.

T

HIS COMMITTEE submitted as information a report on a code of colors for traffic signals in co-operation with the American Engineering Standards Committee. A. H. Rudd (Penna.) represented the Signal Section, while W. J. Eck (Sou.) served as an alternate.

Code on Colors for Traffic Signals

On May 23, 1922, a conference was held in New York attended by representatives of about forty associations, consisting of automobile manufacturers, safety councils, traffic officers, automobile regulating authorities, public service commissions, and the Interstate Commerce Commission. At this conference it was voted that it was advisable to standardize in the United States the signals governing highway traffic.

The outgrowth of this meeting was a decision to refer the matter to the American Engineering Standards Committee. Various sub-committees were appointed; their reports codified and reviewed in 1924 by the Executive Committee of nine, representing the American Association of State Highway Officials, the U. S. Bureau of Standards, insurance companies, Eno Foundation, steam and electric railroads, National Automobile Underwriters Conference, Motor Truck Association of America, Motor Vehicles Conference Committee, and the manufacturers. A preliminary draft of the code was submitted by it to the Sectional Committee for consideration. After a tremendous amount of work, the code was agreed to by the Sectional Committee, submitted through its sponsors, namely, The American Association of State Highway Officials, the United States Bureau of Standards, and the National Safety Council, to the main committee for approval, and on November 15, 1927, was approved by the American Engineering Standards Committee as an American standard, five and one-half years after the first meeting.

The code as originally proposed included non-luminous as well as luminous signs, but, during the progress of the work, it developed that one of the sponsors, namely, the American Association of State Highway Officials, had almost completed its work of standardizing these non-luminous signs, and, as has been previously reported, there was a conflict of opinion between the Committee on the Code of Colors for Traffic Signals and this Association in the matter of the stop

sign. In order to prevent overlapping, the code was, therefore, amended to cover colors for luminous signs. only, thereby supplementing the standards adopted by the American Association of State Highway Officials instead of conflicting therewith. The code adopted as an American Standard follows:

Code on Colors and Forms for Traffic Signals for Highway Vehicles

Section 1. Purpose and Scope

Rule 10. Purpose-The purpose of this code is to provide a standard of colors and methods, as applied to highway traffic and vehicles, which will promote uniform usage and thus decrease the likelihood of accidents, and conserve human life and limb. The approval of specific signs or glasses, as complying with the specifications herein contained should be based upon tests made by competent and disinterested organizations having proper facilities and recognized standing, with final appeal to the United States Bureau of Standards.

Note: To secure the uniform application of this code, enforcing officers are urged to consult the committee which formulated it (or the American Engineering Standards Committee, 29 West Thirty-ninth street, New York) regarding matters of interpretation or items of dispute.

Rule II. Scope-This code is intended to cover the use of luminous signs and signals in connection with highway traffic, including moving, and flashing signals; the use of lights, semaphores, and other signaling devices on vehicles.

Section 2. Vehicle Lights and Signals

Rule 20. Headlights-(a) Headlights shall be white, amber, yellow, or any intermediate hue. If electric headlights are used, they shall conform to the rules governing the approval of electric headlighting devices for motor vehicles, of the Illuminating Engineering Society (tentative standard of the A. E. S. C. D'); (b) No red or green lights shall be displayed upon any vehicle so as to be visible from a point directly in front of it. Rule 21. Tail Lights-Tail lights shall be red, as seen from the rear. (Note: Red has been adopted on account of its present universal use for this purpose. This use is not con

sidered consistent with the general standardization of colors, and if a change is made in any jurisdiction, yellow is reconimended.)

Rule 22. Warning signal lights-(a) Light signals displayed on a vehicle to indicate the driver's intention to reduce speed or change direction should be yellow. They shall not be red or green; (b) Non-luminous mechanical signals displayed to indicate the driver's intention to reduce speed or change direction shall consist of a semaphore which can be projected from the left side of the vehicle. It is recommended that the position of the semaphore when projected be limited to the horizontal. Rule 23. Marker Lights-Marker lights to indicate the dimension limits of bodies or loads should be located as near the It is intended to cover standards for non-luminous signs in a future addition to this report.

upper left front corner as possible and should be yellow and visible from the front and rear.

