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

[jfugate]

Next post on wiring tips coming later today ...

[jaretos]

I was wondering if the topic of wiring tips will cover wiring for accessories such as Tortoise switches? I am planning to upgrade some manual throws to Tortoise and have Digitrax Zephyr system. This info as well as all the other stuff has been fantastic!

john

[jfugate]

John:

In a round about way, yes. There's a lot of hoop-ta-la around finding edge connectors to allow switching out tortoises, but the techniques I cover will make switching out tortoises easy enough. I'll make sure and mention tortoises in this context in the post just so it's clear.

Glad you find the info helpful. Anything to help my fellow model railroaders along!

[jfugate]

Best laid plans ... ah well, looks like I'll finally be able to post the wiring tips over lunchtime tomorrow. [:D]

[jfugate]

TOPIC THIS POST: DCC Wiring tips
These wiring tips for the most part are not DCC specific, but are good general practices for layout wiring, even for straight DC.

Neatness counts when it comes to wiring your layout. Wiring is my least favorite part of the hobby*, so I don't want to have to do it more than once, and I also want debugging to be easy to perform if there is a problem.

We'll assume in this post that you paid heed to the earlier post about bus wire size, and that you're putting in a hefty track power bus. The same goes for feeder wire – because of the amps we want to put through the wiring with DCC, don't scrimp on track feeders wire guage either. I use 18 guage stranded wire for my feeders. Now how do you get the feeders from the power bus to the track?

FROM THE BUS TO TERMINAL STRIP
I use several connectors in my wiring to make it easier to hook things up, and to make it easy to disconnect a track feeder if I need to debug a “mystery short”.



I use terminal strips and spade crimp connectors to hook things up. And I like to use 3M suitcase connectors (officially called “Insulation Displacement Connectors” or IDCs by 3M) to connect in my bus wire drops. I use 12 guage stranded bus wire on the Siskiyou Line and 18 guage stranded track feeder drops. I connect up an 18 guage wire to the bus wire with a suitcase connector, then add a spade crimp connector on the other end, and screw that to one end of a terminal strip on the bottom screw.

Then I take some bare 18 guage solid copper wire and run it from the bottom screw where the spade connector is, to all the bottom screws on the terminal strip to short all the terminal strip connections. Now I can hook track feeders to the top screws on the termial strip and get power to the track. If I need to disconnect a track feeder, it just takes a few seconds with a screwdriver.

I either connect each of the two track feeder buses to their own terminal strip or I split a longer terminal strip into two halves, with each of the two track buses feeding their own half of the terminal strip.

I also connect in an 1156 auto tail light bulb to the bus for short protection, and a toggle switch so I can kill power to the entire “block” -- again all for short management. But we'll save the details of short management for the next post.

FROM TERMINAL STRIP TO TRACK
Once I get several terminal strips with power bus drops along the track, then I start running feeders from the terminal strip to the rail. I prefer to power each 3 ft section of rail with its own feeder and not rely on rail joiners to do the job. On really short sections of track (under 1 foot), I'll usually solder the rail joiners to a longer piece of track that has it's own feeder.

As mentioned, I use 18 guage stranded wire for track feeders. I solder one end to the rail and put a spade crimp connector on the other end and screw it into a terminal strip being fed by the correct power bus for that rail. If you connect a track feeder to the wrong terminal strip connection, you'll get a short, so pay attention!

Here's a photo that shows how this all looks once it's wired up as I describe:

To see a much larger image, click: http://siskiyou.railfan.net/images/wiring.jpg

ATTACHING FEEDERS TO THE TRACK
I'm a stickler for nice looking track, so I don't want my track feeders to show. Take a look at the photo below. Can you see the track feeders at the location of the arrow?


(Click to enlarge the photo)

The feeders are there, they're just on the back sides of the rails. If we put a mirror back behind the track, you can clearly see the feeders (marked with orange circles):


(Click to enlarge the photo)

If you're careful, with a little practice you can attach 18 guage feeders to the back sides of the rails and won't even need a file to keep it clear of the flanges on the one inside rail base.

