Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

LED light making videos in Retail Shop flickers 3

Status
Not open for further replies.

klloh

Civil/Environmental
Apr 11, 2006
4
0
0
SG
WhatsApp_Image_2022-11-17_at_10.56.22_uexpdp.jpg



[URL unfurl="true"]https://res.cloudinary.com/engineering-com/video/upload/v1668656293/tips/WhatsApp_Video_2022-11-17_at_10.42.31_exmsh9.mp4[/url]

Fellow Engineers, need help here.
I am setting up a retail shop, after setting up the LED (4000k) cool white light, we take video using hp and the video flicker as shown. This is bad because many customer would wants to take video in the fashion shop. in real life, the LED is not flickering and they appears ok in our eyes. The flickering only show up in the handphone video. Asking the customers to adjust their handphone camera setting is really NOT an option. can any experts sheds some advice ? TIA
 
Replies continue below

Recommended for you

The problem is caused by a lack of filter in the driver. Sometimes, phosphorescent coatings can be used with the LED's to reduce flicker that is visible to our eyes but this will still get picked up by a camera. You need to find drivers with filters built in.
 
sorry, do u mean that we should not use the Phosphorescent coating type of LED but use LED with driver that have built in filter?
 
Filter = large enough capacitors to ride through the terrible clipping of the input power.

This looks like a cost-cut module - they didn't even bother putting the manufacturer name near the model number, probably to save the fraction of a cent in ink.
 
There is no white LED. The color produced is usually achieved by using a UV (blue, actually) LED source to excite a phosphorescent coating much in the same way a fluorescent bulb works. Only recently has there been a blue LED produced to complete the primary RGB spectrum. These are common on multimedia displays now. I don't know a whole lot about the subject beyond this. There are diffused RGB strips available now. The color rendering index is good indicator of color quality but I don't know of any standard with regards to flicker.
 
Tug; The white LEDs are BLUE LEDs exciting yellow phosphors never UV. UV's are very recent and still fairly gutless and inefficient.

klloh; The flicker as mentioned by Tug and Dave is indeed caused because the LEDs are actually going on and off rapidly due to your driver's technology. There's nothing you can do about it with those particular drivers. You need different ones.

You might check with Lutron or Meanwell. You're looking for 10kHz PWM or faster or drivers that boast high accuracy or flicker-free dimming down to 1% or lower. For them to provide that they have to operate at very high frequencies or are sending out DC.

Alternatively you could go someplace that sells LOTs of lighting. [URL unfurl="true"]https://www.1000bulbs.com/[/url] comes to mind. They have a sales team who no doubt has dealt with this issue many times.

You should be prepared to supply them ALL the details of what EXACTLY you have now so they can easily understand what exactly you may need. There are dozens of LED schemes that can only be driven by certain drivers. You should cobble together every detail you have describing what you've got (data sheets names etc) so they can possibly come up with good alternatives otherwise they'll blow you off or only suggest whole new fixtures.





Keith Cress
kcress -
 
Not sure how the electric grid looks like over there but if you have 3 phases in like 400VAC (3 x 230VAC) you could divide the LEDs in three parts, one on each phase R S and T, then you would need two more LED drivers.
I assume that not all lamps would flicker at the same time then and the video would look normal. [ponder]

“Logic will get you from A to Z; imagination will get you everywhere.“
Albert Einstein
 
Does anyone have an idea of what a LED powered by a square wave would look like? Not a 'smart a**ed' question, just something I was wondering about.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik
 
Without any filtering? Depends on the frequency and duty cycle. Low frequency, you'll see it pulsing on and off. Higher, you won't see this when looking straight at it but will see it if it is moving across your field of view (a generation or two back of GM cars had LED taillights driven by PWM that was on the verge of perception like this, with the Cadillac Escalade being a particularly annoying example). Higher yet, your eyes won't see it but a camera will.
 
The video shows flicker only when the line of sight is moving, so it's the power supply that causing it, since LEDs don't blink on their own.

You've not described how the lights are actually wired, i.e., one driver per light, or ????

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list
 
I recently had to deal with something like this myself on some lighting for high speed parts inspection cameras, here’s an article I found useful to explain it to the end user.


We went to a lamp with a true constant current driver operating at 1.25kHZ, the flicker doesn’t show up on the inspection cameras any more.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
In some cases, the flicker is intentional, to amp up the visual intensity, by overdriving the LEDs briefly, and then allowing them some time to recover; that might be why there's a fan drive on the module. The voltage/current for the LED output suggests a maximum power of up to 35W. But, that's an awfully big module for a single can light

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list
 
We had that problem with a vision camera but it was easily solved by increasing the exposure time.

“Logic will get you from A to Z; imagination will get you everywhere.“
Albert Einstein
 
The reason it's noticeable when moving the camera is that the imaging sensor is scanned from top to bottom (usually along the shorter axis) in what is called "ripple" read, and exposure and since the camera was in portrait mode, which is usually the long axis, any sideways motion will result in non-simultaneous imaging along the movement axis. Note that exposure time is not likely able to fix this, since the illumination of the entire scene is synchronized in flicker, so only by saturating the entire image would the flicker disappear ;-)

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list
 
In this case it was a picture of a none moving object and I assume that the parameter for exposure time treated the image to simulate a shutter on a mechanical camera reading the surface as many times as it could over the set time and then choosing the brightest pixel for every pixel it read.

We don't have any applications with that camera on moving targets so I don't know if it would have worked on that as well.

“Logic will get you from A to Z; imagination will get you everywhere.“
Albert Einstein
 
Status
Not open for further replies.
Back
Top