Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

led driver help needed.

Status
Not open for further replies.

zoulzubazz

Member level 5
Member level 5
Joined
Apr 26, 2012
Messages
87
Helped
3
Reputation
6
Reaction score
3
Trophy points
1,288
Activity points
2,040
Hello everyone,

Attached is a simple circuit that should theoretically perform well theoretically but practically this is giving me all sorts of problems.

IMG_20140529_170548.jpg

there are two switches in the circuit, one is a manual SPDT switch (leftmost) and the other is a mosfet switch (which is on all the time but is included for the sake of completeness). Also there are three of these circuits connected in parallel to the same power supply (supply is 12V from smps supply from an old computer) for three different colored LEDs.
Initially when the manual switch was connected to the 10K pot(ON1) even though the pot was at 0Ohms some of the LED's would turn on, and i measured 1.6V at the gate but both op amp inputs were at 0V wrt ground. When the LM358s were replaced with LM358s from another manufacturer this problem was solved. but now when the manual switch is connected to the bnc although the BNC is not connected to anything all LED's come on and i see about 1.6V at the mosfet gate where as this should be 0V. As expected when voltage on bnc connecter is increased the LED brightness increases but they need to be normally off which is not the case. can anyone shed some light on this phenomenon and how this can be circumvent?
 

Hi,

You need some negative offset in your brightness regulation.

Opamps are no ideal amplifiers.
One disadvantage is that the inputs have a low +/- some mv input offset voltage. If the gate driving opamp has only some mv of positive offset, then even at zero input voltage at + input it wants to regulate. So it drives the fet slightely on so that a small current flows through the leds and cause the some mV on the - input.
This is what happened in your first description.

The next problem with opams is that they can not regulate its output to true 0V, if the lower supply voltage is at 0V.
Rail to rail output opamps are optimized on this.
Therefore the bnc buffering opamp can go down to only some mV on its output. The remainig mV cause the leds to light woth low brightness.

Additionally the human eyes are very sensitive on light. Even if there are only some mA with 1A rated Leds you will recognize it.

To avoid your problems you can add a negative offset to your current regulation. This can easily be done with two resistors.
Place one 100 Ohms resistor in the line from 1Ohms shunt to -input of opamp.
And place a 20k resistor from - input to +5V. Now you have about -25mV offset.

Hope this helps.

Btw. Paralleling of leds without series resistors may cause difference in brightness.
Linear regulation will cause power loss and heating. Switching regulation can imrove on this.


Klaus
 
Hi,


Btw. Paralleling of leds without series resistors may cause difference in brightness.

Definitely NOT recommended - unless they are perfectly matched, one LED will 'grab' more current and burn out. Use resistors on each LED.
 
thanks both, i used lm358 for the very same reason, are there any true rail to rail opamps whose inputs and outputs can reach the rails. thanks again.
 

Hi,

The RR opams i know only work to some mV to negative rail.

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top