Annual Conference at Glasgow
September 4th-8th, 1939

This conference was cancelled due to the start of the Second World War. However the conference programme and papers were printed.

The President's Address was printed in Public Lighting #15.

Note that the programme is incomplete and some pages are damaged or missing. (Adverts are missing for Foster And Pullen, Ltd. and Horstmann Gear Co., Ltd.)

Abstract: Descriptions of lanterns and equipment displayed by Siemens Electric Lamps And Supplies Ltd., The Electric Street Lighting Apparatus Co., The British Thomson-Houston Co. Ltd., The General Electric Co., Ltd, Edison Swan Electric Co. Ltd., Philips Lamps, Ltd., Holophane, Limited, The Brighton Lighting And Electrical Engineering Company, Ltd., REVO Electric Co., Ltd., William Sugg And Co., Ltd., W. Parkinson and Co., Foster And Pullen Ltd., Keith Blackman Ltd., Walter Slingsby and Co., Ltd., Automatic Telephone And Electric Company Limited, Bromford Tube Co., Ltd., Standard Telephones And Cables Ltd., Concrete Utilities Ltd., Whelmer Gas Mantle Co., Venner Time Switches Ltd., Radiovisor Parent Ltd., British Sangamo Co. Ltd., Peebles And Co., Ltd., British Foreign and Colonial Automatic Light Controlling Co. Ltd., Poles Ltd, British Electrical Development Association, C. H. Kempton and Co., Ltd., Engineering And Lighting Equipment Co., Ltd., Metropolitan Vickers Electrical Co., Ltd., British Commercial Gas Association, W. T. Henley's Telegraph Works Co., Ltd., Stanton Ironworks Company Ltd., Gas Meter Company Ltd., Metropolitan Pipe And Pole Company Ltd., Wardle Engineering Co. Ltd., Simplex Electric Co. Ltd. and William Edgar & Son Ltd..

Provisional plans (as outlined in Public Lighting #14 were for the following out-door exhibitions: Foster And Pullen Ltd. (West Campbell Street), William Sugg And Co., Ltd. (West George Street), C. H. Kempton and Co., Ltd. (Vincent Street), W. Parkinson and Co. (Vincent Street), Keith Blackman Ltd. (Shamrock Street), The Electric Street Lighting Apparatus Co. (Argyle Street East), Metropolitan Vickers Electrical Co., Ltd. (West George Street), The Brighton Lighting And Electrical Engineering Company, Ltd. (Bothwell Street East), Simplex Electric Co. Ltd. (Holm Street), The British Thomson-Houston Co. Ltd. (Cadogan Street), The General Electric Co., Ltd (Waterloo Street), Edison Swan Electric Co. Ltd. (Bothwell Street West), Siemens Electric Lamps And Supplies Ltd. (Argyle Street West), Holophane, Limited (Bath Street East), Engineering And Lighting Equipment Co., Ltd. (West Regent Street), REVO Electric Co., Ltd. (St. Vincent Street) and Wardle Engineering Co. Ltd. (Bath Street West)

Dual carriageway lighting displays were also planned for the Great Western Road and Edinburgh Road: The General Electric Co., Ltd (Great Western Road), Siemens Electric Lamps And Supplies Ltd. (Great Western Road), Engineering And Lighting Equipment Co., Ltd. (Great Western Road), REVO Electric Co., Ltd. (Edinburgh Road), Metropolitan Vickers Electrical Co., Ltd. (Edinburgh Road) and Simplex Electric Co. Ltd. (Edinburgh Road)

Adverts: Radiovisor Parent Ltd., Stewarts And Lloyds, Ltd., Siemens Electric Lamps And Supplies Ltd., Amal, Ltd., The Electric Street Lighting Apparatus Co., Peebles And Co., Ltd., The British Thomson-Houston Co. Ltd., Edison Swan Electric Co. Ltd., REVO Electric Co., Ltd., Holophane, Ltd., The Brighton Lighting And Electrical Engineering Company, Ltd., William Edgar & Son Ltd., William Sugg And Co., Ltd., Newport And South Wales Tube Company Limited, Keith Blackman Ltd., Walter Slingsby and Co., Ltd., Automatic Telephone And Electric Company Limited, Bromford Tube Co., Ltd., Falk, Stadelmann and Co. Ltd., Standard Telephones And Cables Ltd., Philips Lamps, Ltd., Concrete Utilities Ltd., Whelmer Gas Mantle Co., Metropolitan Pipe And Pole Company Ltd., Venner Time Switches Ltd., British Commercial Gas Association, British Sangamo Co. Ltd., Stanton Ironworks Company Ltd., Automatic Light Controlling Co. Ltd., Poles Ltd, Gas Meter Company Ltd., W. Parkinson and Co., Measurement Limited, C. H. Kempton and Co., Ltd., Metropolitan Vickers Electrical Co., Ltd., Metropolitan Vickers Electrical Co., Ltd., W. T. Henley's Telegraph Works Co., Ltd., Wardle Engineering Co. Ltd., Simplex Electric Co. Ltd., Engineering And Lighting Equipment Co., Ltd., British Electrical Development Association and The General Electric Co., Ltd.

Provisional papers (as planned in Public Lighting #13):

Tuesday, September 5th, 1939

Presidential Address by Mr. E. J. Stewart, M.A., B.Sc. (Inspector Of Lighting Of Glasgow)

Lighting In Glasgow by Mr. J. M. Ward

Wednesday, September 6th, 1939

Modern Gas Street Lighting by Mr. A. V. Horsfall

Illumination Of Roundabouts, Bollards etc by Mr. K. Sawyer

Lighting Of Bends And Junctions by Mr. F. F. Middleton

Thursday, September 7th, 1939

Lecture Demonstration by Professor S. Parker Smith

Friday, September 8th, 1939

New Installations In Dublin by Mr. F. X. Algar

Some Aspects Of Modern Development In Electric Lamps For Street Lighting by Mr. L. J. Davies (B.T.H Co. Ltd.)

Tuesday, September 5th, 1939

Presidential Address by Mr. E. J. Stewart, M.A., B.Sc. (Inspector Of Lighting Of Glasgow)

Lighting In Glasgow by Mr. J. M. Ward

Wednesday, September 6th, 1939

Modern Gas Street Lighting by Mr. A. V. Horsfall

Engineering Principles In Lantern Design by Mr. J. G Christopher and Mr. J. S. Smyth B.Sc.(Eng.)

Thursday, September 7th, 1939

Lecture Demonstration by Professor S. Parker Smith D.Sc., M.I.E.E., A.M.Inst.C.E.

Friday, September 8th, 1939

Illumination Of Roundabouts, Bollards etc by Mr. K. Sawyer

Lighting Of Bends And Junctions by Mr. F. F. Middleton

Some Aspects Of Modern Development In Electric Lamps For Street Lighting by Mr. L. J. Davies (B.T.H Co. Ltd.)

Professor S. Parker Smith D.Sc., M.I.E.E., A.M.Inst.C.E.

Keywords: Lighting: Physiology, Lighting: Theory

Monday, September 4th, 1939

Abstract: The Professor's paper is just a series of bullet points. These concern the general nature of seeing and ideas on improvements.

Limitation Of Eye

Sudden changes in illumination

Intensity of lighting

Lack of contrasts

Glare or dazzle

Causes Of Confusion

Arising from passing from well-lit to badly-lit streets

Glaring lights from buildings or vehicles

Dark or black surfaces, wet roads.

Invisibility of pedestrians, cyclists and dogs.

Possible Remedies

Ideal lighting

Prohibition of intense lights on buildings or vehicles

Reduction in speed limits in controlled areas after sunset

Raising of motor and dog licences with a view to reducing unnecessary street congestion.

Suggested Experiment In Selected Town

A representative committee should be empowered to carry out all relevant tests.

Mr. E. J. Stewart, M.A., B.Sc.
Inspector Of Lighting Of Glasgow

Keywords: Lighting: Colour, Lighting: Funding, Lighting: Levels, Lighting: Users.

Tuesday, September 5th, 1939

Whole paper (with the exception of the introduction and acknowledgements) published in: Public Lighting, Vol. 4, No. 15. September 1939

Abstract: The paper examines the varying views of the users of street lighting, especially from the point of views of pedestrians and residents. It ranges from how users view different installations and technologies, accidents, and the range and form of complaints.

Street Users As Judges Of Lighting
Persons see differently the same street, by day or by night, because they look for different things.