I crimp the bare end of the 18 guage stranded wire feeder with a pair of needlenose pliers to make the flat like a small spade, then bend just the tip over 90 degrees and solder it just to the base of the rail, as shown here:


(Click to enlarge the photo)

With a bit of practice this is easy to do and I sure like how you can't even see the feeders when they're done like this. When you ballast the track, just paint these feeders the color of your ballast and they'll disappear from sight!

Next post, we'll take a look at how to “short-proof” your DCC layout – in other words show you how to make it so you hardly even notice when someone accidentally shorts the track.

TOPIC NEXT POST: Short-proofing a DCC layout

*NOTE: Why do I dislike wiring so much? I can literally spend days working on the layout and when the guys come over to look at the layout, they comment, “Doesn't look like you've worked on the layout at all this month.” I just mumble to myself once again how much I hate wiring! <grin>

[Anonymous]

Just when I was beginning to suffer Fugate-withdrawal, another great read. Thanks once again, Joe.

John
Austin, Texas

[Anonymous]

Im using Atlas turnouts as well as Atlas under-table switch motors. Is there any way to throw the atlas motors with a zephyr? My plan at this point is to wire them up to toggle switches on several control panels around the room all powered by my old power packs from my pre-DCC days.

[jfugate]

Jaretos:

I promised earlier that I would mention Tortoise machines and how I wire them.

I solder two 18 guage wires to the power feeds on the tortoise circuit board, and then put crimp spade connectors on the ends of the wires, and connect the wires to a small terminal strip that has the machine feeds for that tortoise. I use a toggle switch to throw the tortoise, so the feeds from the toggle go to the terminal strip near the tortoise.

I don't bother with trying to find fancy edge connectors for the tortoise -- just using the wires, spade connectors and terminal strip allows me to make the tortoise easily replaceable if needed.

[jfugate]

JPM:

You can get accessory decoders (as opposed to "mobile decoders" that go in locomotives) to throw solenoid switch machines and to throw "slow motion" switch machines like the tortoise.

I consider DCC control of switch machines to be overkill, because it's rather expensive (over $5 per turnout) and not-as -obvious to your operators how to throw a turnout . A simple toggle switch or push button will do the job and generally costs quite a bit less than $5 per turnout.

I'm a big proponent of walk-around layout design, and the general thinking with this design approach is to put the turnout control on the fascia directly opposite the turnout. That way an operator can see the toggle or button and know exactly how to throw the turnout.

Otherwise your operators need a throttle with extra buttons for accessories, and they need some way to figure out what number the turnout is, and then they need to learn some key sequence for throwing the turnout. It might be cool technology, but it feels more like Star Trek to me and less like railroading.

Real railroaders generally throw some sort of device right at the turnout ... not press buttons on a panel in their locomotive (that is, use buttons on their throttle to throw turnouts). I prefer to keep the "railroad operating" thoughts as similar to the real thing as possible.

In this same vein, I don't like a digital fast clock for a 1950s transition era layout. Give me something with hands, because that's more like what was in use then to tell the time! Digital reminds me of the 1970s, not the 50s! Same idea.

[MisterBeasley]

I'm think of putting panel-mounted meters on my system to monitor loads, particularly as I convert older engines to DCC operations. Will plain DC meters work, or will the DCC signal mess them up? What would be a suggested range for the meters, and should they be center-zero meters like I would use on a DC system?

[jrbernier]

MisterBeasley,

The 'RRampMeter' from Tony's Trains should do the trick. There may be others. Here is the URL: http://www.tonystrains.com/technews/rrampmeter.htm

Jim Bernier

[SpaceMouse]

Could someone smarter than me do a circuit diagram of Joe's work. I'm not getting it.

[jfugate]

Spacemouse:

See: http://siskiyou.railfan.net/model/constructionNotes/wiring.html

That should help ... it includes a diagram. The light bulb and SPST toggle is probably throwing you. I haven't really discussed them in detail yet.

