Oil Headlights - Sourcing Information on Light Output

[Dante Lakin]

Hello all. As of late, I've fallen into a rabbit hole of oil headlights, but even in spite of digging out expired patents and reading period catalogs such as from the Star Headlight and Lantern Co. from 1915, I'm having trouble finding accurate and cited sources about how bright the headlights get at the burners. The Star catalog linked in this post references some states having requirements of headlights able to "distinguish an object the size of a man at a distance of 600ft." But that doesn't tell me how bright the lamp is at the burner, nor what the criteria is for discerning a man at 600ft is. Modern FRA code is very clear about candela requirements at specific distances and angles, but I cannot prove an oil lamp will meet those requirements with the sources I have for peak brightness based on burner/fuel type. Most sources I have are for kerosene lanterns which use flat wicks, whereas headlights of the day tended to use hollow round wicks with a center draft arrangement. Wikipedia has a table comparing peak brightnesses in candlepower, watts and lumens but doesn't cite any source so I am hesitant to use those figures.
If anyone has technical journals or documents, both modern and historic that can provide any more information about oil headlight light output they would be greatly appreciated.

[Benjamin Richards]

What is the goal? Historical curiosity, or using a vintage headlight in service?

The easiest way to demonstrate compliance is to take measurements of a working unit with a lux meter. This skips all the theoretical stuff.

[Dante Lakin]

Using them in service, but I want to at least know I'm in the ballpark for compliance before diving too deep into the "real world" side of things.

[Dave Crow]

Dante,

Try a quick search on Narrow Gauge Discussion Forum; there was a recent discussion about oil headlights vs. electric bulbs, including at least one post by Dan Markoff, the owner of the wood-burning locomotive, Eureka.

Dave Crow

[Dante Lakin]

I believe I found the thread you're talking about. Skimming through it quickly it doesn't immediately show what I'm looking for, but it is interesting to hear about how Eureka's headlight was once lit by the sunlight.

Maybe one day if we have a properly functional and fine tuned oil headlight we can try to replicate that experience

[Bruce Wilson]

Hello Dante. Another interesting subject you are pursuing. I imagine there were court cases involving teamsters or motorists who claimed not to hear a locomotive whistle or see a headlight in time to clear a crossing. Back in the 1970's, I had a few copies of circa 1930's I.C.C. illustrated reports of locomotive boiler explosions. As I recall a case by case study of each explosion was listed. I can't help but wonder if a similar record of headlight deficiencies or failures might have been looked into. Hope this reference to the I.C.C. is helpful to you.

[Dante Lakin]

Thank you for the suggestion, Bruce. I wasn't able to search through the ICC reports as thoroughly as I would've liked. However I did find an answer from a different period technical magazine for enginemen, "Locomotive catechism; a practical and complete work on the locomotive" from 1911. 75 Candlepower peak output (942.75 lumens), which if my calculations are correct show a theoretical output at 800ft as approximately .01 lux, comparable to the light of a quarter-moon on a clear night. This is technically illuminating a man-sized object at the required distance but is so insignificant as to be practically ineffective.

However, the FRA guidelines don't mention lux, and only define required candela at specific angles which I'm assuming is in steradians. If this assumption is correct and I'm not confusing units of light measurement, then the following table shows approximate numbers for Candela at specific angles, calculated as (candlepower*12.57)/steradians

CP |	Lux   |	CD at 7.5d |	CD at 20d
40 |	502.8|	32850        |	5267
75 |	942.75|	70078        |	9876

Now, this hinges on the assumption that the old document is correct in the light output, but any light source with a lumen output of >300 or candlepower >~25 will meet pure candela requirements

[Bruce Wilson]

Great work at finding that information. Sure makes me scratch my head as I struggle with a poor understanding of my high school math!

A few other thoughts. I wonder if you have access to a set of the old Audells publications? There may be some information within one of those technical publications. And lastly, a gentleman by the last name of Dickey made the headlight for locomotive no. 9. As I recall, he worked at the MEC shops in Waterville. He may have passed away, but I believe his son had (or has) a membership in the W.W. & F. Ry. Museum. I'm not sure what craft Mr. Dickey was kept busy at, but he or his son might be able to help you. Good luck.

[Benjamin Richards]

Thank you for the suggestion, Bruce. I wasn't able to search through the ICC reports as thoroughly as I would've liked. However I did find an answer from a different period technical magazine for enginemen, "Locomotive catechism; a practical and complete work on the locomotive" from 1911. 75 Candlepower peak output (942.75 lumens), which if my calculations are correct show a theoretical output at 800ft as approximately .01 lux, comparable to the light of a quarter-moon on a clear night. This is technically illuminating a man-sized object at the required distance but is so insignificant as to be practically ineffective.

