Improved Public Lighting at the University of São Paulo - Brazil

JOSÉ CARLOS MARTINEZ MELERO1, EMERSON ROBERTO SANTOS2, ELVO CALIXTO

BURINI JUNIOR1, HÉDIO TATIZAWA1

1Institute of Energy and Environment, University of São Paulo

Avenida Professor Luciano Gualberto, 1289, Butantã, São Paulo-SP, BRAZIL

2Electrotechnical, Technical School,

Rua Clélia, 2064, Lapa, São Paulo-SP, BRAZIL

Abstract: The University of São Paulo (USP) transitioned the roadway lighting technology at the University

called Armando de Salles Oliveira (USP-CUASO campus) from High-Pressure Sodium Vapor Lamps (HPSVL)

and Metal Halide Lamps (MHL) to the Solid-State Lighting (SSL). So, this paper presents and discusses the

outcomes of the lighting installations and the depreciation of light output from luminaires installed at the Institute

of Energy and Environment (IEE) at USP using field data collected over two years (2018–2023). The analysis

incorporates aspects related to corrective maintenance, predominantly performed last year, necessitated by the

end of the useful life of components such as electronic drivers (66%), 50 W LEDi modules (17%) and others

(17%). The failure rate of the sampled luminaires in the IEE area increased from 10% to 28% between August

2021 and January 2023.

Key-Words: Public Lighting, Solid-State Lighting (SSL), LEDi, light output depreciation, lifetime, corrective

maintenance.

Received: June 22, 2022. Revised: October 11, 2023. Accepted: November 14, 2023. Published: December 31, 2023.

1. Introduction

Road lighting using Solid-State Lighting (SSL)

LEDi technology was installed at the Cidade

Universitária Armando de Salles Oliveira (CUASO)

campus of the University of São Paulo (USP) in São

Paulo (city), Brazil. This process involved the

replacement of High-Pressure Sodium Vapor Lamp

(HPSVL) and Metal Halide Lamp (MHL) lighting

systems to the LEDi (inorganic light emitting diode)

technology as a new system that was implemented in

September, 2013. The project aimed for a minimum

luminous efficiency of 85 lm/W. The initial

illuminance value established in the 2012 project

exceeded the minimum requirements outlined in the

local standard. Additionally, the uniformity index

surpassed the specifications set by the current

Brazilian technical standard [1]. Records obtained for

the parking area in the region known as Polytechnic

School based on data sampled in 2014, indicated an

average illuminance value of 67 lux and a uniformity

index of 0.448 (Emin/Average) [2].

In accordance with the Brazilian technical

standard, the specified minimum requirements are

20 lux for the average illuminance and 0.3 for the

uniformity index [1]. In 2016, 47 lux was presented

for the average and 0.11 for uniformity [2]. In the

lighting project at CUASO, there was no specified

illuminance limit depreciation for maintenance or

replacement of luminaires with devices featuring

higher luminous efficiency.

Projections from the DOE (USA) indicated the

possibility of a rapid evolution in light production

efficiency for SSL technology [3].

Even before the implementation of SSL

technology, CUASO had raised concerns about the

durability of the new SSL system for lighting,

specifically regarding the lifetime of LEDi

luminaires.

Regardless of the light source type used, the

luminous depreciation will occur over a time of use,

however, sampling from the survey by a mesh of

illuminances between poles or contiguous light

points is a task, that requires support staff (both day

and night) that consume consuming a lot of time and,

it is a process not very efficient. In the literature, the

operation method of the LEDi technology is

indicated ("when applicable") as one of the two main

concerns related to failures (USA, 2014) in addition

to the gradual reduction of the light output (luminous

or lumen depreciation) [4]. The IEC has established

two metric parameters for the service life of SSL

(LEDi-based) luminaires.

Both are related to "gradual" and "abrupt" light

output (luminous flux) degradations suggesting the

application of standard set for quantities reported by

the market, such as: "average lifetime" and the

associated "abrupt failure value" [5].

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

229

Volume 3, 2023

This article discusses lighting records of CUASO

obtained in locus and the company report responsible

for implementing and maintaining the LEDi

technology over more than 09 (nine) years (from

November, 2013 to January, 2023) [2]. The CUASO

public lighting system showed a high failure rate in

the Institute of Energy and Environment (IEE-USP)

area after the pandemic period (SARS-COV 2). This

situation required that maintenance actions be

intensified, and such services were developed

throughout 2022 with the team and after the

acquisition of the electronic drivers [6].

