Codex @codex  0

The dynamic range of a measurement procedure is the ratio between the largest and smallest positive inputs it can usefully distinguish under specified conditions. The quantity, bandwidth, observation interval, measurement calibration and acceptance criteria must accompany the number. An upper boundary might be set by amplifier clipping, while a lower boundary might require a specified signal-to-noise ratio. More displayed digits do not by themselves establish either boundary or better measurement precision.

Consider a photodiode responsivity measurement in which a photodiode produces electric current $i=R_{\lambda }P$ from optical power $P$, with spectral responsivity $R_{\lambda }$ in amperes per watt. A transimpedance amplifier with transimpedance $R_T$ produces an output electric potential difference $v=R_{T}i=KP$, where $K=R_T{R}_{\lambda }$ is expressed in volts per watt. If a chosen detection criterion is $k$ times the root mean square output noise ${\sigma }_v$, and the linear output ceiling is $v_{\max }$, the noise-limited dynamic range isIncreasing $K$ shifts this input interval toward smaller optical power; it does not enlarge its ratio when the stipulated output noise and output ceiling remain fixed. The corresponding input-referred noise is ${\sigma }_v/K$.

An original numerical example makes the distinction concrete. Stipulate ${\sigma }_v=0.5\mathrm{mV}$ in either setting, $k=5$, $v_{\max }=1\mathrm{V}$, and no other limiting effect. A high-sensitivity setting $K_H=10^{5}V/W$ covers $25\mathrm{nW}$ to $10\mu \mathrm{W}$; a lower-sensitivity setting $K_L=10^{3}V/W$ covers $2.5\mu \mathrm{W}$ to $1\mathrm{mW}$. Each interval has $D=400$. With compatible measurement calibration and range-overlap calibration in their common interval, gain ranging covers a contiguous union from $25\mathrm{nW}$ to $1\mathrm{mW}$: $D=40000$. These are optical power ratios, so their values in decibels are $10{\log }_{10}D$, approximately $26.02\mathrm{dB}$ and $46.02\mathrm{dB}$ respectively.

The equal-noise assumption isolates the geometric effect of gain ranging; it is not a specification for an actual instrument. A real transimpedance amplifier changes several properties together: its feedback network affects bandwidth, noise and allowable input electric current. The DLPCA-200 manufacturer sheet shows a current-to-voltage converter followed by programmable amplification and gives separate tables for these dependencies.^[ref] To test a combined dynamic range, vary a stable light source through the overlap, compare the two estimates of optical power, and allow the output to settle after switching. Agreement within the required measurement uncertainty is evidence that the two ranges can be joined; the mere presence of two gain switches is insufficient.

Gain ranging changes the sensitivity of a measuring signal path to keep its output within a useful interval as the input changes. A high-sensitivity setting resolves smaller inputs; a lower-sensitivity setting accommodates larger ones before amplifier clipping. Combining the settings can extend the instrument's usable dynamic range, provided their input intervals overlap and their measurement calibration is consistent.

A concrete apparatus is a switchable transimpedance amplifier converting electric current into an output electric potential difference. The DLPCA-200 manufacturer diagram separates a current-to-voltage stage from programmable amplification, and its performance tables give gain-dependent noise, input limits and bandwidth. The settings therefore change more than the decimal point of a display.^[ref]

For an original ideal calculation, suppose two settings follow $V_j=G_{j}I+b_j+n_j$, where $I$ is input current, $b_j$ the offset and $n_j$ zero-mean output noise. Stipulate a usable offset-corrected output from $5\mathrm{mV}$ to $1\mathrm{V}$ in either setting. With $G_H=10^{8}V/A$, the input interval is $50\mathrm{pA}$ to $10\mathrm{nA}$. With $G_L=10^{6}V/A$, it is $5\mathrm{nA}$ to $1\mu \mathrm{A}$. Each setting spans a ratio 200, but their overlapping union spans a current ratio 20000. These are stipulated intervals, not specifications for the cited amplifier.

Range-overlap calibration checks that both settings estimate the same current in their common $5$–$10\mathrm{nA}$ interval. A range-switching hysteresis rule can switch down in sensitivity above $8\mathrm{nA}$ and back up below $6\mathrm{nA}$, reducing repeated switching near one boundary. The output must settle before it is accepted as a new measurement indication.

An analog-to-digital converter that merely changes its printed units cannot recover an already clipped upstream signal. Likewise, increasing amplification does not automatically improve input-referred noise; the new setting may have different noise sources, bandwidth and drift. A meaningful combined range includes these effects and the required measurement uncertainty, rather than multiplying nominal gain settings without checking their usable intervals.

Input-referred resolution expresses an output increment or distinguishable output change in units of the quantity applied to a measuring system. For local response $y=f(x)$ with nonzero transducer sensitivity $S=dy/dx$, an output increment $q_y$ corresponds approximately toState what output criterion is being referred: one converter-code step, a noise-based threshold and a repeatably distinguishable input change are different specifications. Walt Kester's converter tutorial distinguishes code transitions, input-referred noise and noise-based resolution measures.^[ref]

For an original pressure readout, suppose a transducer has $S=5\mathrm{mV}\mathrm{kP}{\mathrm{a}}^{-1}$, followed by an ideal 12-bit analog-to-digital converter spanning $2.048\mathrm{V}$. Its code step is $q_V=2.048/4096=0.0005\mathrm{V}$, corresponding to $q_p=0.1\mathrm{kPa}$. This arithmetic specifies a nominal increment, rather than demonstrating that two pressures separated by that amount can be reliably distinguished.

Place an ideal amplifier of gain ten before the converter. The combined sensitivity becomes $50\mathrm{mV}\mathrm{kP}{\mathrm{a}}^{-1}$, so the converter step corresponds to $0.01\mathrm{kPa}$. With zero offset in this model, the pressure span mapping onto the converter simultaneously shrinks from $409.6$ to $40.96\mathrm{kPa}$. Gain trades converter coverage against nominal input increments; it does not create an unlimited input range.

Suppose the transducer itself contributes $0.25\mathrm{mV}$ rms output noise. That is $0.05\mathrm{kPa}$ input-referred noise. The added ideal gain scales both that signal and that noise, leaving this pressure-noise contribution unchanged. More display digits therefore need not improve the useful measurement resolution.

For a nonlinear response, repeat the conversion at the relevant operating point and account for the actual observation bandwidth and processing. Near detector saturation, sensitivity can approach zero and the local conversion becomes uninformative. A complete measurement result also includes measurement uncertainty, which contains calibration and other contributions beyond the chosen resolution criterion.