Noticeability Of Lighting Differences
The practical question about noticeability becomes: "What is that difference between one lighting installation and another which is obvious to the majority of those whose opinion counts in deciding how a street is to be lighted?" The answer depends, not only on the technical characteristics of the installations, together with the characteristics of the streets, but on the different interests of users of any street, their different previous acquaintance with street lighting etc. The answer depends too, on the different prepossessions, the different mental and emotional contents and reactions of various observers, leaving out the account the many users of streets who never notice the lighting at all.

Obvious Changes Of Appearance
With our close spacing of 75', lamps mounted at 10½' may give complete visibility for at least 100 yards where carriageways, footways and buildings reflect well. Yet the call continues for high mounted lamps. Their power and distriubtion is superior; I believe the demand is not entirely based upon these gains, but upon the different in pole and height which are so much more obvious. Some of the success of discharge lamps is probably on account not only of the actual increase of light but of the notable change of appearance.

Obvious Changes Of Source
Among our 25' or 26' installations are streets which in relation to their form, their surface and their traffic, are well lighted by 200W filament lamps at our 120' spacing. In some others 300W are necessary. In some of these a change from 200W to 300W would satisfy; in others the jump would need to be from 200W to 500W both to give the desirable general brightness and to make obvious that there had been an improvement. One street, with wide carriageway lit passably by 200W lamps near the kerbs, was converted into two separate carriageways. It was not convenient to add rows of lamps towards the centre-line. And increase at the existing points to 500W lamps plus directive apparartus was required to produce distinct improvements by greater brightness of the source and by greater brightness over the road surface. On the other hand even if a more expensive fitting produces a theoretically more appropriate set of distribution curves than another, is it worth while if the resultant difference from a cheaper type is generally not noticed by users of the street?

Noticeable Difference In Height
What difference in mounting height is noticeable by the public? I am doubtful whether a change in height alone from 9 or 10½ to come within the Specification would be noticed. I know examples where, combined with a doubling of light output, it was unnoticed. Generally any changing of height is accompanied by change of candle-power in some direction, even if the spacing remains unchanged.

Spacing
As regards spacing, liability to criticism is closely linked with the lighting output; but while any distance up to 240' or even 300' may pass muster, according to the power and the road, yet when low-mounted units are as much away from our ususal 75' spacing as 120', this is just about like to be noticed, at 150' is likely, and at 180' is very likely - and likely to ensure the request for "big" lamps.

Age
The reduction in light output by age becoems noticeable for filament lamps at about 1000 hours and is the order of 10%. This is noticeable if the lamp is alongside a new lamp. A drop of 15% at 1500 hours may be noticeable along. At what number of hours' burning, corresponding with drop in efficiency, should we remove the various sources? Perhaps when the drop becomes obvious to the user of the street, or the lighting department inspector?

Dirt
A reduction in light output, due to dirt, which is likely to be noticeable to a trained observer, is about 15%, produced in a city atmosphere after about two weeks. Probably distinct noticeability begins at a reduction of about 20%. In Glasgow, all low-mounted gas lanterns and open electric reflectors should be kept reasonably clean. There is no difficulty in noticing that any of these is dirty or becoming dirty. The spread of high-mounted electrical units ahs compelled the adoption of periodical cleaning with regular shifts of men doing this all the time, takign the place of the general purpose lamplighter.

Variety Of Judgment
We generalise about lighting on the basis of the average eye. The same person sees the same thing differently at different times. All those driven or walking past of the end of a series of side streets lit in different ways preferred one kind of source and installation; when they walked into and through these streets the opinion was generally reversed. And the same lighting is seen, used, and appraised by different eyes in very different ways: the adjoining resident, the pedestrian strolling along the footpath, the driver looking for a house number, the driver rushing through, the child running across. More of the population are regarding street lighting from the point of view of drivers; but some drivers are sensitive to the lighting and want it high class in every stretch; others don't worry except at junctions; others are concerned with blind junctions and sharp bends and the entrance to their own garage. Criticisms of the lighting of a whole town may be an enlargement of an aspect of a few streets actually observed.

Likes And Dislikes
The variety in the minds with which observers look at a lighted street makes it often difficult to draw any conclusion about an installation from opinions collected in a questionnaire. Among the factors in the installation affecting liking a colour of light, brightness of source and shape of lantern. The reaction to coloured lighting in many towns is probably different from what itwas five or even two years ago, because more people are accustomed to discharge lighting.

Claimants
The claimant for damages for accident alleged to be due to insufficient lighting have to be seriously considered by the lighting department.

Seeing And Looking
By good lighting we enable people to see; but we cannot ensure that they will even look. To observe requires not only retina but brain; and the brain may be otherwise engaged. Accident may be avoidable by improvement of the lighting; but there are these other factors. So efforts to improve seeing are helped, by road-alterations and, for making the right movement, by various teachers of Safety First to children and grown-ups.

Good Lighting And Other Factors
We thoroughly believe in the value of good street lighting to aid safety and comfort; but we must reject assertions that every night accident is due to lack of light, especially in this city where we are liable to claims for tripping over things, for being run over, for running into things.

Lighting And Accidents
Lighting may be erected to reduce accident but fail according to statistics, from growing traffic and other causes. Yet that lighting may have clearly made the street more comfortable to the user, may have given less chance of accident and certainly less fear of accident - and less opportunity for blaming accident upon the lighting. Analysis of our Chief Constable's report on fatal street accidents reveals that most of these occur in main thoroughfares; but suggestions from police or others than incomplete safety there is associated with poor lighting become few, as to be expected since most of these have been given Group A lighting.

Pedestrians And Residents
A much larger proportion of our complaints and requests in Glasgow is concerned with residents and pedestrians as such in the immediate neighbourhood of their homes than one might expect from a perusal of the usual discussions before technical bodies on the chief factors and requirements in good street lighting.

Complaints

• Drivers: 12 complaints e.g. Difficulty in seeing way, junction, kerb, etc. (4); Not seeing pedestrians (1) ...
• Traffic Constables: 2 complaints e.g. Difficulty in seeing traffic (1); In being seen (1)
• Residents: 38 complaints e.g. Fear of loiterers and attack on women and girls (6); Experience of burglary (3) ...

Cyclists And Other Users
On the carriageway the pedal cyclist is an even better test than a driver or pedestrian than the standard black cat. In Glasgow, the cyclist is often a head-down scorcher. A boy recently cycled in daytime right into the rear of our stationary tower-wagon and apperaed to have his head badly smashed. He turned up at the office, however, the same afternoon, to try to claim a new suit of clothes. The "right of free movement" of the pedestrian has been publically threatened in the name of public safety, and the forecast has been made that he will ahve to conform to certain regulations. A proposal has indeed been published to make it a statutory offence to walk on the carriageway where there is a footpath.

Lanes
Burlaries are sometiems cited as reason for improvement in lighting. We recognise our value to the plice for prevention of several crimes. Sometimes, entry has been obtained at sides of a building where public lighting cannot go. Where there are lanes, we may give a limited assistance to the defence on that front. Lanes, nevertheless, have to take second place after the needs of streets proper, and they are often more troublesome to light and keep lit. Requests for more light in them come from householders who may never have been in their own lane; but who want to feel sure that nobody else is there. Also more of the old lanes are being used by drivers as access to garages, and that raises requests to shift existing lamps, requests for more light, difficulty of placing lamp-posts without danger of damage. Occasionally appeals for lighting, or more lighting, in paths away from streets and in out-of-the-way roads with dark backgrounds are based upon fear of assault on women and children. The gain is perhaps less in the reduction of assults - which may never yet have actually happened - than in removal of the fear.

Glare
We very rarely hear of complaint about glare.

The Unsatisfied
Complaints do come in whcih are unexpected. Mostly they refer to a small part of the street or a court or a stair which is of concern to only a few people; but we try to satisfy them. On the other hand, there are incomplete installations and surfaces and ends and bends whcih disturb the Inspector and some of his staff, and apparently nobody else. The lighting of Glasgow is not yet my ideal, that it is not what I should like it to be, given time and other facilities. It may not be the best that, at the present momment, we can erect or afford.

Stair Complaints
In Glasgow, we have as many complaints from stairs as from streets: that the light is not "bright" enough, or there ought to be extra lights for the dark corners. Economy and the need for uniform practice restricts this.

Visit The Place
Streets and spots must be visited, both for the lay-out of a new installation and for dealing with a complaint about existing lighting. For the latter, the place must be examined at night, because the factors may include some effect of reflection or non-reflection which does not appear on a plan.

