aanndd
LLiigghhttLLaabb
IInntteerrnnaattiioonnaall
4 Lincoln Street,
GLEN IRIS VIC 3146
Tel: (03) 9546 2188
Fax: (03) 9562 3717
Unit 1, Sandown Square,
56 Smith Road,
SPRINGVALE VIC 3171
Tel: (03) 9546 2188
Fax: (03) 9562 3717
RReeppoorrt tt
oonn
EEnneerrggyy SSaavvi iinnggs ss OOppppoorrt ttuunni iit tti iiees ss
i iinn
SSt ttrreeeet ttl lli iigghht tti iinngg
f ffoorr
aanndd
June 1999
SEDA and SEAV Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
File: 244
Contents
EXECUTIVE SUMMARY ……………………………………………………………………………………………………………… 1
Aim…………………………………………………………………………………………………………………………………………………………. 1
Findings………………………………………………………………………………………………………………………………………………….. 1
Strategy…………………………………………………………………………………………………………………………………………………… 1
INTRODUCTION………………………………………………………………………………………………………………………….. 2
Aim…………………………………………………………………………………………………………………………………………………………. 2
Report Organisation ………………………………………………………………………………………………………………………………… 2
Project Timing…………………………………………………………………………………………………………………………………………. 2
STREET LIGHTING IS IMPORTANT……………………………………………………………………………………………………. 3
Function………………………………………………………………………………………………………………………………………………….. 3
Safety ……………………………………………………………………………………………………………………………………………………… 3
Cost………………………………………………………………………………………………………………………………………………………… 3
Environment ……………………………………………………………………………………………………………………………………………. 3
OVERVIEW OF CURRENT STREET LIGHTING INFRASTRUCTURE…………………………………….. 4
THE STREET LIGHTING TASK …………………………………………………………………………………………………………. 4
What Should be Illuminated? ……………………………………………………………………………………………………………………. 4
How are the Street Lighting Areas Classified? ……………………………………………………………………………………………. 4
How Should these be Illuminated?…………………………………………………………………………………………………………….. 4
WHICH LIGHTING EQUIPMENT SHOULD BE USED?……………………………………………………………………………. 5
Overview…………………………………………………………………………………………………………………………………………………. 5
Is Lighting Colour Important? ………………………………………………………………………………………………………………….. 5
Other Equipment Selection Criteria…………………………………………………………………………………………………………… 7
HOW DOES THE PRESENT STREET LIGHTING INFRASTRUCTURE RATE?………………………………………………. 8
Light Colour ……………………………………………………………………………………………………………………………………………. 8
Light Distribution…………………………………………………………………………………………………………………………………….. 8
Efficiency ………………………………………………………………………………………………………………………………………………… 9
Overall Street Lighting Assessment………………………………………………………………………………………………………….. 11
OVERVIEW OF BEST PRACTICE STREET LIGHTING ………………………………………………………….. 12
AUSTRALIA AND NEW ZEALAND………………………………………………………………………………………………….. 12
New Australian New Zealand Standards :…………………………………………………………………………………………………. 12
Lighting for Main Traffic Routes : ……………………………………………………………………………………………………………. 12
INTERNATIONAL…………………………………………………………………………………………………………………………. 13
North America……………………………………………………………………………………………………………………………………….. 13
Europe ………………………………………………………………………………………………………………………………………………….. 14
STREET LIGHTING RECOMMENDATIONS ……………………………………………………………………………. 15
EQUIPMENT ……………………………………………………………………………………………………………………………….. 15
Available Light Sources………………………………………………………………………………………………………………………….. 15
Comparing Light Sources……………………………………………………………………………………………………………………….. 15
Lamps for the Lighting of Residential Streets…………………………………………………………………………………………….. 16
Lamps for the Lighting of Main Roads: …………………………………………………………………………………………………….. 17
Lanterns for Lighting Minor Roads ………………………………………………………………………………………………………….. 17
Lanterns for Lighting Main Roads ……………………………………………………………………………………………………………. 18
Ballasts …………………………………………………………………………………………………………………………………………………. 18
Lighting Control…………………………………………………………………………………………………………………………………….. 19
System Costs and Benefits ………………………………………………………………………………………………………………………. 20
IMPLEMENTATION………………………………………………………………………………………………………………………. 21
Field Trials……………………………………………………………………………………………………………………………………………. 21
Industry Facilitation ………………………………………………………………………………………………………………………………. 21
Representation on Standards Committee…………………………………………………………………………………………………… 21
FURTHER INVESTIGATION……………………………………………………………………………………………………………. 22
The Installation of Lighting Systems :……………………………………………………………………………………………………….. 22
Hardware Choice…………………………………………………………………………………………………………………………………… 22
Maintenance :………………………………………………………………………………………………………………………………………… 23
REFERENCES………………………………………………………………………………………………………………………………. 23
APPENDIX 1: NSW FLUORESCENT LANTERNS………………………………………………………………….. 24
APPENDIX 2: MAGAZINE ARTICLE, NEW STREET LIGHTING STANDARD ………………….. 25
APPENDIX 3 LAMPS COMPARISON……………………………………………………………………………………… 29
APPENDIX 4: COMPARISON OF TWO MINOR ROAD LANTERNS……………………………………. 31
APPENDIX 5 SEMI-CYLINDRICAL ILLUMINANCE:…………………………………………………………… 32
A Personal Note from Kevin Poulton……………………………………………………………………………………………………….. 32
APPENDIX 6: GLOSSARY ………………………………………………………………………………………………………. 34
APPENDIX 7: ABOUT THE AUTHORS ………………………………………………………………………………….. 35
SEDA and SEAV 1 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Executive Summary
Aim
This document aims to:
• identify opportunities to reduce the energy consumption of street lighting, while
improving the quality of illumination, and
• describe actions and a strategy which will result in realising the potential savings.
Findings
The key findings of this report are:
• present street lighting systems are inefficient, in that:
• the quality of illumination of both minor and major roads is much lower than can be
achieved, and
• the energy efficiency is low.
• there is a basic mismatch between the light colour produced by many street lights and the
light colour which the human eye can use under typical street lighting vision conditions.
• the quality of street lighting can be significantly improved, and the energy consumption at
least halved by a combination of:
• more efficient lamps (eg. metal halide and compact / tubular fluorescent).
• more efficient lanterns (reflector design, less light loss in diffuser, more accurate light
distribution without a refractor bowl),
• more efficient ballasts, especially electronic ballasts.
• more accurate control of lighting times (electronic photo-switch rather than the
existing cadmium sulphide cells, to reduce burning time by at least an hour per day (9%)).
• the capital cost premium of energy efficient street lighting is small and is justified by the
very high return1 on the small premium, and:
• the cost of energy efficient street lighting equipment is very likely to fall as the
production volumes increase.
• the cost of upgrading street lighting efficiency is comparatively low now because the
majority of NSW and Victorian minor roads street lights are due for replacement now or
within five years.
• the higher cost of more accurate, electronic photo-switches is justified by their longer
life, with energy savings due to shorter lighting hours being a free benefit.
• mercury vapour is reputed to have low maintenance costs, but this reputation is largely a
result of the practice of not replacing lamps even when light output has fallen excessively
because of lamp fading.
