Controlling the brightness of LED's with a semiconductor - A simple circuit required please

Thread Starter

Neville Thannhauser

Joined Dec 3, 2016
27
Hi All,

I've been making some simple LED based trinkets for the house and for gifts. An example would have, say, 48x amber LEDs which i'd power from a 12V AC-DC supply. The voltage drop of the LED's is 2V and the forward current is 30mA. Because of the number of LED's I've been wiring them in an 8x6 array with a 1ohm series resistor in line with each 6 LED's

What I'l like to do is to control the intensity of the LED's in a simple semiconductor & potentiometer circuit. Can anyone please suggest a circuit for this case and explain the theory (in simple terms) so I can adapt it for different numbers of LED's & supply voltages.

Many thanks.
 

phonic

Joined Sep 29, 2016
37
POST ISSUE 02 of 2017_01_09 PLEASE SEE POST #34 FOR A REVISED VERSION OF THIS CIRCUIT

Hi NT,

Below is a circuit I came up with.

It is a bit more complex than you requested and you will need to arrange the LEDs in a 4 by 12 array, instead of a 6 by 8 array.

phonic

2016_12_03_Iss1_AAC_LED_BRIGHTNESS_CONTROLLER_VER1.jpg
 
Last edited:

wayneh

Joined Sep 9, 2010
17,498
You want to learn about PWM, pulse-width modulation. The circuit above is an example but there are many others. Using the popular 555 timer IC is common and you can find many examples.
 

Thread Starter

Neville Thannhauser

Joined Dec 3, 2016
27
Hi Phonic, Many thanks for the circuit. As you say a bit more complex than I was hoping for, but i'm up for the challenge.
Can you please clarify a few things:

1. The 2K pot. Is it log or lin?
2. Can you give me a part number/name for the N MOSFET
3. Do you have a value for R3 ZR7 (connecting the MF source pin to the 0 rail.

Again, many thanks.

Nev
 

Dodgydave

Joined Jun 22, 2012
11,302
The 2K pot will be linear, and the R3 resistor will depend on what current you want in the leds, try 20ohms that will give you upto 100mA current, the Mosfet try IRF510, or IRF540.
 

dannyf

Joined Sep 13, 2015
2,197
a much simpler approach is to use a comparator like tl431 - it can be driven by a pwm or an on/off signal and has the advantage of being a ccs controller when it is on.
 

Bernard

Joined Aug 7, 2008
5,784
The simplest arrangement might be a pot, 5k to 10k, & a darlington power transistor like N ch D1326 or sim. in TIPxxx or make your own. Connect as common collector with maybe 12 strings of 4 LED's in series with 68 ohm current limiting resistor. Power dissipation of transistor will be around 2W, so a small heat sink is needed..
Advancing the pot. from gnd. to about 8 V gives no light, all of the control is then from 8 V to 12 V. Another advantage of PWM, low wasted power & wide range of control.
 

Bernard

Joined Aug 7, 2008
5,784
' finally found old plot of pot. vs. output of med. power emitter follower. Substitute resistors & LED's for
25.5 ohm resistor as suggested by dannyf.Brightness Control 00000.jpg
 

Thread Starter

Neville Thannhauser

Joined Dec 3, 2016
27
Hi All,

I've ordered all the components from CPC in the UK and will assemble the circuit in the coming days. I'll report back my findings.

Thanks again for all the helpful suggestions.
 

dannyf

Joined Sep 13, 2015
2,197
the mosfet version of bernards circuit, showing the wide range of controls you get by putting the load on the source / emitter.
led control - mosfet.PNG

the advantage of a mosfet is that the loading on the pot (R2+R3) is minimal so you can use high value pot. But the circuit is less efficient than the bjt version.
 

phonic

Joined Sep 29, 2016
37
Hi Phonic, Many thanks for the circuit. As you say a bit more complex than I was hoping for, but i'm up for the challenge.
Can you please clarify a few things:

1. The 2K pot. Is it log or lin?
2. Can you give me a part number/name for the N MOSFET
3. Do you have a value for R3 ZR7 (connecting the MF source pin to the 0 rail.

Again, many thanks.
Hi NT,

No problems

Apologies for not replying to your post above- I have been busy.

To answer your questions:
(1) The 2K potentiometer is linear
(2) The IRF540 NMOSFET will be fine, but it will need a heatsink
(3) The ZR2 resistor is in fact 2R7 which is a 2.7 Ohm resistor.

Did you build the circuit as I posted?

phonic
 

phonic

Joined Sep 29, 2016
37
You want to learn about PWM, pulse-width modulation. The circuit above is an example but there are many others. Using the popular 555 timer IC is common and you can find many examples.
Hi wayneh,

I do know about PWM, but the OP asked for a simple approach and a PWM circuit would not only be more complex but it would be more difficult to understand. And there could well be EMC problems with strings of LEDs forming an antenna.

phonic
 

phonic

Joined Sep 29, 2016
37
Put the LEDs on the source of the MOSFET, pot on the gate, for max control on brightness.
Hi danny,

This is not true. The circuit that I posted gives full and linear control of the current through the LEDs. What you propose exposes the MOSFET gate source to the non linear characteristics of the LED string. The circuit I proposed drives the LED strings with a constant current which once set will be fixed and will not drift with the Vf of the LEDS.

phonic
 

phonic

Joined Sep 29, 2016
37
' finally found old plot of pot. vs. output of med. power emitter follower. Substitute resistors & LED's for
25.5 ohm resistor as suggested by dannyf.View attachment 116618
Hi Bernard,

Your circuit is fraught with problems. For a start you are exposing the emitter base junctions to the non linear characteristics of the LED strings.

Also the non linearity of the two emitter base junctions in series will give an awkward brightness control characteristic.

The emitter resistor just wastes power.

The two base emitter junctions also waste power and add a 1.2V overhead losing 10% of the available 12V.

Also you are not defining the current through the LED strings very well.

Incidentally all linear approaches to LED brightness will result in the controlling element: BJT, MOSFET, IGBT, disipating the same powe, by definition.

phonic
 

phonic

Joined Sep 29, 2016
37
the mosfet version of bernards circuit, showing the wide range of controls you get by putting the load on the source / emitter.
View attachment 116649

the advantage of a mosfet is that the loading on the pot (R2+R3) is minimal so you can use high value pot. But the circuit is less efficient than the bjt version.
Hi dannny,

Your circuit is pretty much the approach that I used but very imprecise and will be temperature dependent and subject to the spreads of individual MOSFETS.

As I have said already, the power dissipation in any controlling device will be the same for any linear approach.

phonic
 
Top