Satisfactory Lighting Brings Discontent
We rarely hear about well lighted streets. Satisfaction is expressed by silence. Lighting needs to be very bad before more than a few percent off all the street users voice or write a complaint. When big lamps have apparently satisfied generally, complainers are ususally interested in one particular spot. For instance, we changed the lighting from one-mantle gas-lamps to Group A, and received strong protest because we had removed a lamp from beside some lock-up garages, so that users could not find the keyholes at night. (However the lockups were soon removed for further building.) After improvement there many be unsolicited comment if the street itself is very prminent one or if a new type of lighting has been installed. Good light is not usually noticed, or it is noticed, not because it is satisfactory, but because it is better than or merely different fron previous or neighbouring installations.

Disposal Of Lighting Funds
Even if we accept that the best way is to light for the most dangerous point in the worst weather for the most negligent user, the problem remains how best to spread the funds over the deserving streets, so as to get reasonably safe lighting. I would rather light one mile so as to give complete visibility 300' ahead without the use of headlights than do only half a mile giving similar visibility for 600' or four miles giving only 75' complete visibility and practically inducing the use of headlights.

The Particular Place
People are still trying to find a satisfactory expression for visibility as an appraisement of street lighting. Is is worth defining or measuring apart from the particular object in question under the conditions in question? We might give a quantitative expression of visibility, especially for the driver; but vision at a distance, especially for the pedestrian, may not come into question in the particular circumstances. The practically important seeing may be of a slight irregularity of surface at the pedestrian's feet. I wish simply that we could have at least the Class E of oru old and unhonoured Specification 307, which is also the 0.1 f.c. minimum of the Fourth Report of the Departmental Committee on Lighting In Factories aaplicable to roadways and other open places in a factory. With the adequate mounting height of the Departmental Committee on Street Lighting and probably the new Specification, and some further provision against glare, the pedestrian ought ot see his way, not should there be excuse for driver to blind him in seeing their way and him.

Mr. J. M. Ward
Assistant Inspector Of Lighting Glasgow

Keywords: Lighting: Columns, Lighting: Control, Lighting: Equipment, Lighting: Legal, Lighting: History, Lighting: Luminaires, Lighting: Maintenance and Lighting: Management.

Tuesday, September 5th, 1939

Abstract: Description of the Glasgow Lighting Department.

The department moved to 20 Trongate in September 1913. Over the years, the building was found to be too small so the Central Electrical Section is at 25 Blackfriars Street (originally a garage for the tower wagons) whilst a rented store is at 1 St. Andrews Street. The Head Office (at 20 Trongate) has 20 staff and includes the Card Index System. Here records of 11,000 street lamps of wattages over 100W, 1,800 traffic signal lamps, and 2,000 traffic lamps (Keep Left and bollards) are kept. One clerk operates the system, issues renewal sheets to the five electric control stations, enters all lamp renewals when made, and enters the day-to-day casualty lamps (those that failed before 1,000 hours).

The Glasgow Lighting Department operates under the Police Act 1866 and subsequent Acts; this lays on the Authority the statutory duty that it shall light the streets, courts and stairs.

The telephone switch board has 15 exchange lines. There are private extensions to 13 depots, and one (Blackfriars) has two lines. One operator is on duty during the day. The telephone system will be changed to automatic in 1940.

The Lampfitters and Pole Erectors Section has 70 employees: one clerk, two labourers with the others engaged on outside work.

Surveyors
There are 11 Surveyors and 10 Sub-Surveyors. They work a 42-hour week in 6 days. Each evening five hours are spent on surveying an aread of roughly 812 acres, densely populated with 3067 street lamps and 7582 stair lights. (Stair-lighting is a peculiarly Scottish undertaking, the tenemental system of dwellings responsible, and in Glasgow there are 94,000 lights under the department's care. 67,000 are gas lights. The type of illuminant is a matter for the proprietor: if they decided to install an electric light, an average 4-flat close can be wired to department's specification for £3, the work being carried out by private contract, but lighting points are fixed by the department's Divisional Superintendent. On completion of the wiring, the department fits switches and brackets, and maintains the installation as public lighting.

All lighters, street and stair, use acetylene torches, previously using oil hand lamps.

There is a joiners workshop with a foreman and 7 joiners. They repair the tower structures of the wagons, make teak boxes for section boxes, upkeep the ladders and other tasks.

The Store occupies the second floor on a level with the Office. Part of the stores was transferred to Blackfriars Street.

The third floor includes the tinsmiths' workshop with a staff of 40 men. All new street gas lanterns, stair and traffic lanterns, are made in the workshop. 8 tinsmiths are engaged only in repair work of lanterns. Another 8 work on the repair of pipes and brackets for stair lamps in the closes and stairs. 6 men handle the burner cleaning and reconditioning, working on 600 burners per week, of the 85,000 burners in use.

There is a Clocks Superintendent who looks after the timekeeping of 96 public clocks. Not all are owened by the Corporation, many being "adopted" for winding and maintenance.

In the preparation of estimates for new work, data is prepared by staff in the Drawing Office from data supplied by the engineers in the Master Of Works Office and Housing Department. Having decided the programme of work, tracings are made from the 5" to the mile map in readiness for the installations. No amount of theorising will ever taken the place of positioning lighting points on the site. Glasgow practice is:

• Preliminary survey undertaken when proposed points listed.
• This is passed for the foreman of the lamp erectors whose knowledge of what lies beneath the footpaths often enables him to suggest necessary alternatives.
• The "next best" positions are fixed and the points plotted on the tracings.
• Feed points are arranged and wiring circuits drawn out.

Year by year evidence is forthcoming that every pound allocated to the testing room is money well spent. The department is far from satisfied with the equipment in this section, but every year sees some addition. Equipment includes a photometer bench with a Leeson disc, two cube integrators (1 metre and 10 ft) and benches for lamp life and relay tests.

Blackfriars Street
This houses the Central Electrical Section and has 136 employees. The section is responsible for the erection of all aerial cables, section boxes, bases, lanterns, switchgear, relays and their subsequent maintenance and repair. The section is manned 24 hours, every day. All "darks" and urgent repairs are phoned in by the divisional superintendents who recevied them from street lighters returning after lighting up. Additional reports are phoned in by the surveying staff who set out the electric control scale time, and spend five hours on survey, and the police are good at passing on any information regarding defective lamps. All reports are logged with the Night Foreman who writes an account of when and how dealt with.

Scrutiny of the "faults" list revealed the necessity of localising damage from "blown" lamps. All lanterns are now fitted with fuses mounted on bridge plates - if there is now space in the fitting when the fuse is mounted on a pole fuse box at bracket level. This work has provided exceptionally beneficial.

10 fitters are employed in this section, reconditioning lanterns, brackts, fitting section while outside squads are engaged in overhauling brackets in continuation of our policy of combating water condension troubles in the lanterns.

A Divisional Station
There are 13 divisional stations in Glasgow. This number was necessary when the lighting was mainly by gas and hand lighting and extinguishing was the order of the day. The present policy, when 20,000 of the 26,000 electric lamps are controlled from district statiosn, is to have 7 electric control stations and 4 are now completed.

A description of Langside is given. Architecturally it is of modern design in silver-grey brick. There is a control room, housing the main switch board, and the street and stair lighting control board; also an office for the Superintendent and his assistant, and a muster hall for the 20 street-lighters and 52 stair-lighters still necessary. At the rear there is the Elecrical Foreman's office, store and spacious workshop while a tower wagon garage and yard completes the station.

The territory allocated to the division covers 2140 acres.

The lamplighter's beat is rapidly changing and very few employees now have the 120 gas lamps which was the former standard.

Historical
Glasgow's first street lamps were erected in 1718 and tallow wicks provided the illumination. These would be feeble and hand lamps or link boys were still used. In 1765, oil lamps were introduced, which were better. The Town Council offered to provide street lamps in 1780 as an inducement to proprietors to lay foot pavements opposite their houses.

In 1818, 1472 street lamps were lit by gas, the Glasgow Gas Light Company, having received the necessary authority to manufacturer gas the previous year.

In 1914, the department was responsible for 27,631 street lamps; 1,667 being electric. There were still 542 flat frame burners lighting the courts; and of the 272,000 mantles used there were 210,000 upright mantles.

In 1939, the department is responsible for 35,631 street lamps; 14,713 are gas and 20,918 are electric. The smallest burner is the swan neck single mantle, and the largest gas unit is the 10-light lantern. Conversion of the remaining single burner lanterns to 3-lights or 4-lights is continuing at 1000 per annum. The change requires lantern alterations and the rate is fixed by the maximum output of the tinsmiths' workshop.