2. Methodology

2.1 Selecting events and dates

After 34,700 hours of operation (August, 2021),

the public lighting system of CUASO using SSL

(with LEDi) had a high failure rate of about 10%,

according to sampling carried out in the IEE area [7].

This indicated the need for corrective interventions to

restore the ordinary conditions of the SSL lighting.

The fact that activities at CUASO were suspended

and re-established after the pandemic period (SARS-

COV 2) required intensified maintenance actions.

These services have been developing with the

acquisition of electronic drivers and their teams.

Currently (January, 2023), it is estimated that the

operation of the equipment is close to 41,000 hours

since the initial energization (September 25, 2013).

The Campus Administration of USP - CUASO

(PUSP-C) records indicate that more than 1,200 light

fixtures were repaired throughout 2022.

A publication of 2014 was presented paired

records on the Correlated Color Temperature (Tcp),

when it was obtained by direct measurement and

calculation using the same instrument (manufactured

by Minolta, model XY-1), which was used to carry

out the field sampling of this work.

Those results were parameterized for different

light sources, such as: (HPS, 03 (three) SBFL, MH

and 02 (two) LEDi). Both values (reading and

calculation) are presented including records (Table 1,

n. order 6) for Tcp carried out on November 11, 2013,

mainly for 11 (eleven) luminaires of CUASO [8].

The calculation methodology used the chromaticity

coordinates to determine the Tcp based on the

algorithm* proposed by Robertson, A. R. in 1968 [9].

In an audit, the objective was to evaluate the

lighting systems comparatively of CUASO. Three

types of light sources were iinstalled at CUASO in

the year 2013 (HPS, MH and LEDi). Parameters

assessed from measurement (reading) and

calculation) are: Correlated Color Temperature (Tcp);

average illuminance (Em); mean luminance (Lm) and

luminance coefficient (q); (ratio of luminance to

illuminance per point) mean and associated bilateral

confidence intervals (ICB).

The average illuminance of typical equipment at

CUASO (manufactured by Peterco, HPS of 250 W

with housing for auxiliary equipment and borosilicate

glass diffuser) was evaluated according to conditions

established in the test field (IEE/USP) [10].

A report of 2019 was considered, containing

records for illuminance obtained by direct

measurement and calculations (average and

uniformity) for different years: 2014, 2016, and 2019,

but it was not indicated, which type of lux meter was

used [2]. In this same article, only the maximum

values of the lane (vehicle) recorded for each of the

07 (seven) grids of sampled points are considered (it

is believed that the measurements were carried out at

the pavement level as observed in 2016 [11]). The

definition of illuminance maximum points is

attributed to the high probability of their occurrence

in the position known as Nadir, which has the

luminaire at the Zenith position, this procedure is

possible to compare samples taken in the IEE-USP

area in 2018 and 2023, which are presented in this

work [12].

The averages have traceability when obtained

from the meshes defined in the measurement report;

the Brazilian technical standard considers 2 (two)

transverse lines normally with higher values [1, 2].

This characteristic can distort the mean raising

its value improperly and for this reason, it was not

considered in this article.

Note: *In the updated CIE terminology (17-258 -

correlated color temperature) there is one note that

tells us, “there is a limitation on the range of valid

coordinates and an indication that the Tcp can be

calculated from a simple search procedure of

minimum value in the determination of the Planck

temperature (black body), which provides the

smallest difference between the input chromaticity

and the Planck locus recommended by methodology

[8]. Also, this signals that some of the values

contained in tables of the cited reference are not

updated and the second abbreviation used as "CCT".

2.2 Parameters considered that infer the

luminous depreciation

The parameters considered in this article are road

plane illuminance (vehicular and pedestrian), the Tcp

(read and calculated from the chromaticity

coordinates, x, y, using software), and the "distance"

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

230

Volume 3, 2023

(Duv) between the considered point and the Planck

locus [13].

To determine the average that was later used to

estimate depreciation between the different periods

(years), the data obtained in the IEE area during 2018

and 2023 and the position called as Nadir, were

stratified by illuminance ranges ≤ 20 lux; between 20

and 30 lux; between 30 and 40 lux; between 40 and

50 lux; between 50 and 60 lux; ≥ 60 lux and < 70 lux;

and ≥ 70 lux.

