AIU Hookup Question

Rod,

I don't see this covered in the 3rd edition of the DCS Companion, so I thought I would ask here.

Take another look!

Is there a way to configure an accessory port to operate a switch track instead, because I am short of available switch ports?

There's a detailed discussion of this topic on pages 93-94 of the DCS Companion 3rd Edition.  

This and a whole lot more is all in “The DCS Companion 3rd Edition!" This book is available from many fine OGR advertisers and forum sponsors, or as an eBook or a printed book at OGR’s web store!

Sorry Barry; I guess I didn't look far enough! Thanks.

Rod

So on page 94 of the DCS Companion there is this statement:

which I take to mean that older Lionel 022 and similar switches cannot be used, for 2 reasons: 1) You can't use the non-derailing feature and 2) you can't also connect the manual switch controller. So I guess that pretty much takes care of any ideas of using the accessory ports for switch tracks!  So be it.

Rod

I haven't been using ACCs for anything.  Seems to me that Barry's comment only applies if the ACC circuits are set for latching---one press for on and one for off.  If they are set for staying on only when button is pressed, I don't see why it shouldn't work.  Problem is that if the wrong setting is used, you're zonked, and it's too easy a mistake to make.  I recommend doing what I did:  I have 3 AIUs connected to 25 mainline of my 50 switches, the others working from the panel only.

Robert,

Seems to me that Barry's comment only applies if the ACC circuits are set for latching---one press for on and one for off.

Yes, they latch. One way (pin "IN" connected to pin "1") when soft key ON is pressed, and they other way (pin "IN" connected to pin "2") when soft key OFF is pressed.

The default upon AIU power-on is pin "IN" connected to pin "2".

RJR posted:

...Seems to me that Barry's comment only applies if the ACC circuits are set for latching---one press for on and one for off.  If they are set for staying on only when button is pressed, I don't see why it shouldn't work. 

Note that this method would require 2 ACC circuits per switch.  Traditional switches with the spring-loaded return-to-center manual lever controller have what amounts to a "center-off" position when no power is applied to either motor/solenoid of the switch.  The ACC circuits in a TIU only have a single relay which will apply power to one side or the other; that's the basis of the first bullet restriction about switches that turn themselves off. 

Each SW circuit in a TIU has two internal relays which allow power to be applied to one side, the other side, or neither side.  And it has built-in timing to only momentarily apply power to the chosen relay.

So in theory you could kind-of-sort-of control 5 additional traditional switches using an AIU's 10 ACC ports with the somewhat onerous caveat of only using the momentary ACC button and the different menu path on the remote. 

To dig a bit deeper, to deal with the timing issue when using an ACC port, there have been occasional threads about methods to limit the amount of time an ACC port is left on.   The classic example is an uncoupler electromagnet under ACC control.  In this case, external circuitry is used to guard against leaving the ACC output latched (left in the ON position).  Similar guard/timing circuitry could be used for the switch application.  It wouldn't be expensive (maybe a few dollars per switch) but you'd need to be a hard-core DIY'er with lots of free-time!

If you have 022 switches or the O72 similar style, you don't have to worry about any guard circuit as they only draw coil current until the switch moves into position.

Agreed, but if I had a nickel for every time a switch was thrown with a mechanical obstruction on the rail...

OOPS!   That would be a potential issue.

Stan, you may recall some months ago we discussed on the forum a device to cut power after a certain number of seconds. I did build such a device, with a total cost of around $10 using Chinese components, but never did install it. It worked fine, but other matters came up and I never did use it. Only one such device is necessary on a layout, placed in the feed to the AIU inputs for the uncoupling tracks or switches, rather than in the individual outputs.

RJR, agreed a single guard circuit is an attractive option. 

But I think we'd need a test-case to work through the scenarios.  For example, while a common timing/guard circuit removes power, it doesn't remove a "faulty" latched connection made by the AIU relay.  That is, what if the user has manual levers controlling a switch in parallel with the AIU relay (to allow manual or remote control)?  Would the latched AIU ACC relay inhibit the operation of the external manual lever?

RJR posted:

Stan, you may recall some months ago we discussed on the forum a device to cut power after a certain number of seconds. I did build such a device, with a total cost of around $10 using Chinese components, but never did install it. It worked fine, but other matters came up and I never did use it. Only one such device is necessary on a layout, placed in the feed to the AIU inputs for the uncoupling tracks or switches, rather than in the individual outputs.

