Constant output video lights

[bghazzal]

Hello all,

 

Looking for a set of video lights fitting my criteria, I've been bobbing around in a whirlpool of brands and specs for the past few days...

 

It's all good, but one important specification which is very difficult to pinpoint is constant output / power discharge for video lights, which requires a control circuit to make sure the light has a stable output.
If not, the light dims as battery drains, which means a steady loss of light power, rather than a timed, but constant output.

 

In practice, this is a rather crucial spec for shooting video, but also one which is generally left out of available documentation (an easy test is to discharge the lights in a tub while taking a picture, using same exposure settings, every 15 minutes, to check if brightness drops - but this requires having the lights at hand...)

 

 

To help future buyers, I'd like to try to sum up information avaialble per brand, so that we have comprehensive list of video light manufacturers (and eventually models) offering constant output video lights, and those that don't.
At first glance, constant output video lights are clearly quite rare - probably because most manufacturers are aiming to keep costs lower by skipping the control circuit, since constant output isn't a feature sought by many users, who simply consider it normal/inevitable for the light to dim as battery power diminishes...
 

Based on what I've gathered online, we can kind of make out the following (please correct or add details as required)

 

 

BRANDS OFFERING CONSTANT OUTPUT VIDEO LIGHTS

 

- Keldan

 

- Light and Motion (Sola)

 

- Sealife

 

- Kraken (output test data given in docs)

 

 

BRANDS OFFERING NON-CONSTANT OUTPUT VIDEO LIGHTS

 

- BigBlue

 

- SUPE

- Scubalamp

- Fotocore

- Divevolk

 

- Nitescuba

 

- Archon

 

- Orcatorch

 

- Xtar

 

- Tovatec

 

- XLight

 

- Almost all the non-branded / semi-branded direct import lights found on Amazon, Ali-Express, Ebay, Shopee, etc. (e.g. LetonPower, Thrustfire, Wurkkos etc)

 

 

TO BE CONFIRMED

 

- Backscatter (MW4300 for instance)

 

- Inon

 

- Weefine

 

- Dive-Adventurer

 

- DivePro

 

- FixNeo

 

- RGBlue

 

- iTorch

 

 

****

 

cheers

ben

 

 

Edited July 29 by bghazzal

[Barmaglot]

FWIW, Kraken and Weefine are the same product, just different badges.

[bghazzal]

49 minutes ago, Barmaglot said:

FWIW, Kraken and Weefine are the same product, just different badges.

 

While this was probably the case in the past, this doesn't actually seem to be the case anymore.
The factory issuing the products is probably the same (likely also covering the X-Adventurer brand) but the product range is quite different now.

 

Here is the Weefine video light range:

 

Smart focus

 

Solar flare

 

And here is the Kraken video light range:

 

Hydra and LTD

 

While some products are still similar - and design seems to indicate that they might be produced all by the same factory - the specs and features are quite different across the two brands (Kraken products are generally heavier, for instance, and Weefine leaning towards a higher price tier)

 

That said, if Kraken lights are constant output, I would imagine Weefine lights are as well, but couldn't find confirmation (Kraken specs are more detailed and precise).

Here's an example of Kraken lumen output test data:

Edited July 29 by bghazzal

[bghazzal]

 
 
As an update Kraken have confirmed that all their lights above 2500 lumens are all COB and are all constant current / output
 
On the Weefine/Kraken front, Kraken also specified that while they did share a factory with Weefine in the past, in the last several years Kraken has had its own factory and make their own lights, which would explain why the product ranges have now diverged so much.
 
As a side note, from looking at quite a few products in the range, Kraken now seems to offer well thought-out products with important features where you want them:  COB, constant output, decent battery design, and importantly tighter beam angles.
 
Aah, beam-angles...
Like constant output, video light beam angle is one of those things where a lack of customer education is letting manufacturers get away with a few cut corners... 

