View Single Post
  #5   Spotlight this post!  
Unread 17-02-2004, 19:07
Dale(294engr]'s Avatar
Dale(294engr] Dale(294engr] is offline
Mentor/Sponsor since '98
FRC #0294 (Beach Cities Robotics)
Team Role: Engineer
 
Join Date: Jan 2004
Rookie Year: 1998
Location: Redondo Beach, CA
Posts: 92
Dale(294engr] is a glorious beacon of lightDale(294engr] is a glorious beacon of lightDale(294engr] is a glorious beacon of lightDale(294engr] is a glorious beacon of lightDale(294engr] is a glorious beacon of lightDale(294engr] is a glorious beacon of light
Re: IR sensor problems

See photo gallery Dale(engr] 2004 robots 2-17,
IR hooded receivers top/bottom (in sockets)

Use of PWM cable requires the Red & Black pins reversed prior to soldering.

For test purposes
construct a special male to male PWM cable
and plug-in the receivers in to test.
I use a PWM "Y" insert
with the 2nd male end flared for observation with the scope probe & gnd.

(a current limited +5v power supply may be substituted for the DIG IN side
power (red/black) set to ~.05A, receivers need ~5mA .005A
for bullet proof protection while testing)

The Vishay 34840 & 4840 receivers do not respond to continuous (DC) IR.

The IR Beacon must pulse at 40kHz +-~2kHz, (50-50 duty cycle).

The receiver very selectively accepts only 40 kHz on=off pulsed IR,
rejecting less than ~38kHz or greater than ~42kHz or continuous IR (DC)

Correct Pulses are provided by the EDU controller with Beacon code running.

The best way to observe proper Receiver operation is with an oscilloscope.
see photo gallery Dale(engr] 2004 Beacon scope waveforms

Probe the 4840 Output pin:
with no pulsed IR present, voltage level is a constant +5vDC,

when 40 kHz is detected, level is 0v (ground) for 1 ms or 2 ms
"this is called active LOW" and what we look for in the code

then back high for the remainder of the period - 9mS or 8 ms,

The cycle repeates every 10mS (100Hz)

The scope is also a good way to view fidelity of the IR signal.
Reflections/weakness often show up as shorter irratic and/or dropout pulses.

Observe care to avoid:
Output pin to +5v shorts when 40kHz IR is active
as this may damage them (we have 3 dead)

(see previous post: the output voltage never goes below ~2.5v

In this fault there is a negative going pulse upon detection, but it's
truncated, i.e. stops short of 0V.
The DIG INP requires the voltage level to go
less than ~.5V [TBD] to signal IR pulses detected.)

Note that a DMM does not give a clear 0V reading in the presence of pulsed IR
due to the short duty cycle, the average is close to +5v (= no IR state)
but a small but discernable decrease from 5v can be seen.

(even with a strong signal the output Voltage is high (+5V)
90% (2ms) to 95% (1ms) of the cycle)