The closeness of spacing of the gas lamps is unequalled in any other large town. 25 yards between posts is quite ususal in many streets. The lanterns are mounted on cast iron pillars 8'9" high; crinoline and lantern give a source light of 10'. All streets lit by multiple burners would come under the "G" class of the existing Street Lighting Specification. Two sizes of lantern house the multiple burners with a 3-light in a No. 00 lantern; and 4-light in a No. 0. 24 w.g. bright copper is used throughout, with the astragals strengthened with 5/16" brass road. The top is hinged giving free access for burner adjustment and maintenance. Porcelain reflectors are still used exclusively. The bottom is of cast brass and fits loosely with one half forming a trap door. The lamps are fitted with pressure governors set to 5" inlet pressure and 26/10" outlet pressure.

The only gas installations apart from standard 3- and 4-light are those in Kelvin Way (10-light lantern mounted on special steel poles giving a mounting height of 22', giving as free a gas flow as possible) and Greendyke Street (4-light lantern installation, harp type, moutned on 12' cast iron pillar, spaced 120').

Glasgow's policy for many years is conversion to electric street lighting and the erection of high mounted electric lamps in important streets. 20,852 lamps are now electrically lit. In the 1920s, for new lighting schemes, a cast-iron pillar 8'9" high with 20" root is used. The swan neck bracket had spigot, fuse box and tube with scroll work. The lantern was locally named the "St. Mungo" and had a copper canopy with vitreous enamel reflector, suitable for 60-100W GLS pearl lamp. There are many thousands of this type of unit doing service. These were included in Riddrie (1920-21), Mosspark (1923-24) and Knightswood (1926-27). The Department laid its own lighting cables, and at fixed points, section boxes were connected to the supplies from the Clyde Valley Electrical Power Company. Hand-switching was provided with lamps lit and extingushed from a set number of points. (Two points for the 120 lamps of Mosspark).

When it was decided that these units did not give the necessary mounting heights, the Department used a steel pole, 28' long, to be used in housing schemes. The light gave the necessary clearance height for overhead cables while a range of mounting heights was obtaintable. The Department has 2,482 28' poles which give a general mounting heigh of 20'. These were fitted with "Glasgow" lanterns taking 150-200W lamps. During the past two years, a simple fitting has been adopted, eliminating glassware.

The present standard practice in improvements and new lighting schemes is the 37' 6" steel pole, lantern mounted at 25', with 300W GLS lamp at 120 spacing - this meets Group A specification. On tram routes, the Transport Department poles have been used, with extension pieces and here the height is nearer 27'6". Another mounting on tram poles is the side bracket with a mounting height of 22'.

Remote Control
Glasgow has worked on a system employing contactors or relays. It was natural with overhead cabling such a scheme should be developed, being capable of operation within our own network, in contrast to the superimposed wave, ripple or bias systems, which entail the placing of control gear in the supply undertaking. 15 years ago a start was made on contactors operating on the cascade principle. During lighting hours the coils (in the relays) must remain energised. This arrangement operated with fewer failures than any other type tried. There are certain disadvantages: The coils are in operation during the full burning period and must be renewed after a definite period; if employed in large numbers there would be a charge for coil consumption and maintenance, cleaning of contacts, and attention to springs; if one relay or circuit fails, all lamps forward in the ring would be left unlit.

The second phase was the switch wire contactor. In this system, the coils were energised during daylight, and de-energised during lighting time. In this system, the advantage was that if any fault developed, the lamps stayed lit, attracting attention and allowign the necessary repairs before lighting up time. It has the disadvantages of coil operation during long hours, coil consumption and need of maintenance.

The third and present procedure is the use of impulse relays. The coil is only in action momentarily at lighting and extinguishing. The energising of the coil pulls up the solenoid, on an extension of which there is a pawl operating a ratchet wheel, which in turn rotates an arm on which the switch contacts are fixed, making and breaking in mercury. Maintenance of this relay is low and consists of a periodic examination of the mercury level in the containers. The coil operation is a matter of seconds and the pilot wires are dead except when the impulse is passing. A defective relay caused the same trouble as the cascade system and now each relay is teed off. The relays are operated in rings running out from the Division Office control board.

Plant And Equipment
This was described in the 1934 Paper. Only alterations and improvements are detailed here.

Tower Wagons
There are 14 tower wagons. On the grounds of stability, for many years we refrained from fitting the towers with revolving platforms, but the last two models include this feature, which is useful for working on span wires (663 in Glasgow).

Pole Carrier
They formally used horse-drawn vehicles as two petrol-electric mobile cranes are engaged on erecting work. The carrier is constructed to take a maximum load of ten 37'6" steel poles.

Cleaning
We had had high mounted units in greater numbers than almost any other authority. Expenditure on cleaning is a very considerable item in our annual budget. There are 14,000 high mounted lamps and the lamp cleaning takes place every 14 days. Our method of cleaning all lamps above 18' is by tower ladders, mounted either on barrows or motor chassis. In the case of the hand tower, two men work together and maintain an average of eight lamps per hour. We insist for an eight-hour day cleaning, 64 lamps are cleaned. On the agreed working of 293 days per year, then 18752 lamps are cleaned per annum per barrow. The cost per clean is 4½d. There are disadvantages in the barrow system of cleaning: (1) Stability in wind is poor; (2) Must be operated in a comparatively flat area; and (3) Lamps must be closely grouped.

The other method is by the use of motor tower wagons. The danger element, except in gales, is eliminated, greater safety is secured in busy traffic, and the cost per clean is 4d. 14 lamps per hour can be cleaned quite comfortably. The difficulty is obtaining funding for the tower wagon fleet. It is hoped to eliminate the hand tower altogether. The tower wagon gives a platform height of 21'6" and it should be possible to clean 11 lamps per hour, giving a cleaning cost of 3d. per lamp.

Clock Lighting
In Glasgow there are 51 turret clocks, and 35 are illuminated. Most have internal lighting but in a few cases, with solid dials, external lighting must be employed. The Tron Steeple has been dealt with by flood lighting, the projectors being fixed to the poles nearby.

Traffic Lamps
There has been very close co-operation between the City Engineer and the Chief Constable in the planning and placing of refuge islands and traffic guides at junctions where accident figures have proved excessive.

The City Engineer's Department has a set of wooden bollards and when a junction is black-listed and its considered that an island would tend to remedy matters, tests are carried out with the wooden bollards. After a test period, and agreement has been reached, a plan is prepared, submitted to the Council, and on approval, is forwarded to the Lighting Department so Pillars of Fire and Keep Left lanterns can be erected and cables laid. A two-phase supply is always used so that the load is divided. There are: (1) Bollard with no Keep Left lantern: 100W GLS lamp; (2) Pillar of Fire (100W GLS) with Keep Left lantern (60W GLS) and; (3) Bollard with large Keep Left lantern: 100W GLS and two 60W lamps.

There are now 145 junctions with refuge islands. The large Keep Left lanterns are made in our own workshop.

Expenditure
The total expenditure on the combined services - Street and Stair Lighting for 1938/39 was £386,000. Wages and illuminants account for a very large part of our annual outgoings.

Decoration
Normally all lighting plant is painted a dark green over the initial two coats of red lead. George Square was considered worthy of something different, and gold was used - but after a few weeks the metallic element had tarnished and the poles had turned a greenish hue. It was assumed the problem was atmospheric and the poles were repainted and given a coat of clear protective lacquer - the same problem occurred. Finally, expert opinion proved that the cause of discolouration was in the medium - to provide a medium of the purity necessary would cost too much. One pole was painted for test with this and still shows no evidence of tarnishing.

One side of the Green was set aside for experimental work and poles have been treated in gold, bronze, aluminium and cream. Several routes to the Exhibition were painted in aluminium with a blue base 6' high. One of the most effective was cream paint which stood up extremely well being effective both by day and artificial light.

An Act To Regulate The Police And Statute Labour Of The City Of Glasgow And For Other Purposes, 23rd July, 1866
SECTION XLVIII: The Trustees acting under the "Clyde Navigation Consolidation Act, 1858, shall light and cleanse their quays, wharves, sheds and streets, and in respect of the cost the Trustees shall be entitled to a deduction from the annual assessments to be levied on them under the Act of a sum equal to 4d. per pound.
SECTION LXVI: In addition to the Officers appointed under "The Commissioners Clauses Act, 1847" the Board shall, appoint the appointment of the following special officers, namely ... and an Inspector Of Lighting.
SECTION LXXVI: The Master of Works and each Inspector appointed by the Board shall in his Department appoint several officers and persons employed therein.
SECTION LXXVII: Every Inspector shall have an absolute power of suspension or dismissing any person employed under him.
SECTION CXLIX: Every person who is guilty of the following shall be liable to a penalty not exceeding £5, or for imprisonment for a period not exceeding Thirty Days: "Every person who wantonly injures any lamp or lamp post." A penalty of 40 Shillings or imprisonment for 14 Days: "Every person who without authority extinguishes the light of any lamp."
SECTION CCCLVII: The Board shall make provision for lighting in a suitable manner the portions of the turnpike roads within the City. This will include the necessary lamps, lamp posts, lamp irons, and other appurtenances. They shall also light the dial plates of clocks. (This does not include the quays, wharves and streets belonging to the Trustees of the Clyde Navigation).
SECTION CCCLIX (359): This dealt with the provision of lamps in private streets and courts, but was repealed by Section VII of the Glasgow Corporation Confirmation Act, 1914.
SECTION CCCLXI (361): The proprietors of every landing having access by a common stair shall provide and maintain suitable gas pipes and brackets, lamps and burners to the satisfaction of the Inspector of Lighting or the Board; or under a penalty of 40 shillings, the Board shall cause htem to be supplied with gas and lighted durign the same hours as the public street lamps, and then pay the Board 10 shillings per annum. This may be recoverable by the proprietor from the occupier in their rents.

