JULIO / DICIEMBRE 2020 - VOLUMEN 30 (2)
/ ISSN electrónico: 2215-2652
Esta obra está bajo una Licencia de Creative Commons. Reconocimiento - No Comercial - Compartir Igual 4.0 Internacional
DOI 10.15517/ri.v30i2.39823
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica
Experimental evaluation of the effect of spatial position
on the power of a 5W LED
Evaluación experimental del efecto de la posición espacial
en la potencia de un LED 5W
María Paula Gutiérrez Domínguez,
Ponticia Universidad Javeriana, Bogotá, Colombia
mariapaula.gutierrez@outlook.com
ORCID: https://orcid.org/ORCID: https://orcid.org/0000-0001-6106-5543
Diana Elizabeth Ramírez López
Ponticia Universidad Javeriana, Bogotá, Colombia
di.ramirez01@javeriana.edu.co
ORCID: https://orcid.org/0000-0002-9503-5462
Ricardo Otero-Caicedo
Ponticia Universidad Javeriana, Bogotá, Colombia
r.otero@javeriana.edu.co
ORCID: https://orcid.org/0000-0002-0358-8538
Recibido: 28 de noviembre 2019 Aceptado: 27 de febrero 2020
_________________________________________________________
Abstract
This experiment evaluates the effect of the spatial position on the power of a 5W LED, which was
obtained using a radiometer. For this purpose, the radiant flow emitted by the LED was arranged and;
taken at different positioning angles and distances from the sensing equipment. Statistical analysis for the
validation of the obtained data was checked against the values provided by the manufacturer according to
the specication sheet. Finally, the increase in tilt angle and distance resulted in a loss of radiant ow emitted
by the LED by 99 %.
Keywords:
Spatial position, LED, power, experimental method
Resumen
En este experimento, se realiza la evaluación del efecto de la posición espacial en la potencia de un
LED de 5W, la cual se obtuvo utilizando un radiómetro. Para tal n se dispuso del ujo radiante emitido por
el LED, tomado en diferentes ángulos de posicionamiento y la distancia del mismo con respecto al equipo
de detección. El análisis estadístico para la validación de los datos obtenidos se cotejó con los valores
Esta obra está bajo una Licencia de Creative Commons. Reconocimiento - No Comercial - Compartir Igual 4.0 Internacional
DOI 10.15517/ri.v30i2.39823
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica
proporcionados por el fabricante según la hoja de especicaciones. Finalmente, se obtuvo que el aumento del
ángulo de inclinación y la distancia generan una pérdida del flujo radiante emitido por el LED en un 99 %.
Palabras clave:
Posición espacial, LED, potencia, estudio experimental
GUTIÉRREZ, RAMÍREZ Y OTERO: Experimental evaluation of the effect of spatial position...
122
1. INTRODUCTION
Progress in solid-state light devices has led to the replacement of incandescent lamps with
Lightemitting diodes, known as LEDs, which have as principle the effect of electroluminescence.
This is based on the light emitted by the device through the transformation of electrical energy
(Rigutti & Tchernycheva, 2013). LEDs have been applied as light sources in countless areas, such
as nanotechnology, computing, bioengineering, medicine, among others (Noori, Mahbub, Dvorák,
Lucieer & Macka, 2018).
LEDs offer a better alternative to traditional light sources, in terms of low cost, smaller size,
longer service life, higher light power, and energy saving (Schubert & Kyu, 2005). Recent deve-
lopments have shown that most white LEDs are typically 30 % efcient, while some have reached
40 %. Although LEDs are known as the most efcient lighting source, a signicant amount of
electrical energy supplied is still converted to heat. Approximately 60 % –70 % of lost energy is
a signicant development and potential for LEDs (Ozluk, Muslu & Arik, 2019). Their lifespan is
due to their compact physical characteristics: LEDs are more durable than other lamps. Incandes-
cent bulbs tend to last 1000 hours, as heat destroys the lament, and fluorescent lamps tend to last
10,000 hours. LEDs can last longer than 50,000 hours or more, and in turn they provide a wider
color temperature range (4500 K – 12,000 K) and a wider operating temperature (20 °C to 85 °C)
(Luo, Hu, Liu & Kai, 2016).
