FasTrack Switch Simulator

Jim, maybe I'm missing something in your wanting to test the controllers without using a remote turnout (switch).  What I do to test the controllers is just connect them to a known good FasTrack switch.  Would this work for you?  If not, it might be helpful to know what you're trying to accomplish by not using the switch to test the remote controllers.

But to answer your question, using an isolated 5VDC source connected to a DPDT switch will allow you to source + or minus 5VDC by simply swapping the polarity.

Thanks Steve ... I do that now.   Problem is manifold ... its large and inconvenient and represents a $100 investment that I would rather have on the layout than on the bench.   My goal is to replace the switch/power track arrangement with a dongle of sorts that will indicate momentary switch position and switch position ... just need to know how to generate the -4.5v.

The circuit above is perfect and the 5 volt wall warts are an easy choice for the power supply. We all have a dozen or so of them in a drawer somewhere.

Thanks for the schematic Steve ... I had thought about that as a solution but believe that the ground is used by the RSC and wonder about it being +5vdc when the switch controller is thrown ... that's the puzzle in my head :-)   The confusion for me is that the momentary switches on the RSC end (i assume but will verify) connect red with black or green with black to send the signal ... the switch machine manages the details and provides led talk-back via yellow.   I'm wondering (and will experiment) on how that can work if the black wire floats between ground and +5v.

The 5V wall wart Chuck mentioned is what I had in mind.  It has a transformer that electrically isolates the secondary (5VDC source) from the other transformer powering the switch machine, and in your case the test dongle.  If you connect the wall wart's output one way you get 5VDC and if you swap the connections, it becomes -5VDC.  Because it's electrically isolated, it doesn't care if one of it's legs is tied to the track power common (black wire on turnout), which is what needs to be done.

The other point you make about the switching for the turnout, yes that switch is a momentary SPDT.  Electrically, what it does is make a brief short between the black ground and either the (Red -  Out or the Green - Through) turnout's terminals to activate a throw of the turnout's position.

The black wire will not float between 0 and +5, the black wire is ground in the switch and will remain at ground. The isolated 5 vdc power supply will float between + or - 5 volts.

Oh sorry, I was writing at the same time as Steve. He already answered your question...

If trying to simulate the switch, I'd think you'd want to simulate it in the entirety (all 4 wires taken together).  I agree $100 is a bit steep to that end.  But what about something like a latching relay that flips one way or the other (and stays) when triggered by the respective momentary switch wire.  The output of the relay selects between +5 or -5V DC to send out on the yellow wire.

If you "manually" select yellow wire voltage using a mechanical toggle switch, it seems to me this divorces the switch-machine action from the feedback.  Of course this may be exactly what you want. 

Stan, agreed that may be a better solution.  Do you have a suggestion for what the circuit you're suggesting might look like?

Steve -  every idea is worth exploring ... and you offered a workable solution.   I had thought about a bistable or latched relay ... in the world of components I can find a 12v triggered latching component or a 5v bistable single channel (pole) component ... its never simple :-)    My concern was with the ground.  and so I'll  plan to use a latching relay to emulate the points and respond to red/green and use yellow to signal position ... i can use the relay outputs for this ... my concern was using the ground to reverse polarity *and* as the "return" for red/green momentary signals.   Part of the fun of this Hobby for me is solving problems ... so thanks for weighing in

Right.  I think it is beneficial to use a single ground/common/black.  So the proverbial black-box above is powered by AC on the left.  It generates +5V and -5V DC using 5-cent diodes, 10-cent capacitors, and 25-cent voltage regulator IC chips.  A $3 "dual-coil" 5V latching relay is set or reset by the momentary RSC (red or green) wires.  And the relay output contacts steers either +5V or -5V to the yellow wire.  Obviously there's soldering and messing with small electronic components...

Thanks so much for the recipe @stan2004 I'm assuming a 1N4001 is sufficient for the diodes ..do you have a suggested value for the Caps?

If you already have 4001 then they're fine. If you need to buy go with 1n4003 which will be same price. If you already have caps they should be 35V rating. 100uF or more should do the trick.  Parts cost are all over the map if willing to wait for Asia shipping.

Arduino with a few LEDs and a sketch ?