The Glasgow Corporation Order Confirmation Act, 1914
SECTION VII: The following provisions of the Act of 1866 are repealed: (A) Section 359 (Proprietors to pay expense of lighting private streets and courts); (B) Section 361 (Lighting of common stairs); and (C) Section 362 (Provision payable by proprietors)
SECTION VIII: The Caledonian Railway Compnay as proprietors of the Forth and Clyde Navigation may light the harbours, quays, wharves and streets owned by them, so can be lighted in a suitable manner to the satisfaction of the inspector of lighting. In respect of the cost incurred, the Company shall be entitled to a deduction from the annual assessment to be levied on them.
SECTION IX: (1) The proprietor of every land or heritage having access by a common stair shall provide, fit up, maintain and renew, pipes and wires, or pipes, or wires, in such common stair suitable for lighting by gas, electricity or other illuminant to the satisfaction of the Inspector of Lighting.
SECTION IX: (2) So soon as such pipes and wires are porvided, and so long as they are maintained and renewed, the Corporation shall, in such common stair, provide, fit up, maintain and renew brackets, lamps and fittings, and other means of lighting and all necessary means of extinguishing the light.
SECTION IX: (3) The Corporation shall light such common stair in a suitable manner with gas, electricity or other illuminant. These will be lighted during the same hours as the public streets.
SECTION IX: (4) Where there is more than one proprietor of a land, they shall pay the cost in proportion to the valuation on their respective lands.
SECTION IX: (5) If any proprietor fails to provide, fit up, maintain and renew such pipes and/or wires then the inpsector of lighting may server a requisition calling on the proprietor to do the work.
SECTION IX: (6) If the proprietor fails to carry out the work required by the requisition then the inspector of lighting may cause the work to be done and the Corporation recover the cost thereof as a debt from the proprietor.

Mr. A. V. Horsfall
Assistant Public Lighting Superintendent - The Gas Light And Coke Company

Keywords: Lighting: Control, Lighting: Luminaires, Lighting: Specifications, Lighting: Theory

Wednesday, September 6th, 1939

The final aim of street lighting is to produce a degree of visibility which enables all road users to see clearly, distinctly, and with comfort, as far ahead as is necessary for safety.

Despite the warning in the M.O.T. Final Report on Street Lighting that the installation with the largest total light output in lumens is not necessarily the most effective, the idea still persists that to obtain the maximum revealing power it is essential to have the greatest numbers of lumens, regardless of colour.

There is a difference according to the colour of the light. This can be demonstrated: take two white surfaces and light each surface to a high intensity with a light of different colour so they are equally matched in brightness. If the light intensity is reduced by the same proportion then the brightness no longer matches, and it appears that one surface is better lit than the other. This fact throws doubt on the general assumption that lumens of white and coloured light are of equivalent value when applied to low levels of brightness over extended fields of view. - [I believe this statement is true and refers to a shift from photopic to mesopic/scotopic - where white light is more adventageous - Simon]

Many modern gas installations give direct evidence that white light gives excellent revealing power at the lower limits of lumen output, provided proper distribution is employed and lamps correctly sited. Similar results can be obtained with white light from tungsten filament electric lamps. So the colour of light employed determines the lumens required to provide adequate lighting at an economic rate - if maximum revealing power is obtainable with the smaller number of lumens of white light, then using a greater number of lumens is just a waste.

Because revealing power is difficult to assess, it is not possible to show this technically. But it is believed the revealing power between a high lumen output colour lighted installation and a low lumen output white installation will roughly be the same. It would seem that white light provides a sharp contrast between objects seen against the brightness of the road surface background. The Electrical Industry is not unawre of the importance of white light as shown by work being undertaking into colour corrected lamps - and tungsten lamps have also been used with uncorrected colour lamps to obtain white light. Given proper distribution and siting, it is not necessary to have large lumen outputs when white light is used: the present M.O.T. reccommendatiosn could be modified to quality further the maximum and minimum lumens required according to the colour of the light used.

Many Councils find they cannot afford even the the existing minimum recommended by the M.O.T., therefore a reduction in this minimum consistent with adequate lighting might enable many areas to improve their lighting.

Modern gas lighting can provide any range of lumen requirements. Installations giving 15,000 lumens per 100 ft. linear of road, which is almost double the M.O.T. lumen figure, and many other installations of lumen output within the 3,000 to 8,000 scale, all using ordinary low pressure gas. Also the gas industry are the only industry who will guarantee in a lighting under penalty of a fine amounting to as much as 5/- per lamp per night, that the illumination will not at any time depreciate by more than 25% (as against 50% with other illuminants).

Gas Lantern Design and Equipment

1. The "Triumph" Lantern
The lantern consists of a rectangular copper housing in which two sets of mirror glass reflectors are arranged with their main (centre) mirrors parallel to the major horizontal axis of the burner, the mantle nozzles of which are arranged in staggered line formation, and are usually 12 in number. The two reflecting systems consist of a set of 3 curved glass mirrors fixed on a sheet metal back plate which is attached my means of hinges to the lantern casing. A simple positive adjuster permits rapid vertical focusssing of the controlled light beams between the angles of 70° and 80° and adjustment can be made to give up to 25° horizontal bias from each set of mirrors, whilst the lantern is designed to give a complete cut-off at 90° with the vertical. The air and gas adjusters are of conventional design, and are accommodated in the roof of the lamps where they are readily accessible. The construction of the lamp is such that complete access to the light chamber for maintenance purposes is obtained by opening the two hinged glass frames which form the lower lantern casing, and there are no interior fittings to hinder lantern cleaning.

2. The "Maxill" Type "G" Lantern
The lantern is of square design, with pyramid top, and is constructed of non-corrosive materials and partly glazed with armour plate glass. There are two separate burners, each with its own gas and air regulator, the mantles being in the form of clusters approximately 10" apart, and varying in number up to seven on each superheater. Just above the mantles and entirely dividing the lantern is a white vitreous enamel reflector which carries two re-designed anodized aluminium reflectors, giving an improved light distribution to those reflectors previous used. These reflectors are independently adjustable in both horizontal and vertical planes, special fittings being available for providing abnormal azimuth bias where required for the proper lighting of sharp bends. The whole of the interior fittings, i.e. burners and reflectors, can be detached from the lantern by undoing five nuts, thereby making maintenance very simple. This lantern is very suitable for midnight reduction and other special features, such as half-night uni-directional lighting.

3. The "Crystal" Lantern
Is designed for use with high pressure gas, the burner being of orthodox type with two high pressure mantles, and the notable feature is the novel type of reflecting equipment which consists of prismatic glass of varying pre-determined lengths suspended parallel to the mantles and at such distance than when the lantern is alight the reflections from the prisms merge together to give a full and uninterrupted flash. The final design of the prisms has not yet been decided.

4. The "Highway" Lantern
The lantern is constructed on modern lines in the form of a heavy gauge top, rectangular in plan, which incorporates teh flue, burner and mantles; an enamelled iron top reflector supports the directive equipment, and the non-corrosive glazed body has four access doors. The reflector system embodies the use of paraboloid reflectors with the object of providing a distribution of light adequately wide for all normal roads, and with a polar distribution particularly suited to the production of high road brightness without excessive glare and giving a high lantern lumen output. The burner is normally fitted with twelve mantles arranged in two rows, and it projects somewhat below the flat top reflector. The anodized aluminium specular reflector units and mountings are each adjustable over a wide range horizontally, so that the direction of the maximum intensity can be set to the best advantage, both on straight and curved roads. The vertical angle can also be adjusted if this is necessary. The reflector unit can be very easily removed and the whole burner system is then detachable for maintenance, etc.