Strategy
This report recommends:
• field trials of a range of street lights,
• facilitating manufacture of energy efficient street lighting equipment, and
• SEDA / SEAV representation on Standards Australia committees relevant to street
lighting.
1 74% p.a. for the street lighting solution described in “System Costs and Benefits” on page 20.
SEDA and SEAV 2 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Introduction
Aim
Terms of Reference
The brief for this project stated that the aims were:
• assessment of the current technology used in providing street l ig ht in g i n Vic to ri a a nd
NSW,
• assessment of interstate and international best practice, and
• recommending methods of reducing energy consumption and maintenance costs.
This report does not discuss the issue of competitive supply of street lighting services, as this
is well covered by Reference 4 (please see page 23).
Project Team Aim
The project team hope that this document will promote discussion and investigation of the
street lighting task and strategies.
Report Organisation
This report is organised into the following sections:
• Street Lighting is Important One page overview of its importance and impact
• Current Street Lighting Infrastructure The street lighting task and how it is now tackled
• Best Practice Street Lighting Providing the greatest? benefits at the lowest cost
Other supporting documentation is in the Appendices.
Project Timing
This is an opportune time to be reviewing street lighting:
• In Victoria, about 200,000 mercury vapour (80 Watt) lanterns were installed starting in
1989. This project saw the replacement of twin 20 Watt fluorescent lamp lanterns. The
new street lights have an estimated service life of 15 years, and so planning of the next
generation of replacement fittings must start now. This will allow adequate time for
design, hardware selection and / or development, and field trials.
• NSW has about 400,000 fluorescent street lighting lanterns2, which are programmed for
replacement now. If NSW were to adopt the same solution to fluorescent lantern
replacement which Victoria took:
_ the capital would be about $M70
_ energy consumption for these lanterns would double, and
_ the recurring greenhouse gas emissions will be 146 kilo-tonnes3 every year.
2 Details of the NSW fluorescent street lighting lantern numbers and costs are in Appendix 1 on page 24.
3 based on 158.9 GWh/year (see Appendix 1) and a greenhouse intensity of 0.92 kg/kWh.
SEDA and SEAV 3 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Clearly, the benefits of the NSW replacement of fluorescent street lighting need to be
maximised while minimising the costs to the community and the environment.
Street Lighting is Important
Function
Good street lighting contributes to the quality of life, by improving personal safety and
perceived safety, and improving the appearance of the local environment. As a corollary,
poorly designed lighting systems can degrade the quality of life, for example, by making
vision more difficult, or through intrusive unwanted light.
Safety
There is a considerable body of knowledge which indicates that the provision of lighting on
roadways has a significant effect on the lowering of accident rates at night.
Therefore the provision of acceptable lighting systems on all roadways is not just a matter of
amenity, but also quite literally a matter of life and death.
For most people, probably the worst fear when venturing out at night is that they may be
accosted by a stranger. If the lighting, and particularly the vertical lighting, is insufficient it
will be very difficult for anyone to make a confident judgement as to whether or not someone
approaching is a friend or foe. We are of the opinion that this widely felt fear is one of the
most significant reasons why people do not wish to go out into public spaces during the hours
of darkness.
Cost
However the provision of street lighting systems comes at a significant cost to the
community.
Street lighting costs the Australian community about $M156 per year in energy alone. The
total cost of providing street lighting, including provision and maintenance of the electricity
distribution network and the street lighting equipment, is about 6 times4 the energy cost
alone. This brings the Australian total to about $M900 per year.
Environment
Street lighting affects the local environment (positively and negatively) by the provision of
wanted and unwanted light. The global environment is affected by greenhouse gas emissions,
and by the depletion of finite fossil fuel and other material resources.
4 Source: Reference 2: Coopers & Lybrand, “Report to IPART on Street Lighting Review” March 1988.
SEDA and SEAV 4 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Overview of Current Street Lighting Infrastructure
The Street Lighting Task
What Should be Illuminated?
Clause 2.1, General Objectives of AS/NZS 1158.1.1, 1997, clearly states :
“To accomplish these objectives, the lighting must reveal necessary visual
information. This consists of the road itself, the course of the road ahead, kerbs,
footpaths, property lines, road furniture and surface imperfections, together with
all road users including pedestrians, cyclists and vehicles and their movements,
and other animate or inanimate obstacles.”
Despite this clear statement, in practice there is a strong tendency of the standard and
designers to concentrate only on the lighting of the road surface and to neglect pedestrian
traffic and precinct areas, etc.
Even a casual reflection on this situation will reveal that this is far from an ideal situation. Of
all the objects which can be illuminated, the horizontal road surface probably presents the
least hazard.
How are the Street Lighting Areas Classified?
The current series of Australian/New Zealand Standards AS/NZS 1158 recognise three
categories of lighting systems, and these may be broadly described as follows :
Category V Applicable where the visual requirements of motorists are dominant
Category B
Category C
For roads and other public thoroughfares where the needs of pedestrians
are dominant
How Should these be Illuminated?
Light direction
The current edition and prior editions of AS1158 have only used horizontal illuminance5 as
the illuminance criteria.
The use of horizontal illuminance is an understandable expediency, as:
• measuring the horizontal illuminance requires only a single measurement at each point,
whereas vertical illumination necessitates also deciding in which direction to make the
measurement, and recording both the direction and the measurement, and
• it is a natural extension of the practice of measuring horizontal illuminance which is relevant
in office and other workstation lighting, where the work (writing, equipment, etc) is
horizontal.
But horizontal illuminance is a very poor indicator of the amount of useful light upon an
object such as the human body. Horizontal illuminance would be a good indicator of an
objects visibility from a helicopter, but not from the viewpoint of a motorist or pedestrian.
Good street lighting requires good vertical illumination, preferably from more than one
direction (please refer to Appendix 2).
5 Please see the Glossary on page 34 for a description of the key lighting terms used in this report.
SEDA and SEAV 5 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Which Lighting Equipment Should be Used?
Overview
The answer to this question lies partly in the answer to the previous one (light direction), and
partly in:
• the desired colour of light (governed by lamp type).
• the desired distribution of light (lantern characteristics:
_ lamp
_ reflector
_ diffuser / refractor)
• the desired brightness
• the desired operating hours (control device, usually a photo-switch but sometimes
methods such as ripple control are used).
• mounting options (usually existing power poles but sometimes dedicated
lighting poles).
• cost factors (energy, maintenance, capital cost, economic life, etc.)
Is Lighting Colour Important?
The colour of light, and especially light from artificial sources, varies widely: from the
“warm” glow of candlelight to the very blue light of the mercury vapour lamp and its
derivatives (tubular fluorescent and metal halide). But not all lighting colours are equally
suitable for all lighting tasks, and this is especially true of most street lighting applications.
To fully understand this matter it is necessary to have an understanding of how the human
vision system operates under various lighting conditions.
The human vision system
The human eye has the ability to adapt to ambient lighting conditions in both daylight and
night-time conditions; from over 100,000 lux to much less than 0.1 lux ( a ratio of more than
one million to one). If there is a lot of available light, for example in a well lit office or out
of doors in the day time, this condition is described as Photopic Vision which means light
adapted. Alternatively if the eye is operating in night-time conditions it is described as having
dark adapted or Scotopic Vision.
