Output of a camera flash in lumens?

StarHalo

Well-known member
Joined
Dec 4, 2007
Messages
10,932
Location
California Republic
I always thought that the Metz Potato Masher units using a fresnel lens did just that.

I've used the Metz 45 and 60 with the fresnel lens for night time baseball games.

That's a thousand dollar flash system though; this is a ~$75 kit you can slap on a ~$70 Yongnuo 560, presto, strobe thrower..
 

ssanasisredna

Well-known member
Joined
Oct 19, 2016
Messages
457
I know this is an old thread, but I've just been thinking about it... if flash pulse is 1/1000s but you only want to shoot at 1/250s that means you only need 1/4 the lumens using a continuous light than you would a conventional flash to get the same amount of light into the camera. So we only need 200,000 lumens? Getting closer....

And how can we not be completely and permanently blinded if a large conventional studio flash puts out 200megalumens?

Lumens does not blind us, lux does :) ... that and the total energy is low and damage mechanisms are typically heat related, so the low total energy can not cause scaring.

One thing of note, 1msec is being thrown around as the standard flash duration. That is true for studio flashes, but small flashes are often much faster. <100usec is not uncommon for the majority of the energy to be dissipated.

I would also take with a grain of salt some of the "tests" that are seen on the web. It's obvious from the pulse shape that they have response time limitations in their measurement equipment.
 

desirider

Active member
Joined
Apr 10, 2012
Messages
25
Lumens does not blind us, lux does :) ... that and the total energy is low and damage mechanisms are typically heat related, so the low total energy can not cause scaring.

One thing of note, 1msec is being thrown around as the standard flash duration. That is true for studio flashes, but small flashes are often much faster. <100usec is not uncommon for the majority of the energy to be dissipated.

I would also take with a grain of salt some of the "tests" that are seen on the web. It's obvious from the pulse shape that they have response time limitations in their measurement equipment.

Andy Gock has done some good measurements: https://gock.net/blog/2012/01/flash-durations-small-strobes/. His photodiode response time is about 5 microseconds for a 6mA collector current.

Duration of speedlights at full power is about 4ms (t.1). Approximating the output waveform of a 75Ws flash as a right-angled triangle with a base of 4ms, the height would be 37500 Watts (37500W*4ms/2=75Ws). Assuming 40lm/W, the peak lumen output would be 37500*40 = 1.5 million lumens.

Desirider.
 

mattheww50

Well-known member
Joined
Jun 24, 2003
Messages
1,048
Location
SW Pennsylvania
I can assure you that those little hot shoe flashes are a whole lot less than 50-100 watt seconds. The big Honeywell units like the 660 and 770 with a KII guide number of 80 were 60 watt seconds, and could literally burn the print off the paper. The little hot shoe units are not anywhere near as powerful. Upper limit is probably about 20 watt seconds. I don't think I have ever seen a 100 watt second unit that is portable. There are some hot shoe units today with kind of power you describe, but you will pay about $500 for one of the big Canon or Nikon units, and they are not exactly tiny
 

Bazar

Well-known member
Joined
Dec 2, 2016
Messages
140
Location
Colorado
OK, pocket cameras.

The Nikon SB800 has a guide number (in meters) of 125 at ISO 100 with its zoom head set to telephoto (105mm coverage). That's a "beam angle" of 20x27 degrees (http://www.scantips.com/speed2.html) and the flash duration is 1/1050th second (same link). 20x27 degrees is about .35x.47 radians or .35*.47/(4*pi) steradians (is that right?) or 0.013 steradians.

The 125 guide number means at 1 meter you'd use f/125 (close to f/128) so this is 4x the exposure of f/64, the
However, because of that 20x27 mm narrow angle, that light is going over just 0.16 or so square meters. So spreading it out to 1 square meter (remember lumens = lux/square meter) we get 9 million * 0.16 or about 1.4 million lumens!

Remember that this is a fairly hefty 4AA-powered flashlight which will give you maybe 100 flashes on full power per set of batteries, or a total runtime of around 1/10th of a second. If you dialed the power back by a factor of 10000 so you were getting just 143 lumens (a more normal 4aa or 2x123 flashlight) the resulting 17 minute (10000 * 1/10th sec) runtime wouldn't be all that impressive, so the efficiency of these flashes is not that high. Of course you can't really dial back the power like that and get the same efficiency, that was just for comparison.

Cool question! Anyone up for checking the answer please do so, I could have easily made a mistake. It was fun researching this.

560 square millimeters is significantly smaller than 1,000,000 square meters, or 16% which is roughly 160,000 square millimeters, not 20x27. Thus making your calculation way less than 6, million lumens.
 

Bazar

Well-known member
Joined
Dec 2, 2016
Messages
140
Location
Colorado
Lumens does not blind us, lux does :) ... that and the total energy is low and damage mechanisms are typically heat related, so the low total energy can not cause scaring.

One thing of note, 1msec is being thrown around as the standard flash duration. That is true for studio flashes, but small flashes are often much faster. <100usec is not uncommon for the majority of the energy to be dissipated.

I would also take with a grain of salt some of the "tests" that are seen on the web. It's obvious from the pulse shape that they have response time limitations in their measurement equipment.

This is the embodiment of why flashes don't have high lumens. Flashes are imperceptible to eyes, and thus make them average out over half to a full second if not at least 10 htz, the frequency of brainwaves, reducing the average luminous output of flashes to fractions so low even 4-10k lumens looks much brighter.

I also don't see any real good solid beginning-to-end math here nor elsewhere on the internet, and I have zero qualifications to do one proof myself. I'd like to reiterate this question, how many lumens are in a flash, and see more answers, as flashlights may use LEDs to burst at outputs comparable to conventional gas powered dinosaur lighting soon, over next 10 years or so.

My question may be; if you counter a paparazzi with a powerful flashlight like the X80gt or even the K75 acebeam, can the light emitted by a flash be so overwhelming that you still pick up the target brighter than the flash from the flashlight?

If the answer is yes perhaps there is a shred of merit to claims that flashes are millions of lumens and lux, but if not, if facial details are lost and bodily orientation is lacking, I'd assume the lumens are actually lacking this ominous omnipotent power.
 
Top