FORUM CLINIC: 12 years using DCC - SIGNIFICANT NEW INFO!

[jfugate]

QUOTE: Originally posted by rrinker

Exactly. When there is no train in the block, the lights do nothing. The unoccupied track is an open circuit.

Bingo!

It's the locos that draw the current. The bulbs really are not a consideration until you get a short or put enough locos on the track that you approach 2 amps. THEN, and only then, do the bulbs enter the circuit as a noticeable load -- and they limit the current to just over 2 amps, and *they become* the load.

It's this current limiting feature of the bulbs that makes them do their magic. For $1 each, it's hard to beat their value for adding load controlling short management to a DCC layout.

[rrinker]

Exactly. When there is no train in the block, the lights do nothing. The unoccupied track is an open circuit.

[Anonymous]

[jfugate]

Ron:

The short answer is: you will not overload your system by adding a dozen or more bulbs in series with your power feeds. It's the locos in the circuit that create the load, not the bulbs.

Now the long answer.

If you design the power feed side of your DCC layout properly, you will typically break things up into two or more power districts, each with their own booster feeding the power. The only exception to this might be the smallest of layouts, like anything under about 50 square feet (the common 4 x 8 is 32 square feet).

By having multiple boosters feeding things, you automatically divide up the bulbs across the layout -- each booster and power district can be thought of as its own isolated system for the most part. The only thing connecting the booster districts together is their common link back to your system's command station. But electrically speaking, each district is its own isolated circuit.

And the bulbs don't do anything until there is current flowing to locos on the track, and while the bulbs are cold, their effect is imperceptible to occupied track. So the real question is how many locos can you load into a single power district? You'll hit the current limit of your booster first from locos on the track -- that's the real concern.

If you wire your layout with train length blocks and one bulb per block, you can have a layout that fills an aircraft hanger and has thousands of train blocks, and you don't need to worry about the bulbs. With a layout that size, you will also have hundreds of power districts -- each with their own booster. And each power district electrically is its own isolated circuit. It's the number of locos typically on the track in a power district that will be your limit, not using 1156 bulbs to do short management.

[n2mopac]

QUOTE: Originally posted by rrinker

QUOTE: Originally posted by selector

Randy, or Joe ( or an electrical engineer who is reading), I understand Randy's response, and it makes sense. Just for my edification, is there an exponential relationship involved? I ask this because the light bulb will be drawing current all the while if it is ever going to draw current. So, is the draw inversely proportional to ....what, resistance? Help me to appreciate what the bulb is doing when it is not protecting the booster and locos during normal running.

Acting like a resistor, a low value one. There are devices called Negative Temperature Coefficient Resistors (NTC) that would work the same as the light bulbs, if the proper values can be found. But they cost more than 1156 bulbs.

--Randy

So, one last clarification, assuming I am not going to put 100 bulbs on my 300 feet of trackwork, is there a point when all of these "resistors" beging to cause a power drain on the system or overload it by their sheer numbers?

Ron

[jfugate]

Great answers, Randy -- thanks!

[selector]

Acknowledged, Randy, and thx.

[rrinker]

QUOTE: Originally posted by selector

Randy, or Joe ( or an electrical engineer who is reading), I understand Randy's response, and it makes sense. Just for my edification, is there an exponential relationship involved? I ask this because the light bulb will be drawing current all the while if it is ever going to draw current. So, is the draw inversely proportional to ....what, resistance? Help me to appreciate what the bulb is doing when it is not protecting the booster and locos during normal running.

Acting like a resistor, a low value one. There are devices called Negative Temperature Coefficient Resistors (NTC) that would work the same as the light bulbs, if the proper values can be found. But they cost more than 1156 bulbs.

--Randy

[rrinker]

QUOTE: Originally posted by Bullitt406

What about Atlas Custom Line versus Super Switch. Any difference between those other than appearance?

