EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Vindhyachal.takniki on February 11, 2018, 09:14:54 am
-
1. I have to control a 3W single led intensity, need to vary intensity linearly from 0 to 100% as user rotates potentioneter of 10K.
2. is there any dedicated low cost asic solution available for this, which can do this.
3. Most of i have checked need a PWM signal to ic like here
http://www.ti.com/power-management/led-driver/led-lighting/products.html#p207=DC (http://www.ti.com/power-management/led-driver/led-lighting/products.html#p207=DC) Linear;DC/DC&p2297=Analog
I am looking for some solution where a potentiometer can do the dimming directly, instead of connecting additional ckt for asic
4. Voltage of Li-Ion cell: 3 to 4.2V
if charging, then voltage go to 5V from main adapter
So input voltage for IC go from 3V to 4.2V
-
A few things.
4. Run the LED circuit from the battery circuit, then the voltage will not go above 4.2V. Obviously you will be using a BMS to charge the LiIon cell, it should limit the voltage battery side to 4.2V.
Current limiting with a POT will run you into issues. Voltage dividing with a POT will run you into more issues.
3W LED which can run off a 3V supply will be pulling around 1A of current. PWM switching 1 amp might require a mosfet driver. You "might" find a PWM chip that will switch 1A, but be careful.
Do also check the forward voltage of the LED. If the forward voltage is 3V and you are supplying 4.2V, then your current limiting resistor will be dropping 1.2V. If you limit the current slightly to say 0.9A (1.3 Ohms) then there will be over a watt of heat being dissipated in the current limiting pot which will melt most small pots.
You will also find the pot less than linear. You can do some things with parralell and series resistors with it to try and make it linear(ish) over a given range, but it's tricky to work out the values. Remember as well that the light intensity of the LED may, or may not be linear with current flow.
Dave did something similar to this on his microscope using a 555 timer chip, but do check your current draw and consider a mosfet to switch the actual current in and out:
https://www.youtube.com/watch?v=OXsu29K_Ap4 (https://www.youtube.com/watch?v=OXsu29K_Ap4)
-
For that kind of approach I would go with some sort of DC/DC switching power supply for LED.
There is some power supplies which can be controlled via external resistor (potentiometer, trimmer).
I know that MeanWell have some LED drivers (ie. LDH series) which can be regulated with 0 - 10V or PWM, but those are with 12V output.
May be there are some series for PCB mount, less power, around 3W.
You need constant current output and you want to regulate current value with potentiometer.
Best approach would be to make some sort of current measurement which will give you feedback signal for voltage regulator.
There are "special" LED driving regulator which can give you constant current output (up to Vmax).
Then you would use potentiometer to set reference point.
Also, look for LM317 (it's oldie goldie linear regulator, so it will dissipate heat, it's not efficient like switching reg.). There are some schematics examples for constant current source. Try to see where can you put potentiometer to regulate current without using potentiometer to actually dissipate due to (V_in - V_load)*I_load.
-
State of electronics today... so easy to cheat. I have a hammer, so everything look like nails.
I'd use an MCU. In fact, I would get a flashlight driver (NANJG 105c) which uses an ATTINY 13 MCU, and I use that hardware as the driver part. It already use PWM. It has all the electronics to use that PWM to control the power to the LED. It just need new firmware to change the pwm setting from on/off switch pulsing to controlled via an encoder or other means (potentiometer).
So, for 0%-100% control, one can (a) just use an encoder or (b) use the potentiometer as voltage divider and feed the center tab of the potentiometer back to one of the unused ADC on the MCU. So this ADC tell the MCU the current potentiometer value and the MCU can set the PWM accordingly.
The good thing with the MCU is, now you can use yet another ADC already on the MCU to determine if your input is from the 5V or the battery (presumably, you have a pair of diode for power-source-selection). The NANJG will handle the 5V just fine. When the MCU sees that the Vin is bigger than 4.3V, it can assume the system is now on 5V power and auto-adjust the PWM down a bit to emit close to the same light on 5V as when it is battery connected.
Possibility is endless. If you run out of room on the ATTINY 13, you can squeeze an ATTINY 85 in place of the ATTINY 13 - 8K programming space is now at your hands. Heck, with 8K, now you have enough room on the damn thing to blink out your national anthem in morse code with your LED (assuming you are not from a country with a long winded national anthem).