Subject:
|
Re: RCX IR Transceiver Circuit
|
Newsgroups:
|
lugnet.robotics.rcx
|
Date:
|
Tue, 28 Dec 2004 17:29:16 GMT
|
Viewed:
|
6024 times
|
| |
|
|
In lugnet.robotics.rcx, Philippe Hurbain wrote:
| |
| |
I managed to get this one drawn today:
http://www.brickshelf.com/gallery/mbellis/Technic/Mindstorms/RCX-Internals/rcx_ir_txrx_circuit.jpg
N.B. I cannot guarantee accuracy as this is my own interpretation of the
circuit of my own IR Transceiver, using data from reference books and
tracing the circuit myself, so I might have made mistakes. If you spot one,
do let me know.
Explaining how it works will take some time and a considerable amount of
text!
The dotted box bottom left is the serial port plug.
The transistors and diode pairs are in 3-pin surface mount packages. The
6Cp transistor, being a higher power device, is in a larger package than the
others. I wasn’t sure if this was a MOSFET rather than a BJT. The range
switch is on the right - contact made for long range, open for short range.
IC2 with the 33k and 100k resistors, capacitor and variable resistor form
the oscillator that applies the carrier wave to the transmitted signal. The
variable resistor in my transceiver had an in-circuit value of 11.6k ohms.
It looks like signals received from the PC on the RxD input are used to
modulate the +5V supply rail and they are then returned on TxD. If the PC
pulls CTS low then TxD will also go low. These return mechanisms allow the
PC to detect the presence of the IR tower.
The LED stays on after transmission finishes, due to the time taken for the
10uF capacitor to discharge through it. I can’t remember if it was possible
to see a 5.1V rating on the zener diode, but the voltage is indicated by te
LED resistor value of 470 ohms - that’s a normal value to use with a 5V
rail.
In long range mode, assuming 2V drop per IR LED and 0.2V C-E saturation
voltage of the 6Cp transistor, there’s 4.8V across a 5.6 ohm resistor,
giving 857mA current - this is quite a lot to draw from a PP3 battery.
There seems to be a residual current through the IR LEDs of 4.4mA via the
6Cp transistor B-E junction (0.6V drop). Therefore remove the battery if
you’re not using the tower for a while.
PLMKWYT. I won’t be able to draw any others till the new year, but this
should give us plenty to discuss till then.
|
Thanks for this one, Mark!
For the sake of completeness, some data on devices used. Note: as there are
“collisions” in SMD device names, these are only “very probable”...
1Kp:
BC848B
3Kp:
BC858B
are NPN and PNP general purpose transistor.
A4p:
BAV70
A1p:
BAW56
A7p:
BAV99
are high speed, small signal silicon dual diodes with different
configurations.
Z1p:
BZX84-C4V7
4.7V zener diode.
6Cp:
BC817-40
Medium power NPN transistor. Provides correct gain at relatively high
current. Interestingly enough, this transistor is rated 500mA continuous / 1A
peak. A bit underrated IMHO considering the high current that can flow
through in long range mode, even if I think that max. current is lower than
the 850 mA figure proposed by Mark (internal resistance of PP3 batteries is
high, and saturation voltage of BC817 is greater than 0.2V at that current).
IC1/IC2:
74HC132
I think that CTS circuitry is there to provide a negative bias for TxD line,
as some RC232 receivers in PCs need to get a signal that swings positive and
negative to overcome their hysteresis and work correctly.
Philo
|
I can verify the data on all but the zener and 6Cp transistor. I intended to
put it in a text file with the diagram, but I’ll add it to the diagram instead.
The 1Kp and 3Kp transistors have equivalents of BC548B and BC558B, which are
common high-gain devices similar to those I’ve use on breadboards.
The diode pairs are all equivalent to two BAW62 diodes.
I suppose it doesn’t matter that the zener is only 4.7V rather than 5V because
the RS232 needs only 3V to work.
comparing my diagram with the schematics on this page:
http://oase.uci.kun.nl/~mientki/LegoKnex/Legoelectronica/IRtower/IRtower.htm#Schematics
perhaps I have the connection of the 1k resistor wrong - it would make more
sense for it to be to the collector of the 6Cp transistor, rather than to its
base. This would remove my previously perceived problem of a residual current
through the IR LEDs. Another look at the circuit board confirms this. I’ll
revise that when I add the data (soon in the new year).
Mark
|
|
Message has 1 Reply:
 | | Re: RCX IR Transceiver Circuit
|
| (...) Indeed, you probably had something wrong there: I checked the battery of a serial tower I didn't use for a very long time, it was still fully charged. Philo (20 years ago, 29-Dec-04, to lugnet.robotics.rcx, FTX)
|
Message is in Reply To:
| | Re: RCX IR Transceiver Circuit
|
| (...) Thanks for this one, Mark! For the sake of completeness, some data on devices used. Note: as there are "collisions" in SMD device names, these are only "very probable"... 1Kp: (URL) 3Kp: (URL) are NPN and PNP general purpose transistor. A4p: (...) (20 years ago, 28-Dec-04, to lugnet.robotics.rcx, FTX)
|
5 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
