DCC-friendly occupany detectors

[Enzoamps]

From May issue, I believe there is a drawing error in the schematic for the simple circuit, figure 4.  As drawn, the relay contacts are across (parallel) the relay coil.  That means when the relay energizes, the contacts close and short across the coil.  That won't work.  Either it would damage the transistor or it might turn the relay into a buzzer.  Or if the normally closed contacts are across the coil, when it tries to energize, the coil is shorted across.  The relay contacts should not be wired to the coil.

The wiring in figure 7 is accurate.

[gregc]

i agree, the relay contacts don't need to be part of the relay driver circuit.   The schematic is drawn incorrectly.

it's also unclear if by -9VDC they mean a -9V source relative to ground or simply the ground side of the +9VDC source.

not sure why there's a capacitor across the relay coil

A think a more interesting design would use a capacitor to determine a reference level due to the ambient light conditions that could be compared to the level on the photo-transitor.   That would avoid the need for adjustment.

[Enzoamps]

From the article, I assumed the -9v meant the negative side of the +9. Not the usual notation. In the article, he mentioned the cap as a bit of delay to prevent the relay from dropping out as the spaces between cars passed over the sensor.

[gregc]

Enzoamps

In the article, he mentioned the cap as a bit of delay to prevent the relay from dropping out as the spaces between cars passed over the sensor.

wouldn't it make more sense to put the capacitor across the photo-transistor which operates with less current instead of the higher-power relay path.   At the very least, a smaller capacitor could be used.

wouldn't it be more conventional to put the relay on the collector side of the transistor, or is it intentional to provide some latching behavior?

Because the relay voltage will vary with the current through it, the turn-on/off voltage at the base will vary.   If the emitter were at ground, like the phototransistor, the turn-on/off voltage would be independent of the current through the collector.

[rrinker]

 Used to be they'd build and test a circuit before publishing it in MR. There are clearly several design issues with this one, perhaps in an attempt to oversimplfy things. Bob Frey's optical detector was about as simple as it gets.

 Yes, the cap should probably be opn the low current said. And yes, an open collector configuration is much more efficient. As drawm it makes a nice buzzer as long as the phototransistor is blocked and until it burns out the driver transistor - the relay will pull in, instantly short, drop open, pull in - over and over. Good thing there's a dioode across the coil. The usualy way to make a buzzer is to run the relay coil through the NC contacts, at which point the relay will buzz at a frequency mainly determined by how fast the contacts move. Running a wire to one of the coil leads (with no diode) will make a nice shocker machine.

                      --Randy

 

[gregc]

an alternate circuit was asked for on a different forum.   I bench tested the circuit below using scrap parts.    It has fewer and less expensive components.

it's too bad that the article both had a misprint and is a more complicated design than necessary.  Such articles can be stepping stones for modelers to learn, build and design their own circuits instead of relying on $$ commercial products.

for me, building such circuits is a significant and satisfying part of building a model railroad similar to building your own turnouts.   When someone suggests to me that there's a commercial product for $$, I think about building by own.

[DigitalGriffin]

Why not just wrap one of the track wires around a magnet coil (transformer) then through a rectifier, then use that lead to feed with a cap in parallel a base pin transitor.  The Collector is connected to the relay power source and the Emmiter to the occupancy sensor or LED.    Just be sure to measure the base pin voltage before you hook it up to the cap or transistor or you may go "BOOM"  This can be corrected with either more or less wraps around the magnet or a voltage divider circuit.

Parts List:

1.  Transistor

2.  Capacitor 

3.  Relay

4.  two resistors (voltage divider circuit)

5.  Wire and metal core to make transformer.

Another popular way is to just put four diodes (two each way in parallel) across one leg of the track for your base source on the transistor.  But I don't like this approach as it causes a 1.4V voltage drop to the track.