However, there is no practical and accurate radiometric analysis to determine the actual light
output that is emitted from LEDs. Light power inuences optimal utilization of the amount of light
that such a device is capable of generating. In some cases, the data provided by the manufacturer
and the power measured in commercial radiometers do not match the power emitted by LEDs, mea-
sured with specialized or high-precision instruments (Price, Ferracane & Shortall, 2015).
Figure 1. 1p-n joint conguration that occurs in light-emitting diodes
Source: Adapted from [4]
The basic structure that allows the phenomenon of electroluminescence in the light-emitting
diode is the p−n joint. A p−n joint is the union of two semiconductor materials, meaning, materials
that are dielectric at rest but under certain conditions, are conductive. Semiconductor type and type
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica DOI 10.15517/ri.v30i2.39823
123
materials are listed as extrinsic semiconductors, because they have been manipulated from their
manufacture to favor a majority of positive and negative loads respectively. The p−n joint allows
an electric eld to be form, as well as a potential difference; therefore, by connecting an electrical
circuit to this joint, it will allow the passage of the current in the sense from positive to negative,
specically in terms of the device: from anode to cathode.
The p−n joint conguration that occurs in light-emitting diodes is what is known by its English
expression as forward bias. As shown in Figure 1, this conguration consists of the majority of
carriers on both sides of the junction crossing the depletion zone and entering their opposite side of
the joint, where they are minority carriers. The depletion zone is one that forms right at the inter-
section of the p−n junction where both types of loads exist and is one that “separates” a material
from another. Minority carriers then become present in the area of majority carriers of contrary
cargo; this is known as the injection of minority carriers (Wilson & Hawkes, 1998). These exceeds
the results in the recombination of excitons, which are the pairs formed both of electrons and by
the absence of these.
According to quantum theory, in energetic terms, when an exciton is recombined, it is because an
electron has precipitated its energy. As is known, atoms of any material, in this case semiconductor
materials, are related by covalent bonds. These bonds are represented in the orbitals or energy scales
that occupy the electrons that make up that atom. When a free electron in the material precipitates
its energy, it is moving from a driving state to a lower state of energy. In the case of recombination,
the electron precisely takes up a space in the valence band or hollow to complete the number of
electrons allowed in that energetic state. Because of this release of energy, the electron emits light.
Therefore, the objective of this type of p−n joint in the emitting diodes is to force as much
recombination as possible, that is, to nd that most electrons can be recombined with so-called
hollows so that this generation of energy can allow light emission continuously (Cabrera , López
& López, 2006).
LEDs have two basic properties: one is the luminous intensity and the other the transmitted
optical power. The luminous intensity is the result of the flow of energy from a solid angle and is
directly related to the illumination on a surface, generally used to express the brightness of an LED.
On the other hand, the transmitted optical power indicates the total energy radiating from that device.
The luminous intensity can be seen in Eq. (1):
(1)
Where
is the spatial angle, and is the radiant flux or power, which can be given from the
energy flow
e
as well:
(2)
GUTIÉRREZ, RAMÍREZ Y OTERO: Experimental evaluation of the effect of spatial position...
124
Where the standard luminosity curve is V ( ), K
m
is the maximum visibility and is approximately
683lm/W to a
=555nm.
The integral of the energy flow
e
in all propagation directions is the transmitted optical power Pt,
and is given by the following equation:
(3)
Where A
max
and A
min
are determined by the sensitivity curve of the photodiode (Komine & Naka-
gawa, 2004).
The radiant flux or power called by the symbol
is the measure of the power of energy that
can be taken at any wavelength per unit of time. It includes infrared, ultraviolet, and visible light.
The unit of measurement is Watt (W). Therefore, if you have a radiation source that has a radiant
flow of 1W, you have that the source emits 1 Joule of energy for every second.
