LEDs not used below 5-10mA

[lilmantis]

                           Hello everyone, I've been looking through some LED datasheets and I have seen in many that the LEDs should not be used below 5mA to 10mA depending on manufacturers. Does anyone know what is the reason for this?

[lilmantis]

Some examples:

http://www.digikey.com/product-detail/en/osram-opto-semiconductors-inc/GW%20JDSMS1.EC-FSFU-5C8E-L1N2/GW%20JDSMS1.EC-FSFU-5C8E-L1N2-ND/4834169

http://www.digikey.com/product-detail/en/osram-opto-semiconductors-inc/GW%20JDSMS1.EC-FSFU-5C8E-L1N2/GW%20JDSMS1.EC-FSFU-5C8E-L1N2-ND/4834169

[Rerouter]

I dont understand what you mean, I had a solid look through those datasheets and many of the performance graphs extended to 200uA

[lilmantis]

The info is at page 15 in the datasheet, on the max. permissible forward current graph.

[amyk]

Probably a "we don't guarantee functionality or any of the specs" thing.

[MatthewEveritt]

Given that it only appears on that one plot I suspect it really means don't use over about 95C, no matter how much you reduce the current.

[lilmantis]

The 5mA is specified over 0 to 100 degrees.

[izx]

                           Hello everyone, I've been looking through some LED datasheets and I have seen in many that the LEDs should not be used below 5mA to 10mA depending on manufacturers. Does anyone know what is the reason for this?

Important to note this isn't a dinkly little If=20mA indicator LED, but a power lighting LED with typical If=120mA. Look at the relative intensity curve on page 13 (reproduced below), and notice that the intensity curve on the Y-axis disappears right around where 5 mA would be on the X-axis. So Osram is saying that typically this LED will have no effective output below If=5mA.

Edit: our regular If=20mA LEDs will happily emit light visible to the naked eye down to at least 0.5-1mA in my experience



BTW, dimming lighting LEDs in general is more commonly done with PWM rather than a linear constant current driver.

[ovnr]

Definitely "We don't characterize this below 5mA, so you're on your own". Osram datasheets are generally terrible in any event.

[izx]

Osram datasheets are generally terrible in any event.

The double-Deutsche everywhere REALLY messes up my normal datasheet speed scan routine

[lilmantis]

                       I was wondering if there is some effect I haven't heard about for certain LEDs. Anyway it is quite an absolute statement instead of "not caracterized below 5mA".

[wraper]

This is a white led for lighting. At the low current, color spectrum will degrade significantly.

[alank2]

I have found that some LED's require you to throw a lot of current at them and others can light brightly on relatively little current.

I've had good luck with these:

APT2012CGCK
APT2012SURCK

[jitter]

                       I was wondering if there is some effect I haven't heard about for certain LEDs. Anyway it is quite an absolute statement instead of "not caracterized below 5mA".

"Do not use below 5 mA."  is definitely a statement that doesn't leave any room for (mis)interpretation.

Look at the end of the datasheet, one of the notes explains that basically, all values are typical statistical values. In other words: every LED is unique and differs slightly from any other LED of the same type. My educated guess is that below 5 mA, there's way too much variation to be able to specify a statistical value.

From the datasheet:

Typical Values:
Due to the special conditions of the manufacturing processes of LED, the typical data or calculated correlations of technical parameters can only reflect statistical figures. These do not necessarily correspond to the actual parameters of each single product, which could differ from the typical data and calculated correlations or the typical characteristic line. If requested, e.g. because of technical improvements, these typ. data will be changed without any further notice.

[mikeselectricstuff]

It means you may see significant brightness variation between parts below the stated current. This is why dimming is done with PWM rather than current control.
At very low currents, <1mA for parts specced at 20mA, brightness variations can be huge, even with parts from the same bin.

[ConKbot]

+1 To what mike says, big variations between LEDs, plus some power LEDs may not even light fully or evenly with too little current. Not talking about multi-chip types either. Some of the Luminus Devices LEDs that run at 9A max, unspecified below 1A, with a warning that it might not even light at 1A. and on candlepowerforums, some folks had issues with this with some multi-mode drivers on the lowest setting.

http://www.luminus.com/products/Luminus_SST90_Datasheet.pdf

"Special design considerations must be observed for operation under 1A. Please contact Luminus for further information."

[retrolefty]

                           Hello everyone, I've been looking through some LED datasheets and I have seen in many that the LEDs should not be used below 5mA to 10mA depending on manufacturers. Does anyone know what is the reason for this?

I do know that most standard arduino boards use 1Kohm series resistors for the power and pin 13 leds which is below 5 ma and they are still quite bright. So generalizations are of little value, just specific LEDs characterizations.

[uncle_bob]

Hi

Sitting here on the shelf, they have zero current flowing. When I unplug the lighting fixture, they have zero current. If "do not use below 5 ma" is taken literally as phrased, then you *must* keep current on them at all times.

A far more accurate phrasing would be "you would be stupid to use this below 5 ma" (which is quite correct and a very similar phrase).

Bob

[starphot]

  Bad phrasing. I noticed color shifts/current on those early dinky T-1 3/4 white LEDs, but I don't think that anyone would operate these ones in question below 5 ma anyway. These are for lighting. PWM dimmers are usually used on these anyway. I got similar high current LEDs at the cabin operating off of a 12v solar system with dimmers.

Joe