The range containing the highest frequency

(primary mode) was used, due to the distribution of

records obtained, which are presented. The two-sided

confidence interval with level of 95% was calculated

and used to quantify the dispersion of the record

concerning the average.

3. Results

Currently, data collected (see Table 4, [6]) for

different types of luminaires indicates that there are

already products on the Brazilian market with:

nominal luminous efficiency in the range from 135 to

170 lm/W; two nominal Tcp; eight for 4000 K and

three for 5000 K; nominal electrical power in the

interval from 47 W to 197 W and relative price in

USD$/klm from 4.36 to 12.88.

Based on a list of 680 repaired items in USP

lighting at CUASO (PUSP-C) from March to

October of 2022 (after 07 months), it was verified

that the highest need for replacement was for

electronic drivers, which represented 67% of

replaced items and together with the replacement of

LEDi modules accounted for more than 86% of the

items already replaced. The smallest fraction of

failure occurred for the surge suppressor item (2.65

%), then for the item called as telegestor

(telemanager) (3.98%) and Y-cable (7.24%) [6].

3.1 Events and dates

3.1.1 Tcp from SSL in 2013

The values presented in Table 1 refer to

measurements carried out at CUASO in 2013 using

different types of light sources; the lines numbered

from 1 to 6 correspond to the average value and line

7 to a single reading/calculation [8].

Table 1 shows the correlated color temperature

(Tcp) values obtained by directly reading the

instrument used [8].

Table 1. Average value determined for 06 (six) types

of artificial electric light sources and single reading

data (in line, number 7). Source: samples measured

at CUASO, USP [8].

In table 1, the HPS and MH installations for road

lighting on a pole with a curvilinear shape present at

CUASO (Av. Prof. Almeida Prado) were measured

on November 07, 2013, and after, when it installed

using SSL (LEDi) luminaries on metal poles at

CUASO (Av. Prof. Luciano Gualberto road for

pedestrians and vehicles) and SBFL - Single Base

Fluorescent Lamp (or compact), whose samplings

measurements were carried out at laboratory [8]. In

addition, the Tcp obtained by direct reading on the

instrument used (Table 1) presents values calculated

using the methodology of Robertson, A. R. and the

chromaticity coordinates obtained by direct reading

using the same instrument [9].

3.1.2 Maximum illuminance, CUASO, USP

Table 2 presents the maximum illuminance for

07 (seven) grids, where measurements were carried

out at CUASO. The data were obtained from the

report delivered to PUSP-C indicating that the

samplings occurred in 2014, 2016 and 2019 [2].

Table 2. Maximum illuminance is measured between

two poles at different points on the road or area grid

of CUASO [2].

Measurement location

and

time of occurrenceb

Maximum

Illuminance (lux)

Year

Identification

2014

2016

2019

Av. Luciano Gualbertoa

Av. Almeida Prado

Av. Lineu Prestes

Rua do Lago (street)

Rua do Matão (street)

Estacionamento FEA

(parking)

106

Estacionamento POLI

(parking)

108

Note: b Maximum values extracted from each

mesh/grid of the report considered [2].

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

231

Volume 3, 2023

3.1.3 Measurements carried out in 2013

The result of the sampling carried out in the test

field (IEE/USP) with poles separated by 35 meters,

the calculated illuminance (mean value and

confidence interval) was 8.2 ± 0.8 lux. The other

mentioned values were sampled on November 7,

2013, in the direction of 45° concerning the normal

direction and origin on the surface of the pavement.

Table 3 shows the results [10].

Table 3. Road lighting technologies at CUASO used

in 2013: average illuminance, average luminance,

asphalt luminance coefficient, correlated color

temperature reading, and correlated color

temperature calculated [10].

Measurement

parameter

MHL c

HPSVL c

LEDi

Average

illuminance

(lux)

13±1

43±12

58±12

Average

luminance

(cd/m2)

0.57±0.05

1.3±0.2

2.8±0.5

Asphalt

luminance

coefficient

(mcd.lm-1)

45±5

34±6

50±6

Tcp (reading) (K)

3600±36

1893

3510

Tcp (calculated)

(K)

4793±43

2220

4235

Note: c Nominal electrical power of 250 W for each

luminaire [10].

3.1.4 Audit carried out in 2016 and 2018

In 2016, follow-up and measurements were

carried out, whose results when in the field showed

an acceptable reading difference between

instruments and points confronted (location: Avenue

Luciano Gualberto) [11]. In Figure 1, 45 illuminance

results were plotted (IEE, in 2018) according to the

luminaire identification (at the Zenith position).