The radiant flow or power is described by the following expression:
(4)
The spectral distribution of the flow is required for the characterization of the response of a
detector based on the incident energy. Since the spectral radiant flow or power, represented by
,
is dened as the flow of energy radiating per unit of time and per unit of wavelength, it is measured
in W /nm as shown in the following expression:
(5)
It is necessary to note that when radiation affects a device that produces a signal such as voltage,
proportional to the incident radiation, the total amount of the flow should be taken as magnitude
instead of the ow per unit area; thus, you must specify the spatial extent of the radiation eld whose
flow is being considered (Gomez, 2006).
For the characterization of a radiation source, spectral irradiance E
considered a magnitude
by describing the incident power that can be measured on an object at any wavelength. Spectral
irradiance is a multivariate function of wavelength , which is position or angular distribution relative
to the source that can beexpressed , (r, , )in spherical coordinates or (X, Y, Z) in a Cartesian
system, Z is the direction of sym metry of the LED and nally of time t. Figure 2 describes the
positions (x, y, z) of the LED.
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica DOI 10.15517/ri.v30i2.39823
125
Figure 2. Cartesian plane in spherical coordinates for Positions (x, y, z)
for the distribution of spectral irradiance emitted by the LED
Different methods have been used in the existing literature to analyze the curve of the radiating
flow emitted by LED. In the study developed by Mendoza, a method was implemented for the
determination of the electric eld model and the transverse oscillation mode of the beam of a laser.
To obtain the necessary information, they obtained the power of the laser radiation, which was
made through a circular aperture to which the radius could be varied (Mendoza & Laboratorio
de Tecnología Láser-ICTM, 2005). The author recommends positioning the detector close to the
source to ensure that measurements are taken with the minimum radius of the laser beam and to
ensure the measurement of the maximum power value. There are also other approximations of direct
calculations in which the efciency of radiation in the LEDs is studied for example, Magalhães et
al. (2016) investigated the effect of different LED arrangements, placed on a disc with a diameter
of 70mm and compared the different congurations such as: radial arrays, diamonds, square radials,
and square diamonds. This study was conducted by optical simulation, in which the researchers
concluded that square radial and radial matrices had better efciency. This conrms that light
efciency,; the relationship between the luminous ux emitted by a light source and the power, has
been improved by achieving greater output power due to phenomena such as the reection of light
and its redirection in the middle through the modication of the geometric shape LED (Jarosz,
Marczynski & Signerski, 2019; Tsai, Chen, Su & Huang, 2010; Chen & Wu, 2010; Chou, Chen
& Yang, 2012).
The objective of this project is to carry out the experimental evaluation of the effect of the
spatial position on the power of a 5W LED and contribute to the characterization of this type of
devices by detecting the power emitted using a radiometer. For this purpose, the measurement of
the power of an LED was available through different distributions, which involve different posi-
tioning angles and variation of the distance of the LED from the detection equipment (Sholtes, et.
al. 2019). The analysis of the obtained data for validation was performed with respect to the data
in the manufacturer’s specication sheet, which allows a comparison between the power or radiant
flow emitted by the LED and the data provided by the manufacturer. Finally, it concludes which
GUTIÉRREZ, RAMÍREZ Y OTERO: Experimental evaluation of the effect of spatial position...
126
are the points of the greatest light power, determining the variation of the accuracy of the device
according to the different distributions.
2. DATA COLLECTION METHODOLOGY.
2.1 Materials
The development of this study used a 5W power light-emitting diode (LED), brandOSRAM
CSLPM1.TG, with a manufacturer-recommended power supply of 2.75V to 3.5V and a maximum
current of 1.4A; a Radiometer (Power Meter), NEWPORT brand, Model 2936-R and an optical
sensor, of the same brand, model 818-UV-L. In order to measure the power generated by the LED, a
power source, ARRAY mark, reference 3631A, which powered the LED, two 360-degree continuous
rotation goniometers (one THORLABS brand (vertical), reference PRM1, and another NEWPORT
(horizontal), reference M-RSP- 2) were used to assemble the LED positioning system; in other terms,
it performs the variation of the angles in the established two axes. Finally, a FLUKE multimeter,
which allowed to visualize the ambient temperature, was also used. All these materials are found
in the laboratory of the Group of Thin lms and Nanophotonics of the Ponticia Universidad
Javeriana of Bogotá.
