My dear Dereelight CL1H, a new way of presenting output

If we consider the light emitted from a flashlight to come from a point source, the light flux (Lux) will be inversely proportional to the square of the distance from the light source. This means that if you measure the flux in one direction, you can based on this model calculate the flux at any distance in this direction.

[FONT=&quot]I have earlier presented measurements of the light flux along at a certain distance (1 m) from the flashlight lens. This light flux can be integrated to give the total light output in lumens. Earlier, I measured this light flux as good as I was able to at 1 m distance. For lights with a very concentrated hot spot, this becomes quite difficult, and also it is a problem that my light sensor becomes quite big compared to the light flux gradients. Now I just use the concept from first paragraph, and can then [/FONT][FONT=&quot]measure the light at any distance, and calculate back to the flux at 1 m. This means that I can use a distance of several meter, and get a good measurement of the spot for instance in my Dereelight DBS. After that, I can measure the spill in a much shorter distance, and combine. The result measured and calculated @ 1 m distance, for my Dereelight CL1H 3SD Q5 SMO looks like this
[/FONT]

This is as I can see it, a good way of presenting the light output from a flashlight. You get both the lumens, the max light flux (Lux) and also the distribution.

Since we here have the light distribution out from center, we can calculate the distance to a certain light flux in any direction. This makes it possible to present the area where the light flux is above a certain level. This looks like this:

Here I have used 1 Lux as one limit. This is may be a limit for usable light (about moonlight). As you can see my CL1H gives this light flux just above 100 m. I have also presented 10 Lux which is a quite good light level.

What do you think about this way of presenting the light output?
If any interest, I can also present the output from my DBS in a similar way.

Regards from
Jens

That is an interesting way of presenting the light output. I'd be interested in seeing the DBS graph. Do you have the smooth reflector or the OP reflector in your DBS?

I have both SMO and OP reflector both for my CL1H and DBS. So far I have only measured the light output from the SMO reflectors.

The DBS has such a small hot spot at 1 m distance, that it would be very difficult to measure the gradients at this distance. I have therefor measured the light in the spot at several distances and calculated back to 1 m. When I directly measure at 1 m, I get a max light flux of about 26000 Lux. When I measure at 2 or more meters and calculate back 1 m, I get about 31000 Lux, quite independent on the distance. This is the most representative number regarding the throw.

The output from my DBS will follow shortly.

Here is the results for the DBS, 3SD Q5 SMO:



As you can see, also some lines from the CL1H is included in the last picture for comparison.

I thought lumens was a measure of total light output, I don't understand what the lumens graph is for?

Lumens is the illuminance (lux) times an area, and gives as you say, the total light output. What i have done to get the lumens curve, is to start at the center and integrate the illuminance (lux) times the delta area which is the delta radius multiplied with 2*pi*r. I have assumed a linear variation of the illuminance between my measuring points just like shown in the curves.

The lumens curve will therefore start at zero in the center (no area) and grow as we move outwards. At the end the curve will show the total light output in lumens.

If you want, you can go in and check how much lumens is output between certain angles. If we take the chart of the DBS as the example: We can maybe define the spot as the area with an illuminance above 2500 lux @1m (corresponds very well to what I would define as the spot by looking at the light beam) and there is also a rapid decline in illuminance at this point. You can then look at the lumens curve at this point. It shows just above 100 lumens. The total light output from this pill in the DBS is just above 200 lumens. This means that my DBS puts out about half of the light in the spot and half of the light in the spill.

If we take my CL1H with SMO reflector, it gives in total a little higher total output, but it puts out less than a third in the spot, and the spot is also bigger.

I hope this was understandable, and also that I was able to demonstrate som interesting points by this way of doing it.

Jens

I'd like to see this type of graph with a few different beam patterns on it, its a bit techie but it might be a great way to compare different lights for group tests e.g. the thrower group test.

It would also answer questions about which is better SSC P4 or XR-E or maybe even SSC P7 vs MC-E etc.

Thank you for your interest. I have several flashlights with very different beam profiles. I find this an interesting way of comparing lights. At this stage, I can not help you so much with comparing different leds, as almost all my flashlights have CREE Q... leds (I have one with a RB100).
I will post comparisons of my different lights, but I think I have to do it in a separate tread, as the title of this one will be misleading.

Those are some really interesting graphs- congratulations

I don't understand the lumens graph at all. If I look at the graph it shows lumens rising as a function of distance. The lumens (total flux or lux output) of a light is a constant representing the source and not a function of distances. It is a finite number, that certainly does not grow to and level off at some distance and again is not zero at the point source. So I would say that the attemp to represent lumens on this chart is simply not correct. Even if measured incorrectly a light that has 200 lumens at the emitter has 200 lumens output, period. How it gets spread out is always the point in question.

I see vendor's of light sources and optics publish graphs of the beam shape, your chart I simply don't understand what the real point is. Lumens can be measured accurately, Lux at a know distance which tells us the intensity of the hot spot can be measured directly.

If you stood 10 meters out from the light and from zero angle you started moving in an arc pattern (maintaining 10 meters using a string) and measured the light as you came off angle and plotted that into a chart, that would tell you something, then you could compare the beam patterns and intensity of various lights. You could do that every 1, 2, 3, 4, 5 degrees, whatever consistent number suited your fancy that would fill in the curve. Why the Lumens numbers are plotted as they are is a mystery to me.

