TCS Trainspeed spedometer — component substitution question

[bbussey]

The TCS Trainspeed spedometer featured in Kim Saign's walk-through module article in the current issue of NSR utilizes a pair of light sensors as triggers and has an LED numeric display built in to their main component board.  TCS makes two versions, a single-track version and a four-track version.  I emailed tech support and asked if infrared sensors could be used in place of light sensors; whether it mattered which sensor was tripped first; and if a remote LED display could be utilized instead.  They responded that no components could be substituted, and that it did not matter which sensor was tripped first.

It seems to me that it shouldn't matter what the input trigger is.  If an infrared sensor is pointed across a number of parallel tracks, only the "single track" version would be required and the sensors would be impervious to the quality of ambient light (or lack thereof).  It also seems to me that there could be multiple pairs of sensors pointing to the same unit, as long as they aren't tripped simultaneously (which of course could be alleviated by using the "four-track" version of the product).  And, as long as a remote LED display was wired/soldered to the connections that the current LED display is, in theory it should be possible to, say,embed it in the fasia instead of having some bulky display protruding from the control panel or fasia board.  Am I missing something, or do they just not want people inexperienced with electronics experimenting with their product?

[peteski]

Am I missing something, or do they just not want people inexperienced with electronics experimenting with their product?

I think that you hit a nail on the head.  Remote display will be easy - just use wires to make the connections between each pin of the display and the solder pad on the PCB.

Multiple sensors attached to the single version of the PCB sensors might be trickier.  Same with substituting them with IR sensors. It depends on the original design.  It also depends on what type of sensors they use in the original design: photo-resistors, photo-transistors, or photo-diodes. either way, you would most likely add some logic for the multi-sensor circuit to work.  That is because on the idle (uncovered state), the sensor has a low resistance. If you hooked them up in parallel then if one was covered the other ones would still create the path of lower resistance.  You could try to hook them up in series but then the total resistance (even in the uncovered idle state) might be too high for the sensing circuitry to read them as uncovered.  Without doing some reverse-engineering it is difficult to answer the sensor questions.

[bbussey]

Understood regarding the sensors.  Given your premise on the difficulty of multiple sensors on a single circuit, is it not reasonable to assume then that using the "four-track" version would allow you to wire four separate pairs of infrared sensors in place of the four separate pairs of light sensors?  And once the item was in hand, is it correct to assume that it would be easy to determine the type of original sensor used, be it photo-resistor, photo-transistor, or photo-diode?

Update - the NSR article states they are photo diodes.

[peteski]

Understood regarding the sensors.  Given your premise on the difficulty of multiple sensors on a single circuit, is it not reasonable to assume then that using the "four-track" version would allow you to wire four separate pairs of infrared sensors in place of the four separate pairs of light sensors?  And once the item was in hand, is it correct to assume that it would be easy to determine the type of original sensor used, be it photo-resistor, photo-transistor, or photo-diode?

Update - the NSR article states they are photo diodes.

Yes, the four-track version would be the way to go if you want to use the speedometer on multiple tracks.  If photodiodes are used, they are naturally very sensitive to IR light already.  If you want to use them with IR LED illumination, just think of the IR LEDs as an additional light source (augmenting the ambient light).  Just aim the IR LED at the photodiode. That way the circuit will work properly when there is ambient lighting in the room or if the room is dark (and only the IR LED is illuminating the photodiode).

[bbussey]

I was hoping though that the diodes could be mounted across the tracks, as opposed to embedded between the ties.  That would make it practical.  Otherwise, it's somewhat more of a novelty if only a handful of tracks can be monitored.

[peteski]

You have few options:

Install an opaque tube in front of the photodiode's lens (the stock photodiode) and then place the photodiode in a light-tight enclosure (or simply paint it black).  The tube can ba a drinking straw made opaque by painting it black or a piece of brass tubing (again, painted black on the inside to prevent ambient light from reflecting inside and reaching the photodiode).  The tube should be long enough to limit the field of vision of the photodiode to the area where the trains will be passing and to be able to see the IR LED.  That way stray ambient light will not interfere with train detection.

You can also place IR filter in front of the photodiode's lens or buy photodiodes encased in IR filter material (which looks opaque black or very dark red under visible light).  But even then you will still have to shield it from ambient light which usually contains a fair amount of IR radiation (especially if using incandescent bulbs).  So some sort of a tube to limit the angle of view of the sensor will be required.

[bbussey]

It was a novelty that was intriguing that maybe I will revisit in the future. I have more important tasks to complete to get trains running. Thanks for the insight.