Flashlight PWM vs Camera Shutter interactions, An Experiment.

[jeff51]

Are Video Reviews of flashlights a good indicator of Pulse Wave Modulation artifacts?
I sometimes see a brief zap of PWM alien death beams in a video review.
Sometimes the reviewer mentions them. Often saying they are not bothersome.
And then followed by "Wow, look at that output, Amazing!"

I decided to take a look at the phenomena and see if I could draw some conclusions.

A little background
PWM is a commonly used method to control brightness. This is done by turning the light on and off quickly. Usually, but not always, at a fixed frequency.
Then the amount of time the light spends in the "on" part of the cycle is varied to change the brightness. This is known as the Duty Cycle.
Hopefully this takes place so fast that it is not noticeable to the user.

Video cameras control exposure by changing the aperture, ISO, and, shutter speed.
The first two are don't matter, except how changing them is related to the shutter speed.
The video frame rate is usually fixed. Typically it is 30 frames per second.
The shutter speed along with the frame rate interacting with moving objects produces visual oddities. Such as when looking at a spinning wheel, at some speeds, appears to freeze or spin backwards.

Adding the On-Off nature of a flashlight PWM into the mix can result in the camera capturing some frames where the light is on and some when it its not.
This shows up as a series of light and dark bands that move across the video frame.
"Alien Death Beams."

Some PWM controlled lights use frequencies and Duty Cycles that can be seen with naked eyeballs.
If poorly executed this can lead to eyestrain, headaches, even nausea.

So, my question is:
Is the absence of PWM artifacts in a video review a good indicator that a light has no noticeable PWM to the naked eye?

This video will look at various flashlight PWMs as they interact with the camera shutter.


The conclusion I reach is that a video review of a flashlight is not a good indicator of the presence of PWM.
Depending on the camera settings and the PWM characteristics of the light, noticeable PWM can be hidden in the video.
Conversely, a light with acceptable PWM characteristics can sometimes show PWM artifacts in a video.

PWM can be measured using a sensor and an oscilloscope.
But at what point does the PWM become noticeable, or more importantly, objectionable?

A moving object is needed to be able to detect PWM both in video and with bare eyeballs.
I will be doing some experimenting to see if I can come up with a good method of doing this sort of test.

Thanks for watching.
All the Best,
Jeff

 

[bykfixer]

I just aim the light at an object like water drops in a shower or fan blades and snap a still shot for definite proof. If you think you see it while "filming" the photo will confirm.
My favorite is a ceiling fan. Snap the photo with a shutter fast enough to give the sense of motion in the blades but still fast enough to see the blades.

Here's a light I knew for certain uses PWM aimed at my ceiling fan.

 

[archimedes]

.... at what point does the PWM become noticeable, or more importantly, objectionable? ....

A brief search here on CPF will reveal multiple threads addressing this particular topic, some quite exhaustively.

Individual perception and tolerance of this phenomenon varies enormously.

 

[jeff51]

I just aim the light at an object like water drops in a shower or fan blades and snap a still shot for definite proof. If you think you see it while "filming" the photo will confirm.
My favorite is a ceiling fan. Snap the photo with a shutter fast enough to give the sense of motion in the blades but still fast enough to see the blades.
Here's a light I knew for certain uses PWM aimed at my ceiling fan.

Very true. Although at faster frequencies and larger Duty Cycles, it becomes difficult to discern in a photo.
Once I have a light in hand, it's easy to measure the PWM with some simple instruments and software.
A sensor, computer soundcard, and free software works quite well.
I did an entirely too long rambling post about that here:
https://www.candlepowerforums.com/vb/showthread.php?461093-Measure-PWM-using-a-Computer

Or, as you say, point it at a moving object.
I was not suggesting using a video to detect PWM.

What I was looking to investigate was lights as seen in video reviews.
I wanted to find out if the lack or presence of PWM artifacts in a flashlight video was indicative of the light itself having PWM.

Often a video review is the only look we get at a light before ordering it.
I have read that some lights light use surprisingly slow Pulse Wave Modulation frequencies.
I have heard HDS uses 454Hz, Malkoff 310Hz, and I think I also read that older Zebra lights were slow also
Not sure if this is a real memory or a senior moment memories creeping into my brain.

Poor PWM is just nasty for extended use IMHO.
For example that camping lantern with the 100Hz PWM. It is just fine to use at full brightness.
But on Low, with the short Duty Cycle, it is intolerably. Even reading, just moving my eyes across the page quickly leads to fatigue. Like playing a video game at 18fps refresh.
The designers should be forced to work under them for a few nights running. Keep the puke buckets handy.

The same with the cheap zoomie. At full bright it's fine. But on low any close detail work, or at moderate distances watching movement, or just eye movement through the field of view is just unpleasant after a while.

The Nitecore TUBE on low at 500Hz is fine for looking for dropped keys or a trip to the bathroom at night. I wouldn't want to use it as my only source of light for extended periods. Just bumping the brightness up a bit switches the PWM to 3000Hz. Which is much easier on the eyes.

Yet I have another $10 light that has a PWM of 13kHz and is perfectly fine to use at any brightness.
So price is not necessarily correlated with crappy PWM.

What I found was that seeing the Alien Death Beams (or not) is not a good indication as to the PWM character of the light.
Although if PWM artifacts are present in the video, further testing should be conducted to verify the usability of the light.

What I think I'm actually saying is that any video review should include shining a light at a moving object, at all the brightness levels, to see if any visible/objectionable PWM shows up.
Or better still, measure it and include that as part of the review.

All the Best,
Jeff