5. The "Alpha" Lantern
The shape and general appearance of the lantern conforms with present-day practice. It is made in four sizes, having respecitvely six, eight, ten or twelve mantles, and the distribution has a wide horizontal spread. The lantern was developed from the well-known Avil "Strip" lantern. An intensive beam candle power is deliberately avoided by using flat glass reflectors which are specially treated by a process ensuring full resistance to conditions met with in use. These mirrors are situated above and below the line of mantles, there being no obstruction of any kind in front of the light source. Positioning of the mirrors is obtained by their being fitted into cast frames of light alloy which lift right out of the lamp for maintenance purposes. The setting of the mirrors according to the spacing of the units and other conditions under which the lamp is to be fitted is usually determined during manufacture. For side lighting the lamp is usually supplied with "Staybrite" steel reflectors attached to the inside of the mirror end frames to increase the beam spread across the road. The body, canopy and ventilator of the lantern are constructed from copper, and both end frames of the lantern are of cast brass. The end frames, together with a bridge piece, give the lamp complete rigidity, and the all-copper casing ensures a long useful life.

Gas Lantern Design With Particular Regard To The Lighting Of Bends
There a two types of distribution given by modern gas lanterns used for main road lighting: (1) Where the mantles are fitted in cluster formation producing beams of roughly circular circular cross section and (2) where the mantles are fitted in a horizontal line producing a wide horizontal spread of light - this may extend to 60° to 70° measured in plan. This is good for bends of not too acute curvature. However, for narrower distributions, it becomes necessary to adjust the normal distribution to project the light beams in the direction of traffic. An azimuth bias adjustment is needed in these cases. With the exception of only one or two gas lanterns, most units of the vertical distribution type used for main road lighting, are not provided with means to adjust in-situ the azimuth bias. (This becomes more important with single side mounting of bends at determined angles of separation as recommended by the Final Report).

Uni-Directional Lighting Single-direction lighting is used on part of the Gas Light and Coke Company's area for several years. Lighting has been by means of dual cluster burner lanterns projecting the light up and down the road from normal lighting-up time until midnight, after which only the burner facing the oncoming traffic is left in lighting. A system of half-night uni-directional lighting can be the means of effecting economy in lighting costs hesides having the practical advantage that at the time when most traffic is using the road, the maximum lighting is given and the light is only reduced durign the period where there's less road traffic. The principle being that less energy is required because all available light produced is projected towards oncoming traffic. With normal street lighting a considerable amount of reflection is provided by light which is projected from the lantern away from the observer, striking objects not necessarily on the road, then being reflected back towards the observer, thereby providing background brightness which cannot possiblity be otherwise obtained and which adds considerably to the general visibility. Slides showing uni-directional lighting and normal lighting show there are advantages with normal lighting. Single direction lighting is only applicable to dual carriageways which, when lit, present a very dismal appearance particularly to pedestrian traffic when looking in the same direction in which the light is projected. Is uni-directional lighting safe lighting under all conditions? There are circumstances which do arise, such as vehicular traffic crossing from one carriageway into the other, pedestrian traffic crossing the road etc., when it is necessary to look in the direction the light is projected. It depends if the economic value of uni-directional lighting deosn't result in the loss of road safety. There are many modern gas street lighting installations which are easily convertible to uni-directional lighting - either for whole time use or as an after midnight reduction.

Mantles Giving Increased Light Output
Investigations are going into flame excited luminescence - it is possible it will provide future means for a considerable increase in light output per cubic foot of gas. Research into the possibility of increasing the efficiency of gas mantles has been carried out and has resulted in the commercial production of mantles giving an increased light output per cubic foot of gas. This result has been obtained by altering the mantle weave and chemical quantities of thorium and cerium with which mantle fabric is impregnated. Test show a constant light output increase of about 20% throughout mantle life with 20-30% decrease in actual time the mantle will remain intact.

Light Operated Series Control System
This method of control has already been applied to lanterns fitted on the Gas Light and Coke Company's area. It has proved so successful that its use is to be extended:

1. All lanterns lighted and extinguished simultaneously. Major control can be actuated either by hand, clock controller or photo-electric cell.
2. Lantersn can be lighted individually for testing.
3. Attention to the control is required only about once per month.
4. All lanterns can be extinguished from one point in case of emergency.
5. The system is independent of the gas supply.
6. Any number of lanterns can be controlled from the one point.
7. Special arrangements such as extinguishing alternative lanterns at midnight can easily be incorporated.

Luminaires: C. H. Kempton Triumph, Parkinson Maxill Type "G", Sugg Highway and Foster And Pullen Alpha.

J. G. Christopher and J. S. Smith B.Sc. (Eng.)

Keywords: Lighting: Design, Lighting: Luminaires, Lighting: Maintenance, Lighting: Materials

Wednesday, September 6th, 1939

Abstract: This paper has traced the development of present practice, and has endeavoured to show how some of the problems which arise have been approached; but no attempt has been made to prophesy the trend of future design or to classify types of lanterns. The rapid progress of modern research will always make dogmatic opinion dangerous; moreover, many desirable developments are precluded at present day by their expense. Netherless, whatever conditions may arise in the future, the broad mechanical requirements of a street lighting lantern will remain unchanged for many years.

Until a few years ago, a street lighting lantern was primarily a cover to prevent rain from falling on the lamp. The optical systems were crude and not always effective. The ill effects of high temperature led to the use of ventilating systems which drew insects and dirt into the interior. The lantern was hung by a hook which allowed it to swing. The wiring was exposed.

The economies effected by the use of modern lamps have answered the popular demand for better lighting while new knowledge of the principles on which the success of an installation depends has made it possible to lay down more rigid limits for light distribution. Increased demand for equipment has made modern production methods essential in manufacture.

The solution is the production of a range of lanterns, each with specific application, rather than a multiplicity of types differing only in their external appearance.

The scope of the paper is restricted to the engineering principles which apply to mechanical design, not to the employment of an efficient optical system.

The Mechanical Requirements Of A Street Lighting Lantern
There are certain fundamental requirements which any satisfactory lantern must fulfil. These are:

A. Mechanical Design
Precise rigid assembly will enable (B) and (C) to be satisfied and which will ensure uniformity of initial performance. All these requirements must be considered simultaneously in design.

The function of the lantern body is to protect and support the optical system and the form of the metal work will thus be controlld by the type of the optical system employed. Some part of the housing, usually incorporating the bracket fixing, will provide a "foundation" on which other components are carried.

B. Service Requirements
1. Resistance to weather.
Any lantern should be able to withstand the worst conditions of tropical rainfall. In an unsatisfactory design: water may enter the lantern and crack the lamp, or in a sudden storm the external glassware of the lantern may fracture due to abrupt cooling. A more insidious trouble is liable to come from condensed water in the bracket arm as a few drops, led down the wiring, may crack the lamp. For this reason, drip proof terminal boxes have been developed for lanterns using top suspension. Some have the additional advantage that they are mounted on spherical seatings on the bodys, allowing the lantern to be tilted by 5° so the lantern can be hung vertically even though the bracket arm is not aligned. The danger of cracked glassware can be practically eliminated in well designed lanterns by the use of heat-resisting glass similar to that used for ovenware. The standard spray test uses a nozzle giving a fan of water spray in a vertical plane equivalent to 10" of rain per hour.

2. Resistance to corrosion.
Bronze, brass and copper, cast-iron, steel and aluminium can be used for the construction of the body of the lantern. Castings in the first three metals are expensive and heavy. Sheet copper is not mechanically strong and sheet steel is liable to rust. Aluminium alloys have light weight and good mechanical properties and are now being wdely used, but some are subject to corrosion and electrolytic action and to avoid these dangers careful selection is necessary. The resistance to corrosion of most alloys can be improved by special finishes; these include special paints and direct treatment of the bare metal by chemical or electrical oxidation. Materials for screws or other components must be carefully selected as aluminium alloys are liable to electrolytic action - this precludes the use of brass or copper but cadmium plated sleet has been found satisfactory. Tests in a salt laden atmosphere are representative of the most severe conditions met in practice. Tests are made on complete lanterns over several months.

3. Ability to withstand shaking and vibration.
There is no fundamental difference between shaking and vibration. In practice the oscillations to which a street lantern is ubject often fall into two distinct classes: low frequency oscillations of large amplitude ("shaking") and high frequency movements of small amplitude ("vibration"). As far as lanterns are concerned, precautions such as the use of special lock-nuts will prevent trouble. "Shaking" occurs most severely when lanterns are mounted on trolley-bus poles. Observations in one particularly bad installation showed that lanterns were shaken with a vertical amplitude of some six inches at about 180 cycles per minute. It is not surprising that sheet metal lanterns sometimes fail under such conditions - so the use of castings for stressed parts of the lantern is preferrable. It is essential to hold every component very firmly - as a large heavy object will soon hammer itself or the lantern to pieces if it is at all loose.
These shortcomings have led to new forms of construction in which the lantern has been made light, rigid and robust. In place of sheet metal, cast aluminium alloys are used. Side entry means that the lantern is supported at its strongest point and the leverage at the fixing is small - no separate drip proof terminal box is required, erection and wiring are extremely simple and the lantern is mounted with no loss of height. The only sheet metal is the spun top, which acts as a cover only, carrying no weight.