Josh

As far as i can tell, exactly the same. When the Super Track Code 83 first came out, the Custom Line was the previous revision, not nearly as nice in appearance as the latest versions are. I think Atlas included a bit of what they came up with for the Super Track when they redesigned the Custom Line stuff - at least the Code 83.
Electrically, they are the same.

--Randy

[selector]

Randy, or Joe ( or an electrical engineer who is reading), I understand Randy's response, and it makes sense. Just for my edification, is there an exponential relationship involved? I ask this because the light bulb will be drawing current all the while if it is ever going to draw current. So, is the draw inversely proportional to ....what, resistance? Help me to appreciate what the bulb is doing when it is not protecting the booster and locos during normal running.

[Bullitt406]

What about Atlas Custom Line versus Super Switch. Any difference between those other than appearance?

Josh

[rrinker]

You can theoretically have as many blocks and light bulbs as you want. Want a block every foot? Go ahead. But the return on investment (not to mention the truckload of light bulbs you would need) isn't going to be very good. Make the blocks too big, and you end up having more trains in the area than the light bulbs support, so they start lighting up and slowign down the trains. That's why the idea of a 'train length' block is as close to ideal as you can get. One train won't trigger the bulbs unless there really is a short.
The output capacity of the DCC system doesn't really have anything to do with how many blocks you can create, only with how many of those blocks can be filled with trains at one time.

--Randy

[n2mopac]

QUOTE: Originally posted by jfugate

Ron:

Yes, that would work, assuming there is only one track feeder to the train block, and it has an 1156 bulb in it.

Great! One more question (I think this was addressed above, but I can't find it). Is there a limit to the number of bulbs you can wire into a layout. I, for example, am presently laying the main that I intend to power with MRC's Prodigy Advance. It comes with a 2.5 amp power supply and I intend to run a 5 amp booster as the main progresses. What would be the maximum number of blocks/bulbs that I could use with such a setup, and how do I figure that number. (I know just enough about electronics to make me REALLY dangerous.)

Thanks again for all your insight. This is all very timely and helpful to me as I am constructing my new layout right now and am venturing into DCC for the first time on it. Thanks again.

Ron

[electrolove]

Thanks Joe for your quick answer. [:D]

QUOTE: Originally posted by jfugate

QUOTE: Originally posted by electrolove

Joe:

Thanks for the great clinic. One question for you. Whalthers/Shinohara DCC friendly turnouts, can they be used right out of the box without problems?

Yes, that's the beauty of all the DCC friendly turnouts listed. You just install them, solder track feeders to the stock rails, and you're done.

[jfugate]

QUOTE: Originally posted by electrolove

Joe:

Thanks for the great clinic. One question for you. Whalthers/Shinohara DCC friendly turnouts, can they be used right out of the box without problems?

Yes, that's the beauty of all the DCC friendly turnouts listed. You just install them, solder track feeders to the stock rails, and you're done.

[electrolove]

Joe:

Thanks for the great clinic. One question for you. Whalthers/Shinohara DCC friendly turnouts, can they be used right out of the box without problems?

[jfugate]

TOPIC THIS POST: All about DCC friendly turnouts

So what's the big deal with DCC Friendly turnouts anyway? Actually, it's all about shorting the track, and the most common place a short can occur on a layout is at turnouts.


(click to enlarge)

A DCC friendly turnout is a turnout with less chance of a short if something doesn't track right. And a DCC friendly turnout can be a key component of good short management, in addition to the 1156 bulb we've already discussed.


(click to enlarge)

Any time I've seen the DCC friendly turnout discussion come up on boards and forums, there's always a lot of talk about the points end of the turnout, but I don't think that's the real issue. More derailments happen at the frog end of the turnout, and it's also the more likely place you'll see a short.


(click to enlarge)

So to focus on the frog end, you have two basic choices with how you handle the frog – live (powered) frog or dead (unpowered) frog.


(click to enlarge)

In both cases, the frog is generally isolated with gaps (unless you use the turnout to do power routing). If you look at a live frog turnout, the polarity of the power to the frog is based on the position of the points.