[jfugate]

QUOTE: Originally posted by jrbernier

MisterBeasley,

The 'RRampMeter' from Tony's Trains should do the trick. There may be others. Here is the URL: http://www.tonystrains.com/technews/rrampmeter.htm

Jim Bernier

This is a great device and if you want to monitor current draw this or something like it is what you will need. A regular meter won't work with DCC.

A poor man's way to measure your system voltage is to use the function leads on a decoder and a standard Volt-Ohm Meter (VOM).

Just hook one VOM lead to the blue common function wire and the other lead to the white function wire, address the decoder address, make sure the direction is forward, and then press F0 to turn on the function.

The voltage you read at the meter for all practical purposes is your track voltage. In my case, it's a perfect 14 volts.

[rrinker]

And don;t be alarmed if you have a digital meter and it reads negative - you just have the leads backwards [:D]

If you have an analog meter, it;s not a good idea to hook it up backwards, the needle can be damaged. The Plus lead goes to the blue wire.

This applies to wiring LEDs and so forth to finction outputs - the blue common wire is PLUS (which can be confusing because people often interpret 'common' as meaning 'ground'). Switching the negative side like this is called 'current sinking', think of it like the drain in your sink (ok, I stole that one from Bruce Chubb but it's a great analogy). When the function is switched on, the 'switch' (which is a transistor in the decoder) closes to connect the line to the negative side of the circuit. This allows current to flow from the blue positive side through the bulb, led, or whatever, and then to negative, making a complete circuit.

FWIW, switching the positive side is called current sourcing. Think of that one as switching on and off the faucet in your sink - you are controlling the SOURCE. Versus opening and closing the drain.

There will be a quiz later. [:D]

--Randy

[jfugate]

Randy:

As usual, a very helpful informative post. Thanks!

--------------------------------------------------------------------

To everyone:

I'm going to be out from tomorrow through Sunday doing train stuff ... the local SP fans are holding a mini-meet including an excursion train behind SD9s down the Tillamook coast branch.

As part of this meet, I'm hosting an op session tomorrow evening and opening for layout tours on Sunday. Saturday is the excursion (got my ticket!) and Friday is an all day session with clinics, swap meet, and other fun goodies.

I plan to take photos of the op session to post for one of my regular op session reports, and I'll be taking photos of the excursion as well!

So the earliest you'll see the next post on this clinic is next Monday or Tuesday.

But at least I have a good excuse! [:D]

[rrinker]

Sounds like fun. Too bad pigs will fly before we see an excursion train around here other than on private tourist lines or the R&N. Between the insane insurance costs and Norfolk Southern's pretty much outright refusal, I'm not holding my breath.

--Randy

[n2mopac]

QUOTE: Originally posted by jfugate

Randy:

As usual, a very helpful informative post. Thanks!

--------------------------------------------------------------------

To everyone:

I'm going to be out from tomorrow through Sunday doing train stuff ... the local SP fans are holding a mini-meet including an excursion train behind SD9s down the Tillamook coast branch.

As part of this meet, I'm hosting an op session tomorrow evening and opening for layout tours on Sunday. Saturday is the excursion (got my ticket!) and Friday is an all day session with clinics, swap meet, and other fun goodies.

I plan to take photos of the op session to post for one of my regular op session reports, and I'll be taking photos of the excursion as well!

So the earliest you'll see the next post on this clinic is next Monday or Tuesday.

But at least I have a good excuse! [:D]

Hope you hage a great time, Joe. We'll look for you next week.

Ron

[jfugate]

Hope to post the next installment on short-proofing tonight.

[jfugate]

TOPIC THIS POST: Short-proofing a DCC layout

Shorts on the layout during an operating session can drive you crazy, so it pays to wire your layout to make it as short proof as possible.

But what's the big deal with getting a short on a DCC layout?

WHY SHORTING THE TRACK IS BAD
On an old style cab-control layout, you cut gaps in the track about the length of a train, and then you wire toggle or rotary switches to each block to allow you to select which “power pack” or “throttle” (called a cab) you use to power the locos.

