Searching \ for '[EE] Outdoor IR object detection' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/irs.htm?key=ir
Search entire site for: 'Outdoor IR object detection'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Outdoor IR object detection'
2007\07\16@231434 by Rosie & Dan Gaudenti

picon face
Hi all,

I have a need to detect an object passing by a phototransistor, 2” to 6” away.  Indoors is easy: I have an IR LED reflecting off the object.  (I can not have the object break a light beam from the LED.  Everything must be contained in a single, small box.)  When the object passes by, I get an increase in voltage.  Outdoors, I expect the object to block ambient light and give me a lower voltage when it passes.  This works most of the time.  Sometimes I get a higher voltage (which is OK) but sometimes I get no or very little change.  I have a narrow tube over the phototransistor about 2” long to make it more directional.  It has to work with the sun in any position.  You would think it would work best when pointed in the direction of the sun (with the object blocking the sunlight), but this is actually one of the worst conditions, especially in the afternoon.  Maybe extra IR is getting in from hot surroundings?

Any idea why I don't always get a voltage change, and how I could possibly make this work?

Thanks much for any suggestions,

Dan

2007\07\17@003805 by Jinx

face picon face
> how I could possibly make this work?

Hi Dan, is the IR modulated ?

What you'd ideally aim for is to make the ambient light level the
same whether indoors or outdoors. Or at least minimise to the
max ambient light outside so the IR can stand out. Do you have
a specific light filter ?



2007\07\17@034722 by Russell McMahon

face
flavicon
face
> I have a need to detect an object passing by a phototransistor, 2"
> to 6" away.  Indoors is easy: I have an IR LED reflecting off the
> object.  (I can not have the object break a light beam from the LED.
> Everything must be contained in a single, small box.)  When the
> object passes by, I get an increase in voltage.  Outdoors, I expect
> the object
/>

You are more liable to get reliable results by sending a signal with
chosen characteristics and detecting its reflection.

A simple method is to apply a constant frequency modulation to
transmitted  IR and look for the reflected signal of the same
frequency. Filtering for the desired signal allows you to do this with
good discrimination. More on this if interested


       Russell

2007\07\19@000832 by Rosie & Dan Gaudenti

picon face
I have thought about putting out a square wave and looking for that with the phototransistor, but wouldn't the sunlight be so much brighter that the phototransistor won't even detect light from an LED?  Indoors is not a problem; when I'm out in the sun I have problems.  I guess the only way to know is to try it.

Yes, if you have more info I am definitely interested.

Thank you for the reply . . .



> I have a need to detect an object passing by a phototransistor, 2"
> to 6" away.  Indoors is easy: I have an IR LED reflecting off the
> object.  (I can not have the object break a light beam from the LED.
> Everything must be contained in a single, small box.)  When the
> object passes by, I get an increase in voltage.  Outdoors, I expect
> the object
/>

You are more liable to get reliable results by sending a signal with
chosen characteristics and detecting its reflection.

A simple method is to apply a constant frequency modulation to
transmitted  IR and look for the reflected signal of the same
frequency. Filtering for the desired signal allows you to do this with
good discrimination. More on this if interested


       Russell

2007\07\19@001616 by Rosie & Dan Gaudenti

picon face
Outdoors, I am using ambient light.  So, no, I haven't tried modulation yet.  I also haven't tried a filter.  Wouldn't a filter reduce both sunlight and LED light equally and still leave me with the same problem?  I'm thinking about using an IR-blocking filter outdoors, but that would also block light from the IR LED.  Maybe I should use a visible light LED with an IR-blocking filter?

Thanks for the reply . . .




> how I could possibly make this work?

Hi Dan, is the IR modulated ?

What you'd ideally aim for is to make the ambient light level the
same whether indoors or outdoors. Or at least minimise to the
max ambient light outside so the IR can stand out. Do you have
a specific light filter ?


2007\07\19@012419 by Brent Brown

picon face
On 18 Jul 2007 at 21:08, Rosie & Dan Gaudenti wrote:
> I have thought about putting out a square wave and looking for that
> with the phototransistor, but wouldn't the sunlight be so much
> brighter that the phototransistor won't even detect light from an LED?
>  Indoors is not a problem; when I'm out in the sun I have problems.  I
> guess the only way to know is to try it.

IR from sunlight is more or less DC (low frequency as clouds vary the suns
brightness), so using AC coupling in your phototransistor amplifier circuit will reject
most of the effects of ambient light. Similarly, IR from incandescant lights will be
100 or 120Hz, so if you choose a suitably high frequency for your signal you can
quite easily filter out everything else.

--
Brent Brown, Electronic Design Solutions
16 English Street, St Andrews,
Hamilton 3200, New Zealand
Ph: +64 7 849 0069
Fax: +64 7 849 0071
Cell: +64 27 433 4069
eMail:  spam_OUTbrent.brownTakeThisOuTspamclear.net.nz


2007\07\19@020749 by Jinx

face picon face
> I haven't tried modulation yet

If you use a standard, eg 38kHz, you can easily find receiver/
amplifiers that will do a lot of the work for you. Pretty much all
the work really. The 1SU60 is readily available, retail/catalogue,
usually around NZ$8

Peter Crowcroft has suggested the Waitrony IR Receiver Module

http://www.kitsrus.com/pdf/pic1018scl.pdf

Another option is the IR receiver from an appliance like a TV
or VCR. These are in a small metal can and are very sensitive
and selective. As you know, a remote control pointed in just
about any direction will still be detected. They are also 3-pin
devices like the 1SU60. Unlike the 1SU60 they can be got for
nothing from a dead appliance. Also the front panel of the
appliance generally has an IR filter in the fascia