Fig. 1: Luminaire (at Zenith position): illuminance at

the Nadir point vs. CUASO SSL luminaires ID

identification number (May 2018).

Fig. 2: Correlated color temperature obtained with an

instrument and calculated vs. CUASO SSL

luminaires Nadir illuminance (May 2018).

The illuminance sampling was carried out for

Av. Luciano Gualberto, between 22h34 min. and

22h46 min. on September 8, 2016. This road is for

vehicles and pedestrians and is located close to the

IEE. Measurements for 6 (six) transverse lines, 1

(one) longitudinal line (pedestrian route), and 10

longitudinal lines (vehicle route) were obtained. For

the 59 measurements in the vehicle, the average and

ICB of 99% and 44 ± 6 lux were obtained; the

standard deviation of 18 lux; ranging from 14 lux to

82 lux; mode of 49 lux, and for minimum

illuminance ratio by the average with the value of

0.32 lux.

For the pedestrian pathway, 6 (six)

measurements were obtained and the average and

ICB 99% of 51 ± 10 lux was collected; a standard

deviation of 10 lux ranging from 39 lux to 67 lux;

and for the minimum illuminance ratio by the

average, the value of 0.77 lux. Based on the results

of a report accessed from a similar location in 2016,

for the 45 records in the vehicle, an average of 41.8

lux was obtained ranging from 13 lux to 77 lux, and

for the minimum illuminance ratio by the average

the value of 0.31 lux [14]. It is mentioned that 15 lux

(minimum illuminance) and 0.20 lux (uniformity)

are expected for the class (V3) considered for the

road [1]. For the pedestrian pathway reported in

2016, 05 (five) measurements were collected and an

average of 45.6 lux was obtained ranging from 36

lux to 56 lux; for the minimum illuminance ratio by

the average value of 0.79 lux[14]. Mention is also

related to 5 lux (minimum illuminance) and ≥ 0.20

(uniformity) expected for the class (P3) considered

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

232

Volume 3, 2023

for the route [1]. The body of the report also

mentions 20 lux (minimum illuminance) and ≥ 0.30

(uniformity) predicted for class (P1) as reference

values for the vehicle parking area [1, 14].

3.1.5 Audit carried out in 2018 at IEE

When stratified values of sampled illuminance

are considered (in 2018) for the condition ≥ 50 lux,

the sample containing 31 results presented the range

from 52 lux to 72 lux, standard deviation of 5 lux, a

mean of 64 ± 2 lux, which corresponds to a

dispersion of 3%. In the condition > 60 lux, the

sample containing 23 results presented the range

from 62 lux to 72 lux and standard deviation of 3.2

lux and a mean of 66.4 ± 1.3 lux, which corresponds

to a dispersion of 2 %.

The Tcp records calculated values (in 2018) and

a sample containing 40 results showed the range from

3615 K to 4309 K, standard deviation of 126 K,

average of 4055 ± 39 K, which corresponds to a

dispersion of 1 %; and Duv records, calculated values

(in 2018), a sample containing 40 results showed the

range from 2.20 to 3.92 (x 10-3) lux, standard

deviation of 0.32 lux, average of 2.75 ± 0.10 (x 10-3)

lux, which corresponds to a dispersion of 3.6 %.

3.1.6 Nadir illuminance, Tcp, Duv (IEE, 2023)

Sampling was carried out on January 9, 2023, in

the IEE-USP area at CUASO. On this date, 11

luminaires without lighting (off) were identified, 01

(one) luminaire with very low intensity and 01 (one)

with intermittent failure resulting in a relative

number of failures (abnormal condition) of 28%.

Even a pedestrian luminaire has two modules

instead of one and a luminaire is mounted above a

roof, both were not considered dark spots. When

sampled illuminance values are considered (in 2023)

stratified by removing the value 113 lux (luminaire

with two 50 W modules, pedestrian route) and

condition > 40 lux, the sample containing 21 results

presented a range from 44.2 lux to 68 lux, standard

deviation of 7 lux mean of 57 ± 3 lux, which

corresponds to a dispersion of 5%.

The Tcp records calculated values (in 2023) and

a sample containing 32 results showed a range from

2256 K to 4130 K, a standard deviation of 324 K,

average of 3975 ± 112 K, which corresponds to a

dispersion of 3% and Duv records, calculated values

(in 2023), a sample containing 32 results showed a

range from 2.50 to 3.40 (x 10-3) lux, a standard

deviation of 0.24 lux, average of 3.05 ± 0.08 (x 10-3)

lux, which corresponds to a dispersion of 3%.

