From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Article discussing the all-important problem of safe driving by night, with special reference to headlamp focus and anti-dazzle beams.
| Identifier | ExFiles\Box 61a\3\ scan0149 | |
| Date | 29th April 1932 | |
| Reprinted from “The Light Car and Cyclecar,” April 29, 1932. Dr. A.{Mr Adams} H.{Arthur M. Hanbury - Head Complaints} Stuart, B.Sc., Ph.D., etc. Discusses the All-important Problem of SAFE DRIVING BY NIGHT With Special Reference to the Need for Correct Focus and to a Lamp which Provides a Diffused Anti-dazzle Beam WHEN a headlamp bulb fails most motorists renew it at the next garage, merely asking for a bulb of suitable voltage, and leave it at that. Very few trouble to look critically at the filament, and fewer still know what to look for if they did. I do not profess to be any better than my fellow motorists in this respect; at least I was not until recently, but during the past few weeks, I have made a considerable collection of headlamp bulbs and headlamps complete and submitted them to very rigorous tests on a photometer bench specially made for the purpose. Many points have been brought out by this work which had certainly not occurred to me before, and I hand them on to my fellows in the hope that, between us, we may improve the breed and get more satisfactory road illumination. My collection of bulbs includes all kinds, British and foreign, ranging in price from 6d. to 4s. Before putting any of them on the bench I made a critical examination of the filaments. I was very much surprised to find that it does not always happen that the filament is fixed correctly along the axis of the bulb. In one, and not by any means the cheapest, I found that the filament was not only badly out of centre but was inclined to the axis of the bulb at an angle of some five degrees. It would, of course, be quite impossible to focus this bulb in a lamp with any degree of accuracy; therefore it is well worth while to look out for this point when making a purchase. Measuring Candle-power. I began my tests by selecting a well-made bulb and mounting it so that it could be controlled by a very accurate voltmeter and a rheostat. The candle power of this bulb was then measured by comparison with a controlled substandard from the National Physical Laboratory and was used as a standard in all subsequent measurements. Before we discuss results it is desirable that we should be quite clear on what is meant by candle-power. The term, without some qualification, is very ambiguous, as the intensity of light emitted by most lamps varies considerably according to the direction in which it is measured. Thus we have the M.H.C.P., or mean horizontal candle-power, which is the mean value for all directions in a horizontal plane, but the more common term is the M.S.C.P., or mean spherical candle-power, which is the mean candle-power emitted in all directions. This value is always considerably less than the maximum value. For example, in one bulb I measured the maximum emission in the horizontal plane was 41 candle-power, while the M.H.C.P. was only 34 and the M.S.C.P. was as low as 13½. All the references to candle-power in what follows are actual candle-power in a specific direction; no mean values are given. The bulbs I have examined may be divided into three groups: (1) very cheap ones, which have a filament in the shape of a V with the point directed forwards; (2) those of the standard pattern with a straight filament supported in the axis of the bulb; and (3) a special bulb with looped filament which gave the most interesting results in the photometer tests. The disadvantage of all filament bulbs lies in the fact that the filament is not a geometrical point, but the more it deviates from a point the greater the difficulty of focusing it in a parabolic reflector. Light Transmission. Testing a bulb of standard pattern, I found that its transmission in a direction at right angles to the axis of the filament was 42 candle-power, while in a direction exactly in line with the filament, it fell to three candle-power. Fig. 1 shows a graph of the results obtained by measuring the candle-power at intervals of a few degrees. It will be seen from this that the power of the bulb lies in its “sideways” light, while it is very weak in its “straight-ahead” light. When such a bulb as this is placed in a parabolic reflector it is easy to see that the bulk of its light falls on the sides of the reflector, while there is very little direct light entering the beam at all. Obviously, as this is the case, the importance of correct focus is greater than it would be if the filament more nearly approximated a point. May I suggest to those concerned that these lamp bulbs would provide a light more in accordance with the needs of motorists if the filament, now placed in line with the axis of the bulb, were fixed in a vertical direction at right angles to the axis? This would give us a beam of light which was strong in its vertical but weak in its vertical distribution. Most motorists will, I think, agree that we need more of a fan-shaped beam, especially when we are turning corners, and while we definitely do not want the very flat-topped beams associated with certain anti-dazzle devices, we could often do with less light 6 ft. above the ground if we could have it where it would be of more use. In my study of this subject I have tested (1) lamp bulbs alone, (2) bulbs in parabolic reflectors but without diffusing glasses and (3) complete headlamps. Test No. 2 gave the most startling results. I found it quite impossible to get a really satisfactory focus, owing to the size of the filament. The result was that the beam was very patchy, portions of it being extremely bright with relatively dark portions near by.