Figure 3. Prototype design at its different angles, used to position the LED
2.2 Methods
2.2.1 Study factors
The test consists of 3 factors: angle of variation on the X axis, angle of variation on the Y axis,
and two distances on the Z axis between the LED and the detector.
The variation of the angles on both the X axis and the Y axis comprised 8 transitions since
measurements were made from 0 ° to 70 °, every 10 °. The distances considered for the Z axis were
taken regarding two lengths between the LED and the detector: a short distance of 15 cm and a
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica DOI 10.15517/ri.v30i2.39823
127
length of 30 cm. Ten replicates were taken for each combination of X, Y, and Z. The study did not
consider the angles of 80 and 90 degrees as they contained a blind spot in the measurement, and
the power was zero.
2.2.1 Experimental set up
The mounting for LED power measurement was adapted on a flat and stable surface. Initially,
the LED was positioned in front of the optical sensor and was fed with the source at 0.500A and 3V,
so these values did not exceed the maximums allowed by the manufacturer, preventing the heating
of the LED, a factor that can affect the optimal operation of the device. The maximum power point
was detected at position 0 ° X, 0 ° Y of the LED to ensure a correct alignment with the sensor, which
collects the beam of light and connects to the Power Meter equipment that measures the power of
the LED. In Figure 4, you can see the assembly made for the measurement of the power of the LED.
Figure 4. Schematic diagram made for measuring LED power
In addition, to ensure optimal environmental conditions during the experiment and to prevent
external factors from affecting measurements, a controlled environment was required. For this
purpose, a closed dark room was arranged, which did not allow the entry of external light. On the
other hand, a multimeter was used with a thermocouple that allowed measuring the room temperature
and ensuring that it remained in a range between 15-20 °C. For experimental design, in order to
have valid and objective conclusions, the response variable is the incident power of the LED per
unit area in units of mW*cm
-2
. As for data collection, the taking of each of the samples was done
randomly with a period of 3 minutes between each measurement to let the LED rest and make
changes in angles and distances.
3. RESULTS
3.1 Statistical analysis
To perform the statistical analysis we used interaction plots of averages from the response
observations. Finally, the statistical test used was a variance analysis (ANOVA) for a factorial design
with 3 factors: the angle in X, the angle in Y, and the distances between the LED and the detector.
GUTIÉRREZ, RAMÍREZ Y OTERO: Experimental evaluation of the effect of spatial position...
128
3.2 Descriptive analysis
For descriptive analysis, data obtained from differences in factors from angles in X, Y and
distances regarding the incident power of the LED are presented.
Figure 5. Average interaction graph of the 10 factor response observations
Figure 5 shows the construction of the average interaction plot of the 10 factor response obser-
vations, using the average of each treatment. It was possible to show that as the angle increases
the averages decrease for both cases of distances; however, the averages obtained in the smaller
distance have a greater variation. Through this chart, it was possible to observe that the power at a
shorter distance and with smaller angles is greater.
By means of the inter-class inuence chart (Figure 6), it was possible to supplement the infor-
mation obtained in the previous graph. In this graph, you can observe through a color scale the
averages of each treatment, which allows reiterating that the greater power is concentrated in the
shortest distance and the smallest angles.
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica DOI 10.15517/ri.v30i2.39823
129
Figure 6. Inter-class influence graph
3.3 Proof of assumptions
With this test, it was possible to conrm the assumptions of the ANOVA through graphical tests,
demonstrating the randomness, independence, homogeneity of variances and the normality of the data.
Kolmogorov Smirnov test for normality results showed (p=0.23). Levene test for homoscedasticity
showed (p=0.42, which conrms the population variances are equal among groups.
