spice models for led

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
*Typ RED GaAs LED: Vf=1.7V Vr=4V If=40mA trr=3uS
.MODEL LED1 D (IS=93.2P RS=42M N=3.73 BV=4 IBV=10U
+ CJO=2.97P VJ=.75 M=.333 TT=4.32U)

*Typ RED,GREEN,YELLOW,AMBER GaAs LED: Vf=2.1V Vr=4V If=40mA trr=3uS
.MODEL LED2 D (IS=93.1P RS=42M N=4.61 BV=4 IBV=10U
+ CJO=2.97P VJ=.75 M=.333 TT=4.32U)

*Typ BLUE SiC LED: Vf=3.4V Vr=5V If=40mA trr=3uS
.MODEL LED3 D (IS=93.1P RS=42M N=7.47 BV=5 IBV=30U
+ CJO=2.97P VJ=.75 M=.333 TT=4.32U)
 
  • Like
Reactions: 1 user
These two LED models come with the library for LTSPice. I don't
quite know what type of LEDs they are, though. The QTLP has
about the same voltage drop as an ordinary red LED, although
somewhat lowish, perhaps. The other one has a much higher
voltage drop. Haven't checked what types they really are since
they are nothing I can buy locally anyway.



.model NSPW500BS D(Is=0.27n Rs=5.65 N=6.79 Cjo=42p Iave=30m Vpk=5 mfg=Nichia type=LED)
.model QTLP690C D(Is=1e-22 Rs=6 N=1.5 Cjo=50p Iave=160m Vpk=5 mfg=Fairchild type=LED)

Edit: I saw Pedja had posted some models in parallel with me.
Maybe they are better. Although obviously intended as generic
models, we at least know what they are supposed to model and
since they are modelled in a different way, they may be easier
to modify.
 
I'm a little down on LEDs for bias setting at the moment. A couple of months back, I took a few "standard" red LEDs from different sources and use a 9V battery, a resistor and a DMM to measure voltage drop. The same part no fro the same manufacurer was pretty consistent. Across manufacturers there was enough difference that I'm not so sure the idea of a "standard" LED (choose a color) is to be depended on. ( I was careful, I thought to avoid "extra bright" or other specialy LEDs.) Thus I think it a little uncertain to treat LEDs as "standard" like can treat 1N4148s as pretty much the same regardless of source. On the otherhand, I don't think the differences are show stoppers, just something to be aware of.

This isn't to say there is anything wrong with LEDs for setting bias, but that it may be a good idea specify manufacurer and part number. Also, I don't find LED data sheets to have the info I want as they are more oriented toward the otical properties - hardly surprising. Anyway, I'm still willing to use them but I measure the drop and make sure they all have the same part number. I'm also thinking of making a survey of the various specilaty colors and types -- maybe someday and extra bright blue will be just what I need.
 
I'm not a device physicist, but as far as I know, zener diodes with low voltage values have real zener breakthrough, which doesn't generate much noise (just a bit of shot noise, I believe), while "zener" diodes with higher values have avalanche breakthrough, which generates enormous noise. Zeners around 5V have a bit of both types of breakthrough.

So yes you can use zener diodes, provided that they have low voltage values (well below 5V; I ususally don't use anything above 3.9V, 3.6V if noise is critical).

Low-voltage zeners have the disadvantage that their differential resistance is rather high. When you bias them at 2mA, the voltage can be hundreds of millivolts below the specified value, which is usually tested at 5mA for low-voltage 400mW zeners.
 
sam9 said:
Across manufacturers there was enough difference that I'm not so sure the idea of a "standard" LED (choose a color) is to be depended on.

I've seen a tendency to use the Red LEDs. The graphic below is from a Fairchild LED datasheet. From it, one can see that the red LED has a steeper If v Vf curve. This is desirable because the voltage drop is less susceptible to changes in current. (In the case below, forward current is rather high: >10mA, but I think this phenomena holds true for the lower current varieties as well.)

fscarpa58 said:
Do you think one can use zener diodes instead of LEDs in real world applications or there are some drawbacks, e.g. noise , etc.?