The ambient temperature and external relative

humidity were sampled at each measurement point

with an instrument manufactured by Minipa, model

MT-241 for a sample containing 30 results. It

presented a range from 20.9°C to 23.1°C, average of

21.8 ± 0.2°C, which corresponds to the dispersion of

0.93% and for the external relative humidity to the

range from 61% to 72%, an average of 67 ± 1 %,

which corresponds to a dispersion of 1.5 %.

3.1.7 Light output depreciation

The reduction of the light output for lamps in the

IEE-USP area sampled in 2018 and 2023, can be used

to estimate the depreciation that was calculated by the

ratio of 57 lux/66.4 lux, which represents 86% that

presented a reduction in light output of 14%

identified when the Nadir point was used as a

reference and the extraction procedure explained

based on the modes. This reduction occurred over 4.6

years. It is possible to estimate an average rate of 3%

per year for the period considered. This value makes

it possible to formulate a hypothesis for the

installation, which in September, 2023 completed 10

years since the initial energization. It may present a

reduction of 30% in the light output that will be at the

level called L70.

Other locations, such as avenues, may show a

more significant decrease in depreciation, due to the

emission of particulates by bus exhaust, for example.

The removal of LEDi modules from luminaires with

two or more modules was observed some in the IEE

area. This fact was also observed with direct impact

on the records of light output at the Nadir point. The

idealized methodology (Nadir measurement) was

used because it is simpler than the IESNA

methodology [15]. The decision to remove LEDi

modules from the luminaires was based on restoring

the maximum possible light output.

Figure 3 shows the data of maximum

illuminance values obtained from the accessed report

for 2014, 2016, and 2019 with 07 (seven) meshes of

CUASO roads and parking lots and average

illuminance from Nadir for the IEE area in 2018 and

2023, including linear regression and equation [2].

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

233

Volume 3, 2023

Fig. 3: Data of maximum illuminance values for

2014, 2016, and 2019 with 07 (seven) meshes in

CUASO roads and parking lots) and average

illuminance of Nadir for the IEE area in 2018 and

2023 included regression and equation [2].

For each set of 03 (three) points maximum, an

adjustment was made using linear regression and the

equations and the value of the quality parameter R2

are presented [2]. It should be noted that the

occurrence of non-conforming records between the

final and previous values (second and third) indicates

an anomaly occurring between each data acquisition

in the field. Usually, a reduction of the light output is

expected; however, in some cases, there was an

increase (see Fig. 3, Rua do Matão (street), for

example). In Figure 4, the points where the poles and

light fixtures are installed in the area of the IEE -

USP, CUASO are presented and identified [16].

Fig 4: Map of the IEE-USP, CUASO, area with the

position of the poles, where the light fixtures with

identification are installed ID [16].

In Fig. 5, IEE-USP area changes are indicated as

proposed in the path of ducts (and light points,

circuits: 002 and 003), due to pedestrian routes not

foreseen in the project (ref. file: <USP- EL-IMP-

R04-21.pdf>, dated May 20, 2013) [17].

Fig. 5: Map, in the plan of the IEE-USP area (in

CUASO), an electrical circuit with a suggestion for

changing the path of ducts (and light points, circuits:

002 and 003), due to unforeseen pedestrian routes

[17].

Some light points planned and implemented, as

shown in Fig. 5 were not included in Fig. 4, designed

to identify the current situation (in 2016) during the

inspection performed.

4. Discussion and conclusion

A hypothesis for the installation, which in

September, 2023 completed 10 years since its initial

energization is the definition of the date, which it will

present a 30% reduction in light output and that will

have reached the end of service life called L70.

Other places, such as avenues, may present a

more significant reduction in light depreciation, due

to the emission of particulates by bus exhaust, for

example.

The removal of LEDi modules from luminaires

with 02 (two) or more modules was observed, some

in the IEE area. This fact was observed by

measurement, as it affected the records made of light

output at the Nadir point. This corrective

maintenance measure may cause a reduction in

illuminance values below the established minimum

requirement.

The idealized methodology (measurement at

Nadir point) was used in the IEE area because it is

simplified concerning the IESNA methodology [15].

It proved to be valuable and capable of

indicating light depreciation and the hypothesis of a

temporal rate.

