I too was referred from Martin Lorton's blog. I am an electronics design engineer and was very impressed with your clear and concise explanation. Using 500kHz switching frequency can cause RFI problems. When finished can you test radio interference in AM and FM bands. I have found using a radio in the same room and sweeping through the bands is very effective in identifying RFI problems. You may have to use lower switching frequencies to control RFI (at the expense of a larger inductor).
Excellent explanation, Steve. Thanks for sharing. I'm driving some 1W RGB LEDs for the first time and your video helped me understand the efficiencies better.
I watched this video when I came from martin's channel. I thought I get back, like and comment on it. Great channel. I watch electronics videos for entertainment instead of TV and there is not enough production. So I'll encourage you in every way I can.
Thanks leppie. It's been a little difficult to judge how technical to go with only small audience figures, but it's good to hear that some of you enjoy the more technical stuff.
SDG Electronics I am big fan of Dave's neglected Fundamental Fridays ;p As a tip, you could do what w2aew does and hide some results so the viewers can think about it while you explain and expose the result (aka "This is baaad!") afterwards.
The overall quality of your videos is astounding... Its an almost an offense that you have so little subscribers...hang on tight!! Content is king and by the persistence you may win
Awesome Job! very well explained. Are you an electronics engineer? I have since graduated and work in power and controls engineering, however, my true passion is electronics. I have come to believe that if you understand electronics at a decent level, picking up any other electrical discipline is easy.
Hey Steve. I wasn't looking for this but it popped up and I want to say thank you. Could I use an atmel 328p for this? I've had a bunch in my drawer and pic costs a fortune now. Well last I checked.
Hi, When you calculate the component values, the output voltage of 3.3V is used to determine the D, which again is used to determine the inductance value. But is the idea, not to set the output voltage? My guess would be that since the duty-cycle is constant to obtain the 0.35A, which is sat by the feedback loop (dsPIC to FET), either you hit the 3.3V or the ripple current in the inductor is not the specified 0.2A peak-peak? Thank you.
+Steve Ku Effectively yes. We're controlling current rather than voltage, so the voltage settles to whatever it needs to be to drive the constant current through the LED. So long as the supply voltage is higher than the LED forward voltage, you don't really need worry about it.
I was wondering if it is possible to dim the LED load by applying a PWM signal to a MOSFET located at the output of a CC LED driver. I was worried that this may overload the driver as it will try to pump (lets say) 600/700 mA through the MOSFET by increasing the voltage above its capacity. Or would i need to apply a MOSFET to the input of a LED driver? I thought that this might mess with the LED driver itself. (Microcontroller start up sequence or whatever after every pulse.).
+ReapingMiner Hi, I will shortly be recording the next part of this series which includes methods for dimming the LEDs. Essentially if using the microcontroller method, you can superimpose a low (~250 to 2000Hz) PWM waveform onto the high frequency switching signal. I've dropped the high frequency switching frequency to around 100kHz on my design. Hopefully I will update in the next month or so.
Hi, maybe it's a stupid question but if you do the current feedback using the capicitor to reduce current ripple, shoud the capacitor be parallel to the Led and sense resistor SERIES? otherwise the feedback would adjust not the current on the LED but the current flowing in the LED//C, right??
Strange duty-cycle value....and I couldn't find any comments about it... Wich is also strange... :-SS guessing I world say Vled/Vdd=3.3/24=D=0.1365 just for completeness..
He used 0.1375. Which is the exact number he used as the efficiency of the linear regulator option 2:20. Total power of the circuit = 24*0.35 = 8.4W. Power of the LED = 3.3*0.35 = 1.155W. 1.155/8.4 = 0.1375.
Sir, Can you suggest me how circuit will be changed if the input voltage is takne as 40V. Will the same circuit schematic can be applicable for high 40V.
I need help by finding correct driver. I need constant output 10A 10V (100W). For input is not important. Output ripple i can deal after ( smoothnes ). You did make super video.
I too was referred from Martin Lorton's blog.
I am an electronics design engineer and was very impressed with your clear and concise explanation.
Using 500kHz switching frequency can cause RFI problems. When finished can you test radio interference in AM and FM bands. I have found using a radio in the same room and sweeping through the bands is very effective in identifying RFI problems. You may have to use lower switching frequencies to control RFI (at the expense of a larger inductor).
Peter Kay Thanks for your comments, yes I can try that if I can find a non-internet radio in the house...
Excellent explanation, Steve. Thanks for sharing. I'm driving some 1W RGB LEDs for the first time and your video helped me understand the efficiencies better.
Appreciate the content and the good mic/audio quality!
I watched this video when I came from martin's channel. I thought I get back, like and comment on it. Great channel. I watch electronics videos for entertainment instead of TV and there is not enough production. So I'll encourage you in every way I can.
Thanks! Any feedback, positive or negative is always appreciated so I can improve.
