Modeling LED in LTspice

[Patrick66]

Hello everyone, I wanted to model the 10W LED based on the datasheet below. In total, I want to use two 10W LED and the total Rd = 5.56, Vd = 21.5V. Can anyone help me to check whether this is the correct method to model this LED based on the current to voltage graph. Thank you in advance.

[Zero999]

I'm not sure if I understand the question.

How accurate does it need to be?

I would just use the built-in diode model.

Looked at the graph on the data sheet.

Set the Vfwd to a bit less than the forward voltage specified when the current is zero.

Calculated Ron by from the voltage vs current slope, where it's most linear.

Set Epsilon to 0.5 because it gave a more accurate curve, than 1, as suggested in the article linked below.

https://www.analog.com/en/resources/technical-articles/ltspice-simple-idealized-diode.html

You can also try using the standard diode model, with many in series and parallel, with a series resistor.

[Jay_Diddy_B]

Hi,

This model uses the diode model D. I played with the parameters N and RS until the VI curve was similar:




Versus the datasheet VI curve:



This should be close enough for most purposes.

I have attached the LTspice file.

Regards,
Jay_Diddy_B

[MarkT]

Model it with what it is, 4-series-4-parallel array of 3.4V 175mA LEDs?

[Patrick66]

Hello Sir, I tried using your method to model an LED based on the datasheet below. Just wanted to know whether this is the correct way of modelling it? Hope to hear from you soon thank you.

[Zero999]

Please post the .asc file.

[Patrick66]

Can I model it in the form of a resistor and voltage source with an ideal diode? Where the resistor is Ron = 6 ohm and voltage source Vfwd = 20.2V when considering two 10 W led?

[Patrick66]

Here is the model for the 3W LED

[Patrick66]

Hello Sir, I have another question that I would like to ask relating to the use of two 10W LED lights. The datasheet stated that the LED produces 10W with ideal current of 700mA. Therefore, I tried simulating an open loop flyback converter with 700mA but the output voltage that I got was around 25.2V with resulting in a 17.64W system instead of 20W system. So for my calculation should I consider my system to be an 18W system that is having an output voltage of 25.2V and current of 700mA or output voltage of 28.57V and current of 700mA?

[Zero999]

Here is the model for the 3W LED

That looks fine.

Hello Sir, I have another question that I would like to ask relating to the use of two 10W LED lights. The datasheet stated that the LED produces 10W with ideal current of 700mA.

10W of what? It won't be 10W of visible light. Probably between 20% and 40%. The rest will be heat.

Therefore, I tried simulating an open loop flyback converter with 700mA but the output voltage that I got was around 25.2V with resulting in a 17.64W system instead of 20W system. So for my calculation should I consider my system to be an 18W system that is having an output voltage of 25.2V and current of 700mA or output voltage of 28.57V and current of 700mA?

It's not a good idea to go open loop. Always include some current limiting.

Regarding the power: it's normally a bad idea to run an LED at its maximum rating. De-rated to 80%. There will be something like a 10% drop in light output, as LEDs are less efficient at high currents, but a much longer life, as it'll run cooler.

[Patrick66]

Hello Sir, can I use combinations of diode, resistor and voltage source to represent my LED? Currently for the two 10W LED, I'm using an ideal diode, resistor of 5.56 ohm and voltage source of 21.4V to model two 10W LEDs. The reason that I use this method instead of the other method of changing the IS, Rs, N value is because I'm able to get a more accurate result using the diode, resistor and voltage source method. Thank you sir for your help.

[Zero999]

Hello Sir, can I use combinations of diode, resistor and voltage source to represent my LED? Currently for the two 10W LED, I'm using an ideal diode, resistor of 5.56 ohm and voltage source of 21.4V to model two 10W LEDs. The reason that I use this method instead of the other method of changing the IS, Rs, N value is because I'm able to get a more accurate result using the diode, resistor and voltage source method. Thank you sir for your help.

That will work, but why not learn a bit more about how SPICE works?

Here are some articles.
https://ltwiki.org/LTspiceHelp/LTspiceHelp/D_Diode.htm
https://www.analog.com/en/resources/technical-articles/ltspice-simple-idealized-diode.html
https://www.analog.com/en/resources/technical-articles/ltspice-combining-multiple-model-instances-into-one-symbol.html

Also note that this really isn't too critical. In real life, the forward voltage varies widely, from part to part, at different temperatures and drive currents. There's little point in perfection.

[Patrick66]

Thank you so much Sir for the help.