Ceiling bounce test accuracy

[Swedpat]

I have discussed this issue earlier and met as well positive and negative responses.
I believe in that shining a light upon the ceiling above a light meter is a very accurate method of measuring percentual differences of total light output. Though likely not completely as accurate as using an IS sphere it's very useful.

The basis for my belief of the ceiling bounce as a decent method is following:

*the total light amount reflexed from the ceiling is the same undependent of the distance I hold the flashlight from the ceiling. (this makes the demand of a clearly limited beam, so that no light spreads out to the walls)

*the received values of lux reduces with longer distances/wider beams, because of that the light from the periphery of beam is more distant from the sensor.

*according to the previous point I therefore always take in consider the beam width of different lights and hold, for example a Maglite, closer to the ceiling than a Fenix. Actually the differences of beam widths will be less important the closer to the ceiling I hold the light. I can also simply hold the light very close to the ceiling and note the brightest available value of the light meter by varying the position of the light.

*If holding different lights at that distance from the ceiling that the projected beam profiles have equal diameter and position above the sensor of the lightmeter (in my case laying on my livingroom table) is the same, the total light amount reaching the sensor has to be very accurate and comparable.

According to optical laws and that above mentioned requirements are fulfilled I can not find any reason that ceiling bounce test would not be a trustworty method to measure differences of the total light output of flashlights.

I welcome commentaries,

Regards, Patric

 

[Flying Turtle]

I always subject a new light to a ceiling bounce comparison to older lights. Even without a meter it seems to be a reasonably accurate way to get an idea how two lights compare in total output.

Geoff

 

[Big_Ed]

I use the ceiling bounce test with my lights. True, it's not as scientific as having numbers from a light meter or something, but my eyes are a more relevant, real-world way of determining brightness.

 

[Patriot]

Except the original poster is talking about using a light meter. ;)


Doing a ceiling bounce test without a light meter is really only useful for telling which light is brighter. If they're close or the tint is much different, even that can be difficult.

 

[milkshake]

i like using frosted light bulbs. place each flashlight to its own bulb fire up and compare. you can really tell the tint differences with the bulbs

 

[Light Sabre]

I don't have an integrating sphere, but I made what I call an integrating "T". It's a white 4" PVC T from the hardware store. I put 2 CDS photocells at the top of the T, one on each side. The flashlight goes into the bottom of the T. You can see the shadow of the photocell in the third picture.

I placed a stop on the flashlight side to that all flashlights inserted are the same distance from the back.

I got the idea for this I think from the guy that originally ran FlashlightReviews.com. I hookup the photocells to a DVM equipped with a RS-232 port and collect data with the program provided. I do runtime tests with this setup.

 

[Swedpat]

Thank you for your replies, and thanks Light Sabre for the pictures!

Earlier I used to compare lights by shining up to the ceiling, and understood that it was a good way to notice total output with my eyes. The eyes can be deceived by different light patterns and widths so therefore it's better to shine up and look down and see how bright the room will be.

But like Patriot mentions here about tints; I found that small brightness differences especially with different tints made me unsure.
I then got the tip of a lightmeter. Using this I can see the result undependent of beam tints, and I also can read the percentual difference.

Our eyes sometimes can be deceived and have hard to estimate what is twice as bright, or what is 4 times, and so on. If the brightness difference is huge, we can hardly estimate if the difference is 20 or 100 times.
Therefore I like the lightmeter.

Regards, Patric

 

[Tom_123]

Well ceiling bouncing is basically just like measurement in a very huge light box.
Of course, you will have many factors that distort the reading but I think it is
sufficient if you want just know if light A is brighter than light B.

On the other hand, if one already got a light meter, it isn’t that much of a hassle
to build a light box or even an integrated sphere to achieve more accuracy.
E.g. to build my light box took me just 2h of work with no cost at all.

Another way to estimate output without a meter might be to build a little scene on a table,
light it up with the lights bounced from the ceiling and take photos
with the same camera settings for the different lights.
I didn’t tried this out yet so I’m not sure if it would work as expected.

regards
Thomas

 

[Big_Ed]

Except the original poster is talking about using a light meter. ;)


Doing a ceiling bounce test without a light meter is really only useful for telling which light is brighter. If they're close or the tint is much different, even that can be difficult.

Oooops! I guess I should have read the original post more carefully.

 

[GTRmiami]

How to do an accurate ceiling bounce comparison between several lights?

Not quite sure I'm doing this right. I have a very basic lux meter app on my phone and am curious to compare the total light output between several lights. So far the only thing I've done is shut the bathroom door, put the phone on the counter, turn on the lux meter app, tailstand the flashlight right next to the phone, and turn it on.

What I'm not sure about is how the bathroom mirror might affect the results or if the results might be inaccurate if some lights have a tighter beam than others and reflect a brighter hot spot back to the phone.

If the mirror has any effect, I can simply move to another room, but if the brightness of the hot spot matters, I'm not sure how to remove that variable so I can get a comparison of the total light output of the lights.

Any suggestions?

 

[Bullzeyebill]

Re: How to do an accurate ceiling bounce comparison between several lights?

If you are aways using the same place to put your lightmeter and flashlight, and you do not change the enviornment in your bathroom you should be ok. I've used my darkest bathroom, small and with a large mirror, to do bouce with a light meter, and have had no problems. Probably best to compare lights with the same or similar reflector/optic.

Bill

 

[reppans]

How to do an accurate ceiling bounce comparison between several lights?

I found bouncing the light horizontally between two walls in a narrow hallway to work better then the ceiling bounce. It's easier to match the total beam sizes between floody and throwy lights, and closing the distance the light travels means brighter samples for insensitive light meters.

I also prefer camera light meter apps (ISO, aperture, shutter speed, EV) to the lux meter apps for the greater granularity offered. Calibrate to and solve for shutter speed and you have you lumen conversion right there. There's also an easy formula to convert EV to lumens for more granularity.