(clcik to enlarge)

But with a live frog turnout, if you run through the turnout from the frog end with it thrown against you, you will get a short. And we don't like shorts on a DCC layout!


(clcik to enlarge)

By contrast, with a dead frog turnout, you expect the other wheels of the loco NOT on the frog to pickup the power. Most modern locos have all-wheel pickup and a dead frog is not an issue. Only if you have extremely short wheelbase locos will you have a problem with dead frog. I model the 1980s SP, so dead frogs work fine for me.

I spoke with Gary Siegel recently (owner of the wonderful L&N layout covered in MR back in the late 90s) and asked him which he was using: live frogs or dead frogs? Like me, Gary started out using live frogs but has now gone to dead frogs.

As to the pros and cons of live frog versus dead frog ...

LIVE FROG
Advantages:

• No dead spots in the track
• When entering a turnout thrown against you from the frog end, the train stops (shorts) preventing a derailment.

Disadvantages:

• Need contacts to route power reliably to the frog, which is more work since using the points alone is not reliable over time.
• A short on a DCC layout during “normal operations” is generally BAD.

DEAD FROG
Advantages:

• Easy to do – no extra wiring
• No short if the turnout is thrown against you – you just go on the ground like the prototype ( which can about 10% of the time cause a short ... so nothing's perfect).

Disadvantages:

• Dead spots in the track can be bad
• May have to cut gaps around the frog – which means more work.

WHAT I PREFER
I prefer dead frog turnouts. Modern all-wheel pickup diesels don't even notice them, although some older steamers without all wheel pickup may have trouble, as well as really short wheelbase locos.

You can get dead frog DCC friendly turnouts from several sources today:

• Atlas code 83 Super -Switch (#4, #6, or #8)
• Peco insulfrogs
• MicroEngineering turnouts
• Walthers/Shinohara DCC Friendly models (although in high demand so are often out of stock)

Older Shinohara/Walthers turnouts need modification to be made DCC friendly.


(clcik to enlarge)

Here's a photo of a Shinohara/Walthers old-style turnout that's been modifed to be DCC friendly (this turnout was recycled from an earlier layout, which is why it looks kind of grungy).

Study the photo and you can see what was done to make it DCC friendly.

• Replace the throwbar with a printed-circuit (PC) board throwbar. Remove the foil from the center of the throwbar so you don't short the two points. The foil under the stock rail helps conduct power to the nearby point so you don't rely on point contact alone for them to get power.
• Attach each point to its closure rail with half a rail joiner. This makes the point the same polarity as its closure rail, which minimizes the chance for a short at the point end if there is ever a derailment there.
• Solder a PC tie under the closure rails, to route power from the stock rails to the closure rails. Remove the foil in the middle of this PC tie so you don't short the two closure rails together.
• Cut gaps around the frog, and fill the gaps with black or gray styrene plastic glued in place with superglue. Leave the frog unpowered.

If you modify an older turnout as shown, even if you get a derailment at a turnout, you will only infrequently also get a short.

WHAT IF I NEED TO USE LIVE FROG?
If you do need to use live frog turnouts, then put an 1156 bulb in the feeder wire to the frog. This way, if you do happen to get a short from running the turnout the wrong way, only the one who gets the short will be affected by it – not the whole power district or layout.

TOPIC NEXT POST: Programming loco decoders

[Anonymous]

This is very good information, and applies to both DCC and non-DCC layouts. Keep it coming! It's especially appropriate for guys like me that are building a new layout and getting ready to lay track by this weekend!

Mark in Utah

[jfugate]

Ron:

Yes, that would work, assuming there is only one track feeder to the train block, and it has an 1156 bulb in it.

[n2mopac]

Sorry I'm a little behind here, but here is one question. Assuming each block is train length, couldn't a person simply wire the bulb into a feeder from the main bus to the track and forgo the 3rd bus wire. I know this may go againsy your standard of supplying feeders to every track section, Joe, but theoretically will this not work?

