An awesome video! I have bookmarked this video and I'm sharing it with my videotography friends. With some of my cheaper LED lights I use in my studio. I would sometimes notice a flickering effect. One day it was very apparent that the flickering would go away when I moveed the camera. I then noticed that it was only a few of the lights that cause the flickering. After a moment or two of trying to figure out what was different. I brought all the lights up to full brightness and the flickering went away. And now your video explains why. Thank you my friend. 👍🎥👍
Thanks Edmund! We're growing slowly but surely, we just need to figure out how to get more love from the RU-vid algorithms. 🤔 In the meantime, we'll keep doing our best.
Thanks - we're trying to increase our quality and get back to our normal quantity but are going through a bit of a dry spell at the moment due to ... life. I appreciate the encouraging words!
There are 2 other large disadvantages of pwm: If it's not a large frequency say over 1000hz it can cause eye fatigue and headaches. Also your video camera might sync up with the pwm frequency causing video artifacts. Sadly most commercial dimmers use something like 250hz pwm
We actually have made some changes in the office that we shoot these videos and were struggling with that exact problem with the video camera. We ended up having to use a less efficient linear dimmer with a large capacitor in parallel for it to stop strobing on the camera. It's funny how (in my opinion) LED technology is so great but the implementation sometimes leaves much to be desired.
cats and other animals have (apparently) shorter persistence of vision therefore they will see the room light flashing at the PWM rate which could affect them, if you could see a light flashing all the time I think you would be affected.
You could do voltage limiting very easily with a 10k potentiometer. Just a suggestion for your video. This would dim the LED to a point. This is basically a voltage divider circuit in this case then. Potentiometers are very cheap by comparison of PWM.
Yep, that's an inefficient but effective method. It does have some non-linear properties but if you have a non-linear potentiometer to offset that, it should work!
Hi, I have a question. Can I turn a light Fixture (chandelier my wife bought) that isn't dimmable into a dimmable Fixture? I would be comfortable doing this, I just don't know how. We put it in the living room and previous light fixture was dimmable but this new one doesn't even with dimmer switch.
I've been doing commercial/residential and industrial electric for the last 24 years and I am currently in college for basic electronics taking technical math and I feel overwhelmed. What I have had to overcome in the field never taught me anything about theory. I don't want to quit but I don't feel prepared for this.
Hey Lance, the math can definitely feel overwhelming. It's also easy to get lost in both the math and the concepts. Just make sure you come up for air every once in awhile and make sure you haven't strayed from the important foundational concepts. If you have good teachers, talk them about your concerns and see if they'd let you know what is the most important things to understand from the lessons so far. Despite this, it will be tough. If you can't get everything done this semester, then you can try again with a huge headstart next semester. It sucks but you wouldn't be the first one to have to do that.
Can you please explain why you need to use pulse modulation or a complex current/voltage control system when I just hooked up an LED straight through a potentiometer and it dims it just fine.
Scaling and efficiency, mainly (and those two are related). As you change your potentiometer, any power that is not being dissipated in your LED is being dissipated in the potentiometer. For a single, small LED, that's totally fine. Once you get much bigger, the potentiometer (which usually ranges around 1/10 watt to 1 watt) will start to struggle with the power. Power that is just being wasted. With pulse modulation, a reduction in brightness also has a reduction in power. Again, for a small LED, it doesn't really matter but it does as it scales. I hope that helps!
Thank you for the video, precise and clear. I am looking for a way to dim a string of led’s smoothly for home a theatre application. Things I have tried so far all seem to dim in steps and then go out suddenly, any recommendations?
If you look at the voltage versus current curve for LEDs (IV curve) there's always a challenge at the "knee" where the current starts to spike up as you adjust the voltage. At this spot, the brightness changes very, very quickly with the smallest change in voltage. If you're manually controlling the voltage via a rheostat or variable resistor, you could potentially put two of them in series - one that has a larger resistive swing and one that is smaller and more precise. That way you can switch over to the smaller, more precise rheostat when it's getting dimmer. Just one thought!