However there is an in-between adaptation situation described as Mesopic Vision, and the
lighting conditions on most of our main roads asks the human eye to operate within this
Mesopic Vision range. The human eye is using mesopic vision when the ambient “light
level” is between about 0.1 lux and 10 lux.
Why is this important? It must be remembered that luminance is a physical parameter which
can be measured by an appropriate photometer, but what the human eye percei ves is the
sensati on of bri ghtness6.
6 Key lighting terms used in this report are defined in the glossary on page 34
SEDA and SEAV 6 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Light colour and vision
The relationship between luminance and brightness is very dependent upon the colour of the
light source.
Under mesopic vision conditions, the luminance (measured value) and the brightness
(perceived value) of a surface under a blue white light source, is much more closely related
than when under an orange-red light source. The eye finds light at the blue end of the visible
spectrum much more useful than light at the red end of the spectrum. Light meters are,
however, calibrated for the eyes sensitivity to different light colours at much higher lighting
levels (ie. under photopic conditions).
Light colour and lighting design
The primary light parameter used in main road lighting is Luminance (candelas per square
metre). Table 2.1 of the Australian Standard AS/NZS 1158.1.1, 1997, Road Lighting, Part 1.1
Vehicular Traffic (Category V) Lighting, specifies values for this parameter for various
classes of roads.
However, the lamp / lantern performance data used as an input to lighting designs, and the
light meters used to assess the resulting lighting installation are based on the total light output
(lumens) and illuminance (lux) assuming that the eye will have the same response to the light
as it would at much higher lighting levels (eg. daylight, which could be over 100,000
brighter). This project team does not agree with such an extrapolation.
Light colour and lamp selection
Light at the blue end of the spectrum is more of an aid to vision under mesopic conditions,
and so this means that the mercury (blue white) based family of lamps is far more appropriate
for than is the sodium (orange red) family of lamps.
Unless the lighting or luminance values are very high and at least twice the values
recommended in Table 2.1 of AS/NZS 1158.1.1, 1997, people will not be using photopic
vision.
At the lower luminance (scotopic and mesopic vision), the relationship between luminance
and brightness will not be the same as during photopic vision. The correlation which exists
for yellow-red light (including high pressure sodium lighting) brightness (visibility) during
the day will not exist at the lower luminance.
In other words, the yellow-red light is less effective in aiding vision at low lighting levels,
and so much of the light produced is effectively wasted.
SEDA and SEAV 7 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Other Equipment Selection Criteria
Lamp
We have described how colour is an important factor in lamp selection. Other factors are:
• efficacy; how much electricity is required to produce the light.
• lumen maintenance; how well is the lamp efficacy maintained throughout the lamp’s life.
• lamp life and reliability.
• initial cost.
• physical considerations; size, robustness, lamp holder type, etc.
Lanterns
We have already described the need for street lighting to illuminate vertical surfaces well,
and this will influence the selection of lanterns. Other factors affecting the selection are:
• efficiency:
_ light output ratio (LOR); what portion of the light produced by the lamp actually escapes from the
fitting,
_ suitable distribution of light, and
_ control of intrusive and upward light.
• service life expectancy,
• maintenance (how easy is cleaning, lamp replacement, photo-switch replacement), and
• capital cost.
SEDA and SEAV 8 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
How Does the Present Street Lighting Infrastructure Rate?
Light Colour
minor roads / residential streets
In Victoria, most minor road lighting is provided by the fluorescent coated ellipsoidal
mercury lamps. Its colour is acceptable, in that its spectral content is mainly at the blue end
of the spectrum.
The fluorescent lanterns used to light NSW minor roads would also be biased toward the blue
end of the spectrum, though because of the greater selection of tubular fluorescent lamp
coatings, the light colour can vary widely.
major roads / traffic routes
In both NSW and Victoria, some major road lighting is provided by mercury vapour lamps,
but the majority is provided by sodium lighting. There is also a continuing trend to replace
mercury vapour lighting with sodium lighting. We strongly recommend against this move.
(Note: While we are convinced of the correctness of this stance, both by
lighting science and our own observations, we understand that other people
in the lighting industry will not be persuaded easily. Therefore, we believe
field demonstrations will be required. This is discussed on page 21.)
Light Distribution
Vertical illumination and efficiency
The present system design and performance is a result of the concentration on horizontal
illuminance. As already discussed, this is not a useful parameter by which to judge street
lighting quality.
On main roads especially, providing vertical illumination while controlling glare is a
balancing act. There are conflicting requirements of controlling glare, by designing the
lantern light “cut-off” and illuminating vertical surfaces. Reducing lantern spacing and
increasing mounting height both help, but both incur additional capital costs.
However, a reasonable portion of the light produced by most minor road lighting
(mercury vapour B2224 refractor lantern or “flower pots” and fluorescent lanterns) does emit from
the lantern as vertical light. However, this is not the result of sophisticated lantern design; it
is the result of almost no design which allows the light to go in all directions. And light going
in all directions has a cost:
• overall efficiency is reduced,
• light pollution (intrusive and upward light) is increased,
and so more electricity is required to achieve the street lighting task.
Glare
The existing minor road lanterns present a significant visual distraction in that there is either
an exposed fluorescent lamp (NSW) or a high brightness refractor (Victoria) in clear view.
SEDA and SEAV 9 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Upward and intrusive light
Upward light pollution makes astronomical observations more difficult, degrading the
experience of both keen and casual observers. While there will also be light reflected upward
from roadways and other objects, reducing direct upward light will clearly assist in reducing
the problem. Also, direct upward light is wasted light, and so reduces the efficacy of the
street lighting system.
The reflector based HID lanterns used on main roads do not present an upward light problem
and do not normally create obtrusive light.
The refractor based mercury vapour lanterns used in Victorian minor roads, and their
f luor escent count er par ts in NSW pr oduce a si gnif i cant am ount of unnecessary upward l i ght.
Efficiency
Lamp efficacy – sodium vapour lighting
Sodium lighting has a high initial efficacy (lumens per Watt) and a high lumen maintenance
(i.e. efficacy stays high throughout the lamp life). However, these lumens are of little real
benefit if they register only on lux meters and not in peoples’ eyes.
When people go out onto the streets at night, they take eyes, not lux meters
Although mercury vapour lighting has acceptable colour, its low efficacy in terms of lumens
per Watt is not acceptable in today’s more energy conscious world. Even when new, the
efficacy of an 80 Watt mercury vapour lamp is under 50 lumens per Watt, and the efficacy
will degrade continuously throughout the lamps service life.
New Technology Fluorescent Lamps
The technology of fluorescent lamps has advanced considerably since the development of the
lamps now used in NSW minor roads lanterns. Even in the decade since Victoria replaced
most of its fluorescent street lighting, the technology improved greatly:
• lamps with a rated life of up to 24,000 hours are now available, compared with 8,000 for
fluorescent lamps of ten years ago and 12,000 for mercury vapour lamps.
• colour rendering has improved.
• efficacy has improved to as much as 100 lumens per Watt for lamps of over 30 Watts.