MrGman said:

If you stood 10 meters out from the light and from zero angle you started moving in an arc pattern (maintaining 10 meters using a string) and measured the light as you came off angle and plotted that into a chart, that would tell you something, then you could compare the beam patterns and intensity of various lights. You could do that every 1, 2, 3, 4, 5 degrees, whatever consistent number suited your fancy that would fill in the curve. Why the Lumens numbers are plotted as they are is a mystery to me.

Click to expand...

That is exactly what he is doing, except instead of using (the radial) angle he is using distance from the center (a.k.a. the radial distance) in meters. He made the radial measurements at a fixed distance of 1 meter from the source. Now, since the flux falls as (1/distance squared), he can extrapolate the beam profile to any distance from the source and that is what he shows in the second plot. Don't confuse distance (which I bolded and changed to red) with radial distance. I do not know what the more accurate term is for the distance from the source. (Maybe axial distance?)

MrGman said:

I don't understand the lumens graph at all. If I look at the graph it shows lumens rising as a function of distance. The lumens (total flux or lux output) of a light is a constant representing the source and not a function of distances. It is a finite number, that certainly does not grow to and level off at some distance and again is not zero at the point source. So I would say that the attemp to represent lumens on this chart is simply not correct. Even if measured incorrectly a light that has 200 lumens at the emitter has 200 lumens output, period. How it gets spread out is always the point in question.

Click to expand...

You are correct, the total output of a light source is a single finite lumen value. What his chart is showing, is a method to calculate the total lumen output. If you integrate the lux as a function of radial distance you get the total lumen output of the light. It is really no different in principle from an integrating sphere.


In a IS, the light is bounced around to make it diffuse, a lux measurement is then taken at a known distance of a known area of the sphere and then the total Lumen output is calculated.

RocketTomato said:

You are correct, the total output of a light source is a single finite lumen value. What his chart is showing, is a method to calculate the total lumen output. If you integrate the lux as a function of radial distance you get the total lumen output of the light. It is really no different in principle from an integrating sphere.

Click to expand...

Yes, the lumen value is neither infinite nor rising infinitely at some slope, it is rising asymptotically to the total lumen value.

I don't understand your graphs at all.

MY CL1H V3 3SD R2 just puts out quite a bit of light in a fairly tight but SOFT edged spot and very useful spill!

It is not the end of my search for a back pocket 2x123 or 1x18650, but is a dang good light in any case...

And so what happens if your beam is offcenter or is not a circular pattern? Your results could vary wildly just by rotating the light 90 degrees. If your sampling points are not high enough you would inaccurately model dark rings, artifacts, or sharp changes in light output which could throw off your measurement a significant amount. Your method works ok as a concept but I don't see it being useful in the real world where 99% of flashlights don't have the perfectly smooth, circular, centered, and blemish free beam patterns that this method would require in order to get consistent measurements.

Thanks a lot RocketTomato you have understood, and explained well.:twothumbs

I don't understand your graphs at all.

MY CL1H V3 3SD R2 just puts out quite a bit of light in a fairly tight but SOFT edged spot and very useful spill!

Click to expand...

With my OP reflector I get quite soft edges. Here I have presented the light distribution from the SMO reflector which have sharper edges.

And so what happens if your beam is offcenter or is not a circular pattern? Your results could vary wildly just by rotating the light 90 degrees. If your sampling points are not high enough you would inaccurately model dark rings, artifacts, or sharp changes in light output which could throw off your measurement a significant amount. Your method works ok as a concept but I don't see it being useful in the real world where 99% of flashlights don't have the perfectly smooth, circular, centered, and blemish free beam patterns that this method would require in order to get consistent measurements.

Click to expand...

I measure the light in several directions out from center, and use a mean value to represent the presented value. I also judge the beam pattern, and try to capture the gradients that I can see (or measures). For most lights, I also measure the beam pattern at several distances as I have explained in first post, and compare the results. Most of my lights are quite symmetrical. As I have explained, it is possible to compensate for a non symmetrical light picture regarding the lumens calculations, but the lux curve and the light throw distance figures would of course become more misleading.
In the figure below, I show measurements for my DBS, made at three different distances (1.2, 2.8, and 3,64 m) from the lense of the light, and calculated back to 1 m as described. The last light blue curve is the one I have choosen to represent my DBS.


The horizontal axis is m and the vertical lux.

The lumens calculations will be quite close no matter which one of the curves I choose (in the region where the curve is represented.)

I hope this gave a little more confidence.

Jenskh, could you overlay a beam shot with your graph of lux vs. radial distance? I think it would make it clearer to everyone what you are actually measuring.

Your multiple measurements overlap quite nicely and it is definitely a good idea to average over a few radial gradients.

Do you have a Malkoff M60 and a M60F? It would be interesting to compare the two to further show the utility of your technique.

Thanks for the advice. I will try to compose a picture like you describe, but I have not had much time to work on it lately.

Unfortunately I do not have any Malkoff modules. I would like to, but the main reason is that I almost only use rechargeable batteries (18650) and I have understood that the Malkoff modules uses 2x CR123 batteries.