Hello, I tried a lot with all the parameters for modeling a colored LED, and went through a lot of threads. I was left kind of frustrated. It seems strange that no curve seems to match a typical colored LED ("knee" around 2V, operating current 20mA, steep curve up, at 3V already around 100mV). In cases where you have to build something based on an existing situation, you are already happy to have this info, and you can not provide the different parameters CircuitLab asks for. I have no idea what some of the parameters mean in the definition of a LED, but many trials didn't lead me to an acceptable Voltage - Ampere curve; the curve always stays rather "flat". And I don't know how to change the knee votage point. That's caused by my lack of understanding. Would someone like to share the parameters to design a LED with a steep curve? The different colored LED's seem only to differ in the knee voltage (between 1.7V and 3V). Thanks a lot. Luc |
by Luc1960
March 21, 2013 |

I'm fairly new to modeling LEDs but I've gathered this so far. Basically you have 3 main parameters to work with. These are: IS: Saturation current N: Emission coefficient RS: Parsitic resistance Modelling diodes and LED's is a two part process. The first being to fit the curve, the second to fit the slope. I recommend dc sweeping one of these parameters in turn while keeping the other two constant for all three and exporting the data into a graphing package (say Excel) to get a feel for their effects. You'll find that as IS is decreased the curve becomes less sharp but operates at a higher voltage. Increasing N increases operating voltage and decreasing RS increases steepness. For RS = 0, the basic function is: ID = IS (exp(VD/(N VT)) -1) A starting value for IS can found with: IS = ID / (exp(VD/(N VT)) -1) where ID and VD are the voltage and current of the LED at the point where the curve becomes linear (I think). Start with N=2... VT = 0.026 (this is a constant). Try plotting this function and curve fitting it in that region (the lower region), to the LED measurement data. Next estimate the other two parameters and run sweeps of your guessed ranges. You may have to re-adjust IS again to perfect the fit. This article is helpful and at the bottom there's a link to a useful spreadsheet which can speed up the process. http://electronics.stackexchange.com/questions/9510/how-do-i-model-an-led-with-spice |
by MartinOnCL
March 21, 2013 |

Note: I'm referring to linear plots in the above rather than log linear :) |
by MartinOnCL
March 21, 2013 |

Please sign in or create an account to comment.

CircuitLab is an in-browser schematic capture and circuit simulation software tool to help you rapidly design and analyze analog and digital electronics systems.