Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Variable power supply (1.8V to 3.4V)

Status
Not open for further replies.

g0rak

New Member
Hi there. This my first post so apologies for a bad description; you might find the diagram clearer... :)

I'm trying to build a variable power supply that has a fixed range between 1.8V to 2.4V. This is to be driving 5 red 1 watt LED's.

Below you can see I plan to use a LM350 (3A version of LM317) and have calculated the bottom resistor value (call it R2) to get both 1.8V and 2.4V from the LM350.

Min brightness - 1.8V output - 106Ω
Max brightness - 2.4V output - 221Ω

I'm trying to replace the single resistor (R2) with a resistor/pot network that will vary the resistance between 106Ω and 221Ω. I've only managed to find a 1K pot (I assumed the lower the better).

I plan to put the pot in series with a 106Ω resistor (giving the minimum resistance when the pot is fully turned clockwise and is effectively ground), and, in parallel with a 130Ω resistor (which when added to the 1k pot gives 115Ω). When the pot is fully turned anticlockwise, the 115Ω + 106Ω series resistor gives the full 221Ω.

The problem (which I'm sure you've noticed) is according to my calculations the pot + 130Ω resistor won't produce a smooth change between 0Ω to 115Ω across its full rotation. It'll be more like a logarithmic scale, and resistance will go up very quickly to start and then barely change. Using a log pot (as opposed to linear) might help a bit but I think I' might be going about this the wrong way.

**broken link removed**

I've not built this yet (LM350's due to arrive) but any pointers before I do would be greatly appreciated!! Thanks ;) Once I'm done I'll post more info as if it works this should be quite a cool project...
 
Yes, it seems like you're doing this the wrong way.

What voltage is your power supply?
 
Thanks for the replies.

The power supply needs to be 12V (for other components in the project) but I'm aware 12V > 1.8/2.4V is probably too much for to voltage regulator to step down. It's likely I'll step the voltage down to about 5-6 volts for this part of the circuit.
 
I don't think you understand what we're saying.

LEDs require a constant current regulator, not a constant voltage regulator.

Assuming a forward voltage of 2.4V and a power dissipation of 1W, the current needs to be varied from 0 to 416mA to go from off to full brightness.

Also bear in mind that 4W will be wasted in the resistor or driver transistor unless you use a switching regulator.

A more efficient option would be to connect four smaller 100mA LEDs in series which should give the same amount of light but use one quater of the power. Alternatively you could use a switching regulator with the current LED which will also be much more efficient.
 
while that is a crappy way to drive an LED, the way to do it with that chip is to put the LED to GND and adjust the resistance between the output and the adj terminal.

you obviously have no clue how much heat you will be generating and how quickly the thing will die. 350ma*5 LEDs is about 2A. minimum voltage across the LM317 is 2V. While you did not mention what you are running this mess off of I will assume it is a car lighter jack giving you a power loss of 9V*2A=18W on a package that can only support 2W with out heat sinking.

the moral of the story: ditch the LM317 and get a LED driver chip to do it right.
 
Thanks for the replies.

Just for the record I was planning on using the LM350 (good for 3A) and using the large metal enclosure for a heat sink. It'll be powered by a wall adaptor (which ideally has to be 12V for the other part of the circuit).

Although having the LED's in series would be best, there is a hardware limitation as the LED's are a string of lights.

Ubergeek63 - from your comments I assume you mean arranging the LM350 for current control, like this:

**broken link removed**

My main concern so far has been I don't want to adjust the voltage/current from 0 to 2.4V/350mA or whatever and need to specify the minimum as well as maximum.

I'll check out some more circuit designs on the internet but meanwhile does anyone have any specific recommendations for driver chips?
 
Thanks very much guys - it would seem that PWM is the way to go.

Btw I've done some more researching and found a really good page that highlights different ways to power and control high-power LED's.

**broken link removed**
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top