For a revers loop, of course you need a DPDT relay. It gets wired just like a DPDT toggle, with the X wiring. The coil is triggered by switch machine contacts. The reason for a relay is that most switch machine contacts aren't rated for the full load of a train. For a frog, it's fine, there can not possibly be more than one loco or lighted car on a frog at any given time so there never can be that much current flowing through said contacts. For the entire power to a complete reversing section - there could be multiple unit locos plus lighted cars, easily exceeding the rating of the switch machine. Plus most have 2 sets of SPDT, or in other words, one DPDT, which means if you use all the contacts for the loop you have none left for signals or frog power.
As for the eyes glossing over, I did say youc ould ignore the terms and just go on with what I explained they mean. No model railroad book will explain any of this. Even basic electricty, model railroad boosk tell you how but do not explain theory. The palce to go would be to find a book on basic electricity. A PSX is just a name of a device, like the AR1 is the Digitrax device, the PSX is Tony's product. A MOSFET - well, it will take a lot more tha a basic electricty course to understand what a MOSFET is and how it works. Suffice to say, they are commonly used as high current solid state switches. Which is what I thought I got across in my explanation. Unlike physical contacts, such as a relay or switch, a semiconductor device will have some measurable resistence between its electronic 'contacts' and resistence is what causes voltage drops. No need to dig further. To inderstand the DETAILS of how one works and what all the parameters are, would require a course in solid state electronics. There's no 'cheat sheet'. Explaining what MOSFET stands for - Metal Oxide Semiconductor Field Effect Transistor - I'm sure makes even LESS sense. Again, the 'cheat sheet' as best as can be is that it is a switch, but an electronic one, not a physical toggle switch. The detailed properties of the device explain why it is the preferred device for some types of circuits whereas others may use an ordinary silicon transistor. Knowing which to pick and when - that's a whole Electrical Engineering degree.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
HaroldAI remember, and not too fondly, the days of an Altas Controller that only worked when I was very profane, or of the days of block wiring when it seemed as if I was forever throwing toggle switches.
HaroldAWhen someone starts talking about PSX's and MOSFETS, my eyes just glaze over because I don't have the understanding.
Well then, I'm afraid that they have baffled you with BS, or at least with more information than you needed. The PSX or AR devices are what I would call black box items. That is, you only need to have a minimum of knowledge to get them to do what you want. In the most basic case there are two wires going in from your power bus, and two wires coming out to the loop. If you managed to get the rest of your layout wired for DCC, I think you will have no trouble figuring out how to make the connections. You don't really need to know what happens inside the device.
There are no relays, contacts, or toggles to contend with. And if there is really some small power loss associated with the device, I don't think you will notice.
Just my opinion.
OK, let's go over this once again.
On a DCC layout with a reversing section (loop), a short occurs but an auto-reverser like an AR1 or a PSX-AR reacts quickly to reverse the polarity and end the short before any harm is done.
It happens so quickly that the momentary short isn't even noticeable.
What's wrong with that?
What's with the obsession to prevent a short from occurring in the first place?
Rich
Alton Junction
Rememebr the DC days? There was never a short, you flipped the toggle BEFORE a short happened. Only more recently did anyone figure out that if you can build a circuit to ract fast enough, you can wait for the short and then fix it. It conflicts with my engineering background that says a short is a fault. Every time.
Plus, I control my reverse loop with a $3 relay and I don't need a $14.95 AR1 or even more expensive PSX-AR. I'm also cheap. I haven't checked for mass quantities from eBay seller in China like for other electronic components, but I suspect I can probably get relays with sufficient current capacity for even less than that.
No doubt the 'black box' of an autoreverser is easier to understand, if treated as a black box, anyway. 4 wires to hook up, done. Relay needs the crossed wiring plus the 4 wires, plus connections to the coil via the switch machine contacts, and a power supply (no big deal since my layou will have an accessory power bus around it to tap off for lights, switch machines, UP5 panel power, etc.). Zero adjustments, and the train will always roll smoothly across the gaps, jut like moving from one piece of mainline track to another, no chance of even the slightest hesitation or stutter or break in sound.
rrinker Rememebr the DC days? There was never a short, you flipped the toggle BEFORE a short happened. Only more recently did anyone figure out that if you can build a circuit to ract fast enough, you can wait for the short and then fix it. It conflicts with my engineering background that says a short is a fault. Every time.
OK, I hear ya. A short is a fault.
But, what's the harm if the auto-reverser immediately corrects the short?
From an engineering standpoint, the idea is to design a system so there are no faults.
Fault recovery is second level, a backup as it were, for a fault creeping in regardless of attempts to prevent them.
Not that it's on the same scale, but, theoretically, if you short a wall outlet in your house, your house won;t burn down because there is a circuit breaker that trips and cuts off the power. It's there in case soemthign you plug in shorts out. Ideally, you never want to have the short in the first place.