RJR; sounds like an interesting addition, certainly for uncoupling tracks. Can you post the schematic for all to see?

Rod

Hopefully RJR will post his final configuration/schematic but I believe this is the thread where you might find related tidbits that could apply to your situation:

https://ogrforum.com/t...g-track-from-the-aiu

Just a FYI here...

Here is a list of recommended PCT fuses put together from input by other forum members that are using them. Might be handy to have for some folks here for future reference. It's in the Electrical Forum in the *Sticky* 'Electrical Materials ad Manuals' thread. If you have corrections or anything to add, please email them to me and I will add/update the list. My emailis in my profile. Thanks

One of the PTCs is for a Fastrack Uncoupler track drawing about 1.8 amps or so.

Here is a link: PTC and TVS - Some Commonly Used Sizes and Part Numbers (from other threads)

I am off to read the thread in Stan's post above to see what RJR ended up doing in the earlier thread...

The last configuration is as shown in the video in the other thread.  Since making that video, I have moved on to other things (like boating). 

I still haven't hooked uo uncoupling tracks to the AIUs, but if I do, I will finalize this unit.  If there were enough demand, I suspect gunrunnerjohn could be induced to produce a few, as appears to be the case in the thread about constant wacthdog signals in DCS.

Because each of the Chinese sensors needs a different voltage, on the "breadboard" are two power supplies.  If I were making this up to use, I'd construct a single power supply, using the schematic laid out a year or so ago in a thread on LED lighting, but have it with two different outputs.

I think it could be simplified a lot with a little work.   I have a couple of the bare induction coil sensors, they would probably be sufficient if you laid out a simple sensor with a power supply, one stop shop.

You also need a time delay relay function, as well as the current sensor.  An electronics whizbang like you could design a nice package, and there probably is more of a market for it than for your watchdog repeater.

Believe me, there's no money to be made for the WD repeater, I'm just doing it for the fun of it.  Perhaps I'll get a little free press, but I sure won't get rich.

The key to your package is to completely define the functionality so you know what you're building.

I really thought that you sales of the watchdog repeater would mean you'd show up for the April 2017 York in a Rolls-Royce

I might be able to get the diecast 1:48 model if I pick on of the cheap manufacturers.

So if it's acceptable to monitor the problem (rather than fix it), I think RJR's approach would be an inexpensive way to control 5 additional switches per AIU (using 10 ACC ports).  In other words, once the monitor circuit detects a latched ON relay, it signals the user, and the user must navigate to the offending or latched-on ACC relay and clear it.  As I figure it, one would assign a switch to two ACC ports calling them, say, Switch Straight and Switch Turnout.  Then select one and press ACT button (not ON button).

This would allow both AIU control and simultaneous manual control using existing lever-switches which the OP suggests is a requirement in one of his posts.

There is a somewhat humorous TV commercial from Lifelock security company along the lines of "Why monitor a problem if you can fix it."

As I see it, since it would be extremely difficult to reverse-engineer the TIU/AIU to automatically fix (clear) an ACC latched relay, it would have to be done with external circuitry probably including new relays.  Again, not particularly expensive (using $1 per relay eBay modules), but somewhat a hassle to implement and an AIU is "only" about $100 and gets you 10 more switch ports.

I think Stan's point is right on, why not just add an additional AIU?

Actually, Stan, since a relay turns on the buzzer, there's no reason why it also can't use a relay to run off the power, and stay turned off until reset.  I only used the buzzer as an indication the relay was pulling in for testing.  Of course, one can always get a howler.

Agreed.  Many ways to surround the AIU with circuits/relays to use ACC ports to operate switches.

But I think we're in agreement that the answer to the OP's original question,

"Is there a way to configure an accessory port to operate a switch track instead, because I am short of available switch ports?"

is "no" for traditional swtches. 

I can imagine how MTH could change the software in the DCS Remote menu.  So during ACC set-up you configure the AIU to pair ACC 1 and ACC 2 to make a SW.  And pair ACC 3 and ACC 4 to make another SW, and so on.  Then the Menu would show the proper icons for Straight and Turnout and perform the correct SW momentary timing so no latched relay condition.  The wire connections to the ACC ports on the AIU terminals would be slightly different than to an SW port but no extra external components.  So the AIU would become 10 SW and 10 ACC...or configurable to 11 SW and 8 ACC, 12 SW and 6 ACC...up to 15 SW and 0 ACC.  Again, that's imagination!