Online, most mid-range video light manufacturers (and their favorite influencer-mouthpieces, such is the twisted world we now live in...😅)  will tout their super "wide beam" (typically 120°, sometimes even wider...) as a positive feature (perfect for video! will cover a wider area!)  - when this is actually a by-product of poor reflector design / choices, and more often than not a way to lower costs...
 
Most people here on WaP will already be aware of this, but since this thread is about important video light specs left out or, in this case, misrepresented by manufacturers, here's a reminder:
for video - in most real world applications - you actually want a tighter beam (say 90°, 100° angle...) as this will simply give you more more light for a given lumen output.  

As often stated here, for the same lumen output a 90° beam angle will give roughly twice as much light falling on the subject than a 120° beam.
And with the typical 2-light setup used in UW videography, two 90° beam lights will actually provide full coverage for lens with less than a 130° field of view (which goes for action cams, rectilinear lenses and WWL/WACP type lenses etc...).
Meaning that for video lights, that super-duper "extra wide" beam is actually more a design defect than a feature... 😉

I would say constant output and beam angle are the two most commonly misunderstood / misrepresented video light specs.  

We could go even further and offer that beyond wonky-named manufacturers dreaming up fantasy-world lumen outputs for their wonky-buttoned lights, in the slightly more reputable world of the branded mid-range, it's probably this double-whammy of dimming-on-discharge + diffusion by wider beam angles which accounts for a lot - if not most - of the "this video light is not as bright as what it should be given stated lumens" complains often heard about mid-range products...

This is probably not what your pusher outlet's infomercial review or shill ambassador-influencer's predicable video will be discussing when you end up clicking on a that click trolling engagement-generating thumbnail showing an unavoidably puzzled facial expression, but hey, that's what prehistoric relics such as forums are for, right? 😎
 
Returning to constant current/discharge/output issue at hand, I've asked Weefine the same question, not finding the data online, and waiting for an answer. If anyone has info on this, please share.
 
cheers!
ben

 

Edited July 30 by bghazzal

[DreiFish]

On 7/29/2024 at 2:50 AM, bghazzal said:

 

In practice, this is a rather crucial spec for shooting video, but also one which is generally left out of available documentation (an easy test is to discharge the lights in a tub while taking a picture, using same exposure settings, every 15 minutes, to check if brightness drops - but this requires having the lights at hand...)

 

 

Careful with doing this test in air, as it may lead to more powerful video lights significantly overheating -- which may also lower lumen output I guess? It's a test best done in a water tank.

 

I have the Divepro G18 (18k lumen video lights). I'll see if I can test the light output and whether it decreases with battery discharge at a later date.

 

On 7/29/2024 at 10:07 PM, bghazzal said:

 
Most people here on WaP will already be aware of this, but since this thread is about important video light specs left out or, in this case, misrepresented by manufacturers, here's a reminder:
for video - in most real world applications - you actually want a tighter beam (say 90°, 100° angle...) as this will simply give you more more light for a given lumen output.  

As often stated here, for the same lumen output a 90° beam angle will give roughly twice as much light falling on the subject than a 120° beam.
And with the typical 2-light setup used in UW videography, two 90° beam lights will actually provide full coverage for lens with less than a 130° field of view (which goes for action cams, rectilinear lenses and WWL/WACP type lenses etc...).
 

 

 

Amen. 90 degree beam angle is very desirable. It's one of the few things that has left me scratching my head about Keldan's otherwise sound design choices -- why, oh why, did they go with a 120 degree in air / 110 degree in water beam angle?

 

Good information as usual, Ben. Thanks for this.

[bghazzal]

As an update, Weefine has confirmed that their Smart Focus light range does have a control circuit and is constant output / discharge.

They also indicated that Kraken lights were previously produced by Weefine, meaning that Kraken lights - which have now evolved into distinct, independently manufactured products - are based the same design.

I'm waiting for Weefine's test data for the Smart Focus lights.

