Searching \ for '[EE] IR detector interface' 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: 'IR detector interface'.

Exact match. Not showing close matches.
PICList Thread
'[EE] IR detector interface'
2005\11\28@134904 by Jamie Jensen

picon face
I have a circuit im breadboarding so I am transmitting a IR puslse to a open collector type reciever, being used to align two devices  If I connect the emitter to ground, then have a 150ohm pullup on the collector, I can see the voltage on the PIC input range from no IR signal of 5V, and down to about half a volt when i position the IR tx diode just right.

 So, question is, in order to increase the range can I simply burst pulse the LED with the max current (right now 30mA but it can accept up to 50mA) or change the pull up resistor to change the sensitivity?  Does my question make sense?  In other words, I see a range from 200 to 255 but I'd like to see the transistor turn on harder to pull it more to ground so which makes sense to do.

 JJ

               
---------------------------------
Yahoo! Music Unlimited - Access over 1 million songs. Try it free.

2005\11\28@144147 by Neil Baylis

picon face
Are you trying to use the collector voltage as an analog signal that
indicates the degree of alignment of the two parts?

If so, I suggest you look at using the TAOS TSL230BRP detector. This will
give you a digital output that you can connect directly to a PIC pin.

Using a simple photodiode and phototransistor in this mode will be
difficult. In any case, your collector resistor could be much higher..
perhaps 10k.

Are you trying to measure degree of alignment, or simply a binary signal..
aligned, or not aligned?


Neil

2005\11\28@160304 by Spehro Pefhany

picon face
At 10:48 AM 11/28/2005 -0800, you wrote:
>I have a circuit im breadboarding so I am transmitting a IR puslse to a
>open collector type reciever, being used to align two devices  If I
>connect the emitter to ground, then have a 150ohm pullup on the collector,
>I can see the voltage on the PIC input range from no IR signal of 5V, and
>down to about half a volt when i position the IR tx diode just right.
>
>   So, question is, in order to increase the range can I simply burst
> pulse the LED with the max current (right now 30mA but it can accept up
> to 50mA) or change the pull up resistor to change the sensitivity?  Does
> my question make sense?  In other words, I see a range from 200 to 255
> but I'd like to see the transistor turn on harder to pull it more to
> ground so which makes sense to do.
>
>   JJ

You can increase the value of the pullup resistor but it will make the
response proportionally slower. The LED will 'age' (decrease in brightness
with use) faster with higher current. Generally speaking, the 'sensitivity'
will not be very stable with time or temperature unless you close an
independent
feedback loop around the LED. Silicon photodetectors are relatively stable.

>Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spam_OUTspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff


2005\11\28@160327 by Mike Hord

picon face
What's your end goal?  Position/distance sensing?  Data
transmission?  Alignment detection?  Beam breaking
detection?

At any rate, your LED will almost certainly live far beyond
50 mA, if you're pulsing it.  Many LEDs will be quite happy
with short (say, 10 us) pulses above one or two amps.

You might need to be thinking about active amplification
beyond just a pullup.  Your open-collector sensor could
be used to drive another transistor, giving you a good
deal more gain, or an opamp based circuit, or any of a
number of solutions...

Mike H.

2005\11\30@042515 by Bill & Pookie

picon face
If you do pulse the LED, then the input on the pic pin will also be pulsed.

Why have you not already tried this?

You will need to decrease the value of the curmit limitting resistor to
increase brightness when you pulse LED.

What would happen if you used 2 pic pins to drive the LED.  One on steady
and the other pulsed.  Then would have best of both worlds.

Bill



{Original Message removed}

2005\11\30@043823 by Jinx

face picon face
> the emitter to ground, then have a 150ohm pullup on the
> collector, I can see the voltage on the PIC input range
> from no IR signal of 5V, and down to about half a volt
> when i position the IR tx diode just right

Isn't 150R a little low ? The receiver may not have enough
gain to completely overcome that much pull-up current


'[EE] IR detector interface'
2005\12\09@090329 by Jamie Jensen
picon face
Hi guys

 Ok, now back in town again I can get back to this project.

 Neil had asked if I was trying to use the collector voltage as an  analog signal that indicates the degree of alignment.  Its more  of....is it there or not there, like within its field of view.   Using something like the light-2-freq chip wont work as well because I  have designed it to be looking for either no IR reflection (black),  most IR reflection (white paper) or in between the two (open to ambient  light).

 Mike H asked the end goal...its a combination of things to detect  relative position and alignment detection.  I have connected the  LED thru a NPN transistor so I can pulse it on and off from the PIC, so  adjust the current limiting resistor should be able to give me  "brighter?" IR pulse?  How exactly does that work.  If I  pulse the IR with more current, does it increase the sensitivy of the  reciever circuit?  I've thought of a simple transistor amp but  also need to keep the costs down so trying it right now.  I can  also deal with slower response time with a higher pull up if that will  increase the range detection from lower to higher.  Oh, and the  LED is NOT driven directly from the PIC, but via the transistor for  swithing the ground pin on them, via the PIC.

 Jinx did ask if the 150ohm was low for a pull up on the collector  feeding the PIC.  Maybe it is,  but how do you calculate the  best value for the rx?  Obviosly I would like to have a lot higher  value resistor because this will be battery operated and I want to keep  the power consumption as low as possible.

 Again, thanks to everyone for the ideas and suggestions

 -JJ




                       
---------------------------------
Yahoo! Shopping
Find Great Deals on Holiday Gifts at Yahoo! Shopping

2005\12\09@101855 by Mike Hord

picon face
>   Neil had asked if I was trying to use the collector voltage
> as an  analog signal that indicates the degree of alignment.
>  Its more  of....is it there or not there, like within its field of
> view.   Using something like the light-2-freq chip wont work
> as well because I  have designed it to be looking for either
> no IR reflection (black),  most IR reflection (white paper) or
> in between the two (open to ambient  light).