EDIT: I originally thought that a Latching Dual Coil DPTD circuit using a single 5VDC power supply might be possible and tried to conceive one that would work.  After discussing it offline with @gunrunnerjohn, the design I had would have been problematic in providing switched power to the relay coils.  He has a better solution than mine using a Boost Converter with + and - 5VDC outputs from a single 4.5VDC power source.  His solution is a couple replies down the page.

@SteveH posted:

I'm wondering if a small electrolytic capacitor push-pull pair might be a good idea in this circuit to ensure the relay coils are energized long enough to fully change state, and if so, the resistors would limit capacitor discharge current.

The relay should change states in a few milliseconds, I see no reason that you'd need a "boost" to make that happen.  The K.I.S.S. principle clearly applies here.

From the data sheet for the EC2/EE2 series of relays.

Sure seems fast enough to me.

@gunrunnerjohn John, Thank you for looking at my preliminary design and offering your valuable input on it.  I'm going to take a step back and see what other ideas might be presented here.

Here's one concept for the job.  I used a dual polarity supply as I believe that makes this job much more simple.  You could also substitute a single bi-polar red/green LED for the two LED's and eliminate one more component.

Thanks everyone, this is really good.   And as always thanks for your contribution @gunrunnerjohn ... I'll need to make my way over to Henning's  one day soon ... I'm just not that far away - Jim

Well, I'm not normally at Henning's, just go there to drop of repairs and products and hopefully pick up checks.   But you could come visit me at my place, I'm close to Henning's.

Will do.  Never been to Hennings, Nicholas Smith gets most of my money :-)

Bring some money to Henning's, I need them to stay healthy so my products sell!

@SteveH posted:

EDIT: I originally thought that a Latching Dual Coil DPTD circuit using a single 5VDC power supply might be possible and tried to conceive one that would work.  After discussing it offline with @gunrunnerjohn, the design I had would have been problematic in providing switched power to the relay coils.  He has a better solution than mine using a Boost Converter with + and - 5VDC outputs from a single 4.5VDC power source.  His solution is a couple replies down the page.

I'm not sure where you guys are going?

I thought the objective, per the topic title, was to come up with a FasTrack Switch Simulator ... since using the real McCoy is not practical given its price tag.  In other words, I figured this was one of those "walks like a duck, talks like a duck, etc." situations.

So if this Simulator is to debug/troubleshoot RSC controller wiring and such, it stands to reason it should be wired like a FasTrack Switch.  In other words, it should be powered with whatever the FasTrack Switch would be powered by...presumably 14V AC Accessory or 18V AC Track...and with a common ground/black wire to the RSC controller wiring.

That's why I'm not clear on why you'd use a 5V wall-wart.  I can see value from a Forum Discussion point of view and was about to comment on some theoretical issues therein; the circuit has/had a so-called race condition that might lead to its demise but I see the diagrams have vanished.

@LIRR-Jim posted:

Thanks so much for the recipe @stan2004 I'm assuming a 1N4001 is sufficient for the diodes ..do you have a suggested value for the Caps?

So are you good to go?  I see the most recent posts show how to add red/green LED direction indicators which I presume the actual switch-machines have (separate from the RSC LED indicators).  That makes sense to make the Switch Simulator look/behave more like the prototype.

Valid points Stan, I think we lost sight of that objective.  However, I believe the object of the exercise was to simply test the Fastrack controller, that was what I was shooting for.  I think the last circuit I posted would accomplish testing the controller.

@stan2004 Stan, please accept my apologies for the confusion my attempts at a circuit design and subsequent retraction of those preliminary ideas may have caused.  I think your suggestion does accomplish the goal of emulating "the Real McCoy" using 18V track power.  Jim also mentioned in his original post the possible option for:

@LIRR-Jim posted:

Thank you again .. The question is how does one create a circuit that delivers -4.5v (from either 18v track voltage or a dedicated ~5v power supply)  ..

Your suggestion delivers on the 18V power option and your recommendation for the Latching Dual Coil DPDT 5V relay started me thinking that maybe I could combine that relay with my earlier DPDT switch diagram as the foundation for a 5v sourced solution.  The further I got into it, the more I realized it wouldn't be as simple (for me) as I'd hoped.

Since John was online this morning (5/6), I reached out to him for his advice on my concept.  During the course of that conversation, he suggested using a Boost Converter that would provide +/- 5VDC from a 4.5V source.