Edited August 2 by bghazzal

[eocean-eu]

I think the specification of constant output is hard to find simply because most if not all LED video lights have constant current outputs due to the electronic needed to power up the LEDs and provide power dimming. This is what people into electronics call "led driver". In our battery-powered devices, the driver will make the best use of the batteries so that the output current stays constant. So the LED in opposition to the old filament bulbs is not expected to naturally dim. It might and will dim at some point because the electronic says so to protect batteries or extent battery time, or it will simply start to have a visible strobe effect or just shut off. 

 

I've played with some of those electronic modules in the past. Some have the ability to dim the light by lowerings output amps but often the dimming is achieved through "PWM" (Pulse Width Modulation). The LEDs is then powered by a high frequency square current with for each cycle a fraction of time at full output (light on) and the remaining time it's off. The eyes are unable to catch those fast changes but PWM is sometimes responsible for image flickering when frequency is too low and does not match camera settings. 

 

Also, dimming might affect the quantity but also the spectrum of the light right from the LED. LEDs have an optimal range in terms of volts, amps and core temperature (which requires adequate external cooling). 

 

 

[bghazzal]

Interesting thanks - I don’t have enough electrical knowledge to pinpoint the exact mecanism at play (maybe others can chime in), however one thing is clear: 

- the vast majority of LED dive lights / video lights do not have constant light / lumen output, which means that most dive / video lights will dim as the battery runs down.

Schematically speaking, variable output lights dim to preserve battery, whereas constant output lights remain at the same brightness until the battery dies.

This can be checked by empirical testing with the lights, and also by looking at available test data.

Overall, it's pretty clear that unless the manufacturer states that the light has constant output (or even better, provides burn tests), it’s safe to assume it doesn’t and will dim as the battery runs down - incidently, most lower to mid-range manufacturers do not mention the output or back it up by tests, whereas in the higher end segment this is more common.

 

As mentioned, is pretty easy to verify by running the light at maximum output in a tub of water and taking pictures at the same exposure settings, to see if there is a change in light intensity (our eyes adapt too easily to light intensity, whereas a camera sensor will be a more impartial judge...)

If the light has constant output, the intensity will remain the same same until the battery can no longer provide enough power to do so, and then it will drop quickly.

This feature is important, as it means that this will give the videographer a set window of usage, during which the light will provide a stable light, and the lumen output it is supposed to.

 

Otherwise, we also have the data from more scientific lumen output tests.

The Kraken burn test included in their docs for instance, show a rather constant output over the burn time:



And this is corroborated by more slightly more independent testing, such as this site (owned by Light and Motion, but test methodology is given in the FAQ).

Here is an output test on a Weefine model:

 

https://www.wetestlights.com/test.html?docId=0uB0YFsc1Z1e5xw7OX0E

 

Or on a Light & Motion / Sola light, here:

 

https://www.wetestlights.com/test.html?docId=5I5Z4eZ8Y8bA8OFfNja2
 

However, unfortunately, this important feature is not commonly found in most LED dive lights.

And if the light is not designed with to ensure constant ouput, then the light will start dimming as soon as the battery starts discharging.

While I don't really know enough to understand the exact mechanism at play (constant voltage driver? I've read that the issue might be linked to not including constant output emitters - maybe others can offer explanations), the results are quite clear.
 

Without the type of control circuit required to ensure a constant output as shown above, the burn test curve will look like this test of Big Blue TL2800P light here:

https://www.wetestlights.com/test.html?docId=9pPEJyoiJaUNjmRz6KJ1



or this Orcatorch D550 test here:




Such curves are well, much less sexy...  And empirically, taking pictures at a regular interval at a set exposure will show the light dimming right from the begining of the test...
As an example, here are pictures of such a bucket-test of a Big Blue VTL9000P light showing the dimming...
 

Unfortunately for us end users, this is actually the most common design found in underwater LED video lights (and also dive lights), especially in the lower to mid-range.