3.4 Variance analysis
The ANOVA variance analysis was used to examine the effects described above in the response
variable, which for this case was the incident power of the LED. The results recorded in Table 1
GUTIÉRREZ, RAMÍREZ Y OTERO: Experimental evaluation of the effect of spatial position...
130
indicate that the p-value of the iteration between the factors was 0.4308, so it is proved that it was
statistically signicant.
The results on Table 1 show that the factors had a signicant effect on the response variable
(p-value <0.0001). Using smaller angles and smaller distances increases the irradiated power of
the LED as is shown in the previous chart. A Tukey-type Post-Hoc test was also used to know if
statistically the three factors (X-angle, Y-angle, and distance) are different from power, demonstrating
that 98 % of classes are statistically different. However, some classes are statistically the same
because their p-value is greater than 5 %.
Table 1. ANOVA Table
ANOVA
Df Sum Sq Mean Sq F value Pr (>F)
Angle Y 7 170067 24295 38982.99 < 0.0001
Angle X 7 123933 17705 28408.11 < 0.0001
Distance 1 186034 186034 298500.45 < 0.0001
Angle Y - angle X 49 12983 265 425.14 < 0.0001
Angle Y - distance 7 28071 4010 6434.47 < 0.0001
Angle X - distance 7 16987 2427 3893.81 < 0.0001
Angle Y - angle X - distance 49 1797 37 58.86 < 0.0001
Residuals 1152 718 1
4. DISCUSSION AND CONCLUSIONS
This study assessed the effect of variation in the positioning of an LED with power. The results
show that, according to existing research, the variation in positioning affects the distribution and
uniformity of the factors of light emission including power, which increases the measurement error
when there is greater distance and does not allow an optimal value of the power emitted by the
light-emitting diodes.
Ingeniería 30 (2): 120- 132, julio-diciembre, 2020. ISSN: 2215-2652. San José, Costa Rica DOI 10.15517/ri.v30i2.39823
131
In addition, it is also necessary to take into account disturbances that can occur in the operation
of the LED. This prevents its optimal performance such as the generation of harmonics, temperature
variation, and voltage drops in the devices with ashing or ashing in the LED, affecting power and
light intensity. ANOVA results shows that at greater distances (30 cm) their power values were low
due to interference in the environment, This generated that the LED lighting created a diffraction
of light, reducing its ability to penetrate and lowering its power.
In accordance with the results derived from statistical analysis, especially with the variance
analysis, the research concluded that for the data taken at the distance of 15 cm there is a greater
inuence between classes, resulting in greater power emitted at the points between 0 ° and 50 ° of
inclination. This makes it possible to determine that at a shorter distance (15cm) there is a unifor-
mity of emission, and a value of greater potential is detected, highlighting that the evaluation was
done in a controlled environment based on the manufacturer’s specications so as not to alter the
optimal performance of the device.
Finally, the power is correlated with the angle variation, since increasing the inclination of that
factor decreases the power, having as a maximum point the value of 77.73 mW*cm
-2
for an angula-
tion of 0 ° X, 0 ° Y of inclination and a minimum point of 0.86 mW*cm
-2
. This determines that the
detected power decreases by approximately 99 % when there is a variation at larger angles.
5. RECOMMENDATIONS AND FUTURE WORK
In future work, other factors should be considered that could affect the optimal performance
of the LED or other congurations that allow to optimize and that do not affect the data analysis of
the response variable. It should be noted that the data obtained in the p-value for this study might
have been affected by external factors that were not taken into account in this study such as the
humidity of the environment and the fluctuations of the electrical voltage. Measurements for this
study were taken only with an LED, but according to research conducted to increase the power emi-
tted by the device, other factors besides the angles and distance can be added such as the number
of LEDs that can form a matrix and its shapes.
REFERENCIAS
Cabrera , J., López, F., & López, J. (2006). Optica Electromagnetica: Fundamentos (Segunda ed., Vol. I). Madrid:
Universidad Autónoma de Madrid.
