Simulate LED circuit

Hi everyone,

I am trying to build a circuit to recreate the graph (forward current of LED vs forward voltage) shown in the Luxeon rebel color data sheet p.14.

I looked up the specs for that particular model (green)
.model LXML-PM01-average D(Is=8.0330e-8 Rs=.4818 N=7.0005 mfg=Lumileds type=LED)
and added the Is, Rs and N parameters to a D1N4148 in pSpice.

The data sheet states that the typical forward voltage of the green LED is 3.05V at a current of 350mA, I might be completely misinterpreting something, but I thought that the forward voltage of the LED would be (350mA x Rs).

I used a supply voltage of 10V and calculated my series resistor from (10-3.05)/R =350mA

The circuit is then a 10V supply with a 20 Ohm resistor and I did a DC sweep from 0-10V and plotted I(D1) Vs. V(D1), which gives me a graph that does not look very like the one on the datasheet.

Would anyone be able to help me by telling me what I am doing wrong? Am I misinterpreting something completely?

Any guidance would be greatly appreciated!

Hi there,


First off, you dont get a forward voltage based on Rs alone. The diode part (N and Is) contribute to the voltage drop too, and quite a bit.
Where did you get the Is, Rs and N parameters from, and where did you find the data sheet?
It is usually possible to emulate an LED using the spice "D" general model. If you got those three parameters from a reputable source then maybe something else is wrong, or else those parameters need adjustment for your exact LED.
BTW, how far off is it?

Hi, I actually got the parameters from the Luxeon website so they should be accurate enough. I think it was the default parameters for some of the other parameters in the diode that may have altered the results a bit. Although even after I changed those, the parameters on the website don't seem to be 100% spot on, but it doesn't matter too much for my measurements!

Hi again,


If you want to supply a link to the data sheet i'll take a look and see if we can adjust the param's a bit to match the data sheet, if that's what you want to do.
LED curves on data sheets are not perfect anyway. Usually what you have to do is assume a wide enough characteristic voltage to cover any LED you buy with that part number. You test your circuit with different characteristic voltages to make sure it can handle any of those LEDs because even the same part number can vary from piece to piece somewhat.

Here is how LTSpice models a similar Lumileds LED:

Hi Guys,

I am having the same problems with the same parameters listed above.
What exactly is the relationship between the:
.model LXML-PM01-average D(Is=8.0330e-8 Rs=.4818 N=7.0005 mfg=Lumileds type=LED) and the
.model lxlh-bw02 D(Is=4.5E-20 Rs=0.85 N=2.6 Cjo=1.18n Iave=400mA Vpk=5 mfg=Lumileds type=LED)

with LXML-PM01 Adding Vfwd~3. to the parameters is really giving some strange results its almost like the diode gains lots of resistance from it

Any Ideas?