What 12VDC PS (or filter strategy?) guarantees perfectly smooth truthfully flicker free LED strip lighting?

Thread Starter

LiteBrite

Joined Aug 13, 2020
1
I just finished building and installing LED strip cove lighting in my new jewelers studio. Though I do not see any flickering, to my refined "Spidey Senses" something feels distinctly "off" about the quality of light bathing me- and I'm guessing it's the switching PS. For comparison's sake, I have precisely zero problems at all with the high CRI LED strip lighting running off my Ford Transit high top van's 12V DC car battery power.

I've worked to educate myself on the subject of flicker and different style Power Supplies, but still feel unclear on the best PS solution, so anyone's help'd be most welcome!!! (Yes, I read and did my best to make sense of the one post here on this site.) My studies have taught me that "flicker free" is a relative term. That yes, certain higher frequencies of "visibly indetectable" flicker are acceptable to digital videographers, for example- yet what I am determined to have here must be beyond that. In this particular installation, I don't need dimming,. so that simplifies things. ANNNND I'd like to have that conversation another time- as I would like my next project to be 100% smooth AND also dimmable.

My why: We're vibrational beings, and I'm convinced some vibes have a disturbing and unhealthy effect on at least some of us. Those rare times when I have gotten an ocular migraine, my vision strobes painfully. These eyes of mine easily see even very fast visual flicker/strobing on the increasingly ubiquitous car tail lights and commercial/home LED applications- GRRRRRRRR!!!!
I've stopped patronizing some restaurants just for the flickering. (Yes, I let them know why and asked them if they'd be willing to upgrade!) Cheap LED Christmas lights running on AC power have become a seasonally nauseating annoyance. Austin Texas has a wildly popular annual Christmas "Trail of Lights". It's final feature is a huge exit tunnel of blue LED string lights maybe 30 yards long. Made me dizzy and sick. I chose to interview folk as they exited this feature asking them what they felt coming out of the tunnel- over 30% of the people asked reported being dizzy and/or nauseated. HOW I LONG FOR THOSE GOOD OL' ANOLOGUE & INCANDESCENT DAYS!!!

That said, I'm running almost 50 feet of this LED strip light attached to aluminum C channel for thermal cooling. I've got two strand 12 guage power wire running in a full UNADJUSTEDNONRAW_thumb_176.jpgloop with periodic jumpers over to the LED strip to minimize voltage drop and keep the light levels even. This is my first post here atv All About Circuits, BTW. One thing some of you might like to know about me is my username is my Biurning Man Playa name- which I got for decor lighting our camps over the years. I have built some very interesting and otherworldly beautiful
LED light sculptures. I so appreciate your good thinking., and a big thanks for reading!

LED Type: SMD 5630
CRI RA 97-99
Working Input Voltage: 12V DC
Output power: 5 Meter ---tested about 40-60Watts(depends upon of the real sonsumtion)
Daylight White -- 5500K-6000K


My temporary PS is the popular MEAN WELL LRS-350-12 DC Switching Power Supply 12V 29A 350W. https://www.meanwell.com/Upload/PDF/LRS-350/LRS-350-SPEC.PDF

 
Last edited by a moderator:

ElectricSpidey

Joined Dec 2, 2017
1,236
You shouldn't see any flickering using that PS. And I doubt you could perceive the 150mV ripple.

I would do a blind test using a linear supply, just to be sure.

In the future when you plan a dimmable system, I would stay over 5k Hz PWM, or if that is still a problem I would advise a linear dimmer, if you can afford the wasted wattage. (a simple emitter follower would do the job)
 

DNA Robotics

Joined Jun 13, 2014
580
Try powering it with a car battery. That would be zero flicker.
If you can tell the difference, A large capacitor will smooth your power supply output to zero ripple. But not so large that it abuses the power supply on startup.
 

MisterBill2

Joined Jan 23, 2018
7,009
To accurately check to see if there is any measurable flicker you can try using a photo-diode or a photo-transistor illuminated by the light in question and observe the output with an oscilloscope. Of course the simple measuring circuit will need to be totally ripple free, but for the short time you can power it from a battery. If there is a variation of intensity this method will show it.
One possible way to reduce any supply ripple for the LED strings would be to add a linear regulated supply in parallel, if you do observe ripple. And certainly 150 mv of ripple may show up in LED intensity variations at some operating conditions. LEDs are diodes, and diodes are non-linear.
 

ErnieM

Joined Apr 24, 2011
8,055
The ripple spec of the power supply is basically meaningless since they add an extra 47uF filter cap and don't say what current they test at, though is is probably decent enough for lighting. Hopefully it's 29A rating is well above what your strips consume.

I do hate your color, way to harsh for my eyes.

Our eyes share the ability to see flashes where other do not. I find car tail lights especially distracting as I sense movement where there is none just by looking one side of the road to the other.
ElectricSpidey: I will keep your 5K number in mind. I have an RGBW controller I built that runs at 1K and I do occasionally see some flicker effects. Next time I work the code I will push up that number.
 

MisterBill2

Joined Jan 23, 2018
7,009
The comment about the color of the light is funny! There are so many folks who much prefer the failing fire-light color because it tends to promote sleep, while the light with a blue component promotes wakefulness and higher energy. So probably for sleeping quarters that "warm white" color is better, while for work areas a more wakeful blue content is better. We do not pay folks to be sleepy on the production floor!

And like I stated in a previous post, using a light sensor with good frequency response will allow you to see any intensity variations on an oscilloscope screen. That removes human perception from the evaluation process.
 
Top