Chen, H.-C., & Wu, G.-Y. (2010). Investigation of irradiance efciency for LED phototherapy with different arrays.
Optics Communications, 283, 4882–4886.
Chou, H.-Y., Chen, C.-C., & Yang, T.-H. (2012). Maintenance of stable light emission in high power LEDs. Microelectronics
Reliability, 52, 912-915.
Gomez, E. (2006). Guia basica de conceptos de Radiometria y Fotometria. Sevilla: Universidad de Sevilla.
Jarosz, G., Marczynski, R., & Signerski, R. (2019). Effect of band gap on power conversion efciency of single-junction
semiconductor photovoltaic cells under white light phosphor-based LED illumination. Materials Science in
Semiconductor Processing, 107, 1-5. doi:10.1016/j.mssp.2019.104812
GUTIÉRREZ, RAMÍREZ Y OTERO: Experimental evaluation of the effect of spatial position...
132
Komine, T., & Nakagawa, M. (2004). Fundamental Analysis for Visible-Light Communication System using LED
Lights. IEEE Transactions on Consumer Electronics, 50(1), 100-107. doi:10.1109/TCE.2004.1277847
Luo, X., Hu, R., Liu, S., & Kai, W. (2016). Heat and uid ow in high-power LED packaging and applications. Progress
in Energy and Combustion Science, 1-32. doi:10.1016/j.pecs.2016.05.003.
Magalhães, T., Weig, K., Costa, M., Werneck, M., Barthem, R., & Costa, C. (2016). Effect of LED-LCU light irradiance
distribution on mechanical properties of resin based materials. Materials Science and Engineering C, 63, 301-
307. doi:10.1016/j.msec.2016.02.060
Mendoza, Omel; Laboratorio de Tecnología Láser-ICTM. (2005). Estimación de Modos TEMmn Laguerro-Gaussianos.
Revista Cubana de Física, 22(2), 153-156.
Noori, A., Mahbub, P., Dvorák, M., Lucieer, A., & Macka, M. (2018). Radiometric analysis of UV to near infrared
LEDs for optical sensing and radiometric measurements in photochemical systems. Sensors and Actuators B:
Chemical, 262, 171-179. doi:10.1016/j.snb.2018.01.179
Ozluk, B., Muslu, A. M., & Arik, M. (2019). A Comparative Study for the Junction Temperature of Green Light-
Emitting Diodes. IEEE Transactions on Components, Packaging and Manufacturing Technology, 2024-2035.
doi:10.1109/TCPMT.2019.2929172
Price, R., Ferracane, J., & Shortall, A. (2015). Light-Curing Units: A Review of What We Need to Know. Journal of
Dental Research, 94(9), 1179-1186. doi:10.1177/0022034515594786
Rigutti, L., & Tchernycheva, M. (2013). Characterization of Semiconductor Heterostructures and Nanostructures.
Torino: University of Torino. doi:10.1016/B978-0-444-59551-5.01001-3
Schubert, F., & Kyu, J. (2005). Solid-State Light Sources Getting Smart. Science, 308, 1274-1278. doi:10.1126/
science.1108712
Sholtes, K., Keliher, R., & Linden, K. (2019). Standardization of a UV LED Peak Wavelength, Emission Spectrum,
and Irradiance Measurement and Comparison Protocol. Environmental Science & Technology, 53(16).
Tsai, P., Chen, W., Su, Y., & Huang, C. (2010). Enhanced light output of InGaN LEDs with a roughened p-GaN surface
using different TMGa ow rates in p-AlGaN layer. Applied Surface Science, 256, 6694–6698. doi:10.1016/j.
apsusc.2010.04.072
Villamarín, A., Ferrero, A., Pons, A., Campos, J., Rabal, A., Hernanz, M., . . . Corróns, A. (2012).
Angular and spectral radiant intensity distribution of high brightness white LEDs. Óptica pura y aplicada,
45(2), 131-136.
Wilson, J., & Hawkes, J. (1998). Optoelectronics: An Introduction (Tercera ed.). Englewood Cliffs: Prentice Hall PTR.