Section 3. Highway Traffic Signals

Rule 30. Significance of colors-(a) Red shall be used as an indication to stop and for other purposes where required by law. (Note: The use of red is proper as an indication to stop and to then proceed if conditions are favorable, as, for example, when "stop and proceed" requlations are in effect.)

(b) Yellow shall be used as an indication to exercise caution. (Note: Yellow is appropriate when caution is to be exercised without stopping, as for partial street obstruction, so as to reserve red for a stop signal.)

(c) Green shall be used as an indication to proceed. Rule 31. Significance of form-Where definite forms of luminous signs or signals are used either alone or in combination with colors to give the indication specified in Rule 30, the following will apply:

(a) Stop shall be indicated by having the greatest dimension horizontal, as with a horizontal string of lights. (Note: Where the word STOP is used to direct traffic, it should be in red letters arranged horizontally.)

(b) Caution shall be indicated by having the greatest dimension at an angle of 45 deg. with the horizontal when an inclined string of lights is used.

(c) Proceed shall be indicated by having the greatest dimension vertical, as with a vertical string of lights. (Note: Where the word GO is used to direct traffic, it should be in green letters arranged vertically.)

Rule 32. Significance of flashing-Flashing luminous signals shall conform to the use of colors prescribed in Rule 30.

Section 4. Signals at Railroad Grade Crossings Rule 40. Aspect-An electrically or mechanically-operated signal used for the protection of highway traffic at railroad crossings shall present toward the highway when indicating the approach of a train the appearance of a horizontal swinging red light and/or disc. (Note: This covers the use of so-called wig-wags and of alternately flashing red lights, and the use of these devices should be restricted to the purpose of indicating the approach of a train.)

Rule 41. Location-The railroad standard highway crossing sign and the signal shall be mounted on the same post.

Rule 42. Operating time-Automatic signal devices for indicating the approach of trains shall be so arranged as to indicate for not less than 20 sec. before the arrival at the crossing of the fastest train operated on the track. Local conditions such as three or more tracks, bad approaches, etc., should be allowed for by increasing the operating time, bearing in mind that too long an operation by slow trains is undesirable.

Rule 43. Flashing light type-(a) Height. The lamps should preferably be not less than 6 ft. nor more than 9 ft. above the surface of the highway. (b) Width. The two lamps shall be mounted horizontally on 2 ft. 6 in. centers. (c) Flashes. Lights shall flash alternately. The number of flashes of each light per minute shall be 30 minimum, 45 maximum. (d) Hoods. Lamp units shall be properly hooded. (c) Range. When lamps are operated at normal voltage, the range, on tangent, shall be at least 300 ft. on a clear day, with a bright sun at or near the zenith. (f) Spread. The beam spread shall be not less than 3 deg. each side of the axial beam under normal conditions. This beam spread is interpreted to refer to the point at the angle mentioned where the intensity of the beam is 50 per cent of the axial beam under normal conditions. (g) Lenses or roundels. The size shall be 53% in. minimum, 83% in. maximum.

Rule 44. Wig-Wag type-(a) Length of stroke is the length of chord which subtends the arc, determined by the center of the disc in its extreme positions, and shall be 2 ft. 6 in. (b) Disc. The disc shall be 20 in. in diameter. A red lens or roundel should be placed at the center in front of the lamp. (c) Number of cycles. Movement from one extreme to the other and back constitutes a cycle. The number of cycles per minute shall be 30 minimum and 45 maximum. (d) The lamp, with which the disc shall be equipped, shall be lighted when the disc is swinging.

Rule 45. Approach signal-Advance warning signals which indicate approach to a railroad crossing, and not the actual approach of a train, shall conform to the provisions of Section 3, Rule 30.