And alas, manufacturers are also not really transparent about this...
This is the answer I got today from Big Blue regarding the output of their COB video lights:

"For our products, we have built-in control circuit which the light will dim constantly once the battery drop to keep one and half burn time" which is, well, not the clearest I've read...
But correlate this with the BigBlue light test above gives an idea of what to expect on lights of their range...
Another user had an exchange with Big Blue regarding this issue as well - Scubaboard thead here

In the video-light segment - where this feature is quite important as it allows us to shoot reliably at constant exposure - makers offering confirmed constant output lights are as follows, for the time being:
 

- Keldan

- Light and Motion (Sola)

- Sealife

- Kraken
- Weefine

There probably are others, but it's not easy to have access to this type of product info, which is regrettable...

And the situation is similar for standard dive lights actually, where most low to mid range lights will be non-constant output, but higher end models (Apeks for instance, and others, Dive Rite, Dive Gear Express see here) offer constant output - which an important feature to ensure a consistent light intensity over a given run-time in artificially-lit overhead environments such as cave or wreck diving...
 

Edited August 2 by bghazzal

[Chris Ross]

I suspect the explanation is quite simple.  LED brightness is proportional to current, and this will drop as voltage goes down.  As the battery discharges the voltage drops and the LED dims.

 

The cheap lights don't dim to preserve batteries, they dim because the output voltage drops as the battery discharges as they don't have the extra circuitry to keep the constant current required to maintain brightness.

 

Brightness is generally controlled by PWM (pulse width modulation) which switches the LED on and off rapidly to change the average current.  I would think the cheap lights just have the PWM circuit to provide the preset brightness levels and as the battery voltage drops so the current drops and the LED dims.  There are a few ways to keep the current constant - a DC-DC constant voltage converter that supplies a constant voltage until the battery voltage drops to a pre-determined level.  Another way is a constant current power supply. 

[eocean-eu]

The thing is most people find video lights expensive and stops mostly at the number of lumens you get but quality components inside are expensive. Cheap lights have generally one or several issues: 

- bad quality battery = short runtime

- badly cooled LED = short life for the product

- bad driver = short runtime * little possibilty of power adjustment * unsteady power * short life 

and then there's bad design e.g. the light will leak after a few dives like for many lights that are sold on Aliexpress

 

I've tried to repair a few divelights from friends and most of the time, the electronic was just cheap and bad or the LED was not even powered sufficiently. A good driver is expensive because it monitors the batteries, temperature, acts as a step-up voltage converter with constant current output and offer dimming. That's a lot of things to deal with. The circuit will naturally heat and if not properly cooled like the LED  it will malfunction, Components may even unsolder themselves or fry. 

 

 

[bghazzal]

59 minutes ago, Chris Ross said:

I suspect the explanation is quite simple.  LED brightness is proportional to current, and this will drop as voltage goes down.  As the battery discharges the voltage drops and the LED dims.

 

The cheap lights don't dim to preserve batteries, they dim because the output voltage drops as the battery discharges as they don't have the extra circuitry to keep the constant current required to maintain brightness.

 

Brightness is generally controlled by PWM (pulse width modulation) which switches the LED on and off rapidly to change the average current.  I would think the cheap lights just have the PWM circuit to provide the preset brightness levels and as the battery voltage drops so the current drops and the LED dims.  There are a few ways to keep the current constant - a DC-DC constant voltage converter that supplies a constant voltage until the battery voltage drops to a pre-determined level.  Another way is a constant current power supply. 

Thanks for the clarification Chris - this is what I had initially understood as well, which is why Big Blue's and Orcatorch's answers to my enquiries seemed a little off...
 

48 minutes ago, eocean-eu said:

The thing is most people find video lights expensive and stops mostly at the number of lumens you get but quality components inside are expensive. Cheap lights have generally one or several issues: 

- bad quality battery = short runtime

- badly cooled LED = short life for the product

- bad driver = short runtime * little possibilty of power adjustment * unsteady power * short life 

and then there's bad design e.g. the light will leak after a few dives like for many lights that are sold on Aliexpress

 

I've tried to repair a few divelights from friends and most of the time, the electronic was just cheap and bad or the LED was not even powered sufficiently. A good driver is expensive because it monitors the batteries, temperature, acts as a step-up voltage converter with constant current output and offer dimming. That's a lot of things to deal with. The circuit will naturally heat and if not properly cooled like the LED  it will malfunction, Components may even unsolder themselves or fry. 