Therefore, judgement of modern fluorescent lighting should not be based on memories of
fluorescent lamp technology of the 1960s and 1970s, or even the 1980s.
In other words, old NSW fluorescent street lighting should not be compared with modern
alternatives without also considering modern equivalents of tubular fluorescent and compact
fluorescent lamps. This is discussed further on page 15.
SEDA and SEAV 10 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Time control
Street lighting should operate when it can make a contribution to visual amenity, and not
when it can’t. Street lighting control is achieved by photo-switches in Victoria and a mix of
photo-switch and electricity mains signalling (“ripple control”) in NSW. There is a trend to
replace ripple control with photo-switch control, and we expect this trend to continue.
The standard street lighting photo-switch uses a cadmium-sulphide cell. This device can be
installed in the casing of a single lantern or can be used to control a group of lanterns by
switching the power in a dedicated street lighting conductor or “fifth wire”.
The cadmium sulphide photo-switches have several disadvantages:
• they have a rated service life of only 7 years.
• they have a switch-off illuminance of three to five times the switch-on illuminance.
Thi s me a ns t h at i f t he d e si re d s wi tc h -o n il l umin a nc e is se t c or re ct l y, t h e swit c h- of f t ime wil l b e
s ig ni fi c an tl y l at er th an ne ed ed (a dd i ng a bo u t 15 mi nu te s 7 to t he da il y o pe ra t in g time).
• the switch-on setting is typically in the range 30 – 60 lux, which is much higher than the
ambient illuminance at which the street light can make a contribution to vision.
This unnecessarily adds about a further 20 – 30 minutes3 per day to burning time.
• the switch on setting drifts by about 10% per year, causing the lights to switch on even earlier
and off later. By the end of the 7 year service life, the switch-on point and switch-off point
will both have doubled. Again, this will add to unneeded burning time (about another 15
minutes3 per day, averaged over the service life of the photo-switch).
• they consume about 2 Watts when the lights are turned off. This is not a large power draw,
but it is unnecessary.
There is an alternative: electronic photo-switches, which are discussed on page19:
7 the times described in this section are indicative, being based on readings in Melbourne only, and only during
May. A more definitive analysis of the effects of control changes would require further simulation and
calculations, and is beyond the scope of this report. However, a more definitive study would not alter the
conclusions in this section of the report.
SEDA and SEAV 11 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Overall Street Lighting Assessment
Service quality
Minor road lighting (mercury and fluorescent) generally has acceptable colour and a
reasonable portion of the total light is delivered as vertical illumination
Major road lighting generally has a red-orange colour which is unsuitable for night vision.
Efficiency
The efficacy of mercury vapour lamps is unacceptably low, and this is further degraded by
the inefficiency of the lantern – refractor.
The LOR of most major road lanterns is acceptable, but the overall effectiveness of the
lighting is low, because of the poor response of the eye to the light colour.
The lighting system is also degraded by its unnecessary operation, caused by inaccurate
photo-switch control.
Potential
These shortcomings provide the potential to improve street lighting quality (safety, amenity,
etc.) while also reducing financial and environmental costs. A win-win situation. The fact
that these improvements can be made now, when significant capital expenditure is required
anyway, makes the potential easier to achieve.
SEDA and SEAV 12 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Overview of Best Practice Street Lighting
Australia and New Zealand
New Australian New Zealand Standards :
A new draft of an Australian/NZ Standard for residential areas has recently been prepared
and is presently under public review. It is called AS/NZS 1158, Part 2, Pedestrian Area
(Category P) Lighting. This draft differs from the old AS 1158.1-1986 in that all lighting for
residential roads, pathways and circulation areas will be designated Category P, replacing the
older B and C Categories.
A copy of an article by Fisher and Rogers printed in the May 1999 edition of “Lighting,”
Vol.19, No.3, discusses the changes from Categories B and C to Category P, and is appended
(Appendix 2) for your reference. As stated by the authors in their summary, this is a major
revision of the Standard in relation to lighting for pedestrian traffic.
The major change is the inclusion of a vertical illuminance parameter, but while this is a
useful criterion it is also very simplistic. In our opinion the inclusion of semicylindrical
illuminance (Ehz) would have been far more meaningful for situations involving the
modelling of people. (Appendix 2 refers).
Another concern we have is that the illuminance values quoted in the table of “Values of
Light Technical Parameters for Roads in Local Areas and for Pathways” are very low, and
we have grave doubts as to the easy availability of photometers to accurately measure these
small values.
This is particularly so because the currently used HPS lamp is strongly biased towards the
orange-red end of the spectrum, while in a similar manner the metal halide lamp is strongly
skewed towards its blue end. These are the colour bands in which the greatest errors will
occur if a conventional photopic-calibrated lux-meter is used.
Lighting for Main Traffic Routes :
The current road lighting practice as set out in the Australian New Zealand Standard AS/NZS
1158.1.1-1997 (Category V) follows European practice rather than the IESNA (North
American) methodology.
The European practice has evolved from a framework developed over many years by the
Commission Internationale De L’Eclairage (CIE), much of it under the guidance of Dr. A. J.
Fisher of Sydney, who is the current Chairman of the Standards Australia Road Lighting
Committee LG/2.
The European Committee for Standardisation (CEN) produces European road lighting
standards which also follow the CIE guide-lines, and so it would be very difficult to make any
major break from the principles of AS/NZS 1158. Minor changes depending on local factors
are possible, but probably not any national change.
SEDA and SEAV 13 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
International
This sections gives a brief description of street lighting practice and trends. A more detailed
description is in Reference 5 (page 23).
North America
Summary
The trend in North America is toward street lighting standards based on performance,
assessed according to the visibility of a standard small “target”8. We agree with this move
from an assessment of lighting based on simple and inappropriate parameters such as
horizontal illuminance.
Description
The current IESNA9 recommended road lighting design method is based on the use of either
illuminance or luminance10, but this body is now developing a standard based on visibility.
Of course, this means that the road lighting design aim will be more closely aligned with the
goal of road lighting: to make objects visible. This is a very sensible aim, but a very big task.
With a system based on horizontal illuminance, the focal point of creating the standard is
deciding on a value for that single parameter, for a limited number of situations. However,
with standard based on assessing visibility afforded by the lighting system it is necessary to
decide:
• which factors will affect visibility, and
• how these will be assessed and incorporated into the standard.
The factors identified by the IESNA are:
• adaptation luminance,
• contrast between target and background,
• veiling luminance,
• target size,
• observation time, and
• observer age.
These factors each hide a wide range of options. For example, observation time will clearly
be dependent on the speed of the observer and the “target”. And if the observer is a motorist,
the greater the speed, the more warning is required, and so the target needs to be visible from
a greater distance. Target size could represent a black cat, a child, an adult, a cyclist, or some
other object, or could represent range of targets. Observer age is also a sweeping
simplification; not everyone of the same age has the same visual ability.
8 “target” refers to a visual target, but given that the device represents a pedestrian, it is a poor word choice.
9 Illuminating Engineering Society of North America
10 please see the glossary on page 34
SEDA and SEAV 14 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
There are other factors which could be considered such as:
• the difficulty of the driving task (some roads present more hazards and distractions than
others).