Section 5. Specifications for Colors

Rule 50. Definition of Colors--Red, yellow and green as used in this code are intended to mean:

(a) The colors resulting from transmission of the proper light through the proper glasses and having the characteristics described in Rules 51 and 52.

Rule 51. Qualitative definition of colors for luminous signals —(a) Red. The spectrum of red shall contain both red and orange but not more than a trace of yellow and no green, blue,

or violet. The most desirable hue is entirely free from yellow which means that the glass does not transmit the yellow light from a sodium flame. (b) Yellow. The spectrum of yellow shall contain red, yellow and green, with but little blue and no violet. The most desirable hue is entirely free from blue and might be designated a light amber. (c) Green. The spectrum of green shall contain yellow, green, blue and violet, with only a trace of red and orange. This hue is known as "admiralty green" and has a bluish tint when observed by daylight.

Rule 52. Quantitative definition of colors for luminous signals-The colors red, yellow and green shall have the following characteristics: Dominant wave-length

Red Yellow

Millimicrons

Purity Per cent

Integrel transmission of glass Per cent

.Not less than 624.... Not less than 100.... Not less than 10
Not less than 592

nor more than 600....Not less than 97.... Not less than 24 Green Not less than 496

nor more than 536....Not less than 45....Not less than 11 These values are determined by the transmission of light from a source at the color temperature of 2,360 deg. K (practically that of the acetylene flame or present type of vacuum tungsten lamp at normal voltage) through the respective glasses. They are based upon spectral transmission measurements and upon computations carried out in accordance with the methods and data described in the Colorimetry Report of the Optical Society of America.

Note The light and dark limits of the glasses on which the above values are based have the following relative transmissions on the scale of the American Railway Association:

We take the position that in certain crossings with light traffic, the railroad has done its duty when it indicates that there is a railroad there. The United States Supreme Court recently has adopted our view.

In the matter of grade crossing protection, we hope some day our government will be as liberal as the Dominion's. In New York a bill has been passed, providing for a large fund-$400,000,000-the state borrowing the money and loaning part of it to the railroads. The railroads pay 50 per cent of the cost of grade crossing abolition, then amortize the loan and pay the interest.

In the state of New Jersey the railroads pay it all. A bill is pending now providing for adequate protection on the basis that the railroads have not made the crossings more dangerous but that the increased number of automobiles has. In the state of Pennsylvania we have arrived at an agreement with the Highway department, and approved by the Public Service Commission, that when we protect crossings on state highways the highway department pays half and the railroad half.

[Mr. Rudd then read the report of the Special Committee on Highway Crossing Protection].

Mr. Rudd: Since this report was written, I was a member of a committee of railroad men appointed by Mr. Hoover to prepare a standard ordinance for cities and towns which will contain traffic rules, laws, signals and signs, so that we may standardize throughout the United States. I was a member of the Committee on Signs and Signals. I found early in the game that the American Engineering Council, as an adviser of that committee, was at variance with some of our ideas.

Shortly after, Mr. Aishton wrote me asking me if I would serve on the American Engineering Council, because that body did not have any railroad representatives. They have recommended for adoption at the National Conference on Highway Crossing Protection either the wig-wag or the alternately flash ing red light as an indication of the approach of a train. That is how far we have gone in standardization. Our recommendations have been adopted by the American Railway Engineering Association, and by the Code on Colors of Traffic Signals, thus making it an American standard and carrying with it the approval of the National Association of Highway Officials and the American Engineering Council. Undoubtedly it will be adopted by the Third National Conference on Highway Crossing Protection.

The committee submits this for letter ballot approval for inclusion in the Manual. (Mr. Rudd's motion that the limits set forth in the report be accepted for submission to letter ballot for inclusion in the Manual was then put and carried.)

W. P. Borland (I. C. C.): Has the committee given any consideration to any other form of protection to augment or supplement the standard which has been adopted, such as gates or something of that kind?

Mr. Rudd: Yes sir, we have. The question that was referred to us was the automatic protection of the crossing. That is what we have handled.

Mr. Borland: We are interested in that particular because it is before us now in several forms.

