circa 1875 locomotive headlights

[theswitchman]

I posted this question on Trains.com forum and received no replies, perhaps this is the right place.
Kerosene powered.
This a generic question. What is the candlepower of these family of headlights?
More specifically, one manufactured by the U.S.Headlight Co., Buffalo, NY somewhere between 1873 and 1905. We know that this particular headlight was onboard the V&T Empire between 1905 and 1920.

[mersenne6]

I think you may be asking for a level of precision which may not have been applied to the lamps of the day. The unit of candlepower was defined in 1860 and it was based on a measurement of the light produced by a pure spermaceti candle weighing one sixth of a pound, burning at a rate of 120 grams per hour. The vagaries of the production methods of the day insured that it would be a poorly reproducible measurement and it wasn’t until 1909 that attempts were made as standardization on the international level. Even with a standard candle and a calibrated light meter there is the issue of where to take a measurement. The value measured depends on where in the beam you take the measurement (the highest value found is what is normally used), and on how well the beam is focused. Because of this it is not unusual for candlepower values to vary greatly from unit to unit on otherwise identical lights.
Candlepower specifications are really only useful in comparing lights if the lumen values are also provided. If two lights had similar lumen values and one had a higher candlepower value it would mean is that one light has a more tightly focused beam.
From Trains and Technology we do have the fact that the “kerosin” lamps of the 1860’s were “expected to show in a dark night sufficient light to enable the time to be seen on a watch, the observer standing 1000 feet in front. Even at this range, it is unlikely that an engine running at speed could stop in time before striking any obstruction that might be revealed by the light (to the eye of the engineer).” The lamps themselves were not provided by the locomotive manufacturers – they were sold by independent suppliers to the railroads. Given the cautions listed above you would probably be better off trying to build a model of one of the lamps of the day (there are lots of blueprints available), make the measurements yourself, and extrapolate to what might have been.

[theswitchman]

mersenne6,
I have been continually researching this question all week. You have beautifully summed up what I have only been able to find bits and pieces. I saw somewhere where in 1905 the ICC required headlights to be able to illuminate a man at 1000 feet. This was for electric lights which were invented in 1881 I believe. No where do I find a candlepower, candela, lumen, etc. requirement. FYI, the original question was posed by a third grade student who was in a group that I was leading on a tour of the Nevada State Railroad Museum.
Thanks very much for your excellent reply.

[Anonymous]

QUOTE: Originally posted by theswitchman

mersenne6,


FYI, the original question was posed by a third grade student who was in a group that I was leading on a tour of the Nevada State Railroad Museum.
Thanks very much for your excellent reply.

I would say that this is a very bright (pun intended) [:)] young man.

This would be an interesting project for a railroad museum, having the locomotive headlights, to conduct.

[theswitchman]

I think that we are going to have to commence a Science Project in order to determine some approximation of candlepower for any given headlight.
We will not reveal the results of this test to the public, as they have not been too enthusiatic about helping out.

[mersenne6]

Actually, with blueprints in hand and with computer firepower being what it is today you could probably do some very nice simulations with respect to modeling old headlights. From the standpoint of a science fair project there would be a lot of reading of the basics of optics and mirrored surfaces. There would be the problem of computer coding, and there would be a need for investigating certain aspects of statistics with regard to random variation. All in all it would make for a very interesting project.