• the position(s) of the target with respect to the observer (on the road, on the kerb, etc.).
Once such a standard is written, performance of road lighting systems need to be assessed
according to the standard. This can be done using:
• computational methods,
• measurement using CCD technology developed for video cameras.
Relevance
We believe that the trend towards this performance assessment based design strategy will
continue in North America and be adopted in other countries. Australia needs to watch and
participate in this process.
Europe
Developing European street lighting standards involves ensuring that the standards are
relevant and acceptable to the member countries. Within this environment, the process
focuses on producing a framework for designing lighting systems appropriate to local needs.
The four main activities are:
• classifying the lighting situation based on factors such as:
• weather types in that location,
• type of road (eg. divided carriageway, number of lanes),
• vehicle speed,
• background visual complexity.
• deciding on the performance required of the lighting system for each road classification.
Lighting performance for main roads is based on the silhouette principle, and horizontal
luminance is still the main parameter.
For pedestrian areas, more complex and subtle performance measures are also specified,
including semi-cylindrical illuminance (see page 32) hemispherical illuminance and vertical
illuminance. This is particularly relevant to some countries where very low mounting height
lanterns are used, and achieving horizontal illuminance between the lanterns is almost
impossible. With the trend to under-grounding of electricity cables, this will become more of
a consideration in Australia.
• calculation methods, and
• assessment methods.
SEDA and SEAV 15 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Street Lighting Recommendations
Equipment
Available Light Sources
The progression of electric lamps for street lighting application has broadly followed the
historic development of the electric lamp. The first lamp type to be used was the incandescent
filament lamp, and this was followed by the early mercury MA and MB type lamps and the
low pressure sodium lamps.
Later replacement of the incandescent lamps with tubular fluorescent lamps for both main
roads and residential streets was the next development, and in more recent times these
fluorescent lamps have again been replaced by mercury vapour (in Victoria) in lamps of a
variety of wattages, and in many main roads by high pressure sodium lamps.
The most recent developments in lamp technology, apart from the Sulphur or Induction
lamps, has been in small watt metal halide and compact fluorescent lamps, but to date neither
of these lamp types have had wide application in street lighting.
This wide variety of available lamps suggests that it is not unreasonable to question whether
or not one type of lamp is better than another.
Comparing Light Sources
As shown in Appendix 3, there is a number of discharge lamps available which have a lumen
output approximating that of the 80 Watt mercury fluorescent lamp. All of the lamps shown
in Appendix 3 may be considered as long life, when compared to the life of GLS
incandescent or Tungsten Halogen lamps. Both of the values; life and lumen output must be
viewed with a certain degree of caution, since some manufacturers tend to be conservative in
their claims in order to minimise criticism of the performance of their products, while others
are more optimistic.
The technical reason for this is based on an assumption about when a batch of lamps is
considered to be at the end of its life. The conservative school of thought claims the average
life is when 20% of the lamp batch has failed, while the optimistic school says that this point
is reached when 50% loss is noted. It is therefore obvious that the conservative group will
show a shorter lamp life when compared with the latter group.
Similarly there are several methodologies used for determining the lumen output of lamps so
the two groupings of conservative and optimistic are also evident here, but general consensus
takes the lower (conservative) figure of approximately 3400 lumens for the average 80 Watt
mercury vapour fluorescent lamp.
SEDA and SEAV 16 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Each of the lamp types listed in Appendix 3 has weaknesses, as shown below :
• The single ended mercury Halide Group 1 lamps have G type bi-pin bases which do not cope
well with rough service or vibration.
• The Group 2 HPS lamps in general, but especially the 50 Watt version, have poor colour
renderings and tend to badly distort all colours and especially skin colourings.
• The CFL Group 4 lamps tend to be physically large, and again use pin type electrical
connections which suffer from vibration unless fitted with specially designed bases to support
the lamps. While this is acceptable, it makes lamp replacement a slower process.
Another lamp that has been recently introduced to Australia is the 16mm diameter tubular fluorescent
lamp known as the T5. A similar 14 Watt version (549mm long) produces 1220 lumens which is
approximately 10% more than that from the old 26mm, 18 Watt lamp for a lower Wattage.
The main disadvantage of the tubular lamp for street lighting is that the lumen output of the
lamp is considerably reduced when the ambient temperature is low, and this could be of
serious concern in Tasmania, some parts of Victoria, and in other “high country” areas under
consideration. However, the 14 Watt T5 tri-phosphor lamp is understood have overcome this
problem, and will soon be trialed in NSW.
Lamps for the Lighting of Residential Streets
Recommended
Considering the requirements of efficacy, colour, and life, it is our considered opinion that
the most appropriate light sources for residential streets are:
• the 35 Watt metal halide lamp with a blue white spectrum, and
• compact fluorescent lamps, also with a blue-white spectrum.
The most appropriate for a given situation will depend on factors such as mounting
restrictions (eg. existing poles, new installation, spacing).
A single 35 Watt metal halide lamp in a lantern with appropriate photometric characteristics
will replace a standard 80 Watt mercury vapour lantern. This will result in improved
illumination quality and a power saving of at least 54% (details on page 20).
Compact fluorescent street lighting lanterns are discussed on page 17.
Watching brief
Another lamp type which is very promising is the 50/35 Watt Osram Citylight. This is an
acceptable lumen package which gives good colour rendering, good lumen maintenance, and
the option to switch to a lower wattage (from 50 to 35). However, this lamp type is so new
that we recommend further experience be gained with the lamp before large scale
installations are contemplated. Small-scale trials however are recommended (please see page
21).
SEDA and SEAV 17 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Lamps for the Lighting of Main Roads:
In general and at this time, the most commonly used ranges of lamps for main roads are the
high pressure sodium or mercury vapour 250 and 400 Watt units with occasional use of the
150 Watt lamp. As shown in Appendix 3, the metal halide lamp has more than a 30% higher
efficacy than the mercury fluorescent lamp and the use of this lamp would give a 30%
increase of light on the road surface.
A change from mercury vapour to metal halide lighting will result in a significant reduction
in energy consumption, e.g.: by replacing a 400 Watt mercury lamp lantern (19,000 lm and
circuit power of 425 Watts) with:
• a single 250 W metal halide lamp (20,000 lm) 268 Watts total saving 37%
• 2 x 150 mWetal halide lam(p2s4,000 lm)336 Watts tsoatvailng 21%
The savings with a single 250 Watt lamp are higher, though there may be other advantages in
using two 150 Watt lamps, either in a single lantern or as two single-lamp lanterns, such as:
• more even illumination,
• less glare, by facilitating closer lamp spacing or greater mounting height and greater light “cut
off”,
• simple “lumen set-back” by switching one lamp,
• greater reliability, as half lighting will be available in the event of one lamp failing or the arc
extinguishing.
Lanterns for Lighting Minor Roads
Metal halide
We have modelled the performance of an efficient minor roads lantern based on photometric
test results for a GEC / SLI11 brand “Urban Minor” model lantern. This shows that the
combination of this lantern and a 35 Watt metal halide lamp will produce better illumination
than the present GEC model B2224, 80 Watt mercury vapour lantern which is used throughout
Victoria.