This gives you a form of independent control of your train, and if you happen to short something (like run through a turnout from the frog end when it's thrown against you) -- the only one who notices is you. All the other trains are connected to other cabs, so they don't even notice.

On a DCC layout, you can literally connect two wires to the layout and run all the trains (but if your layout is much larger than a 4x8, you'll have voltage drop problems, so you'll need to run a bus wire and install rail feeds around the layout as described in earlier posts to avoid this).

But since everything is being powered by one “power pack” (called a booster in DCC terminology) if you short something, *everybody* is affected!

On a DCC layout, when someone shorts the track somewhere, it's common to have everyone else in the room holler, “Hey, who shorted the layout?”


SHORT ISOLATION: MORE BOOSTERS
The first thing you can do to ease the problem with shorts on a DCC layout is to follow the basic notion that made cab control not have the short problem, and that's to gap the track and install more boosters. These subdivisions are called “power districts” in DCC terminology.

Creating power districts isolates the short to just that district. All the other power districts keep on running just like nothing happened. However, power districts are generally much larger than a train length because boosters aren't cheap.

The most common place shorts happen is at turnouts, so look for locations with lots of turnouts and put them in their own power district (like a yard or a major switching district). This will isolate the shorts to that location and will allow the other areas of the layout (like the mainline) to keep on running even if some guy in the yard runs a turnout thrown against him.


BEYOND SHORT ISOLATION TO SHORT MANAGEMENT
Getting a short on a DCC layout causes other bad things to occur, like scrambled decoders and damaged wheels or track. The reason is because with DCC, you have a constant high voltage (14-16 volts is typical) on the track with a current potential something like 5-10 amps, unlike old style cab control.

With the new crop of high inrush current sound decoders from QSI, shorts cause even more problems because after the booster cycles from the short, it sees the high inrush current of the QSI locos in the power district, and gets tricked into thinking the track is still being shorted, even though the short may be removed.

As a result, the booster may not come back on after a short. That power district will go down and just stay down when that happens. Not good.

The key to solving this problem is to move beyond power district short isolation to full fledged short management. With short management, the booster never sees a short, the layout “absorbs” the short in effect, and everything keeps on running except the train that caused the short, even in the same power district.

It turns out the way to get this sort of behavior is to use the humble old 1156 auto tail light bulb. If you wire a bulb in series with one of your track feeders, as long as the bulb stay cold (and does not glow) it has virtually no effect on the circuit. However, if you short the track, the bulb will suddenly glow and limit the current flow to around 2 amps.

This means the booster will think you suddenly put a lot of locos on the track, but it will not register as a short at the booster.*

If you gap the track into train length blocks (like you did for cab control) and have an 1156 bulb in series with the bus feeding that “train block” you get the ultimate in short management on your layout.

What happens is you can short the track in that train block, and all the other trains in the same power district will keep on running! The other trains in the same power district may slow a bit, but they won't stop running.

Here's a video clip from my upcoming Siskiyou Line Video volume 3 demonstrating how this all works.


To watch this video clip, click here: http://mymemoirs.net/preview.php

TOPIC NEXT POST: Wiring details for 1156 bulb train blocks

*NOTE: You need your booster to have a delay of about 500 milliseconds to allow the bulb to “kick in” and glow when a short occurs. This means, for example, you need to set your Digitrax booster delay setting to 500 ms to use the bulbs for short management.

[jfugate]

Not a lot of comments on this thread since the last post ... did it just stun everybody? Can you play the video clip, and does using the bulbs for short management make sense?

Any questions?

[simon1966]

Hey Joe, were you telling jokes before the video? Rick looks like he is going to burst out laughing at any minute!

Actually, this is really clear. I will be getting some bulbs shortly!!

[Bikerdad]

Joe,

Looks like a really good idea, although its unclear how the lightbulb is mounted/secured/electrically isolated (UL approved fixture, nah...). How does this compare in functionality and reliability with something like the Digitrax PM42?