Ron

[ereimer]

QUOTE: Originally posted by jfugate

Before we leave the 1156 light bulb short management, have we answered everyone's questions? Do the two posts make it clear why and how?

i think so Joe , even i understand how the bulb works now [:D]
now we just need someone to test the block detector issue to see if they will work

[rrinker]

You certainly could - just use scraps of dimensional lumber fromt he benchwork to make 'stands' to lift the bulbs out of direct contact. Or you could (at great expense) buy actual sockets for the bulbs which would put the base of the bulb near the benchwork and the globe well above it.
Keep in mind in their intended application, these bulbs are inside a tightly closed plastic space - and I don't hear very many reports of the taillights in cars catching fire. Given that a 12v short in a car is far worse than even a 10 amp booster shrting out, I don't think there's much issue there.

--Randy

[Bikerdad]

Yup, makes things clear. Thanks a lot. Any thought to perhaps creating a little more "standoff" for the bulb to reduce fire risk from "microscopic" to "infantesimal"?

[jfugate]

Before we leave the 1156 light bulb short management, have we answered everyone's questions? Do the two posts make it clear why and how?

[jfugate]

Selector:

It took one other guy and me 3 eight hour days to rewire the Siskiyou Line to be set up this way. That was back in the summer of 2000 and it was the best thing I've ever done wiring-wise to the Siskiyou Line. It was time *very well* spent.

[selector]

All very clear, and clearly a lot of work!!![:O] I can see that you get great benefits from all that work, but....whew..that must have taken many days to do, Joe. I'm glad that my 'salvation' will be done in about two hours.

[jfugate]

TOPIC THIS POST: Wiring details for 1156 bulb train blocks

Here's a typical DCC layout with three power districts, each powered by their own booster.


(click to enlarge)

Let's zoom in on the wiring for a single power district and then see how we “upgrade” it to add the 1156 bulb short management.


(click to enlarge)

Here we are zoomed in on our power district. Let's go through the steps to add the 1156 bulb short management protection.


(click to enlarge)

To add the 1156 bulb short protection to the power district, you first need to cut gaps in both rails to break the track up into train length blocks. Here we can see this gives us 8 train blocks.


(click to enlarge)

Next, we need to run a third bus wire for each train block thereby creating an 1156 protected bus feed for each train block. For each block we add a feed from the main bus to the third bus, through a SPST toggle switch and an 1156 bulb. If you wire a SPST toggle into each train block bus feed, you can kill the power to the block for debugging purposes, like when looking for mystery shorts. This is what I have done on the Siskiyou Line and it has come in handy more than once.


(clcik to enlarge)

You can now add the toggle and 1156 bus feed to each train block bus, and run track feeders to each rail section for a given train block from its associated third bus and the block will be protected from shorts by the 1156 bulb.


(clcik to enlarge)

Once you add all the toggles, bulbs, and track feeders for your power district, here's what you get. This is the complete wiring to add 1156 bulb short management to your layout.


(clcik to enlarge)

HOW THE WIRING ACTUALLY LOOKS ON THE LAYOUT
Here's how I wire in the bulbs. I solder wires to the terminals on the bulb, add crimp spade connectors so I can wire the bulb into a terminal strip. I mount the bulb using a plastic 3/4” cable clamp and a drywall screw.


(clcik to enlarge)

And here's how it all looks under the benchwork at a train block location on my Siskiyou Line.


(click to enlarge)

TOPIC NEXT POST: All about DCC friendly turnouts

[jfugate]

All:

I hope to post the wiring installment tonight ... that should help more of you get an "ahah!" from all this light bulb discusssion.

[jfugate]

Selector:

You got it! Should work just as you describe.

[SpaceMouse]

Okay, Joe, now I gotcha. I can split the inner loop from the outer loops with no problems, which is consistant with usage and should ONLY TAKE ABOUT A DAY to do.

But then again this layout is my learning layout so I'll do it.