Compact fluorescent
In order to select a lamp configuration for minor roads, we have relied on first principles. The
calculated results can then be used as a “market pull” to encourage manufacturers to supply
appropriate lanterns. Given the size of the market and discussions with manufacturers, this
approach should be successful.
11 Sylvania Lighting Industries
SEDA and SEAV 18 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
What light output is required from the street-light’s lamp(s)?
The required lamp light output can be calculated by:
• starting with the light output of present street lighting lamps, and
• reducing this figure according to the higher efficiency of modern street lighting fittings in
delivering the light from the lamps to the area requiring illumination.
Light produced by an 80 Watt mercury vapour lamp 3400 lm
The existing minor roads lanterns using a refractor glass diffuser with
inherent light loss, and an overall indicative LOR of
60%
And so the amount of useful light which leaves the lantern is 3400 x 60% 2040 lm
We expect a modern, efficient lantern with a clear lens and efficient reflector
to have a LOR of:
80%
And so the next generation of minor roads lanterns will require a lamp output
of: 2040 lumens divided by LOR of 80% 2550 lm
Referring to the table in Appendix 3 (page 29), at the lowest power, a suitable fluorescent
lamp for a lantern to directly replace the current 80 Watt mercury vapour, is a single 36 Watt
compact fluorescent lamp (2800 lumens).
Lanterns for Lighting Main Roads
Because of relatively small changes in lamp technology in recent years, there has been only a
small change in the development of lanterns for main road lighting.
Similarly there are only a limited number of suppliers in this country at present, but perhaps
if restrictions were eased there could be a market opened up to overseas suppliers, some of
whom have already shown interest.
Ballasts
Street lighting should use electronic ballasts, because of the higher efficacy and the
lengthening of lamp life. However, there are no reasonably priced and readily available
electronic ballasts available for discharge lamps. This is a chicken and egg situation: suitable
electronic ballasts are not available because there is insufficient market demand, and the
demand is low because of pricing and product availability. Pricing is also directly related to
sales and production volumes.
The large scale re-equipping of minor roads street lighting in NSW and Victoria brings the
opportunity to overcome this barrier. We recommend that SEDA and SEAV facilitate a local
manufacturer to equip for the volume production of electronic ballasts to suit the compact
fluorescent and HID lanterns described in the preceding pages.
SEDA and SEAV 19 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Lighting Control
Electronic (also called solid “state”) photo-switches with silicon diode sensors are preferable
to cadmium sulphide cells (described on page 10) because they:
• have a rated service life of 12 years, reducing both labour and vehicle costs, and improving
lighting system reliability.
• have a switch-off illuminance which is independent (including lower than) the switch-on
illuminance.
• are available with a standard switch-on settings of 10 lux, reducing lighting times by about 20
minutes per day compared with a 30 lux level. Further, a lower switching level could be
ordered for purchases of large batches.
• have a switch on setting which is stable over the life of the photo-switch.
• consume negligible power.
Electronic photo-switch cost – benefit
The maintenance savings alone justify the price premium of electronic photo-switches.
Assuming that replacing a photo-switch costs $70 ($50 for truck and labour, $20 for call
centre, purchasing and other administrative costs), the annual cost of standard and electronic
photo-switches is:
Photo-switch
Type
Initial Cost Life Total replacement
cost
photo-switch
maintenance cost
$ years $ each time $/year
Standard $9 6 $79 $13.17
Electronic $15 12 $85 $7.08
So electronic photo-switches are justified even without considering the benefits of energy
savings and increased lamp life.
An electronic photo-switch will reduce lantern burning time by an hour per day, and so will
increase the interval between lamp replacement by about 10%.
SEDA and SEAV 20 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
System Costs and Benefits
Minor roads
The following is a comparison of the costs and benefits of a standard 80 Watt mercury
vapour street lantern and the proposed 35 Watt metal halide light:
Before After
Fitting B2224 “Urban Minor”
Lamp type mercury vapour metal halide
Capital Cost $65 $80
Photo-switch upgrade $0 $6
Ballast upgrade $0 not included or estimated12
Total capital cost $65 $86
Lamp Watts 80 35
Ballast Watts 16 8
Circuit Power Watts 96 43
Energy Use
Burning time hours/year 4,335 4,000
Lantern kWh/year 416 172
Photo-Switch kWh/year 9 0
Total kWh/year 425 172
Reduction in energy use n/a 60%
Energy cost / year $/year $34.00 $13.76
Energy Saving $/year $20.24
Maintenance
Initial cost premium ## $15.00
Annual return on initial premium % p.a 74%
Appendix 4 shows that the Urban Minor also delivers more even illumination than does the B2224.
Notes:
# We have estimated a medium term electricity price (weighted average peak and off-peak) of 8 cents
/ kWh, including energy and DUOS and NUOS charges. This differs from the IPART approach of
assigning a value to each component (Reference 2).
## The cost premium does not include the additional cost of an electronic photo-switch, as this
premium is already justified by the lower cost of maintaining the photo-switch.
We have not included the benefit of longer effective lamp life. The mercury lamp has a
shorter effective life than a metal halide lamp, because it will reach an unacceptably low
efficacy sooner. Of course, the mercury lamp could be left in service for longer than a metal
hal ide lamp, as it wi ll oft en last many years without fading completely. However, it’s light output
at that t ime wi ll be unacceptably low and the quali ty of street light ing unaccept ably poor.
12 The price will be very dependent on production volumes and the street lighting market could change the
production volume significantly. There is the potential to reduce electronic ballast prices sharply with street
lighting projects.
SEDA and SEAV 21 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Implementation
We do not expect or anticipate that this brief overview will, by itself, dramatically change
Australia’s one billion dollar a year street lighting industry. Indeed, we expect that our
conclusions will be challenged by many people in the industry, and we hope that this will
prompt further discussion and exploration of what the community expects of street lighting,
and how this can best be achieved.
We are aware of at least some undertakings which will assist this process:
Field Trials
We strongly recommend that field trials of various street lighting lanterns be conducted with
the cooperation of suitably interested electricity distribution businesses and local
government. Such trials will serve to:
• allow people to judge street lighting with their own eyes,
• provide an opportunity for innovative manufacturers to demonstrate their wares,
• refine new hardware, including modifying design with the aim of minimising installation
time, etc
• render tangible some of the abstract concepts described in this report, and so democratise the
process of developing new street lighting systems by facilitating comment by lay people as
well as lighting professionals,
• facilitate the involvement of appropriate groups (eg. IESANZ, road safety and research,
electricity industry, local government, Standards Australia, etc).
Similar trials / demonstrations were conducted in Sydney in the 1960s, and were known as
the “Beecroft Road experiments”.
Industry Facilitation
We recommend that SEAV and SEDA consider facilitating Australian industry’s
development of efficient street lighting lanterns which “push the envelope”. This need not be
an expensive undertaking, indeed we do not recommend any direct subsidy. Instead we
envisage acting as a catalyst, by bringing together potential equipment purchasers (local
government and electricity distribution network companies) and manufactures, to form
buying contracts and volumes which will make the energy efficient equipment viable for both
groups.