 

Totally agree for the cheap lights, they are cheap for a reason, and have a very short lifespan due to poor design and components...

 

But just to be precise, non-constant output light are not limited to such disposable, non-branded semi-direct imports sold on aliexpress/alibaba/amazon/ebay plateforms.

 

Non-constant output lights are also the most common in the mid-range, including commonly found video light brands sold by reputable retailers such as:

 

- BigBlue

- SUPE / Scubalamp / Fotocore / Divevolk

 

These are widely-used mid-range products - not exactly falling in the cheap light category -  and are also not constant output lights - as are models by many other light common brands of variable reputation ( Orcatorch, Archon, Nitescuba, Xtar, Tovatec, XLight...)

 

It would be interesting to know if Backscatter's video lights, such as the one included in the Backscatter Hybrid Flash or the MW4300 have the circuitry required to provide constant output or not - I couldn't find test data.
The MW4300 - which I use - is a well designed light, with nice features and a tight 85° beam in wide mode, but also some surprising design choices such as low CRI LED modules (CRI of 71.1 in wide and 70.3 in spot mode), so it's not very clear what to expect in this respect...
 

Edited August 2 by bghazzal

[eocean-eu]

3 hours ago, bghazzal said:

It would be interesting to know if Backscatter's video lights, such as the one included in the Backscatter Hybrid Flash or the MW4300 have the circuitry required to provide constant output or not - I couldn't find test data.
The MW4300 - which I use - is a well designed light, with nice features and a tight 85° beam in wide mode, but also some surprising design choices such as low CRI LED modules (CRI of 71.1 in wide and 70.3 in spot mode), so it's not very clear what to expect in this respect...
 

 

I have also a MW4300 I use for snoot lighting and a luxmeter. I can do some light measurement at 2 to 5 min intervals 😉 and see how constant it is or not. 

 

 

[Davide DB]

Well, the topic is complex and we should reverse engineering those lights to understand exactly the technique used to drive and dim the led. AFAIK the PWM method cited by Chris is not well suited for video or photo professional lights. 

Despite this, it is perhaps the most widely used method of adjusting the brightness of LEDs even though in public and private lighting, caveats are used and intl. standards are followed. Then there are other methods similar to switching power supplies that work at much higher frequencies.

However, I believe that a light designed for video makes use of constant current drivers.

If you are passionate about the subject, I will give you some links.

 

The problems of PWM with video footage:

 

https://www.ledsmagazine.com/specialty-ssl/article/14299599/video-lighting-iso-settings-and-led-dimming-impact-video-quality

 

https://ledlightstation.medium.com/why-do-led-light-flicker-on-video-and-camera-9922bb8aabe9

 

A not-too-technical roundup of ways to drive an LED in a flashlight

 

https://budgetlightforum.com/t/understanding-the-difference-between-linear-buck-boost-and-direct-drive-drivers/28594

 

https://lygte-info.dk/info/DriverTypes UK.html

 

Bye

[eocean-eu]

Thanks for the links. The last one concludes 

 

'If you have a flashlight with a single LED single lithium battery then get a Linear driver.
If you have a flashlight with one or two NiMH/Alkaline batteries, then you need a Boost driver.
If your battery voltage is much higher than the LED voltage then get a Buck driver '

 

A recurrent design for lights with several LEDs (not the single COB design) of the same color is to assemble them in serial. It has several advantages in terms of light quality, effeciency, safety, design but with a major disadvantage that when one led breaks, all the matrix is dead. A diode might break when cooling is not sufficient as it may break or unsolder. If one looks at a high power LED datasheets, there's a fair amount of data and information regarding cooling. The thing is the added voltage of all LEDs in series in divelight in many cases exceeds the total voltage of the battery pack. This puts boost/step-up drivers as the preferred choices for big divelights.