@stan2004 posted:

I'm not sure where you guys are going?      ...and was about to comment on some theoretical issues therein; the circuit has/had a so-called race condition that might lead to its demise but I see the diagrams have vanished.

I'm a bit embarrassed to have overlooked the race condition in my preliminary design in the first place.  There might be a way to make that circuit design functional by relocating one small RC circuit each, across the coils, but the point seems moot now...

Since your track power option provides one requested solution and  John's fulfills the other ~ 5V option more elegantly than mine would have, even with corrections, I decided to withdraw my suggestion.  If you'd like for me to send the revisions on that design to you via email, please contact me via the email address in my profile.

Again, please accept my apologies for my mistake and confusion I may have caused.

Steve

No apology required.  This is after all a discussion forum!

In re-reading the original post one could indeed infer that the only matter at hand was how to generate -5V for the yellow wire output.  In other words +5V already exists.  I didn't track down the exact dual-output converter GRJ suggests but if +5V already exists, another commonly used method is the isolated DC-to-DC converter module.

These are 4 terminal devices.  2 wires in, 2 wires out.  5V DC input, transformer isolated 5V DC output (there's a transformer in the module).  Then simply connect the + output to the - input and you now have 5V and -5V referenced to a common ground.  These modules have been around forever...much longer than the insanely low-priced DC-to-DC voltage converter modules out of Asia.

Note that the quantity 1 DigiKey pricing is quite reasonable.

For a tool, intended for only occasional use, I think I'd just put three AA cells in series rather than fussing with wall warts and converters. Connecting them between black and yellow in both directions would do the trick.

Is there truly a reason for the involvement of an existing 18 VAC or 5 VDC supply?

@PLCProf posted:

For a tool, intended for only occasional use, I think I'd just put three AA cells in series rather than fussing with wall warts and converters. Connecting them between black and yellow in both directions would do the trick.

Is there truly a reason for the involvement of an existing 18 VAC or 5 VDC supply?

You know us, why have a simple solution when you can have a complicated and convoluted one.

However, the "full" solutions presented also test the actual switch continuity when operated.

@gunrunnerjohn posted:

You know us, why have a simple solution when you can have a complicated and convoluted one.

However, the "full" solutions presented also test the actual switch continuity when operated.

Well, in that spirit, you could add the Universal MP3 sound card to the latching relay scheme. That way, you could put the thing at the far end of the layout, near the switch location, and it would audibly announce the simulated switch position, as well as sending the +/- 5V for the indicator. Might be handy for those larger layouts.

@PLCProf posted:

Well, in that spirit, you could add the Universal MP3 sound card to the latching relay scheme. That way, you could put the thing at the far end of the layout, near the switch location, and it would audibly announce the switch position, as well as sending the +/- 5V for the indicator. Might be handy for those larger layouts.

I like it, and I know just where he can get a sound board!

And for the sake of completeness, I was surprised to find the 7660 voltage inverter IC chip is still alive and kicking!  This is one of those chips like the 555-timer IC chip that has been around forever.   

With the 7660 which came as an 8-pin DIP (like the 555), you simply add 2 inexpensive capacitors and you have a positive-to-negative voltage inverter.  So put 5V DC in, you get -5V DC out.  Put 4.5V DC in, you get -4.5V DC out.  It uses so-called charge-pump or switch-capacitor (NO inductors/coils) design and cannot deliver a lot of power but plenty for driving the -5V for the yellow wire output.

I see you can even buy a 3-pin module.  Though with the simplicity of the circuit, if soldering up a circuit anyway, this might be one of those unusual situations where you can build it cheaper from parts than you can buy the completed Asia module! 

Again, this addresses the limited interpretation of the OP's question as to how to generate -5V (along with +5V) to drive the yellow-wire.  So summarizing:

1.  5V wall-wart with DPDT switch to flip the polarity

2. Half-wave rectified AC with a -5V regulator IC chip

3. Three 1.5V batteries with a switch to flip the polarity

4. Boost step-up switch-mode converter generating +/- outputs

5. DC-DC isolated converter, + of floating DC output tied to common

6. Charge-pump switched-cap voltage inverter

7. Use an actual FasTrack switch

I think we've got this one covered.

So, there is seven ways to do it.   You're right Stan, got it covered!

No potato battery?

@cjack Good one!  That was funny!

@cjack posted:

No potato battery?

If you insist!

Yeah!