However, the FRA guidelines don't mention lux, and only define required candela at specific angles which I'm assuming is in steradians. If this assumption is correct and I'm not confusing units of light measurement, then the following table shows approximate numbers for Candela at specific angles, calculated as (candlepower*12.57)/steradians

CP |	Lux   |	CD at 7.5d |	CD at 20d
40 |	502.8|	32850        |	5267
75 |	942.75|	70078        |	9876

Now, this hinges on the assumption that the old document is correct in the light output, but any light source with a lumen output of >300 or candlepower >~25 will meet pure candela requirements

We have to be careful here. The FRA requirements call out candela values measured at a specific angle from centerline. This is not the same as simply taking lumens divided by said angles. A perfect point light source placed at the focal point of a perfect parabolic reflector produces a cylindrical beam of relatively constant intensity (candela), with everything outside the beam dropping immediately to zero. Introducing a non-point light source creates some beam spread with non-constant intensity, as does moving the light source off the focal point. This is how maglite flashlights work.

In layman's terms, the regulations want a light source which deviates from the ideal parabolic model in some specific way, such that most of the beam is on the tracks, but some of it does spill onto the sides. This is why I suggested field measurements as the most practical path to compliance. Attempting to model the luminous characteristics of an oil flame well enough to accurately predict the real-world behavior seems daunting, and I have formal training in this stuff.

You might get away with approximating a tungsten filament as a point source. Probably not an oil flame.

As an aside, you can at least get some bounds on things by assuming a perfect model. I need to sit down and go through this rigorously to provide more details, though.

[Dante Lakin]

Introducing a non-point light source creates some beam spread with non-constant intensity, as does moving the light source off the focal point.

Very true. In my research I've found the common diameter of round wicks for headlights is generally 1 1/2", so the light output would be all ~3/4" out of focus.

[Dante Lakin]

Some more notes on this topic

A. This image is what started this whole search. Our #9 in a past life as SR&RL #6, with a round oil headlight. Picnic basket on the running board would make a nice touch for a future event, too


B. The headlight most suitable for what I'm trying to recreate is marked as being from WW&F 3, however it doesn't match historic pictures I've seen so far of #3 with a round case oil headlight. The builder's plate for the headlight as well identifies it as a Glazier headlight, which according to brief research merged into Star in 1890.

C. It is in rough condition, and the work needed to repair and modify it into a representation of what was on SR&RL 6 would destroy the main body of the headlight as it sits now. With it being from the 1880s or earlier and with enough scars from a working life, it almost goes against the idea of preservation to restore this one as would be required for operational purposes.

Now, it stands at approximately the right height, depth and width to stand in for the pictured headlight, but so many details are noticeably incorrect (lack of number boards, different chimney case profile, straight base instead of curved, etc). My question pivots to this, does anybody have access to blueprints for either Glazier headlights or Star headlights? Star classifies this style of headlight as a "Wabash Style" and I believe the size I'm looking for is 16". In the previously linked catalog, it would be Style 22.

[Philip Marshall]

As an aside, that's a great picture I don't recall seeing before. This is a consist that could be recreated, because in addition to SR&RL No. 6, both passenger cars are also currently present at Sheepscot: combine 14 (shown with its original clerestory roof) and of course the Rangeley. (But what was the Rangeley doing in Bigelow?)

[Dante Lakin]

This is a consist that could be recreated, because in addition to SR&RL No. 6, both passenger cars are also currently present at Sheepscot: combine 14 (shown with its original clerestory roof) and of course the Rangeley.

That's another reason I was so drawn to this image. It would be so easy to set up at Alna Center if the stars align, but The Rangeley being well, The Rangeley makes it difficult. We can dream though

[Keith Taylor]

I am not convinced that the headlamp shown lamp in the above photo is an “oil” lamp.
Whe the locmotive was buillt it hhad a box shhaped oil lamp. Why would they upgrade to just another oil lamp when by that time acetylene lamps were avialable?
Keith
2

[Dante Lakin]

Keith, from my understanding, it seems kerosene was really the preferred fuel for headlights until electric really took over. Even in the 1915 Star catalog linked in the initial post, kerosene was still the standard with options for acetylene or electric.

I believe the switch from the box headlights to the round case was a matter of trying to match newer equipment. Builders' photos of the Eustis locomotives show the same style of headlight as the picture of #6. Later still, most SR&RL engines have the same headlight style that 9 wears now. It seems that the Phillips shop boys enjoyed having at least some standardization on their operating fleet.