Representation on Standards Committee
We believe that SEAV and SEDA should lobby Standards Australia to influence a future
revision of AS/NZ Standard 1158.
SEDA and SEAV 22 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Further Investigation
The Installation of Lighting Systems :
Currently roadway lighting systems are very stereotyped and there appears to be very little
room to move as far as change is concerned.
Some common European practices such as Catenary lighting13 have never been adopted in
this country, and we would suggest that it might be time to revisit this particular
methodology.
With the easy availability of large Elevator Platform Vehicles the opportunity for greater
mounting heights and wider spacing of lanterns is now possible, and we believe that these
should certainly be considered. Obviously this practice would reduce the number of
poles/lanterns required per kilometre, and so achieve a greater saving of energy. The
appearance of the urban streetscape could also be enhanced.
Since few manufacturers presently produce lamps with a Wattage of between 400 and 1000
Watts, the use of multiple heads of lamps eg. 3 x 250 Watts or 2 x 400 Watts may be worth
considering.
In our opinion, opportunities for savings in energy and capital installation costs will
come through the reduction of the number of poles and lanterns provided in any given
installation. This concept can be achieved by significantly increasing the presently used
mounting heights of the lanterns, and we believe that this recommendation should be
fully explored.
Hardware Choice
In current practice the number of types of lanterns used is restricted on the grounds of
limiting the number of spares to be held in store. Similarly HPS and MV lamp types and their
Wattages have been restricted to a narrow range in the pursuit of standardisation.
While these policies may have had some justification in a large organisation as seen in the
older State Electricity Commission, we believe that in the smaller organisations in vogue
today in the form of the Distribution Businesses, greater flexibility should be possible.
Distribution Businesses claim to have policies with regard to added service and value adding,
so the general public should therefore be able to expect better quality service and more
evident concern regarding environmental protection and the need for the conservation of
energy.
Demonstration of these concerns could be shown by the purchase and use of better quality
photoelectric daylight switches for use with the present lanterns, so as to avoid their burning
long after they are required.
It is hoped that with contestability will come not just choice by price or tariffs, but also the
offering of a wider choice of equipment in such things as lamps and lanterns. It is also hoped
13 Catenary lighting means lighting provided by lanterns attached to a horizontal suspension cable.
SEDA and SEAV 23 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
that this contestability will bring about a reduction in maintenance costs and an improvement
of services offered to the general public.
Maintenance :
We understand that the decision as to who is to be responsible for the maintenance of the
public lighting system has not been made and is still under discussion. In the years following
World War 2 the street lighting administration for the Melbourne metropolitan areas was
centralised at the Rooney St Richmond Depot, and patrolling officers from this depot were
responsible for the maintenance of this system.
It is clear that this system has now gone forever, but given the importance of street lighting
within the community it is essential that a similar service level should be resumed. If the
general public was encouraged to report malfunctions of the system to a toll free telephone
number, surely in these days of the all embracing computer a return to such a service level
should not be too difficult.
References
1 Electricity Supply Association Australia “Electricity Australia 1998″
2 Coopers & Lybrand in association with Worley Consultants
“Report to IPART on Street Lighting Review” March 1988
3 Australian / New Zealand Standard AS/NZS 1158.1.1, 1997 Public Lighting
4 Sustainable Energy Authority, Report: “Street Lighting and Contestability” Author Paul Rogers, January
1999.
5 Lighting Magazine, September 1998, Article “Lighting for Safer Roads on Three Continents” pp 40-45.
Lewin, Simons and Grundy
SEDA and SEAV 24 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 1: NSW Fluorescent Lanterns
Lantern inventory
According to Reference 2, the number of fluorescent street lighting lanterns under the control
of the NSW electricity distributors is:
Electricity Distributor Advance
Energy
Australian
Inland
Energy
energy
Australia
Great
Southern
Energy
Integral
Energy
North
Power
Total
Total Lanterns 30,612 3,089 343,850 38,566 173,379 60,493 649,989
Fluorescent portion of
total lanterns
43% 56% 69% 40% 57% 51% 61%
Fluorescent lanterns 13,163 1,730 237,257 15,426 98,826 30,851 397,253
The fluorescent lanterns are programmed for replacement now. If NSW adopts the same
solution to fluorescent lantern replacement which Victoria took, the capital and recurring
costs will be about:
Capital cost
Cost each Total
Fluorescent lanterns 397,253
Replacement lantern cost (e.g. B2224 MV 80 Watt) $65 $M 25.8
Installation cost (Labour, truck, etc). $100 $M 39.7
Total cost, supply and install $175 $M 65.5
Recurrent cost
Based on replacement of each fluorescent lantern with an 80 Watt mercury vapour fitting
(circuit power 100 Watts), the recurrent energy consumption and total cost using
recommended (Reference 2) IPART electricity pricing would be:
Distributor Advance
Energy
Australian
Inland
Energy
energy
Australia
Great
Southern
Energy
Integral
Energy
North
Power
Total
Fluorescent
lanterns
13,163 1,730 237,257 15,426 98,826 30,851 397,253
Installed demand,
based on 100W each
MW 1.32 0.17 23.73 1.54 9.88 3.09 39.7
Energy use,
based on 4000 hr/ yr
GWh/yr 5.27 0.69 94.90 6.17 39.53 12.34 158.9
Lighting Charge $/MW.h
14
$210 $210 $210 $210 $240 $240
Annual Charge $M/year $M1.11 $M0.15 $M19.93 $M1.30 $M9.49 $M2.96 $M34.9
14 The NSW street lighting cost recovery method converts all fixed and operating costs to an “equivalent” charge
per unit of energy consumed. While the project team does not agree with this methodology, it is used here for
consistency.
SEDA and SEAV 25 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 2: Magazine Article, New Street Lighting Standard
The following article is reproduced from the May 99 edition of “Lighting” magazine.
SEDA and SEAV 26 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
SEDA and SEAV 27 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
SEDA and SEAV 28 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
SEDA and SEAV 29 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 3 Lamps Comparison
Lamps for minor roads / residential streets
Group Type & Nomenclature Power
(Watts)
Brand
(example)
Output
(lumens)
Efficacy
(lm/Watt)
Osram 3400 42
1. Mercury MBF/U 80
Thorn 3850 48
a) Single End HCI 35 3400 97
b) HQI 70 5500 79
50/35 4500 /
2500
90 /
71
2. Metal Halide
c) City Light DS
80/50
Osram
6100 /
3400
76 /
68
a) NAV 50 50 3500 70
b) NAV 70 70 5600 80
c) NAV 50 Super 50 4000 80
3. High
Pressure
Sodium
NAV 70 Super 70
Osram
5000 71
a) DULUX T 26
(2 x 26)
1800
(3600)
69
b) DULUX T/E 42 3200 76
4) Compact
Fluorescent
Lamps
c) DULUX F 36
Osram
2800 78
a) Standard
26 mm diameter
18 Osram 1100 61
5) Tubular
Fluorescent
Lamps b) tri-phosphor “T5″
16mm diameter
14 Osram
“lumilux plus”
1220 87
SEDA and SEAV 30 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 3 Lamps Comparison (continued)
Lamps for major roads / traffic routes
Group Type & Ellipsoidal Lamp Tubular Lamp
Nomenclature
Power Output Efficacy Output Efficacy
(Watts) (lumens)(lm/Watt) (lumens)(lm/Watt)
1. Mercury Deluxe 250 14,000 56
400 24,000 60
2. Metal Halide (Coated) 250 19,000 76 20,000 80
400 32,000 80 42,000 105
3. High Pressure Sodium 250 25,000 100 30,000 120
400 47,000 117 54,000 135
SEDA and SEAV 31 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 4: Comparison of Two Minor Road Lanterns
The following charts show the calculated isolux plots for two street lighting lanterns using
the same lamp and the same spacing. The lanterns are:
• the B2224 “flower pot” which is the standard minor roads lantern in Victoria, and
• the “Urban Minor”
B2224
Urban Minor
The comparison shows that the Urban Minor provides more even illumination.