The reference Tcp is considered as a parameter,

the average is calculated and presented (in Tab. 1,

line 6 of 4230 K) for the sample obtained in 2013 and

the average values for 2018 (4055 K) and average in

2023 (3975 K), there presented a reduction of - 4%

(in 2018) and reduction of -6% (in 2023) according

to samples used.

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

234

Volume 3, 2023

For the Duv parameter, taken as reference, the

average calculated for 2018 (2.75 x 10-3) and the

average in 2023 (3.05 x 10-3) presented absolute

increase of 0.30 x 10-3 and a relative 11% according

to the samples used.

It was verified that the need to replace electronic

drivers represented 67% of the items replaced and

that, together with the LEDi module, they

corresponded to more than 86% of the items already

replaced [6].

Data collected for different types of luminaires

indicated that there are products in the Brazilian

market with nominal luminous efficiency in the range

from 135 to 170 lm/W with a relative price range of

USD$/klm from 4.36 to 12.88 [6]. These data, among

others, should be considered in an evaluation of the

best maintenance of components, as has been carried

out (in 2022) by the CUASO.

Administration or the complete replacement of

the luminaires and in a procedure called "in the

group".

Considering the mean value of illuminance (of

13 lux) and nominal power (of 250 W), the accessed

MHL light point (Table 3) proved to be

uncompetitive concerning the HPSVL (of 250 W,

manufactured by Peterco, with 8.2 lux) having agreed

to deactivate both lamps.

Without the application of any correction, due to

different spectral emissions of the sources, the SSL

system (LEDi) showed significantly, higher

illuminance and luminance (2.8 cd/m2) compared to

the HPS system (1.3 cd/m2). The luminance

coefficient indicates the possibility of good

performance for the system with SSL technology and

suggests the possibility of reducing the light output.

In the comparison of the systems (Table 3) with

a correlation between luminance and illuminance,

luminance coefficient showed a higher value

(50 mcd/lm) for LED technology of CUASO

concerning MHL and HPS and the condition

established for comparative analysis between

technologies. In a previous work and methodology

(in 2002 and luminance meter manufactured by

LMT, model L 1009), luminance coefficients of 96

mcd/lm were obtained for a system carried in France

and 77 mcd/lm carried out in São Paulo, Brazil, for

the HPSVL technology [18]. In the present analysis,

note that the ability of the pavement to return incident

light is considered. When lighting is analyzed based

on the concept of illuminance, the focus is

exclusively on equipment. It is imperative to include

the floor and the user as part of the lighting system.

Both for the capacity to produce light per electrical

power of the installation (HPSVL in France with 150

W and in Brazil with 250 W, nominal) and for

average luminance produced under nominal

conditions, the results are similar and above 3 cd/m2.

Based on a period of about 4 (four) and a half

years, an average rate of 3% per year for light

depreciation was estimated in a region that does not

have vehicle traffic whose particulate emissions can

be significant. This value made it possible to

formulate a hypothesis for the installation, in general,

that in September 2023 (10 years from the initial

energization) could present a reduction in light output

of 30% or reach the typical end-of-life level called

L70. The present result aligns with the depreciation

rate for the accumulation of dirt of 3% per year, as

fixed by literature [19]. This suggests the

incorporation of cleaning practices as a maintenance

procedure by PUSP-C in CUASO.

The use of LED luminaires in public lighting is

a reality and its expansion requires case studies that

can reveal the useful life of this equipment since this

information is fundamental for the financial models

adopted. This article aims to contribute to these

studies, providing data that can be compared.

Acknowledgements

To Eng. Leonardo Brian Favato, Campus

Administration of USP - PUSP-C, for spreadsheets

and information on maintenance at CUASO, and City

Hall of Santo André, São Paulo state, for budget

information to the acquisition of SSL luminaires

(LEDi) in the Brazilian market.

References

[1] ASSOCIAÇÃO BRASILEIRA DE NORMAS

TÉCNICAS NBR 5101, Iluminação pública —

Procedimento (Public road lighting —

Procedure), segunda edição, 35p., 2012.

[2] Alper Energia, Relatório luminotécnico:

Medições USP, São Paulo, p. 25, 2019.

[3] Department of Energy, Solid-State Lighting

Research and Development: Multi-Year Program

Plan, p. 68, Figure 5.5: White Light LED

Package Efficacy Projections for Commercial

Product, April, 2012.