FWIW: I've read (but not seen data) that LEDs are quieter than zener diodes. I would like to find actual noise data for LEDs.


JF
 

Attachments

  • vf curves.jpg
    vf curves.jpg
    21.9 KB · Views: 1,850
johnferrier said:


I've seen a tendency to use the Red LEDs. The graphic below is from a Fairchild LED datasheet. From it, one can see that the red LED has a steeper If v Vf curve. This is desirable because the voltage drop is less susceptible to changes in current. (In the case below, forward current is rather high: >10mA, but I think this phenomena holds true for the lower current varieties as well.)



FWIW: I've read (but not seen data) that LEDs are quieter than zener diodes. I would like to find actual noise data for LEDs.


JF

John,

please forgive the ignorance of a non-native speaker, but
what colour is "HER"??? :confused:

Regarding noise, I think there once was a claim posted on
the fourm that green LEDs are the lest noisy ones. There
was no motivation or reference, though, as far as I can
remember.
 
For protecting mosfets (zener as a shunt across gate and source) I have used with good success a reverse biased base-emitter junction of generic BC550 or BC560. Capacitance is low and the zener voltage is quite the same across different devices.

Cheers,

Gerhard
 
johnferrier said:

From it, one can see that the red LED has a steeper If v Vf curve. This is desirable because the voltage drop is less susceptible to changes in current.

My measurements are similar.
I once measured the red, green and yellow ones I had.
Biased them from 1.0mA to 2.0mA, with steps of 0.1mA and put the voltages accross them in a spreadsheet.

Red was best, then green and yellow had a considerably less steep curve.

Note, this is with my LEDs, YMMV of course.
 
Out of curiosity I checked the datasheets of some LEDs from
Everlight, which is what the local supplier here sells. I basically
just checked a couple of 3mm LEDs all having their typical
values rated at 20mA. Their mileage does seem to vary indeed.

All diodes in the range from yellow to yellow green had an
almost linear characteristic, while the red ones had a more
normal exponential type curve and the blue ones were rather
inverse exponential. I checked the voltage variation
between 10mA and 30mA for these and there were no big
differences, but yellow, sunset orange and yellow green seemed
the to have the smallest voltage variation, ie. the steepest
curves in this range.
 
diyAudio Retiree
Joined 2002
She LED me on........

"I would like to find actual noise data for LEDs."

This was discussed on the forum. Try site search engine.

I believe John Curl did say that there was a drop in noise as the bias current exceeded about 2 mA. The comparisons of dynamic impedance between colors is interesting but it would seem that the size of the LED would be more important. For any more investigations into this the impedance of the LED in the 1 mA to 5 mA region would be of the most interest, since that is the range in which they will likely be used for voltage references (for high supply rejection and low noise). The rejection of noise from the supply may be more important the voltage noise generated by the device itself, since the noise from the supply appearing across the LED will often be the greater voltage. This is definitely the case for resistor bias of the LED and CCS bias is highly recommended. It seems that LED size would be more important with large and possibly high brightness LED having the lowest impedance.
 
When using as diode of any kind in a CC source it's pretty common to bypass with a 47uF cap primarily (I think) to shunt AC from the rails to ground. This leaves me wondering whether as a secondary effect it also serves as a snubber for the diode. If that's the case the concern with the diode as a noise source would be reduced.

This is idle speculation. I'm not asserting this as something I know.
 
Re: She LED me on........

Fred Dieckmann said:
"I would like to find actual noise data for LEDs."

This was discussed on the forum. Try site search engine.

Hi Fred,

Yes, I was aware of those discussions. There was no data. The most conclusive thing I read was that MAT03 datasheet shows a super low noise amplifier that uses an LED for a voltage reference. So it was implied that if it was good enough for PMI/Analog Devices, then it must be low noise. (Additionally, yes, Walt Jung uses an LED in his low noise regulator as well.)

Former thread that mentions LED noise:
http://www.diyaudio.com/forums/showthread.php?s=&postid=13794&highlight=#post13794

In the past, I've google search for "LED" and "noise". I found no LED datasheet that indicated a noise spec.

Though it sounds as if others are interested too, it's just a matter of personal curio.


JF
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.