{R.W. Bailey - Chief Engineer} I came to the conclusion that the function of the diffusing glasses now almost invariably used is to even up the beam. In my opinion the solution of the motorists’ problems of road illumination will not be reached until more attention is paid to the production of a bulb in which the filament more nearly approaches a point. In this connection I was very much interested in a bulb designed by Mr. W. H.{Arthur M. Hanbury - Head Complaints} Lund, the patent of which is owned by the Safe-beam Lamp, Ltd. It is made by the B.T.H. concern and marketed by Desmo, Ltd., of Birmingham. This bulb was designed to meet the demands of a special optical system incorporated in the Safebeam Antidazzle Road Lamp, but the light given by the naked bulb gave very interesting test results. The bulb is of the normal size, but the conductors carrying the filament are parallel and lie in a horizontal plane. The coiled filament is hung in a small loop from the ends of the conductors and the bottom of the loop is arranged to fall on the axis of the bulb. Thus it is easy to place the loop so that it is nowhere very far removed from the focus of the reflector and as the hottest—and therefore brightest—point of the loop is at the bottom the efficiency of the focus is further increased by placing this point at the focus. A test on the naked bulb showed an almost uniform candle-power of from 65 to 70 distributed over an angle of 170 degrees in the horizontal plane and over an angle 240 degrees in the vertical plane. I found no other bulb with a record comparable with this. I might add that during this test the lamp was taking slightly under 3 amperes at 12 volts, and as it is rated at 36 watts and would doubtless take its full load on a 12-volt battery which was “on charge,” as it would be under driving conditions, these results appear to me to be quite outstanding. A Special Reflecting System. I then placed on the photometer bench a Desmo Senior Safebeam Lamp complete. This lamp, also of Mr. W. H.{Arthur M. Hanbury - Head Complaints} Lund’s design, has a modified parabolic reflector with a portion of a conical reflector superimposed on the forward portion of the upper half. The reflecting system is developed on sound optical principles. The object aimed at in the lamp has been to produce a beam of light which, instead of having the common cylindrical form, would have a section similar to a broad triangle with the corners rounded off. In other words, the designer has attempted to concentrate all available light where it is most needed. The width of the beam at eye level is sufficient—but not more than sufficient—for driving purposes, while it widens out at lower levels until at ground level it more than covers the whole width of a normal road. In addition to giving the driver a better light the lamp has the advantage of being of a truly anti-dazzle nature. Claims Substantiated. That the designer’s aims have been achieved is amply proved by the polar graphs shown in Fig. 2. The graph A shows the distribution of light in a horizontal plane along the axis of the lamp, and B is a similar graph referring to the vertical plane. Note the wealth of light which is thrown on to the road surface. One point, a small one maybe, which very much pleased me in this lamp was the fact that the bulb holder is carried in a simple frame, which can be adjusted by a screw at the back of the lamp. It is a perfectly simple matter to focus this lamp, one-handed with no other tool than a coin. It may be recalled that I pleaded for this concession in a recent article on parabolic reflectors. A fairly lengthy road test of the Safebeam lamp has of 230 degrees in the horizontal plane and over an angle 290 degrees in the vertical plane. I found no other lamp with a record comparable with this. I might add that during this test the lamp was taking slightly under 3 amperes at 12 volts, and as it is rated at 36 watts and would doubtless take its full load on a 12-volt battery which was “on charge,” as it would be under driving conditions, these results appear to me to be quite outstanding. Image Captions: [Image of Car Headlamp]: A Desmo Safe-beam lamp mounted for test purposes just above the near-side dumbiron of an Austin Twelve-Six car. [Fig. 1]: Fig. 1. — A polar graph showing the variation in candle-power of a head-lamp bulb when measured at intervals of a few degrees from “straight ahead” to “side-ways.” [Fig. 2]: Fig. 2.—(Above) This polar graph shows the distribution of light in the Desmo Safebeam lamp in horizontal and vertical planes as explained in the text. (Right, top) A cheap 'V' filament bulb. (Centre) The standard type. (Bottom) Dr Stuart's suggested modification. | ||