Maker Pro
Maker Pro

IR LED direction accuracy

P

Pimpom

Jan 1, 1970
0
A couple of weeks ago, I posted a question in s.e.b. about
sending a trigger pulse over some hundreds of metres. I apologise
for not responding to each reply. The reasons were social
obligations of the season and the death of my mother-in-law.

I'd previously rejected the idea of sending the pulses by IR, but
suggestions in seb made me consider it again. I've thought of
some techniques to extend the range and preliminary calculations
indicate that it's doable.

I've already been using IR links for other purposes over
distances of ~10m, but while experimenting to see how much
distance I can get without any special technique, I ran across an
unexpected snag: I found that the no-name IR LEDs from the local
market focus their beams in directions way off the LED axis.

So I've been looking for sources where I can get specific models
of IR LEDs with datasheets. I've tentatively selected the Vishay
TSAL6100 with a half-power beam width of 10 degrees, but it will
take at least 2 or 3 weeks for me to get the devices. Does anyone
here know if such branded LEDs can be relied on to project their
beams accurately along their axes?
 
P

Pimpom

Jan 1, 1970
0
John said:
Most lens-front LEDs do this, fire light in random directions.
Chip
placement isn't very accurate. The best way to focus LED light
is to
use a flat-face chip and add your own lens.

I was not really thinking of focussing the LED output. My
approach (so far theoretical) is to use an array of LEDs for
higher intensity. Their combined IR emission would be impossible
to focus into a parallel beam with a single lens. The lens will
rather be at the receiving end.

My concern with the direction of emission has to do with ease of
aiming the emitter at the receiver. According to the datasheet,
the Vishay TSAL6100 I mentioned has high intensity due to the
narrow angle: 130 mW/sr typ compared to 20-40 mW/sr for wider
angle types, and still has 90% intensity at +/- 4 degrees. 4
degrees is 7' at 100', or 28m at 400m. It shouldn't be difficult
to aim the beam with that degree of accuracy, provided that the
direction of peak emission is approximately along the axis of the
device.

However, if the direction of peak radiation for individual diodes
is unpredictable, it will also partly defeat the aim of combining
the IR outputs from several devices.
No. Even though the directionality graphs seem to peak at 0
degrees,
real parts will usually fire in odd directions. More datasheet
creativity.

Why not use a laser, maybe a laser pointer?
I've also been considering that and have not completely rejected
it. The only laser device I can get hold of without too much
hassle is a pointer.
 
P

Pimpom

Jan 1, 1970
0
Bill said:
It might be worth thinking about a packaged IR laser diode.
They cost
the earth, but the built-in optics offer a fairly narrow beam.
This is
an example that I found by a bit of googling - I've no idea if
you can
actually buy it, or how much it costs. Ostensibly similar parts
cost
around a hundred euro.

http://www.worldstartech.net/pdf/UT5-70G-830.pdf

Interesting. But I might have given the wrong impression by
mentioning the beam width of the TSAL6100 LED. There's no reason
to confine the beam to a narrow angle except insofar as to
concentrate the intensity at the receiver and thus extend the
usable range. The TSAL6100 has a stated radiant intensity of 130
mW/sr at 100mA compared to 20-40 mW/sr for wider angle types.

I was thinking of using several LEDs, something like 25 of them
in a 5x5 grid, and a lens at the receiver. But the effect will be
partly cancelled if they don't all radiate in the same direction,
at least within a very few degrees.
 
H

Hammy

Jan 1, 1970
0
Interesting. But I might have given the wrong impression by
mentioning the beam width of the TSAL6100 LED. There's no reason
to confine the beam to a narrow angle except insofar as to
concentrate the intensity at the receiver and thus extend the
usable range. The TSAL6100 has a stated radiant intensity of 130
mW/sr at 100mA compared to 20-40 mW/sr for wider angle types.

Look at the Data-sheet those LED's are meant to be pulsed and are
rated as such.Radiant intensity at 1A pulses goes up to 650 to
1000mW/sr.

What you could try is a light pipe this could be as simple as a bic
pen with the ink part removed.This would help to increase the focus
of the beam and improve the range, other then that you need a lense.
 
P

Pimpom

Jan 1, 1970
0
Jan said:

Ah yes, a burner laser. I'd forgotten about them. Thanks for
bringing it up. My son and I started a project using those last
year, but never finished it because other things took priority
over it. I wonder how practicable is the idea of projecting a
laser beam onto a receiver from 400m, especially since the system
has to be set up quickly, used for a day and then dismantled. An
advantage of using IR is that the receiver can be a standard
remote control receiver module.
 
J

Jan Panteltje

Jan 1, 1970
0
Ah yes, a burner laser. I'd forgotten about them. Thanks for
bringing it up. My son and I started a project using those last
year, but never finished it because other things took priority
over it. I wonder how practicable is the idea of projecting a
laser beam onto a receiver from 400m, especially since the system
has to be set up quickly, used for a day and then dismantled. An
advantage of using IR is that the receiver can be a standard
remote control receiver module.

Pointing the laser will be difficult.
But sending IR to a remote control receiver over 400 m will be next to impossible,
at least I had big problems with much shorter distances.
At least with a laser in the visible range you can see where it hits (points).
I have tried no more then 50 meters with a helium neon laser though, YMMV.

Are you using some kind of telescope at the receiving end, does it pass IR?
 
J

Jan Panteltje

Jan 1, 1970
0
PS
there is also a safety issue with lasers, the risk of
shining somebody in th eyes, somebody may topple your transmitter,
anything, aircraft..
I would use a radio link if I were you.
Cheap about 430 MHz modules in a free band are available from many places,
with digital IO too.
Most optical link announcements I have seen over time always stayed just that: announcements.
 
P

Pimpom

Jan 1, 1970
0
Hammy said:
Look at the Data-sheet those LED's are meant to be pulsed and
are
rated as such.Radiant intensity at 1A pulses goes up to 650 to
1000mW/sr.

The datasheet does not directly give a duty cycle vs. pulse
duration vs. current rating. The radiant intensity figure for 1A
is only for a single 100us pulse. The current vs. pulse duration
curves are drawn up to 1A for duty cycles up to 0.05, so I assume
that's the maximum permissible level. I'd like to use a duty
cycle of at least 0.25 with 25usec pulses. Interpolating the
curves implies that this would be safe only at around 0.25A max.
What you could try is a light pipe this could be as simple as a
bic
pen with the ink part removed.This would help to increase the
focus
of the beam and improve the range, other then that you need a
lense.

The problem with trying to focus the beam with a lens is that I
don't think a single LED will be powerful enough to cover the
distance, and a single lens will not be able to focus radiation
from multiple LEDs into a parallel beam. OTOH, a lens at the
receiving end will be able to focus the radiation from several
LEDs onto a single receiver.
 
P

Pimpom

Jan 1, 1970
0
Jan said:
Pointing the laser will be difficult.
But sending IR to a remote control receiver over 400 m will be
next
to impossible, at least I had big problems with much shorter
distances.
At least with a laser in the visible range you can see where it
hits
(points).
I have tried no more then 50 meters with a helium neon laser
though,
YMMV.

Are you using some kind of telescope at the receiving end, does
it
pass IR?

When I first started thinking seriously about this project, I
also quickly discarded the idea of using IR at such distances. I
felt that RF was the only realistic option. But then I got to
thinking about IR again and came up with this:

I've been using cheap no-name, no-datasheet receiver modules and
LEDs (see my opening post) successfully at distances of around
10m outdoors under bright sunlight. With high-intensity LEDs and
sensitive receivers, both with known characteristics, I hope to
get at least twice that range, i.e. 20m. The target distance is
20 times that or 400m.

An array of, say, 25 LEDs, will give 25/20^2 = 1/16th of the
radiant intensity at 400m.

The receiver's sensitive area is about 0.2" in diameter. A 2"
lens will catch x100 radiation. This gives a theoretical received
radiation of 100/16 = 6.25 times that with a single LED at 20m.

Factors that will inevitably reduce that figure in practice are -
1. Atmospheric absorbtion: No idea except a wild guess. At least
there's only minimal pollution at the site. There may be slight
seasonal haze.
2. Partial opacity of the lens glass to IR: Most sources found
with a quick google search for ordinary glass gave only vague
statements like "fairly transparent to near infrared", "passes
near IR very well", and so on. The one curve I found for window
glass gives a transmissivity of about 0.86 at 940nm. (I wonder if
a cheap magnifying glass will be better than a coated lens).
3. Imperfections in alignment and focus: If the Rx is somewhat
offset from the focal plane of the lens, it will reduce
efficiency, but will be more tolerant of alignment errors.

If all these factors reduced the received radiation to 0.2 of the
theoretical figure, it will still be in the same range as that
for a single LED at 20m. Now please feel free to point out any
flaws in my reasoning.
 
T

TTman

Jan 1, 1970
0
When I first started thinking seriously about this project, I also quickly
discarded the idea of using IR at such distances. I felt that RF was the
only realistic option. But then I got to thinking about IR again and came
up with this:

I've been using cheap no-name, no-datasheet receiver modules and LEDs (see
my opening post) successfully at distances of around 10m outdoors under
bright sunlight. With high-intensity LEDs and sensitive receivers, both
with known characteristics, I hope to get at least twice that range, i.e.
20m. The target distance is 20 times that or 400m.

An array of, say, 25 LEDs, will give 25/20^2 = 1/16th of the radiant
intensity at 400m.

The receiver's sensitive area is about 0.2" in diameter. A 2" lens will
catch x100 radiation. This gives a theoretical received radiation of
100/16 = 6.25 times that with a single LED at 20m.

Factors that will inevitably reduce that figure in practice are -
1. Atmospheric absorbtion: No idea except a wild guess. At least there's
only minimal pollution at the site. There may be slight seasonal haze.
2. Partial opacity of the lens glass to IR: Most sources found with a
quick google search for ordinary glass gave only vague statements like
"fairly transparent to near infrared", "passes near IR very well", and so
on. The one curve I found for window glass gives a transmissivity of about
0.86 at 940nm. (I wonder if a cheap magnifying glass will be better than a
coated lens).
3. Imperfections in alignment and focus: If the Rx is somewhat offset from
the focal plane of the lens, it will reduce efficiency, but will be more
tolerant of alignment errors.

If all these factors reduced the received radiation to 0.2 of the
theoretical figure, it will still be in the same range as that for a
single LED at 20m. Now please feel free to point out any flaws in my
reasoning.

Outdoor 'laser' Paintball by any chance ????
 
J

Jan Panteltje

Jan 1, 1970
0
When I first started thinking seriously about this project, I
also quickly discarded the idea of using IR at such distances. I
felt that RF was the only realistic option. But then I got to
thinking about IR again and came up with this:

I've been using cheap no-name, no-datasheet receiver modules and
LEDs (see my opening post) successfully at distances of around
10m outdoors under bright sunlight. With high-intensity LEDs and
sensitive receivers, both with known characteristics, I hope to
get at least twice that range, i.e. 20m. The target distance is
20 times that or 400m.

An array of, say, 25 LEDs, will give 25/20^2 = 1/16th of the
radiant intensity at 400m.

The receiver's sensitive area is about 0.2" in diameter. A 2"
lens will catch x100 radiation. This gives a theoretical received
radiation of 100/16 = 6.25 times that with a single LED at 20m.

I have read of some amateur project that used a parabolic reflector
and visible light (light bulb) on the transmission side,
telescope on the other end.
It is possible to slowly modulate a light bulb, or perhaps faster
with some LCD shutters in front (from some LCD 3D glasses).
Then you could use a powerful car headlight :)
LCD glasses can do 50 on / off flips per second at least mine can.
Kerr cell ? IIRC.
Hey I just invented that LCD thing.
Signal lights like that are as old as the world,
and with a normal telescope can bee seen far away,
and are not dangerous.

Factors that will inevitably reduce that figure in practice are -
1. Atmospheric absorbtion: No idea except a wild guess. At least
there's only minimal pollution at the site. There may be slight
seasonal haze.

http://en.wikipedia.org/wiki/File:Atmospheric_electromagnetic_opacity.svg
IR has more attenuation than radio, but it could work.


2. Partial opacity of the lens glass to IR: Most sources found
with a quick google search for ordinary glass gave only vague
statements like "fairly transparent to near infrared", "passes
near IR very well", and so on. The one curve I found for window
glass gives a transmissivity of about 0.86 at 940nm. (I wonder if
a cheap magnifying glass will be better than a coated lens).
3. Imperfections in alignment and focus: If the Rx is somewhat
offset from the focal plane of the lens, it will reduce
efficiency, but will be more tolerant of alignment errors.

If all these factors reduced the received radiation to 0.2 of the
theoretical figure, it will still be in the same range as that
for a single LED at 20m. Now please feel free to point out any
flaws in my reasoning.

Sounds OK, but then again, the proof of the pudding is in the eating.
My experiences with IR LEDs are not that hope giving, but it is
many years (30 or so) ago.
I was interested in this for transmitting simultaneous translation
to a group of people with receivers from a big IR LED panel.
Only indoors.
These systems exist.
But outdoors you will also get a lot of noise from daylight and other sources.
Give it a try, I am curious.
Anyways, for outdoors we went to radio.
 
T

TTman

Jan 1, 1970
0
SNIP

Sorry for the, perhaps, silly question. But why not use red LEDs? At
least then you can see them.... use your eyes to line things up. And
though it may not be important, you can turn the red on and off faster
than the IR diodes.
Is there some advantage to the IR diodes I don't know about? Are they
brighter?

Because IR detecters are the 'preferred' reception device....
You can use 632nm laser, (class1e.g. ) and a fibre optic array to collect
the beam.... as used by LaserRunner systems.
 
S

Spehro Pefhany

Jan 1, 1970
0
There's no spec for accuracy of beam alignment. Getting the body
aligned would require snug holes in a plate or something like that.
Sorry for the, perhaps, silly question. But why not use red LEDs? At
least then you can see them.... use your eyes to line things up. And
though it may not be important, you can turn the red on and off faster
than the IR diodes.

Is there some advantage to the IR diodes I don't know about? Are they
brighter?

George H.

Or just throw a 10W or 20W white LED at the problem (modulated and
with higher peak power), or perhaps a 3W with a lens.
 
P

Pimpom

Jan 1, 1970
0
George said:
Sorry for the, perhaps, silly question. But why not use red
LEDs? At
least then you can see them.... use your eyes to line things
up. And
though it may not be important, you can turn the red on and off
faster
than the IR diodes.

Is there some advantage to the IR diodes I don't know about?
Are they
brighter?
Don't know if they're brighter or not. But IR has at least one
advantage: I can choose from a range of receiver modules that
already incorporate amplification, AGC, interference suppression,
etc.

If necessary, I guess I could use a digital camera to monitor the
IR from the transmitter at the receiver's location and direct an
assistant to adjust the transmitter's direction. However, I don't
think that will be necessary as even the narrow-angle TSAL6100
LED outputs 90% of the axial intensity at 4 degrees off center. 4
deg is 7' at 100'. Yesterday, I marked out these distances in my
compound and decided that it shouldn't be difficult to aim the
transmitter with that level of accuracy.

All of this, of course, is *IF* the LED's peak radiation is along
its axis, at least approximately. Which was really my original
question.
 
C

Charlie E.

Jan 1, 1970
0
There's no spec for accuracy of beam alignment. Getting the body
aligned would require snug holes in a plate or something like that.


Or just throw a 10W or 20W white LED at the problem (modulated and
with higher peak power), or perhaps a 3W with a lens.

You have to be careful with massed IR LEDs, as they can cause eye
damage pretty easily! Be very careful with them...

Charlie
 
S

Spehro Pefhany

Jan 1, 1970
0
You have to be careful with massed IR LEDs, as they can cause eye
damage pretty easily! Be very careful with them...

Charlie

Yes, at least you can see how (literally) blinding a 20W white LED is,
and avoid it, as when looking at the sun.
 
N

nospam

Jan 1, 1970
0
Charlie E. said:
You have to be careful with massed IR LEDs, as they can cause eye
damage pretty easily! Be very careful with them...

And you know that because?
 
J

Jasen Betts

Jan 1, 1970
0
It might be worth thinking about a packaged IR laser diode. They cost
the earth, but the built-in optics offer a fairly narrow beam.
an example that I found by a bit of googling - I've no idea if you can
actually buy it, or how much it costs. Ostensibly similar parts cost
around a hundred euro.

spend a few bucks for a "keyring" laser pointer.
visible's light's much easier to aim than IR.
 
P

Pimpom

Jan 1, 1970
0
Rather than brute force this design with power, have you
considered
modulating the led rather than just a pulse, and then improve
the
detection by using a filter on the received signal. The essence
of any
communications scheme is apriori knowledge of the signal. Say
the goal
was a 10uS pulse. Suppose the beam was modulated with a CW
(plus DC
offset to keep the led on) at 1MHz. On the receiving end, set
up a
detector with some basic AGC and then a filter at 1Mhz. Do an
energy
detection on the filtered signal, which could be as simple as
rectification and DC filtering.

My intention, if I do decide to use IR, is to use a standard
remote control receiver module. I have some experience with this
and it already has all the features you mentioned -
amplification, AGC, frequency tuning, noise suppression. What
needs to be tackled is the problem of making the system work at
long distance.
I wouldn't go pointing lasers into the aether unless you know
it is
safe. [The military targets with a class IV laser, which sounds
kind
of dangerous to me, until you consider what is coming next.]

There's that to consider, of course.
 
Top