AFAIK receiver outputs are normally high and go low when they
detect 38kHz. ISTR that you need a data signal impressed on
the 38kHz to keep the receiver output going. The receiver strips
the 38kHz out of the signal, leaving the data. For example 1kHz
modulated with 38kHz, such as

http://home.clear.net.nz/pages/joecolquitt/irtx2.gif

with 1kHz on the 'data in'. In Fig2b you'll see a Zetex transistor.
ZTX are par excellence for driving IR LEDs. They switch very
quickly and cleanly and have low saturation, unlike other transistors
of similar current rating (BC337, BC639, TIP etc) which may
require a lot of base drive

> I also haven't tried a filter.  Wouldn't a filter reduce both sunlight
> and LED light equally and still leave me with the same problem ?

The difference is that the IR in sunlight is longer wavelength than
the IR you'd be sending. You also have a carrier frequency not
present in sunlight or indoors (lighting) that can be amplified after
a bandpasss filter (in the receiver modules above) has isolated it

It's been said that a blank exposed and developed film frame,
as you might get at the end of a roll (or if you left the lens cap
on, d'oh !) makes a good filter for this, but I've not tried it

> I'm thinking about using an IR-blocking filter outdoors, but that
> would also block light from the IR LED.  Maybe I should use a
> visible light LED with an IR-blocking filter?

Nah, stick with IR. Long ago I made a remote for an old camera,
using a small motor to depress the shutter release. With 3 x 5mm
IR LEDs and the receiver down a short tube, it worked fine up to
about 15m outside. IR LEDs can be driven very hard, a few amps
in pulses, and the light is detectable at quite a distance

2007\07\19@043852 by Alan B. Pearce

face picon face
>Outdoors, I am using ambient light.  So, no, I haven't tried
>modulation yet.  I also haven't tried a filter.  Wouldn't a
>filter reduce both sunlight and LED light equally and still
>leave me with the same problem?

The answer is 'it depends' ;). The problem may be that the sunlight is so
bright it saturates the sensor, so no matter what change in light you make,
the sensor sees no change in detected light, unless you can reduce the
ambient background light level.

>I'm thinking about using an IR-blocking filter outdoors, but
>that would also block light from the IR LED.  Maybe I should
>use a visible light LED with an IR-blocking filter?

Hmm, don't know if you can get an IR blocking filter, unless you use lenses
from sunglasses, that may do it. The normal way of doing it is the opposite,
block visible light and pass IR.

But using bursts of 38kHz modulation is the standard way of increasing
sensitivity, because it is then possible to AC couple and amplify with a
bandpass filter to select the 38kHz signal. Because it is AC coupled and
bandwidth limited, it is possible to use a large amount of amplification to
recover the desired signal from high ambient light levels.

2007\07\19@093326 by Recon

picon face
Rosie & Dan Gaudenti wrote:

>I have thought about putting out a square wave and looking for that with the phototransistor, but wouldn't the sunlight be so much brighter that the phototransistor won't even detect light from an LED?  Indoors is not a problem; when I'm out in the sun I have problems.  I guess the only way to know is to try it.
>
>Yes, if you have more info I am definitely interested.
>
>Thank you for the reply . . .
>
>
>
> > I have a need to detect an object passing by a phototransistor, 2"
>  
>
>>to 6" away.  Indoors is easy: I have an IR LED reflecting off the
>>object.  (I can not have the object break a light beam from the LED.
>>Everything must be contained in a single, small box.)  When the
>>object passes by, I get an increase in voltage.  Outdoors, I expect
>>the object
>>    
>>

About 20 years ago I worked for company that had a lot of glass jars
(400/min.) going down a conveyor. At one section they went  past a
window.  the varying light caused all kinds of problems.  at first a
simple card board box set over the sensor point solved the problem.  We
later installed  a detector but it used a special reflector.  The light
beam went through a polarizing lens when it hit the reflector it would
some how turn the polarization 90 deg. the photo transistor had a lens
in front of it also. The range of this was about ten feet from sensor to
reflector.  If I remember right it was also modulated with several
settings so multible sensors could be used near each othe with out
interference.

Where this might not work for your application have you considered ultra
sonics.  A lot of people have used the unit out of old polaroid
cameras.  or you can buy a unit here
http://www.maxbotix.com/Maxbotix__Buy_Now.html

Do a google search for polaroid sonar.

RECON


2007\07\19@145110 by Richard J. Pytelewski

picon face
Indoors may not be as easy either.... with the increased use of plasma TV's
and fluorescent lighting, these may cause interference with any IR by
swamping out your signal in "noise" from these sources .... I have had this
occur and I had to use a "tuned port" (actually a old plastic film canister)
to act as a "shield" for these sources of IR for the detector.

Rich

{Original Message removed}

2007\07\25@175000 by Barry Gershenfeld

face picon face
...7/16/07... Dan Gaudenti wrote:
>Outdoors, I expect the object to block ambient light and give me a lower
>voltage when it passes.  This works most of the time.  Sometimes I get a
>higher voltage (which is OK) but sometimes I get no or very little change.
>  ...
>Any idea why I don't always get a voltage change, and how I could possibly
>make this work?

For whatever reason--transmission, reflection, geometry or whatever--if
sometimes you get a higher voltage and sometimes you get a lower voltage,
then it follows that sometimes the two situations will combine and you get
something in between; namely, no voltage change.


More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...