[IRONROOSTER]

QUOTE: Originally posted by jfugate

Not a lot of comments on this thread since the last post ... did it just stun everybody? Can you play the video clip, and does using the bulbs for short management make sense?

Any questions?

I haven't been commenting because I have don't have DCC. But I have been following along because I have been thinking about getting a DCC system. This has been very instructive and yes it is clear.
Thanks for sharing
Paul

[jwr_1986]

Joe, every item you've discussed makes good sense to me except this one. Circuit breakers are the solution to go for here because they stop all current flow not just some. I saw a highresistance load short on a bulb protected block and melt an engine once. While the breakers may be considerably more expensive they are the only safe way. My opinion of course. Thanks for the other good tips though.

Jesse

[jfugate]

Simon:

Yes Rick and I were cracking jokes with the cameraman before the clip was shot. But we've also been told we look too deadpan in front of the camera so we've been working on trying to be more *alive* in our expressions and delivery. That's part of it too.

Glad you're finding the info useful.

[jfugate]

BikerDad:

You ask about the power shield / circuit breaker devices ... they don't always seem to work like you hope. Here's a posting on the DCC4everyone Yahoo list made by Marcus Ammann from Australia. He was experiencing problems with DCC shorts and booster recycling with lots of sound decoder locos on the track in his modular group, and stumbled upon the light bulb option.

After trying it out, he's totally sold on the light bulb approach, and has been expounding it to everyone who will listen. Here's some of what he's said recently:

QUOTE: Comments by Marcus Ammann posted on the DCC4Everyone Yahoo list
Back in July last year while building a demonstration DCC module I came
across this problem, more than one Soundtraxx equipped loco in the same
power district and my NCE EB3 circuit breaker would not reset.

I looked on all the lists and found on the Digitrax list that some
modellers were having this problem. I posted on the NCE, Soundtraxx and
QSI list my test results and asked for some help.

I looked up Tony's site and under his auto reversers, that since has
been updated, a statement about Sound Considerations at:
http://www.tonystrains.com/technews/dcc_autoreversers.htm
and it said that the reverser had problems resetting with Soundtraxx
locos in the same power district.

Mark Gurries on the NCE group suggested I wire a 12 volt lamp in
parallel with my NCE EB3 and I tested with different wattage lamps and
was able to get the results I wanted with 5 - 10 watt lamps.

Tony's Power Shields, Digitrax's PM42s and PM4s as well as the NCE EB3
were having the problems along with some power boosters. I helped a
couple of U.S. modellers get their layouts running properly, but they
were happy to fit the "fix" if it meant be able to have normal trouble
free sound loco operation, One was a Digitrax user and the other was an
NCE user.

Many modellers with BLIs were very disappointed with the booster
inability to reset after a short. I have only Soundtraxx units and I
have the problem.

I had the same problem when I connected my EB3 directly to my Command
station/booster track outputs and only one piece of track on my work
bench, so in my case nothing to do with my layout and wiring.

About August and September last year, many modellers on the QSI group
voiced their frustration about this problem and I am sure that is all
DCC system are suffering the same problem.

...

My own layout is wired up using light globes as the short protection and
power division device. When wiring my layout for DCC, I like many others
had lots of questions and used the Internet and found Allan Gartner's
Wiring for DCC at HYPERLINK

http://www.wiringfordcc.com/track.htm"http://www.wiringfordcc.com/track.htm

invaluable with so many answers to the question I had. At Joe’s great layout web
site he also has a reference to light globes for short protection and
now a 3 minute video clip where Joe explains about light globes.

I did not want my layout to completely shut down when an operator
created a short at a turnout so I wired in light globes. Making the power
district as small as possible at approximately ‘one train length” as Joes
does removes the chance that two trains can be in the same power district
and with as many as 4 to 6 locos, some with sound, exceeding the light globe
maximum operating current before it has an adverse effect on loco
performance.

I have gone one step further than Joe shows in his video, I have mounted
the 1156 in a line side building, so that when the train stops, the building
lights up bright, not really very prototypical but a very good visual warning.