[4] Systems Reliability Consortium, LED luminaire

lifetime: recommendations for testing and

reporting, Third Edition, September, 32 p., 2014.

[5] Lighting Europe, Evaluating performance of

LED based luminaires, 20 p., 2018.

[6] E.C. Burini Junior, J. C. M. Melero, L. B. Favato,

I. K. Otero, L.O. Assaf, I.L. Sauer, Iluminação a

Estado Sólido (SSL-LED) na CUASO-USP após

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

235

Volume 3, 2023

38 kh da migração tecnológica, XVI Congreso

Iberoamericano de Iluminación LuxAmérica

2022 – Chile, p. 315–323, November, 2022.

[7] E.C. Burini Junior, I.L. Sauer, J.C.M. Melero,

Manutenção Corretiva para a Iluminação Viária

na Cuaso. 2021, Anais. São Paulo: Instituto de

Energia e Ambiente, Universidade de São Paulo,

2021. Available in:

https://prceu.usp.br/congresso/2021/11/25/manu

tencao-corretiva-para-a-iluminacao-viaria-na-

cuaso/. Access in February 5, 2023.

[8] E.C. Burini Junior, Radiometria, colorimetria,

fatores e respostas, Revista Lumière (Printed) ,

v.16, n.192, p.88-96, 2014.

[9] A.R. Robertson, Computation of Correlated

Color Temperature and Distribution

Temperature. J. Opt. Soc. Am. v. 58, p. 1528,

1968.

[10] E.C. Burini Junior, J.I.B. de Moraes, E.R.

Santos, Avaliação em sistemas para iluminação

viária na Cidade Universitária Armando de Salles

Oliveira, XII Congreso Iberoamericano de

Iluminación LuxAmérica 2014, p.44-48.

[11] E.C. Burini Junior, E.R. Santos, H.A. Furuya,

Light Output Measurement of Solid State

Lighting Technology. In: 2019 Fourth

International Symposium on Instrumentation

Systems, Circuits and Transducers (INSCIT).

IEEE, p.1-6, 2019.

[12] Design Manual - Street Lighting. IOWA,

SUDAS, Section 11B-1 - Luminaires, Figure

11B-1.04: Luminaire Cutoff Diagram, p.8,

Revised: 2020 Edition.

[13] Color Calculator Software, OSRAM Sylvania,

Inc., Version 7.77, September 2019. Available at:

<https://www.osram.us/cb/tools-and-

resources/applications/led-

colorcalculator/index.jsp>. Access in December

10, 2023.

[14] Alper Energia, Lighting report: Measurements

USP, São Paulo, p.18, 2016.

[15] IES LM-50 Guide for Photometric

Measurement of Roadway Lighting Installations,

2013 Edition, August 27, 2013. IES Guide for

Photometric Measurement of Roadway Lighting

Installations. Journal of the Illuminating

Engineering Society, v. 18, n. 2, p. 136–138,

1989.

[16] T. Migliorato, Partial map prepared, 2016.

[17] Missive to Eng. Enea Neri e Eng. Paulo Strazzi,

Proposal for changes to the IP project for the

IEE/USP area, (the figure is based on the drawing

on sheet 21, file:<USP-EL-IMP--21.pdf>, date

from 20/Maio/2013), 7 p., São Paulo, August 2

2013.

[18] E.C. Burini Junior, Refletância, Visibilidade e

Luminância como Fatores para Melhoria da

Iluminação Pública, Universidade de São Paulo,

Instituto de Eletrotécnica e Energia, 320p., São

Paulo, SP., 2002.

[19] R. Gibbons, M. Palmer, J. Meyer, T. Terry, IES

RES-1-16 Measure and report luminaire dirt

depreciation (LDD) in LED luminaires for street

and roadway lighting applications, Final Report

prepared for the Illuminating Engineering

Society of North America, p.73, Tab. 7. January

2016.

The authors equally contributed to the present

research at all stages, from the formulation of the

problem to the final findings and solution.

No funding was received for conducting this study.

Conflict of Interest

The authors have no conflicts of interest to declare

that are relevant to the content of this article.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

DESIGN, CONSTRUCTION, MAINTENANCE

DOI: 10.37394/232022.2023.3.22

José Carlos Martinez Melero, Emerson Roberto Santos,

Elvo Calixto Burini Junior, Hédio Tatizawa

E-ISSN: 2732-9984

236

Volume 3, 2023

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself