ITADs and Making Your Own

John H posted:

...  I thought I told you I don't know what I'm doing.

Actually, that's a good idea to use 2 parallel modules so that either module can trigger the relay to handle inter-car gaps.  As it turns out, this module is of the type that you can "short" two outputs together without ill effect,

But in the break-the-beam method where both beams are in parallel and close together, the output of the two modules provides redundant information.  So you only need one output.  But, yes, it will still work in parallel - no harm, no foul.

So for simplicity, the one across the track would just need power and ground and the near one would not need the emitter?

By jove, I think he's got it!

And as those IR modules are 60 cents each (free shipping), my obsession with "wasting" half of each module is moot.

Stan, your card above should become some kind of a symbol around here, maybe like the mascot or something? Perfect! 

And a lot cheaper than staying at a Holiday Inn Express

By Stan, I think I got it!

Are there other ways to trigger these 5V relays besides the opti-sensors?

Isolated rail sections. Gunrunnerjohn has a new device in the works that would be neat for isolated rails. I'll go look for a link to the thread. Arduinos may be an option too, isolated rails could probably used to input to the Arduino. Another possible use for GRJ's new device.

Stan will probably have a lot more and better ideas as well.

Here's the link to GRJ's Isolated Rail Relay Sensor. I think he is close to having some made, if he hasn't already gotten some.

John H posted:

Are there other ways to trigger these 5V relays besides the opti-sensors?

Is this for occupancy-detection? 

These relay modules are triggered by a voltage so pretty much any sensor that can generate the required voltage can be used.  So in addition to direct voltage itself (such as from an insulated rail) there are all manners of sensors/transducers that can generate a suitable trigger voltage in response to some physical phenomenon...optical, pressure, magnetic field, sound, heat, etc.

Are you abandoning the optical method and if so why?

I guess my question should have asked what voltage can be used. Does it have to be 5vdc? My isolated rails would be 18vac. I have three more relays on the set I have and have a few isolated rail drops already built into the track where I planned on signals. If I have to electronically change the isolated rail voltage to use the "blue" relays, I might as well just use a regular relay in those places. I still have places for the optical sensors, although I haven't figured out how to use a pair at each end of a section where I will have a gate or signal to cover trains going both ways. So far the single sensor doesn't work well enough to cover the dark cars and gap between  all cars, so if I have a connected beam at both ends, breaking only one won't trigger the relay. Is that thinking correct? I'm still waiting on some different sensors as the ones I used had to be almost 90 degrees to the train and an inch away tops when using only one. That makes it harder to place them with any kind of disguise.

In re isolated-rail triggering, the 5V DC modules can be use if you connect the common of your 5V DC supply to the outer-rail common.  The 5V DC supply MUST BE from a wall-wart; if this requirement is not crystal-clear, ask for clarification!  Your 5V relay module is triggered on a "low" signal; the isolated-rail section provides such a "low" signal if the outer-rail common is connected to the 5V DC supply's common.

In re bi-directional crossing, if you can get the reflective-method to work then the two sensor modules can be tied together (in parallel which you've done previously) and either or both will trigger the relay.  However, if you use the break-the-beam method then the "logic" is reversed.   That is, you were using the opposite relay contact (the NC).  So your accessory would only be activated when both beams were broken which is not what you want.  To un-reverse the "logic" you would need to fuss with external components - albeit nickels and dimes so cost is not the issue - but small components and possibly a bit of soldering.  This is the "classic" issue of ITADs at both ends of a crossing - the ITAD relay outputs can "fight" each other so additional components are typically added to make them play well together.

Stan,  In re bi-directional crossing ................."To un-reverse the "logic" you would need to fuss with external components". It would be great if you could post schematic on how these external components would be connected.

Your help with the timing module for Pennsy seven LED signal circuit was very helpful to me in making  roll your own signals.

I have the commons connected, but my 5vdc supply is a PC power supply tap. I use the 12vdc for signals, gates and buildings with led strips, and the 5vdc for this relay.

John H posted:

I have the commons connected, but my 5vdc supply is a PC power supply tap. I use the 12vdc for signals, gates and buildings with led strips, and the 5vdc for this relay.

The PC power supply is a big wall-wart!   

I should have said that to connect the commons between the 5V DC source and the AC track supply, the two must be galvanically isolated.  But that's just techno-babble!  For practical purposes it just means the DC supply needs to have it's own AC-wall plug...whether it be a wall-wart, PC power supply, separate HO transformer, etc..

SGMret posted:

Stan,  In re bi-directional crossing ................."To un-reverse the "logic" you would need to fuss with external components". It would be great if you could post schematic on how these external components would be connected.

I can do so but let me know how far along you are - or just thinking of dabbling with roll-your-own ITADs.  Do you actually have components or modules in-hand?

John H is using a 5V relay module that is triggered by a "low" signal...in conjunction with a bounce-the-beam (aka reflective) IR module that happens to put out a "low" signal when triggered.  So this logic reversal was a consequence of modifying the operation to break-the-beam (aka transmissive).

But there are 12V relay modules.  There are relay modules that can be triggered by "low" or "high" signals.  And perhaps the biggest leverage when rolling-your-own is having the time/interest to mess with components and do some soldering.  Hence I ask where you're "coming from" so I can tailor the suggestions.  Do you buy-in (have space on your layout) to use the break-the-beam approach? 

Do you need time-delay like an off-the-shelf ITAD?  Note that for bi-directional operation with two ITADs one function of time-delay is when running a short consist that could be between the ITADs not triggering either.

John H posted:

I have the commons connected, but my 5vdc supply is a PC power supply tap. I use the 12vdc for signals, gates and buildings with led strips, and the 5vdc for this relay.

The PC power supply is isolated, but take note that the ground lead is common to all of the power outputs of the supply.  This may not be an issue in your application, but it's something to be aware of.

While there are many ways to invert the voltage, I think one of the simplest would be with a transistor and a couple resistors if you only need 1 or 2 such devices in the same location.  Using a common, dirt-cheap, 2n3904 or PN2222a transistor would allow an input voltage of a couple volts up to about 40VDC.  One would connect the emitter of the transistor to ground.  The Collector would be connected to the input to the relay module.  The collector would also be connected to a 10K ohm resistor with the other end of the resistor connected to the power source of the relay module (usually 5 or 12 VDC).  Last the output of the IR module would be connected to a second resistor, likely a value of 1K to 2.2K Ohms would be good here, and the other end of this resistor connects to the Base of the transistor.  

JGL

Finally received what I called the type B sensor module from Asia.  It is absolutely unsuitable for O-gauge occupancy detection as it only handles digital data rather than simple on/off occupied/unoccupied detection.  So I ordered a type C module and will wait a few weeks and report back.  Duplicated photo below of the various modulated IR modules I found on eBay:

I've been meaning to post this video showing a type A sensor module.  As previously mentioned I do NOT recommend this module because you have to mess with interacting adjustments of modulation frequency and LED output power.  But it does show the "technology" should work even for detecting dark (black) surfaces.  Midway through the video I crank up the ambient lighting to show how detection performance is not materially affected if you use modulated IR.