Wiring a layout for DCC now has become a little more than that “one
liner” about one of the advantages of DCC is “simple wiring, just 2
wires the track, that’s all there is”. But just like everything in life,
nothing is easy and for better operation (no complete shut downs of
layout with a short) come at a price, time effort and a few dollars.

The booster not being able to reset after a short circuit with a few
sound locos in the same zone does not happen when using light globes as
Joe also suggested, because the booster never shuts down if everything
is working correctly EXCEPT when you are operating near the booster’s
current limit, and now is the time to add a second or third booster.

Thanks a lot Joe.

Marcus Ammann
Model Railroading in Australia

[jfugate]

QUOTE: Originally posted by jwr_1986

Joe, every item you've discussed makes good sense to me except this one. Circuit breakers are the solution to go for here because they stop all current flow not just some. I saw a highresistance load short on a bulb protected block and melt an engine once. While the breakers may be considerably more expensive they are the only safe way. My opinion of course. Thanks for the other good tips though.

Jesse

Like I said in the beginning of this FORUM CLINIC, you may not agree with me, and that's fine. We can continue to disagree and that's okay. If we all agreed on everything, life would be pretty booring!

In theory, Jesse, I agree with you. Stopping the current rather than limiting it would certainly be the perfect solution. In practice, there are several considerations that make the ideal not as perfect as it might at first appear.

1. Cost
It's hard to beat $1 per train block for short protection. At $30 each for a Tony's Power Shield, it will cost you a lot more per train block to get a layout that absorbs shorts and doesn't affect anyone except the guy who caused the short.

2. Simplicity
The non-bulb electronic circuitry solutions don't always work as you would hope in all situations, like on booster reset after a short. The humble light bulb does what it does without any fancy circuitry, so it's actually more predictable. You throw the load in the block to about 2 amps or more, and the bulb's going come on and limit the current - guaranteed. You can't get more simple than that.

3. Practicality
For all practical purposes, limiting the block current to a maximum of 2 amps no matter what is as good as dropping the block current to zero when you get a short. The net effect is the same: no short at the booster and instant resets once the short is removed. While the electrical result is very different, the practical result on your layout is the same, and that's what matters to the guys operating your layout.

Electronic gadgets are cool and to a purist they are not as "sloppy" or as "crude" as an 1156 bulb. But if you turned it all into a black box and never explained to those operating the layout what made it short proof, the result is the same. They just know they could run and didn't have to worry about shorts shutting things down.

Sometimes the low tech solution is actually more elegant and I think the humble 1156 bulb fits this characterization well. Also, I ask those who disagree to point to their example layout that "absorbs shorts" without incident the way the Siskiyou Line does and then we can discuss the pros and cons.

As it is, the 1156 bulbs have been field tested on large layouts (mine, Marcus's modular group [see previous post], and my friend Charlie Comstock's layout, to name a few) and work very well -- and they didn't break the bank doing it.

[ereimer]

Joe , in your "DCC and Realistic Operations on the Siskiyou Line" DVD from the clinic you gave at the 2004 NMRA convention , one of the attendees raised the question of how does the 1156 bulb affect block occupancy using the current detection method . your answer was you hadn't gotten that far (in your detection / signalling project ) , but that the 1156 might cause a problem . any further info on that yet ?

btw the dvd is great , i found it very helpful in explaining DCC , also it's a good intro to your other videos , which i'll be ordering when my paychecks catch up to my paypal account !

[jfugate]

QUOTE: Originally posted by jwr_1986

I saw a highresistance load short on a bulb protected block and melt an engine once. While the breakers may be considerably more expensive they are the only safe way.
Jesse

Jesse:

Let me add that in 5 years of operating the Siskiyou Line about once a month with the bulbs for short protection, I've never seen any damage done to locos, rolling stock, or track from a short.

For every short protection device we might use, you will be able to find a case where it failed. There is a saying in legal circles that "deciding a case based on the exception is bad law."

You need to decide based on what's typical. And based on what's typical from several people's experiences, the light bulbs do the job for far less money than more expensive devices.