The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: firstname.lastname@example.org (Paul Mathews)
Subject: Re: LED Pulsing - Apparent Brightness?
Date: 8 Jan 2003 09:12:51 -0800
NNTP-Posting-Date: 8 Jan 2003 17:12:51 GMT
Kevin McMurtrie wrote in message news:...
> In article ,
> email@example.com (Paul Mathews) wrote:
> >Kevin McMurtrie wrote in message
> >> In article ,
> >> Mark Klinger wrote:
> >> >I am experimenting with wireless communication via LEDs, and have a
> >> >simple question of apparent brightness vs. actual light output.
> >> >Keeping in mind that my receiver will be a photo detector (Fairchild
> >> >L14N2) I'm trying to understand brightness.
> >> >
> >> >At DC, 20mA is the max drive for my LEDs and they are extremely
> >> >bright. If, however, I put them in my circuit and pulse them with a
> >> >width of around 1mS, they are barely even illuminated.
> >> >
> >> >Will the distance of the transmission suffer since they are
> >> >"apparently" not as bright, or are they actually the same brightness
> >> >as at DC to the photodetector, but my eyes just aren't seeing that?
> >> >
> >> >Ultimately what I am wondering is if what the photodetector sees is
> >> >effected by the short pulse duration, or is it equally as bright at
> >> >1mS as it is at DC?
> >> >
> >> The pulse durration matters. A reduced duty cycle makes the pulses
> >> harder to see above noise. If bandwidth isn't an issue, you can make a
> >> receiver that's narrowly tuned to those pulses. Over time it will be
> >> able to find the signal.
> >> Think of this as an analogy: If somebody tapped together drumsticks in
> >> a crowd once every minute, it would take you a long time to notice it.
> >You see the average power for short pulses. You can pulse most LEDs
> >at higher currents than rated....easily 100 ma for 1ms pulses,
> >provided that the average is within ratings. Peak intensity may even
> >be slightly higher in the pulsed case if the LED is used with a
> >heatsink and the average current is lower than 20 ma. This is because
> >LEDs have higher output at lower temperatures.
> >I don't agree at all with the 'drumsticks' analogy. If you 'tune'
> >your receiver to be especially sensitive for the particular pulse
> >shape you use, you can easily exceed the detection performance for DC
> >excitation. This is because the ambient environment is mostly steady
> >illumination...brief events do, in fact, stand out if you know how to
> >tune for them.
> >Many experimenters are using IR remote chips to do this type of
> >detection. You can find a wealth of info by searching on phrases like
> >"IR remote". Throw in a few more keywords like "schematic" or "robot"
> >to narrow down the results. There's more info here:
> >Paul Mathews
> I didn't mean to imply that DC was better. The sensor has to operate at
> a frequency higher and narrower than other light sources. A 50% duty
> cycle would be best.
> (Funny story - I was once getting an interference pattern on my TV that
> corresponded to the IR remote Power command. Every time I walked in
> front of the TV, it turned off!)
Contrary to popular belief, 50% duty factor is not necessarily best.
Most commercial high sensitivity photelectric sensors operate at very
low duty factors and high peak intensity. Again, this is because the
ambient environment has predominantly low frequency energy, so
shifting the modulation spectrum upwards pays off.
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup