

kFeedbackResistanceDefault

template<typename SampleType = float>

double pulp::signal::FeedbackClipperT< SampleType >::kFeedbackResistanceDefault = 51000.0

staticconstexpr

Feedback resistor: sets the linear gain Rf/Rin and, with Cf, the in-loop knee corner.

[design parameter] default 51 kΩ, range 10 kΩ .. 200 kΩ.

kInputResistanceDefault

template<typename SampleType = float>

double pulp::signal::FeedbackClipperT< SampleType >::kInputResistanceDefault = 4700.0

staticconstexpr

Input resistor.

Rf/Rin is the stage's linear gain. [design parameter] default 4.7 kΩ, range 1 kΩ .. 100 kΩ.

kKneeCornerHzDefault

template<typename SampleType = float>

double pulp::signal::FeedbackClipperT< SampleType >::kKneeCornerHzDefault = 720.0

staticconstexpr

In-loop knee corner, fc = 1/(2π·Rf·Cf).

Above it the capacitor shunts feedback current around the diodes, so high harmonics see less clipping-loop gain — the mechanism behind the smoother, brighter-forgiving character of this topology. The default sits in the documented behavioural neighbourhood, not on any schematic. [design parameter] default 720 Hz, range 100 Hz .. 5000 Hz.

---

The documentation for this class was generated from the following file:

• /home/runner/work/pulp/pulp/core/signal/include/pulp/signal/distortion.hpp