SDG Electronics Can you send me an email www.microlab.info/contact. I want to send you an email and your link on sdgelectronics.com is not working.
Microlab I get (reason: 550 relay not permitted)
Very clear and well explained. Keep them coming. Thanks
Short & sweet, right to the point. Thanks a lot for a great presentation. :)
Subscribed via Martin's channel, great videos! Looking forward to seeing more.
DextersLab2013 Thanks for subscribing. I have lots more content to come. Should be a part 2 of this shortly.
SDG Electronics Same here. Really like the more theoretical bits covered. +1
Thanks leppie. It's been a little difficult to judge how technical to go with only small audience figures, but it's good to hear that some of you enjoy the more technical stuff.
SDG Electronics I am big fan of Dave's neglected Fundamental Fridays ;p As a tip, you could do what w2aew does and hide some results so the viewers can think about it while you explain and expose the result (aka "This is baaad!") afterwards.
leppie Good plan, I might give that a shot.
The overall quality of your videos is astounding... Its an almost an offense that you have so little subscribers...hang on tight!! Content is king and by the persistence you may win
Awesome Job! very well explained. Are you an electronics engineer? I have since graduated and work in power and controls engineering, however, my true passion is electronics. I have come to believe that if you understand electronics at a decent level, picking up any other electrical discipline is easy.
Great job, thanks Steve.
Wow, man, you're awesome beyond words, thanks for the video, very appreciated, liked and subscribed at the speed of light
Just want to mention how nice your handwriting is, it's pretty but still legible xD
great video!
mancio92M Thank you!
Hey Steve. I wasn't looking for this but it popped up and I want to say thank you. Could I use an atmel 328p for this? I've had a bunch in my drawer and pic costs a fortune now. Well last I checked.
Hi,
When you calculate the component values, the output voltage of 3.3V is used to determine the D, which again is used to determine the inductance value. But is the idea, not to set the output voltage?
My guess would be that since the duty-cycle is constant to obtain the 0.35A, which is sat by the feedback loop (dsPIC to FET), either you hit the 3.3V or the ripple current in the inductor is not the specified 0.2A peak-peak?
Thank you.
Subscribed!
Great video!
Is the output voltage of the driver being controlled by the Vf of the LED?
+Steve Ku Effectively yes. We're controlling current rather than voltage, so the voltage settles to whatever it needs to be to drive the constant current through the LED. So long as the supply voltage is higher than the LED forward voltage, you don't really need worry about it.
+SDG Electronics Thank you!
I was wondering if it is possible to dim the LED load by applying a PWM signal to a MOSFET located at the output of a CC LED driver. I was worried that this may overload the driver as it will try to pump (lets say) 600/700 mA through the MOSFET by increasing the voltage above its capacity.
Or would i need to apply a MOSFET to the input of a LED driver? I thought that this might mess with the LED driver itself. (Microcontroller start up sequence or whatever after every pulse.).
+ReapingMiner Hi, I will shortly be recording the next part of this series which includes methods for dimming the LEDs. Essentially if using the microcontroller method, you can superimpose a low (~250 to 2000Hz) PWM waveform onto the high frequency switching signal. I've dropped the high frequency switching frequency to around 100kHz on my design.
Hopefully I will update in the next month or so.
Hi, maybe it's a stupid question but if you do the current feedback using the capicitor to reduce current ripple, shoud the capacitor be parallel to the Led and sense resistor SERIES? otherwise the feedback would adjust not the current on the LED but the current flowing in the LED//C, right??
so constant current is not the same as linear driver like lm338? lm338 = no flicker no pwm?
Can you explain please, difrance between current source and voltage sources and applications of that,I can't understand that
What's the range on the adjustable voltage?
Strange duty-cycle value....and I couldn't find any comments about it... Wich is also strange... :-SS guessing I world say Vled/Vdd=3.3/24=D=0.1365 just for completeness..
He used 0.1375. Which is the exact number he used as the efficiency of the linear regulator option 2:20. Total power of the circuit = 24*0.35 = 8.4W. Power of the LED = 3.3*0.35 = 1.155W.
1.155/8.4 = 0.1375.
Sir,
Can you suggest me how circuit will be changed if the input voltage is takne as 40V. Will the same circuit schematic can be applicable for high 40V.
+shameem hussain The same topology is suitable, however you need to take care of the regulator for supplying your SMPSU controller.
can u design for me a circuit for continuous work laser pointers for sensor illumination
I need help by finding correct driver. I need constant output 10A 10V (100W). For input is not important. Output ripple i can deal after ( smoothnes ).
You did make super video.
Champions league!
I can get the schematic as pdf
Part 2??
jpalm32 Coming this weekend - apologies, work has been really busy this week. Thanks for watching and for your interest!
Hello I need this for Arduino and 100W LED (boost) can u help? I pay you haha
Please send code
Ja.