Mr. Rudd: In the opinion of the committee, not expressed officially, but for the information of the representatives from the Interstate Commerce Commission, gates are a good thing to protect a train against a foot passenger if he does not get under it and a good thing for horse-drawn vehicles, but it is just an annoyance to an automobile. The automobile comes up to the gate and just brushes it aside. There are more gates broken every year than there are watchmen killed. We are taking out gates wherever we can, wherever the public authorities permit us, and putting in the automatic protection. We are taking off watchmen where we can and putting in automatic protection, because we think it is more reliable.

The

T

HIS COMMITTEE submitted five revised drawings, recommended that six drawings be made obsolete and submitted four new draw ings. Those suggested as needing revision included: (a) A. R. A. Sig. Sec. 1070B-covering a change in screw thread standard for A. R. A. binding posts and nuts, from 10-30 to 10-32 thread; (b) A. R. A. Sig. Sec. 1084B-covering a revision to make the assembly drawing show the use of the improved type of top roller pin for mechanical plant pipe carriers; (c) A. R. A. Sig. Sec. 1085B-covering a similar change in drawing of pipe carrier details to incorporate the improved type of top roller pin for pipe carriers used to mount mechanical interlocking pipe

lines; (d) A. R. A. Sig. Sec. 1504B-regarding the change of the size of the pin hole from 1⁄2 in. to 33/64 in. in the lever number plate for mechanical interlocking machines. The added clearance is recommended to prevent breakage of the pin when applying the number plate; (e) A. R. A. Sig. Sec. 1399B-Low Target Stand-Details and Assembly, revised to make it interchangeable, right or lefthand.

The drawings to be rendered obsolete include: R. S. A. 1001-Wire Adjusting Screw-Details and assembly, R. S. A. 1226-Stuffing Box for Wire, R. S. A. 1237-Adjustable Lock Rod-Details and assembly, R. S. A. 1350-Horizontal Chain Wheels.

R. S. A. 1351-Vertical Chain Wheels, and R. S. A. 1352-Vertical Chain' Wheels.

Four new drawings, A. R. A. Sig. Sec. 1614A— Base for Ground Signal Masts (superseding Drawing 1034), A. R. A. Sig. Sec. 1615—Foundation for Mechanical Dwarf Signals (superseding R. S. A. Drawing 1106), A. R. A. Sig. Sec. 1597A-Concrete Battery Box (superseding A. R. A. Sig. Sec. 1343), and A. R. A. Sig. Sec. 1598A-Concrete Battery Box-Details (superseding R. S. A. 1342), were submitted for acceptance for submission to letter ballot. Electric Lamps for Signals

A total of 167 varieties of lamps are being used by signal departments for the lighting of signals and other signal and interlocking apparatus. Of these, 143 are termed special.

We have been advised that upwards of 215,000,000 lamps are produced annually for commercial purposes; of these there are 87 varieties, and only about 215,000 produced for signal lighting; of these there are 143 varieties; the price to consumers has a direct bearing upon quantity demand for each kind. For example, the list price of a lamp, as shown on A. R. A. Sig. Sec. 1544, having tolerance of plus or minus 3/64 in. on axial alignment and light center length, is 50 cents and a lamp with a tolerance of 1/32 in. is 75 cents each. The "Precision" lamps are selected from among the regular lamps.

Other advantages, even more important than first cost to the user, will be found in standardizing on fewer varieties, eliminating the confusion and errors in replacement of outages, in ordering, and in the reduction of stock.

Lamp characteristics necessary to meeting signaling requirements are: 1, long life; 2, low voltages; 3, precision of filament to secure accurate focus with lens or reflector; 4, good socket contact; 5, as low wattage as practicable; and 6, convenient size bulb. 1. Lamp Life. The life of all lamps is based on burning continuously at rated voltage. If burned intermittently, the added life is said to be so little that it may be neglected. Under voltage burning gives reduced candle power and longer life, but the entire theoretical life cannot be expected as service conditions and other factors affecting the life are not alike.