SEDA and SEAV 32 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 5 Semi-Cylindrical Illuminance:
In 1992 the CIE published a document entitled “Guide to the Lighting of Urban Areas”,
Publication No.92. This recommends that semi-cylindrical illuminance be the preferred light
technical parameter to adequately illuminate vertical surfaces, and in particular, the human
face.
Semicylindrical illuminance is also the preferred parameter in some Nordic countries where
lanterns with low mounting height or bollards are extensively used. These types of lanterns
tend to have a poor distribution of horizontal illuminance, but quite satisfactory vertical
illuminance.
Earlier studies undertaken in Russia and more recently in Australia, have shown that semicylindrical
illuminance is a much more acceptable indicator than horizontal or vertical
illuminance of the adequacy of lighting in non task areas such as public spaces, railway
stations and public buildings.
We believe that this matter of vertical illuminance versus semicylindrical illuminance as
indicators for the true modelling of the human body and in particular the human face, should
be taken up with Standards Australia. Perhaps it should be considered by a joint meeting
between the Standards Australia Technical Committees LG/1 and LG/2. Should you wish us
to be a contributor to such a meeting we would be happy to oblige.
A Personal Note from Kevin Poulton.
I have taken a great interest in alternative illuminance parameters since the early 1960s when
the late Dr A. Dresler introduced mean spherical illuminance to us as his students at the
RMIT. This was a concept which he developed as an alternative to horizontal illuminance
during his research days at the Berlin University in the 1930s. In the 1960s the Russian
researcher M. M. Espaneshnikov published extensively on the application of the parameter,
mean cylindrical illuminance and showed that it was a more adequate indicator for revealing
the true shape of the human form than either horizontal or vertical illuminance. Later in the
early 1980s Stockmar and Haeger published an article on a calculation method for half or
semi cylindrical illuminance and this has since been adopted by both the Danes and the
Dutch for sports lighting and the lighting of pedestrian areas.
During the preparation time of the 1990 edition of AS1680, Interior Lighting standard I
carried out a series of studies to discover what was an acceptable value of ambient light in
non task interior spaces. Using mean semi cylindrical illuminance as the main parameter I
found below 100 lux was too low and above 700 lux was too high. Unfortunately the
Standards Committee would not accept this study and went back to the conventional
parameter of horizontal illuminance.
Thus historically there has been much research and support for the concept of cylindrical
illuminance but for some unexplained reason the Western world has been reluctant to adopt
this concept.
SEDA and SEAV 33 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
The difference between semi cylindrical and vertical illuminance is a subtle one, semi
cylindrical illuminance has a significant “side lighting” effect which the vertical illuminance
has none. Semi cylindrical illuminance brings out the roundness, the three dimensionality of
the human form. It is easy to calculate and to measure.
Theoretically any visual environment has two components of illuminance, that is, there is a
direct component and an indirect or diffused component. The direct component come directly
from the lantern/s. The second component is the diffused component which is reflected from
adjacent surfaces or objects. In many interior spaces the diffused component is equal too, or
approaching the direct component, hence shadow patterns are very soft or even non existing.
However, in most exterior situations the diffused component is often negligible or
approaching this, hence as a consequence the dominant illuminance is very directional and
thus the strength of the light and shadows patterns is very noticeable.
Vertical illuminance values are very dependent upon the orientation of the subject, especially
the human head, with respect to the light source. This means that the vertical illuminance of
one face of a cube will be very high but on another face at ninety degree rotation could be
approaching zero. Yet these readings do not necessarily reflect the visual reality. In the case
of the human face, vertical illuminance does not take into account the roundness of this
object.
SEDA and SEAV 34 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 6: Glossary
Key lighting terms
The following key terms used in the report are arranged in a logical (rather than alphabetical
order)
lantern the term traditionally used for a street lighting luminaire, i.e. a complete
“lighting fitting” including the lamp, lamp holder, body, lens / diffuser, and
control components including ballast, power factor correction, and possibly
photo-switch.
illuminance a measure of the amount of light flowing from a light source incident on a
given area.
expressed in lux, where 1 lux = l lumen per m_ .
horizontal
illuminance
the illuminance measured by a meter in the horizontal plane, facing directly
up. This gives an indication of how much light is falling on a horizontal
surface.
vertical
illuminance
the illuminance measured by a meter in a vertical plane, facing in a specified
direction. This gives an indication of how much light is falling on a vertical
surface.
semicylindrical
illuminance
please see explanation in Appendix 4
luminance a measure of the amount of light reflected from an object, and so is
dependent on both the amount of light incident upon it (illuminance),
characteristics of the incident light (eg. colour) and the reflective properties
of the object (colour, gloss, surface texture).
brightness a term to describe the apparent or perceived amount of light on an object, and
so is dependent on:
• the illuminance incident on the object,
• the luminance of the object,
• glare, contrast, and other environmental factors,
• the position of the viewer,
• the viewer’s adaptation to the ambient light,
• the quality of the viewer’s vision.
Brightness is subjective as well as being influenced by measurable
parameters.
glare Unwanted light; light which interferes with vision or visual comfort.
LOR Light Output Ratio. The portion of the light produced by a lamp which
escapes from the lantern, expressed as a percentage.
SEDA and SEAV 35 Street Lighting Efficiency
file: 244 © Genesis Automation and Lightlab International June-99
Appendix 7: About the Authors
Geoff Andrews
Geoff is the founder and Engineering Manager of Genesis Automation, a company
specialising in reducing energy consumption while improving the of the energy services
delivered. He has specialised in the implementation of energy saving projects, including
overcoming the practical and institutional barriers which often impede the full achievement
of energy efficiency opportunities.
These services have been applied in a wide range of organisations including many local
governments, government, commercial and industrial customers.
Before Genesis Automation, Geoff worked as an in-house energy manager and then as an
energy efficiency consultant.
Kevin Poulton
Kevin is well known in the lighting industry, both as a professional lighting practitioner, a
teacher, researcher, and a leading light in the Illuminating Engineering Society of Australia
and New Zealand.
He has qualifications as both an architect and an electrical engineer.
Kevin is the founder of LightLab International, an Australian independent photometric
laboratory and consulting company providing services and equipment to many local and
international clients. This role includes designing and testing luminaires.
His street lighting design experience includes working for the SECV15 as a street lighting
designer.
15 The former State Electricity Commission of Victoria.