It does follow through with the rest of my layout building - the booster has a circuit breaker, and I do have downstream circuit breakers. But I srive to make my trackwork as buleltproof as possible, so nothing ever deraisl adn NEEDS that circuit breaker to act. SO far that's worked out pretty well, when I can shove cars through the trackwork at warp speed and they all stay on the rails, I know it's going to be pretty good for runnign at proper speed, and that's proven out across the two layouts I've built since going DCC. I expect my next one to be no different, even though it is going to be by far the largest layout I've ever built, AND i will be using a track brand that I have personally never used before. But with slow, careful work and making sure all straights are straight and not all kinked, and all curves flow smoothly with easements, and the roadbed nice and even withou humps to put vertical kinks in the track, and grade transitions made very gently and smoothly, I expect nothing less than the same high reliability I've gotten from my previous layouts.
I guess I am just picky. I'm not very good at scenery, but dang it, my track and electrical work is as bulletproof as I can make it. There may be no pretty scenes for the trains to run through, but run they do. With my usual slow rate of construction, I may dabble in some of my own control circuits for this stuff, but it's hard to justify other than for satisfying a curiosity since the commercial products are relatively inexpensive and I can just buy something and hook it up and know it works, without trying to put my own code into my own hardware and test and test to validate it all.
Randy, thanks for that follow up. It makes for an interesting discussion and for some difficult considerations in layout building
If your layout calls for one or more reversing sections, I see three options.
One, just use commercially available auto-reversers.
Two, avoid shorts and build in DPDT toggle switches to reverse polarities manually before the short occurs.
Three, design a relay system to automate the reverse polarity process before the short occurs.
With the use of DCC I would think that using automatic reversers like the AR1 or PSX-AR would be the only way to go. using a DPDT reverses the polarity on the main line while the AR units reverse just the reversing section.
retsignalmtr With the use of DCC I would think that using automatic reversers like the AR1 or PSX-AR would be the only way to go. using a DPDT reverses the polarity on the main line while the AR units reverse just the reversing section.
I am not aware of any problems or damage that an auto-reverser can cause or any problem caused by the occurrence of a short due to reverse polarity.
I hope that no one on this thread is suggesting that the use of an auto-reverser is a bad thing.
You use the DPDT switch on the Reverse Loop when using DCC - It acts exactly like the AR1 (seeing as how it is using a Mechanical Relay).
Don't get the 2 technolgies mixed up (DC & DCC) the way they handle Reversing Loops is way different!
BOB H - Clarion, PA
richhotrain Randy, thanks for that follow up. It makes for an interesting discussion and for some difficult considerations in layout building If your layout calls for one or more reversing sections, I see three options. One, just use commercially available auto-reversers. Two, avoid shorts and build in DPDT toggle switches to reverse polarities manually before the short occurs. Three, design a relay system to automate the reverse polarity process before the short occurs. Rich
I have been following this discussion. I would add that a couple of determining factors for deciding which approach to use might be a person's 'comfort level' with one method over another and maybe individual knowledge. I am perfectly comfortable using an auto reverse unit but don't ask me to build a relay system. Wouldn't have a clue.
I also appreciate what Randy has been saying. It's great to know someone who takes such pride in building an error free layout. I did see a comment he made about running a common power bus for turnouts, lights, etc. To some degree I have done that but since I am planning some modifications this winter, that one is now in the cooker.
There's never time to do it right, but always time to do it over.....
Bob nailed it - if you should want to use a DPDT toggle for the reverse loop, it gets wired like the AR, not like you would wire a reverse loop direction switch in DC, so it flips the loop, under a moving train.
WHat I suppose confuses people is that it would work FINE wired like a DC reverse switch. You'd be flipping the phase under EVERY train except the one on the loop - but this is fine.
The relay method automates it with no special circuitry. As the train runs around the loop, you have to change the turnout from allowing access to teh entrance of the loop to conencting the exit of the loop, otherwise you will have a derailment on your hands. So the relay gets controlled by the motion of the switch machine. Sure, you can ALSO do it with som sort of IR beam breaker detection, like IRDOT or something, so that as the train passes a certain point, it flips the relay, but this gets right back to being complicated.
What might be missing - a decent relay for this purpose has a low current coil - 30ma or less. In one position of the turnout, the relay will be on, solid. But they are deisnged for this, and also draw 30ma or less, so it's like a stall motor switch machine. No fancy arrangments of relays to emulate a latching relay, or a real latching relay, are needed.
rrinkerWHat I suppose confuses people is that it would work FINE wired like a DC reverse switch. You'd be flipping the phase under EVERY train except the one on the loop - but this is fine.
I guess that I question this. Let us suppose that one train is entering the loop and there is a following train. Are we saying that when the first train totally enters the loop and the toggle switch is thrown to change the phase under the following train there is no effect on the following train during the phase change?
And in another scenario let us assume that a single train enters the loop and the toggle is thrown to change the phase on the track entering the loop. Let us further assume that this section of track eventually encounters a boundary with another power district. Certainly the phase change will cause a problem at that boundary similar to what would occur if the track in question had been wired incorrectly to begin with. Is that not correct??
Well, if you wire a toggle in DCC the same way it's wired in DCC - you switch ALL of the layout, except the loop. In DC the extra reversing toggle comes before ALL block toggles, the equivalent would be havign the switch before ANY breakers dividing the layout into power districts.
CHanging the phase under a moving train has no effect - it's exactly what an autoreverser does.