2. Low Voltage. As almost all signal apparatus is designed to operate with low-voltage current, the electric lamps ranging in voltage from 2% to 16 meet present day requirements best.

3. Filament. While various shapes of filaments for lowvoltage lamps are now in use, the C-21, C-3, and rounded C-2, and PS-16 are best for signaling; of these the C-3 shown on A. R. A. Sig. Sec. 1544 is the least efficient and the rounded C-2 the best. The C-2, when properly focused, gives more beam candle power than the C-21, due to the absence of dark spots caused by anchorage. It can be produced by automatic machinery. Single filament lamps are recommended by the manufacturers instead of a multiple filament (both burning), because the single filament is more rugged in consrtuction resulting in longer lamp life; they are also easier to make. The multiple filament is not insurance against failures, since failure of the one. throws excess voltage on the other and it soon fails also, and the excess voltage of the remaining filaments makes it difficult to detect the failure of the one.

4. Bulb. The S-11 plain pulb for the voltage range above, appears to be best and is recommended by the manufacturers. It can be made quite uniformly as to size and shape; its extensive use on automobiles gives it good quantity production; it is convenient to handle in the factory, field and stock room. As tests indicate, the corrugations on the bulb reduce the beam candle power. The plain bulb is recommended. The S-14 bulb should be discontinued as it is expensive, obsolete and not suitable for precision, or proper exhaustion.

5. Base. The single-contact bayonet candelabra base is now

standard, has been proved in service, and, of course, should be continued as standard for all signaling except for track diagram where the miniature screw socket is regarded as

necessary.

6. Precision. Since the filament of a low-voltage lamp produces a small light area, it is important to have the light center in focus. Variations in the location of filaments with respect to the base pins or with the theoretical correct light center and axial alignment causes poor signal lights. It is essential that variations be as small as practicable, particularly in lamps with rounded C-2 filaments. The manufacturers' representatives advise that improved methods enable them to make "regular" lamps of this type to meet a tolerance of plus or minus 3/64 in., and that precision lamps will be furnished to meet a tolerance of plus or minus 2/64 in. in axial alignment and light center length.

The possibility of using the S-11 rounded Č-2 filament lamp for day indications of light signals was discussed. The signal companies' representatives asked for 1/64 in. tolerance; they were told that such accuracy could not, at present, be assured, but that the tests would be made in the laboratory and in the field. Some field tests have been made near Nela Park on the New York Central but records of the results are not available for this report. It is therefore thought best to continue the use of the rebased P-16 lamp with the doublet lens pending further tests, and that the signal companies endeavor to provide interchangeability of the rebasing. At the present, it is in an unsatisfactory state for the railroads having light signals of more than one make and consequently a number of roads have abandoned rebased lamps, using instead the S-11 lamps and adapter.

Our conclusion from the foregoing is that the following schedule of electric lamps is best suited for signal and interlocking requirements, except position light signaling,

for which it appears to be necessary to continue temporarily the PS-16 rebased lamp. While some signal engineers continue using the CT-8 lamp with Lebby mirror we believe the S-11-C-2 lamp will be found to be a satisfactory substitute.

Schedule No. 1 for Electric Lamps
For Night Signals

[His motion to this effect was carried.]

Mr. McKeen: The committee presented the new drawings which were submitted for acceptance for submission to letter ballot.

[His motion for acceptance was carried, J. C. Mock (M. C.) then presented a progress report on the subject of electric lamps for signals.]

C. H. Morrison (N. Y., N. H. & H.): I am wondering if the S-11 bulb is to be used. I notice in the table no higher wattage than the 10 to 18, while it is common practice on the New Haven to use 27, 40 and 50watt bulbs. We believe that where electric energy is secured from the service line, the expense of electric current is so small compared with the amount of benefit which the electric light gives that we believe a higher wattage giving a more powerful light is more beneficial and will take care of the extra expense. I do not believe an S-11 bulb is of sufficient diameter to take care of 27, 40 or 50 watts. We use these in semaphore signals. I note that the report refers to daylight signals. We also use a 40-watt bulb in the daylight signal. Therefore a provision should be made for a bulb of a size that will be large enough for 40 watts.