4. Satisfactory operating temperatures.
High temperature shows itself in the lamp and the glassare. Internal connections and magentic deflector coils may also give trouble unless asbestos covered cable is used. For new designs, it's often necessary to make up an experimental lantern on which measurements can be made. Excessive lamp temperatures will adversely affect the lamp performance and may cause difficulties with lamps and holders. Internal reflectors can be used as a heat baffle. The lamp cap temperature can be reduced by several degrees if the inside of the lantern is painted black; the outside surface finish also has an appreciable effect. When lamps are burnt horizontally, high lamp temperatures usually do not occur.

A range of glasses is available having varying degrees of resistance to rain-splash. With silvered glass, even when protected from the weather, excessive temperatures may cause failure of the silvering, though the glass itself may be unaffected. Temperature measurements are bests made with thermocouples. To avoid random colling, tests are most conveniently made in a draught-free enclosure. Tests are made with the lamp running on overload to ensure the conditions are more severe than in practice.

C. Maintenance Requirements
1. Simple erection and wiring.
Arrangements which appear simple on a bench in the workshop may become infuriatingly complex on top of the tower wagon. Erection should therefore be made as simple as possible. Many lanterns are fitted with detachable couplings, which are first fixed to a threaded bracket arm; the lantern can then be attached and the wiring completed without twisting the cables. The side entry type of mounting avoids the need for a detachable coupling.

2. Ease of cleaning and freedom from dirt collection.
The best way to ensure easy maintenance is to make the optical system as simple as possible and to make the lantern totally enclosed. Dirt will then collect only on the outside which can often be made smooth, the prismatic surfaces being on the inside. Reduction in light output due to absorption in the additional glassware and glazing bars for a totally enclosed lantern is only some 5% to 10% and this loss is usually outweighed by reduced depreciation. As much of the optical system as possible should be kept smooth.

The principle of total enclosure has recently been extended to lanterns which have not been available in this form i.e.e the "cut off" lantern for horizontally burning electric discharge lamps, which has usually been constructed with exposed reflectors mounted in metal housings. The new form of the lantern (GEC Blown Cut-Off) the glassware is blown as a complete globe, the upper pars of which is silvered to form the reflecting surfaces. On the lower portion a suitable diffusing finish is moulded in order to break up striations. There is the further feature that the silvered surfaces are permanently fixed relatively to one another. No trouble can ensue from inaccurate assembly of the reflectors.

Total enclosure may give rise to difficulties if suitable precautions are not taken. Increased temperatures must be expected since there is no free circulation of air around the lamp. There is also the need for a satisfactory gasket to form the seal between the components of the lantern. So heat-resisting glassware may be necessary and asbestos-covered wire may have to be used for magnetic deflector coils. Rubber is not satisfactory as it will perish under the influence of heat and ultra-violet radiation. More satisfactory is the thick felt gasket, cemented permanently into a recess provided in the lantern body.

3. Unvarying performance without adjustments.
The initial cost of a street lighting installation is about 15% of the total cost. Glass refractors and glass/metal reflectors are widely used for the optical system. Refractors have the advantage of permanence and silvered glass reflectors of high reflection factor and permanence if of good quality. Anodically brightened and oxidised aluminium reflectors now used have much greater powers of resistance to atmospheric corrision than earlier types. The permanence and ease of cleaning glassware swings the balance in its favour. The optical system should aways be kept as simple as possible and should be held firmly in the lantern body. Wherever possible it is preferable to avoid adjustable lamps or optical systems. Where a focussing adjustment is necessary it should be simple, and some positive indication of the correct setting should be given.

D. Appearance
1. Unobtrusive and attractive.
Street lighting and other installations should be as unobtrusive as possible. This can only be done by avoiding heavy ornate columns nad ugly lanterns which obstruct both the footpaths and the view. It is the optical system which controls the general form of a lantern. It is reasonable to give the optical system the main emphasis and make the metal work as unobrusive as consistent with the strength and balanced apperance.

Luminaires: Holophane Bi-Way Directional Refractor, GEC Difractor (Z8128), GEC Horizal (Z8425), GEC Blown Cut-Off (Z8435), BTH Mercra "H" and GEC Uniway (Z???)

References:
[1] English, Ill. Eng. Soc., Vol 27, 1934, p352
[2] Wilson, A.P.L.E. Conf, 1937
[3] Beggs And Wilson, G.E.C. Journal, Vol 6, No 3, August 1935, p127
[4] Maxted, Electrical Review, Vol. CXXII, May 13th, 1938, p678
[5] Dix And Bowman, Symposium on Corrosion Testing Procedure, 1937, p57
[6] Edwards, Transactions Of The I.E.S., Vol XXXIV, No. 4., April 1939, p427
[7] Christopher, Smyth and Waldram, G.E.C. Journal, Vol. X., No. 3., August 1939, p205

Francis F. Middleton

Keywords: Lighting: Specifications and Lighting: Theory.

Thursday, September 7th, 1939

Abstract: Instead of projecting high intensity beams towards approaching traffic, they should be directed to those regions forming an important porportion of the field of view, where normally minimum brightness is apparent.

The success or otherwise of a Street Lighting installation depends upon the ability of the Lighting Engineer to utilise his skill and judgement in:

• Type of light source
• System of control or distribution
• Location in relation to the backgrounds he will illuminate

The bright region ("T" shape) size and width is subject to considerable variation: condition of the road, wet or dry, state of repair, degree of polish, structural composition, camber and longitudinal profile of the road.

When a surface is uniformly illuminated from a single source, it is the proportion of favourable facets in one region to the fewer number of favourable in another region that determines the variation in apparent brightness. Therefore, we can so design the distribution that it projects a higher intensity of light to the fewer number of favourable facets located in the normally less bright regions, and a lower intensity to the greater number in the brighter region to create a more uniform appearance.

Installations, where lanterns are expressively designed to increase the brightness of the brighter region, are unsatisfactory, particularly for junctions, for they provide efficient contrast only on a limited area, and permit little or no tolerance on the siting positions.

The theory, is that instead of projecting high intensity beams towards approaching traffic, they should be directed to those regions forming an important proprtion of the field of view, where normally the minimum brightness is apparent.

A lantern so designed for either central or side mounting at heights of 25' are now generally adopted. These will offer decided advantages on bends and junctions over the type which produces increased brightness between observer and the source. Lanterns can be deisnged to deal adequately with an area 120' long by 50' wide with a light output within the range 3600 to 9600 lumens. It is suggested to standardise two types based on this principle: one for side of road mounting 'S' type (for bends) and the other for central mounting 'C' type (for junctions). By skilful judgment in selectign the siting positions for these two types a multiplicity of designs, with increased manufacturing cost, will be avoided.

Proposed siting arrangements for these two types of lantern are then given.

K. F. Sawyer, B.Sc., A.M.Inst.Gas E.
Watson House Laboratories, Gas Light And Coke Company

Keywords: Lighting: Distibution and Lighting: Theory.

Friday, September 8th, 1939

Abstract: From an experimental scale model of a roundabout, the effects of lanterns with different distributions and sitings was studied. The conclusions drawn from the experiments was applied to a real roundabout.

Describes experiments made on a very simple form of roundabout.

The experiments were designed to show what arrangement of lantern positions and what type of candlepower distribution were most effective in providing:

• (1) A clear indication to the driver approaching the roundabout of the route throught the junction; the nature and curvature of any bends; and the presence of any objects ahead.
• (2) Easy perception while the driver is negotiating the roundabout of the objects in his vicinity and at the ends of the roads leaving the roundabout.

The requirement that the lighting installation should also warn oncoming traffic of the existence of the roundabout was not considered of primary importance; it was felt that street lighting could not replace the warning signs normally employed.

The manner in which objects are revealed differs considerably according to whether the observer is approaching the roundabout or is travelling in the encircling roadway. As approaching at speed, the road surface is a narrow field of view and objects are seen in silhouette at a distance. On the roundabout, the speed is reduced, direction of travel constantly changes, and attention is given to a wider field of view much nearer. Verges, footpaths and the buildings beyond will generally become more important that then road surface as a background and the angles at which they are seen preclude any extensive use of specular reflection. Horizontal and vertical illumination have more significance than in normal street lighting.

