DPRG
DPRG List  



[DPRG] fairchild QRB1133/QRB1134

Subject: [DPRG] fairchild QRB1133/QRB1134
From: Dale Wheat dale at dalewheat.com
Date: Fri Sep 5 12:00:02 CDT 2003

Chris,

> I would like to use a few Fairchild QRB1133/QRB1134 reflective object
sensors
> as a line sensor.
>
> (1) What interface circuity do I need to build?

I use the QRB1134 parts as a line sensor.  You need at least two different
circuits to use these devices.

Emitter:  You need a driver for the infrared LED.  It can be as simple as a
fixed resistor, or as complicated as a constant current source, which I
recommend, since it eliminates weird readings when your battery voltage dips
(like when your motors run).

For a fixed current limiting resistor, subtract the forward voltage of the
IR LED from the supply voltage, then divide by the desired current.  The
specs sheets list 50mA as the "absolute maximum" current, but I recommend
20mA-25mA, since I burned up two of these things in about 30 minutes running
them at 40mA.

For example, if using a 4.8V battery, subtract the forward voltage (under
load, not according to any spec sheet or "diode tester"), which is typically
1.2V, leaving 3.6V.  This is the voltage across the resistor.  Divide by the
desired current in amps (20mA = 0.02 amps), which gives you a resistor value
of 180 ohms, which just happens to be a standard value for 5% carbon
resistors.  If you end up with a non-standard value, go up to the next
higher value.

Next calculate the power flowing through the resistor, by multiplying the
current (0.02 amps) by the voltage across the resistor (3.6V), to obtain the
power rating in watts (0.072W).  Use something rated for substantially more
power, as a safety precaution.  In this example, a 1/8W (0.125W) resistor
would be sufficient (on paper), but would be running at almost 60% of
capacity, and would therefore be warm to the touch.  I would suggest a 1/4W
resistor in this case.

Connect one end of the resistor to your positive supply, and the other to
the anode of the LED.  Connect the cathode of the LED to ground.

A simple current limiter would be more robust, as it would keep the IR
illumination steady even during voltage dips.  The easiest circuit is to use
a three pin, adjustable voltage regulator like the LM317L with a floating
feedback resistor.  A technically simpler circuit uses a single transistor
configured as an emitter follower flowing through a fixed output resistance,
which regulates the current by setting the voltage at the base.  I say
"technically" simpler as the total circuit contains fewer components (if you
count all the internal components of the three terminal regualtor), but I
suggest using the regulators, as they also have overvoltage, overtemperature
and overcurrent protection built in.  Also, they're about $0.25 in small
quantities.

Detector:  The detector is usually configured with the emitter shorted to
ground, and the collector tied to a pullup resistor.  The output is the
voltage at the collector.  It will vary from one diode drop from your
positive voltage to ground, with more illumination driving the voltage
lower.  You should then condition this signal with a comparator with a small
amount of hysteresis.  You can use a simple comparator such as the LM339 or
almost any op amp.  I have a couple of different circuits that work very
well.  The trick is to limit the amount of hysteresis so as not to have the
output signal get "stuck".  Adding a comparator also lets you have an
adjustable "threshold" for your light/dark detection, which is very helpful,
even necessary.  I suggest adding an LED to the output so you you can
directly "see" what the sensor is "seeing".

You can get some good performance out of these parts.  I've found them to be
highly immune to ambient light.  We'll see how well they perform against the
Dreaded Wall of Light at the Science Place!

Thanks,

Dale Wheat
http://dalewheat.com
(972) 486-1317
(800) 330-1915, access code 00


More information about the DPRG mailing list