PWM dimming does nothing for LED efficiency (lm/W)

(I think)

So a lot of people here speak of dimming their COBs to get more efficiency (lumen/watt or lm/W) and while it is true that lowering the Vf (or current) of an LED will increase its efficiency like this:



the typical dimming circuit incorporated in many power supplies is a pulse width modulation circuit (PWM) which dims the LED by turning it on and off repeatedly. This circuit does not vary (very much) the actual Vf and current it sends to the LED it just does it for a fraction of the total time. Thus there is no "moving the efficiency curve" to higher and higher lm/W numbers with PWM dimming. You’d have to lower the actual current/Vf sent to the LED not just the energy they receive over time. So with PWM dimming you get the same lm/W but only a fraction of the time = less total photons hitting plant, so you are just giving the plants less of what they need with no real increase in efficiency; you might as well run fewer COBs full on.

Of course there will be less heat (and light) with a PWM dimmed LED so that may help them stay cooler and gain some efficiency by being cooler but there is no such thing as a free lunch with PWM dimming on a properly cooled LED.

By current (of Vf) limiting/dimming your LEDs you can increase their efficiency (lm/W) as the above curve shows but not (I believe) by PWM dimming them.

By all means tell me if I am wrong!

Yep, that's the basics.

Meanwell and many other drivers companies 3 in 1 function uses constant current dimming no matter if you signal it via potentiometer, analog voltage, or PWM. So you can get PWM precision, and still gain efficiency in your output.

Sorry, no time for long explanations, but Wiki will help you!
https://en.wikipedia.org/wiki/Buck_converter
This is what happens. (The inductor is the key element here)

The real benefits of PWM are twofold, a time based square wave maintains the full-forward current through the LED current, with analog dimming your changing the current which shifts the colour, the 2nd benefit and yet to be fully realised here is the digital control of led systems to mimic the real world grow world outside your room
Cheers
Mark

Mark, you are right on about the digital controls not fully being utilized. I have been saying this for a long time. I wish I had the energy to learn programming and all the other parts of it. I have started one project with a computer cooling control panel with temp sensors and 4 fan control, just to play with. Tie this in to a control system? Oh the possibilities.

nevergoodenuf said:

Mark, you are right on about the digital controls not fully being utilized. I have been saying this for a long time. I wish I had the energy to learn programming and all the other parts of it. I have started one project with a computer cooling control panel with temp sensors and 4 fan control, just to play with. Tie this in to a control system? Oh the possibilities.
View attachment 3701452

Click to expand...

yes im thinking mimicking all the cycles of the sun, temp sensors, humidty control, motor control on the rigs, the limitless opp's digital brings , we need to crowdfund an Arduino/Rapberry wiz kid
Cheers
Mark

Downloadable light cycle from a library, for each strain.

welight said:

yes im thinking mimicking all the cycles of the sun, temp sensors, humidty control, motor control on the rigs, the limitless opp's digital brings , we need to crowdfund an Arduino/Rapberry wiz kid
Cheers
Mark

Click to expand...

Let me finish the enail and the led controlled rig first. I know I am not the only one that plays with microcontrollers