Tool · Laser characterization

Wall-Plug Efficiency Calculator

What fraction of the electrical power delivered to this laser leaves it as light, and at what drive current is that fraction largest? From pasted LIV data (current, voltage, optical power) the calculator computes the wall-plug efficiency at each measured point, reports the highest value in the set, and plots efficiency against current. Background: wall-plug efficiency, LIV curves, and slope efficiency.

Data sets

The two example sets are generated, not measured: a straight L-I line above threshold and a voltage V₀ + R·I, with the parameters in the first comment line of each. Each preset also sets the wavelength; the default data take 1310 nm.

Input data

Three columns. Comma, space, or tab separated. Lines starting with # are ignored.

Peak WPE

η_WPE: 34.15 %
at I: 40.0 mA
V at peak: 1.230 V
P_out at peak: 16.80 mW
Points used: 10

Where the power goes, at the peak

Emission wavelength λ (nm)

Threshold I_th (mA)

Voltage factor hν/qV: 76.9 %
Threshold factor 1 − I_th/I: 81.7 %
Differential efficiency η_d: 54.3 %
Product: 34.15 % = η_WPE

The address holds the data rows, the wavelength and any entered threshold.

η_WPE = P_opt / (I · V) = (hν/qV) · (1 − I_th/I) · η_d, exactly, with η_d defined as above. Peak is the maximum of the discrete dataset. For data with significant scatter, smooth the LIV before computing WPE or fit a polynomial near the peak. Runs entirely in your browser. Nothing you enter is uploaded or stored.

Wall-plug efficiency vs. drive current

Wall-plug efficiency at each measured row, joined by straight segments; the open circle marks the peak reported in the results.

Computed values

I (mA)	V (V)	P (mW)	P_elec (mW)	WPE (%)
8.70	1.050	0.000	9.13	0.00
12.00	1.070	2.450	12.84	19.08
16.00	1.100	4.550	17.60	25.85
20.00	1.120	6.650	22.40	29.69
30.00	1.180	11.900	35.40	33.62
40.00	1.230	16.800	49.20	34.15
50.00	1.280	21.400	64.00	33.44
60.00	1.320	25.700	79.20	32.45
80.00	1.400	33.400	112.00	29.82
100.00	1.480	39.600	148.00	26.76

Methodology

For the full WPE definition, peak-finding methodology, and worked example, see the article Wall-Plug Efficiency Extraction from LIV Data. For the underlying slope efficiency, see Slope Efficiency Extraction.

The model

At each row of the LIV data the electrical input power is the product of drive current $I$ and terminal voltage $V$ , and the wall-plug efficiency is the measured optical power $P_{opt}$ divided by it:

P_{elec} = I V

η_{WPE} = P_{opt} / (I V)

With current in mA and voltage in V the product is in mW, so no unit conversion is needed when the power column is also in mW. The peak reported is the largest value among the rows entered; no curve is fitted, so the true maximum can lie between two measured currents.

The shape of the curve follows from two simple approximations. Above threshold the output is close to $P_{opt} \approx η_{s} (I - I_{th})$ , and the voltage is close to $V \approx V_{0} + I R_{s}$ . Near threshold the current that produces no light dominates and the efficiency is low; at high current the resistive term and the heating it causes lower the slope, and the efficiency falls again. The result is only as good as the power column: if it holds front-facet or fiber-coupled power, the efficiency is for that output, and it never includes the power drawn by a thermoelectric cooler or the driver.

The tool splits the efficiency at the peak into three factors whose product is exactly $η_{WPE}$ :

η_{WPE} = \frac{h ν}{q V} \cdot (1 - \frac{I_{th}}{I}) \cdot η_{d}, η_{d} = \frac{P_{opt}}{(h ν / q) (I - I_{th})}

The first factor is the voltage defect: of the energy $q V$ each electron delivers, at most $h ν$ leaves as a photon, and the rest is dropped in the series resistance and the heterobarriers. The second is the fraction of the current above threshold. The third is the differential quantum efficiency averaged from threshold to the operating current, the photons emitted per electron injected above threshold, which absorbs internal loss, light leaving through other facets and any droop from heating. The photon energy is $h ν = 1239.84 / λ$ eV with $λ$ in nm. Unless a value is entered, $I_{th}$ is the current-axis intercept of a least-squares line through the three lowest-current rows with nonzero power.

Worked example, from the default data. At 40 mA and 1.23 V the electrical input is 49.2 mW and the output is 16.8 mW, so $η_{WPE} = 16.8 / 49.2 = 0.3415$ , which the tool reports as the peak of 34.15 %. At 20 mA the ratio is 6.65/22.4 = 29.69 %, and at 100 mA it is 39.6/148 = 26.76 %, so the efficiency falls on both sides of the peak even though the output power keeps rising to the last row. The default data do not state a wavelength; taking 1310 nm gives $h ν$ = 0.946 eV. The rows at 12, 16 and 20 mA lie on a line of slope 0.525 mW/mA that meets zero power at 7.33 mA. At the peak the three factors are 0.946/1.23 = 76.9 %, 1 − 7.33/40 = 81.7 % and 16.8/(0.946 × 32.67) = 54.3 %, and their product is 34.15 %. The threshold costs less than the voltage defect here, and the largest loss is in the differential efficiency.

References: L. A. Coldren, S. W. Corzine and M. L. Mashanovitch, Diode Lasers and Photonic Integrated Circuits, 2nd ed. (Wiley, 2012). The companion article Wall-Plug Efficiency Extraction from LIV Data covers the measurement.