Experimental Equipment
A 1/20th full scale model was constructed. The reflecting properties of all the surfaces concerned were made to match those with occur in practice.

Four types of candlepower distribution were used:

• (1) Asymmetrical with maximum candlepower at 80° and with the candlepower above this angle well maintained up to the horizontal. This is called the "high beam" distribution.
• (2) Symmetrical with no marked directional properties. Similar to that produced by a suspension gas lamp or electric lantern provided with horizontal enamel reflectors. This is called the "general" distribution.
• (3) Symmetrical with cut-off reflector. The candlepower was greatly reduced above 70° with no light emitted above the horizontal.
• (4) ????

Experimental Procedure
Using each of the first three candlepower distributions, various groupings of the lanterns on and around the centre island and in the approach roads were bserved. Records were made visually and photographically.

These were studied extensively with the road surface dry then, after selecting the most promising, they were again investigated with the road and footway wet.

Visual observations were made from a number of viewpoints, but to reduce the extent of photographic work, photographs were taken from three fixed viewpoints only e.g. approach to roundabout, entering roundabout and leaving roundabout.

To provide pictorial evidence of the degree of visiblity provided by the various layouts, three test objects were used: (1) a pedestrian and his dog; (2) a cyclist and (3) the cyclist at a different position.

Siting Of Lanterns In Approach Roads
Last lantern sited on the right would allow the first lantern of the opposite road to show objects in silhouette on the approach to the roundabout. This would not be the case if the lantern were sited on the left. This arrangement also showed some advantage for objects leaving the roundabout. In view of these experiments, the positions of the lanterns at teh ends of the approach roads were fixed for all the subsequent work.

Installations Using High Beam Candlepower Distribution
The first layout had a simple lantern mounted in the centre of the island with a four-way distribution mounted at 25'. The installation provided good road brightness in the approach roads - but the extent of the central island and the road around was not well defined.

Four lanterns were mounted over the kerb of the central island at points opposite the approaching traffic. They provided a well-lighted background and removed all troublesome shadows from the roadway. The lanterns were first arranged so the beams were directed tangentially to the centre island (A) and then turned through 90° so that maximum candlepowers were directed along the approach roads (B). Axial distributions were used for the lanterns on the island, and the usual non-axial distribution were used for the approach roads. With their power and mounting height they produced reasonable results merely on account of the large amount of light. In (A), there is little to recommend it as most of the redirected light falls outside the carriageway.

In (B), the grouping of the three lanterns quite near the centre of interest gave rise to disability(?) glare.

As far as the oncoming driver is concerned, this arrangement is quite satisfactory approaching the roundabout. But negotiating the roundabout is not so satisfactory. When the lanterns on the central island are comparatively close together, the vertical illumination is relatively high and the the vision of objects on the roundabout is fairly good. During wet weather or when the lanterns are more widely spaced, vision of objects, particularly those near the outer kerb, become very indifferent.

Lanterns in the next installation were mounted on the outer kerb of the encircling roadway at points midway between the road entries. This siting provided a general improvement in the distribution of illumination and resulted in a considerable improvement in the ability to detect objects near the outer kerbs. It contributed little towards providing a high surface brightness in the approach roads.

Installations Using A General Candlepower Distribution
The experiments confirmed what had already been anticipated: the approach roads and the encircling roadway would require different treatment and that is would be impossible to make the lantern which adequately fulfils one function to fulfil the other to the same extent.

The effect of using lanterns having a general instead of a directional with the same siting was investigated. It resulted in a considerable reduction in glare and the brightness of the approach road, but on the roundabout itself, the lighting suffered from the same defects as the high-beam lighting.

Installations Using A Cut-Off Candlepower Distribution
The four lanterns on the centre island were replaced with lanterns having a cut-off distribution. The whole of the centre island and the roadway surrounding it were raised to a high level of brightness which extended well into the encircling roadway. Visibility of objects from both distant and near viewpoints was better than previous installations. The glare was much reduced.

Cut-off lanterns on the centre island could only deal with the lighting of the encircling roadway Non-cut-off lanterns were retained for lighting the approach roads. The bright areas of the encircling roadway contracted noticeably in the wet - this was less serious when the lanterns were centrally suspended. There was no doubt that this installation was superior to any others previously investigated.

Additional Installations

(a) Lanterns mounted centrally on island

In the interests of economy roundabouts are frequently lighting by a single centrally mounted lantern. High beam, symmetrical and cut-off lanterns were tried. A cut-off lantern is best to light the roundabout and the encircling road, whilst a general distribution of moderate horizontal candlepower is best to warn oncoming traffic of the existence of the roundabout. When the approach roads were lighted, experiments indicated that the symmetrical direction distribution was most satisfactory as a single central source.

(b) High Mounted Cut-Off Distribution

The defects of a single lantern mounted in the centre of the island were mitigated by adopting a high mounting height with a slight increase in the amount of light. This method is frequently used on the Continent for lighting large open spaces. Four cut-off lanterns were mounted in the centre of the island to a height of 40'. The general appearance was rather pleasing and the ease of vision better than any other central source. However the performance was still inferior to more previous, more orthodox installations.

(c) Short Column Lanterns

This use has been advocated from time to time. It consists of eight lanterns mounted on columns approximately 12' high and arranged at equal intervals on the kerbs of the central island. The general effect was rather like that of a birthday cake. It suffered from an unexpected degree of glare and confusion. It was disappointing both for traffic approaching and negotiating the roundabout.

Conclusions From Experimental Work

(1) For the driver negotiating the roundabout itself, the most satisfactory installation for both wet and dry surfaces was that of symmetrical cut-off lanterns mounted on the outer kerb midway between the road openings.

(2) For the driver approaching the roundabout, lanterns having a general distribution mounted over the kerb of the central island at the points opposite the approaching traffic lanes was the most satisfactory under wet and dry conditions.

(3) A combination of (1) and (2) gave the best performance i.e. cut-off lanterns on the outer kerbs and general lanterns opposite the approach roads.

(4) High beam lanterns mounted over the outer kerbs were quite inappropriate since the redirected light passed usually beyond the roundabout.

(5) Lanterns mounted centrally on the island without supporting lanterns on the outer kerb were very unsatisfactory. Certain defects could be overcome to some extent by increasing the mounting height and the amount of light provided.

(6) The installation of a number of short column diffusing lanterns at frequent intervals on the kerb of the island was less satisfactory than a smaller number of lanterns mounted at 25'.

(7) The position of the last lantern in the approach road, where staggered formation is used, could with advantage be on the offside kerb as seen by the approaching driver.

(8) When the lanterns are arranged as in (7) and the roads on opposite sides of the island are in the straight line, its is possible to use the lantern at teh end of one approach road to produce road brightness in the opposite approach road, provided high beam distributions are used and the central island isn't too large.

(9) When asymmetric distributions are used in lanterns situated opposite the driving lanes of the approach roads, the main beams should, be directed at right angles to the kerb, so that the redirected light is used to produce surface brightness in the approach roads.

Practical Interpretation Of Experimental Results
These conclusions refer strictly only to the experimental roundabout. In practice much of the symmetry and simplicity of that layout is lost and cost becomes a consideration. The economical approach of a single lantern in the centre, instead of those on the inner kerb, was investigated and experiments showed it gave reasonable performance if the island did not exceed the dimensions from the experiment. For larger and more unbalanced layouts, a single central source can never be sufficient.

Lanterns having an asymmetrical high beam distributions are frequently used for the lighting of roundabouts in spite of their being inappropiate.

Experiments have indicated the great importance of some sort of fence surrounding the central island. It plays a very important part in the revealing of traffic crossing the end of the approach roads.

Application To An Existing Roundabout
These principles were applied to a roundabout on one of the western main roads out of London. It was chosen because it was irregular and presented two islands side-by-side. There was no symmetry. No lantern on the larger island can, therefore, serve more than one approach road and the lantern at the end of one approach road can contribute little to the lighting of any other. Adequate lighting necessarily involves the use of a considerable number of lanterns.

In all, the scheme requires the use of eight cut-off lanterns, four symmetrical lanterns and four high beam lanterns in the approach roads. This replaced a scheme of eleven lanterns.

What may therefore appear to be a somewhat extravagant scheme requiring 16 lanterns and its approaches is only a slightly more satisfactory application of the lighting already supplied. Five additional lanterns provide the more even distribution characteristic. The installation cost would be higher but the running cost would be substantially the same - but the more certain visibility, especially under wet conditions, would more than justify the initial expense. Some criticism may be levelled at the use of lanterns having three entirely different types of candlepower distribution - such a practice is only a logical outcome of applying the light available.