Mr. Mock: I understand that S-11 is not recommended for anything above 32 volts and that the committee does not even want to use it for 32 volts.

Mr. Morrison: We use 8-volt, 40-watt lamps. A. W. Fisher (U. S. & S.): The S-11 lamp is limited to 25-watts.

[H. H. Helmbright (G. E.), confirmed that information. When any requirement of wattage is greater than 18, the committee' favors 25. Twenty-five watts would be the limit in the S-11 bulb. The larger bulbs that are needed to take care of that wattage would probably be the size of the PS-16, perhaps the G-181⁄2 or the G-162. The G-18 is generally used now and the P-16 is being used for daylight signals of 10 volts, 40 watts and of 10 volts, 30 watts.]

Mr. Morrison: I am of the opinion that those using 40-watt lamps would find it advisable to go to the standard Edison screw-base.

Mr. Fisher: I believe the paragraph you just read answers Mr. Morrison's question about providing screwbases for 40-watt lamps. If you are using low-voltage lamps your filaments are small, and you want greater accuracy in location than is possible with a screw-base lamp. If you want to maintain an accurately located base and an accurately manufactured lamp, it should be of a bayonet type. If you go to a larger bulb than the S-11 with 40 watts, the G-18 lamp is the only one used in the candelabra base, it being so much larger in diaeter than the base that the lamp is weakened. It can be twisted off the base much easier than the S-11.

Mr. Mock: The New York Central Lines have been making some tests, and have inspected all their lamps. The number of incandescent lamps inspected exceeded 1,000,000, and the rejections amounted to about two per cent. The inspection cost was approximately 124 cents per 100 lamps.

J. E. Saunders (D. L. & W.): How many of these are commercial lamps, that is, used in other lines of work?

Mr. Mock: These lamps are all in use now and are all standardized.

B. J. Schwendt (N. Y. C.): Some references have been made to tests on the New York Central. We were unable to get enough bulbs to put on an extended test, so confined it to a local test, comparing a signal equipped

justed to make sure it was the best signal that could be obtained from such a bulb. We took about 36 lamps selected at random all with S-11 bulbs and inserted onehalf of them to compare them with the PS-16 bulb. The result was favorable to the S-11 bulb. The S-11 has one advantage, which is quite noticeable. The commercial tolerance of the S-11 bulb, 2/64 in., is twice as great as in the PS-16 bulb, but due to its shorter light center length it doesn't require such great tolerance.

C. A. Dunham (G. N.): The Great Northern has 1,400 miles of color-light signals, and we have standardized practically on two makes of lamps. One is an 8-volt, 5-watt, and the other an 8-volt, 10-watt lamp. Even a low wattage lamp, 5 watts, for instance, if used with accuracy of adjustment, is sufficient for the purpose.

Chairman McKeen: The desire of the committee is to boil down the 57 or more varieties to as small a number as possible in order that we can get away from the precision work and obtain a lamp that can be put on a production basis. It is doubtful whether in the 40watt lamp, the S-11 type is necessary.

The committee also presents revised drawings on a low target stand. This drawing is submitted for acceptance for submission to letter ballot superseding subject matter in the manual.

[The chairman's motion to this effect was carried.] Chairman McKeen: The committee presents new drawings for a concrete battery box and concrete battery superseding subject matter in the manual. box details for acceptance for submission to letter ballot

R. B. Ellsworth (N. Y. C.): Referring to Plan 1597A, this plan, or a similar one, was objected to previously and referred back to the committee. I am disappointed to see it here again, with the recommendation that it supersede the present standard of the section, which is much superior to this plan from an economical point of view, in that it provides a lining to protect the batteries and particularly the primary batteries in cold temperature.

Chairman McKeen: The committee recommends that drawings 1597-A and 1598-A (concrete battery box and details) be submitted for acceptance for submission to letter ballot superseding subject matter in the manual. [His motion was put and carried.]

Early Days in Railway Signaling