So this is a reflective sensor?  Be certain to include some
provision for shielding the receiver from the transmitter- best
is a piece of metal between the two.  Some apparently opaque
plastics are quite transparent to IR (ask how I know).

>   Mike H asked the end goal...its a combination of things to
> detect  relative position and alignment detection.  I have
> connected the  LED thru a NPN transistor so I can pulse it
> on and off from the PIC, so  adjust the current limiting resistor
> should be able to give me  "brighter?" IR pulse?  How exactly
> does that work.

Think of the transistor as a current-controlled current source.
If you connect a PIC pin through a resistor to the base, then
the emitter to ground, your current into the base will be
(approximately) (Vcc-.6)/Rb= Ib.  For, say, a 2n2222, you
can expect the collector emitter current to be approximately
200 times that.  If you choose Rb such that the current into
the base is 1 mA, you'll get 200 mA through the collector.
If your LED is forward biased between Vcc and the collector,
that 200 mA will flow through it, too.  The appropriate forward
voltage drop will occur across the LED, and the remaining
voltage will form Vce.  Don't exceed your maximum power
rating on the transistor (Vce*Ic) and you'll be fine.

>  If I  pulse the IR with more current, does it
> increase the sensitivy of the  reciever circuit?

No, it increases the amount of light received.  That will
increase the response in a given situation, but not as much
as increasing the sensitivity of the receiver would.

> I've thought of a
> simple transistor amp but  also need to keep the costs down so
> trying it right now.

A good way to "read" that photodiode is shown on page 253 of 2nd
ed. Horowitz and Hill.  Basically, connect the photodiode such that
the anode (pointy end of the arrow) connects to the - input of an
opamp, cathode to the + input, connect the + input to ground thru
4.7 MOhm, and 4.7 MOhm negative feedback resistor from - input
to output.  Your output will be 10V/uA.  If you don't have photodiodes,
per se, you can use one of your IR LEDs as one.

Of course, remembering that the amount of light hitting that
photodiode (or phototransistor, or whatever it is) controls the amount
of current through it, and remembering that an NPN transistor can
work as a current-controlled current source, should be enough to
give you ideas....

Mike H.

2005\12\09@102931 by Neil Baylis

picon face
In this application (sensing presence of an object), you'd be wise to
use an integrated IR receiver. This is a 3 pin device that will output
a logic signal when your object is present. It's exactly what's used
to receive signals from a TV remote. These devices are very common,
and very cheap. This willl eliminate all problems with ambient light.
You will need to switch your LED on and off at about 40kHz.

Neil

2005\12\09@110859 by olin piclist

face picon face
Mike Hord wrote:
> Basically, connect the photodiode such that
> the anode (pointy end of the arrow)

No, the pointy end on the schematic is the cathode.

******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products

2005\12\09@125416 by Mike Hord

picon face
> > Basically, connect the photodiode such that
> > the anode (pointy end of the arrow)
>
> No, the pointy end on the schematic is the cathode.

Crap.  I have a heck of a time with that.  I thought I
had it figured out:  cations are positive, anions are
negative.

Wait, I see the problem.  Thanks to Ben Franklin,
current flows from the positive terminal to the
negative terminal, but that really represents the
opposite of the electron flow.  The cathode is the
device from which electrons flow, hence the more
negative terminal in fact.

Maybe now I can remember.  And if that's the only
gripe you (Olin) can find in an e-mail that I wrote
that's as long as that one was, I'll sleep a little
better tonight.  ;-)

Mike H.

2005\12\09@133658 by William Chops Westfield

face picon face
On Dec 9, 2005, at 8:09 AM, Olin Lathrop wrote:

>> the anode (pointy end of the arrow)
>
> the pointy end on the schematic is the cathode.

Arrgghh.  Ambiguity alert!  A diode symbol has a bar with a lead
at one end (this is the cathode.)  The other end has a triangle
with the other lead coming out of the base of the triangle.  that
lead is the anode.  The pointy end of the triangle touches the bar,
so it points toward the cathode, but the triangle is the anode side.

BillW

2005\12\09@154357 by Jamie Jensen

picon face
oh...I took the suggestion of someone  else...and I am using an integrated IR unit the QRD1114.  I hope  that doesn't change the conceptual idea?


                       
---------------------------------
Yahoo! Shopping
Find Great Deals on Holiday Gifts at Yahoo! Shopping

2005\12\09@160516 by Mike Hord

picon face
>  oh...I took the suggestion of someone  else...and I
> am using an integrated IR unit the QRD1114.  I hope
>  that doesn't change the conceptual idea?

No, but it may change the practical implementation.

Many integrated sensor units have extremely short focal
distances- sub-cm range.  The datasheet for your
device suggests that its peak response occurs at about
30-40 mils, and drops off to zero at about half an inch.
Pretty miserable, although they don't specify what they
were illuminating, so it could be a piece of black
construction paper for all we know (but I doubt it)(likely
you'll never see a response curve as good as they
show).

What range were you hoping for?  What are you trying
to detect?  I've had excellent luck using an SFH229FA
paired with an SFH485 LED- the 229's visible light
rejection filter is top notch.  With those, I've been able
to reliably detect objects over an inch away, and often
much further depending on the reflectivity of the object.

Mike H.

2005\12\09@195801 by Bill & Pookie

picon face
Electrons flow into the arrow.  Electricians and dog catchers have warped
views of the world.

Pookie

{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...