Beginner here. I'm working on a project with a LG LED 55 inch 55UK6090PUA. I got it free because the screen stopped working but not the backlights. I'm using it to make a Movie Poster backlit frame. Question. How would you go about powering just the backlights? I need to be able to dim them as well because it is REALY bright and I no longer have a way of seeing the settings menu obviously. I'm hoping to disconnect the Main board and just keep the power supply. Or if it makes more sense, get an external power supply. Thanks
Cool project! I'm not sure how you could control the main board but if you find the LEDs themselves and how to power them, you could get an external power supply and then control the brightness quite easily. A lot of LED drivers can just use a screwdriver to adjust the current flow (and thus brightness).
Technically yes but they are much more powerful than the LED used as an example and the significantly higher current demand complicates things in terms of finding the right parts and dealing with excess heat.
Thanks, so is it safe to assume that most LED controllers (eg. Zigbee controlled ones) use pulse with modulation? that would explain why my manual voltmeter did not detect any voltage/current changes
Yeah, I personally assume that they use PWM unless they say otherwise. It's the worst in terms of quality of light output but best in terms of cost and design.
Just flip it on its head! If you want to go from dim to bright, just start out with a higher potentiometer setting or a lower duty cycle on your PWM. If you want something brighter than the max brightness of the LED, the only good option is to get a bigger LED. You can overdrive LEDs for awhile and they're pretty tough but you'll definitely be shortening their life and risking a small yet catastrophic failure. Bigger LEDs have better heat dissipation methods as, though they're MUCH more efficient than incandescent, they definitely still produce heat.
Yep, they should be electrically the same from your perspective. Unless the LED strip is made up of programmable LEDs, which could complicated things. If it's just a standard dimmable strip, then you should be totally fine.
@@CircuitBread i meant will the dimmable bulb of the same wattage and voltage work with slight error margin with pulse width modulation ? Secondly , is emi caused for 20 Khz for the same modulation of the bulb and will this effect nearby wifi signals ?
Yeah, as long as the bulb and the strips are both driven directly by you, it should work just fine with PWM. As for EMI, at 20kHz, you may produce some EMI (which should be reduced if your wire from your controller to the bulb is short - less of an antenna) but Wifi is 2.4GHz so, even with harmonics, you're in a completely different band. It shouldn't have any affect on Wifi.
Your theory doesn't sound right to me. I don't think its is fully on and fully off in those pulses. I don't think our eyes would see that as something being half lit. I think it is more likely the LED materials speed of reaction is such that a 50% on pulse would behave the same as 50% current, and therfore the led would only shine half as bright.
Hey Mark - glad we could help! I am going to disagree with the LED materials speed of reaction idea, though, and only because LEDs can turn on and off very quickly - that's why they're sometimes used in communications (like remote controls) because they can do clean and quick infrared pulses. The best way to test this it to get an emitter and a detector and then hook the detector output up to an oscilloscope - you'll see that the incoming signal is very clean. It does blow my mind sometimes, though, that that's the way our eyes work...
It has to do with something called perceptual inference. Our brain makes assumptions based on last state and present state based on vision being 30 frames per second at best.
There is way too much propaganda here. It is sufficient to say that less power, a lot less) is consumed and that the LED bulb does last a LOT longer whuch means a lot less changing light bulbs in those overhead fixtures. And anyway, at one time rheostats were used to dim incandescent lights.
Yeah, the constant voltage dimmer is nearly always better when it comes to performance, it's just a bit harder to create yourself. We just updated the office where we shoot the videos and had to switch drivers for some LEDs to a constant voltage dimmer (and I threw a large capacitor in parallel on the output as well) because they were flickering badly on the camera.
@@CircuitBread agreed 💯. Do you have the link for that video? It definately looks tricky to make, I'm just lucky it was prefabbed. Do you know if just a potentiometer would work for dimming? I have a funny feeling about it lol
Sorry, I didn't see this but yes, a potentiometer would be incredibly inefficient yet very smooth. We don't have a link for the video because it was just something we were doing behind the scenes for my office (that has been completely overtaken as a studio now...) We've shot a couple videos with the new backdrop where this came up but they're all still being edited.
@@CircuitBread makes sense, that would be too easy lol. Definately not optimal for a group of LEDs. And gotcha, I'll keep my eye out! And no worries thanks for the reply!
