Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

How to Debug Short Circuit in DCC

16747 views
18 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2001
  • From: US
  • 406 posts
How to Debug Short Circuit in DCC
Posted by donhalshanks on Thursday, September 24, 2009 11:55 AM

As I added track to my layout, I always tested and any problems were easily traced and corrected to what I had added.  It has ran flawlessly.  Well now I have about 20' of layout and yesterday when brought power up the Digitrax showed a short circuit and my turnout console lights were blinking.  It had been a couple days and I had done some scenery etc.  I went down all the track and checked for stray spikes, things across the rails etc, and all looked ok.  I removed all engines and running stock which didn't fix it.

How do I go about finding where the short is happening?  I have a multimeter and one of those DCC measuing meters, but all areas of the track show nothing on the meters because of the short.   I guess I'm asking what disciplined process do I go through to get to the problem area?

Thanks for the help, Hal 

 

 

 

 

  • Member since
    January 2001
  • From: US
  • 406 posts
Posted by donhalshanks on Thursday, September 24, 2009 12:55 PM

David:  I have the Zepher Digitrax DCC.  I do have a diagram of my track plan.  I hand laid and built all of my turnouts (using Tony Koesters"s(sp ?) method in a MR handbook for laying track etc.)  I wired off of the tortoise LED lights on my console which show which way the turnout is thrown, as well as the switch lever for throwing the turnout.  In some places I have Caboose hand throws to which I wired LED lights on the console.

 Hal

  • Member since
    February 2007
  • From: Christiana, TN
  • 2,134 posts
Posted by CSX Robert on Thursday, September 24, 2009 1:03 PM
Do you have your layout divided into sections? If not this is one reason for doing that. If it is divided into sections, you can unhook sections until you find the one with the short to narrow down the location.
  • Member since
    October 2006
  • From: Western, MA
  • 8,571 posts
Posted by richg1998 on Thursday, September 24, 2009 1:25 PM

 Put your multimeter in the ohms, resistance mode. You did not specify if you used the ohms mode for your testing. Disconnect the DCC controller. Connect each probe to a rail. You should see about zero ohms or very close to that value with a short. You will have to start disconnecting feeders to the added section. Most probably a turn out short.

Rich

If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.

  • Member since
    September 2007
  • From: Charlotte, NC
  • 6,099 posts
Posted by Phoebe Vet on Thursday, September 24, 2009 2:25 PM

Step one is always to make sure that you really have a short.  Disconnect the command station from the track and turn the track power on.  If it still indicates a short then the problem is in the command station, if not then it is in the track someplace.  Check all the track including sidings, for tools and debris and reconnect the command station.  Throw and close all your switches.  A bad switch can cause a short when it's position is changed.  Throwing a switch, under some circumstances, can also engage a reversed section you didn't realize you had created.  That's why others have asked you to post your track diagram.

If you have done nothing to the track since it was completed and working, then you have shorted the track working on the scenery.  If you have changed something in the track, start undoing the changes, starting with the most recent.

If you have added or modified any rolling stock, remove it from the track.  If you work backward from your last change, you will eventually find what you did.

Dave

Lackawanna Route of the Phoebe Snow

  • Member since
    January 2001
  • From: US
  • 406 posts
Posted by donhalshanks on Thursday, September 24, 2009 3:18 PM

Rich:  I did not do an ohms mode test as you described and thanks for suggesting it.  Didn't realize there was this kind of a test.

CSX Robert:  I did not divide what I have so far into sections.... thinking when I do the next expansion it would be the 2nd section.  Maybe a bad choice.

Phoebe:  After doing the ohms test, then I'll follow your suggestion to back track.

My track plan may not be in the form I can easily put up to show.  It is pretty much the end of a short line terminal yard with several spurs and leads into a turntable and back out.  There are 5 turnouts for the sidings ands spurs etc.  I'm thinking the short is in one of the turnouts.

Thanks for the help.  Hal

 

 

  • Member since
    October 2005
  • 1,033 posts
Posted by betamax on Thursday, September 24, 2009 3:39 PM
One simple thing to do is a quick visual check of the wiring. Make sure that somewhere in there you didn't reverse the connections and cause a short. It is easier to do that than many people realize.
  • Member since
    February 2007
  • From: Christiana, TN
  • 2,134 posts
Posted by CSX Robert on Thursday, September 24, 2009 3:44 PM
Depending on how you did your turnouts(I have not seen the book you mentioned, so I am not familiar with his method), the turnouts may create a short on the leg that they are not aligned for if you do not have insulating gaps on them. Depending on how the layout is wired, these shorts may not always show up either. For example, if you have a passing siding where neither the siding nor the mainline have feeders between the turnouts, you will not see a short if both turnouts are aligned for the mainline or if both are aligned for the siding, but if one is aligned for the mainline and the other for the siding, you will see a short.
  • Member since
    September 2007
  • From: Charlotte, NC
  • 6,099 posts
Posted by Phoebe Vet on Thursday, September 24, 2009 4:01 PM

If I may add one line...

That is why you should use a consistent color code in all your wiring throughout your layout.

Dave

Lackawanna Route of the Phoebe Snow

  • Member since
    December 2004
  • From: Bedford, MA, USA
  • 21,351 posts
Posted by MisterBeasley on Thursday, September 24, 2009 4:54 PM

Contrary to popular belief, things like this seldom break by themselves.  My guess is, you were under the layout for whatever reason, and you unknowingly caught a loose wire and pulled it out, so that it fell back and made contact with something of the opposite polarity.  Check under your layout.  Think about where you might have caught something.  Don't forget your control panels.

Do you have a reversing section?  If so, try disconnecting that and see if you can bring power up.  That's going to be an isolated piece of track, even if you don't think of it as a "block."  A programming track is also an isolated section you can disconnect.

Did you install any bumpers on the ends of sidings?  Some of these are all-metal, and if you connect both rails they will be a short.  You need to gap one or both of the rails with these.

Have you removed all the engines and rolling stock?  Very often, shorts are caused by wheels.  If it's a car or engine causing the problem, it's easy to figure out which one by removal.

One forum member had a problem like this once, and he was kind enough to share his story.  It's one I've remembered, because it's such a simple thing to do, and so easy to overlook.  He had been adjusting couplers, and he left his Kadee coupler gauge on the tracks.  It's all metal, and a dead short.  But, it looks like it belongs there, so as he scanned for misplaced wires and tools, his eyes passed right over it.

 

It takes an iron man to play with a toy iron horse. 

  • Member since
    August 2006
  • 1,517 posts
Posted by trainnut1250 on Thursday, September 24, 2009 5:23 PM

I'll add one more.  Check your gaps at turnouts etc  Make sure that none have closed up due to expansion or other weather/scenery related causes.  I have been bitten by this one on occasion.

 

Guy

see stuff at: the Willoughby Line Site

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, September 24, 2009 5:25 PM

donhalshanks

As I added track to my layout, I always tested and any problems were easily traced and corrected to what I had added.  It has ran flawlessly.  Well now I have about 20' of layout and yesterday when brought power up the Digitrax showed a short circuit and my turnout console lights were blinking.  It had been a couple days and I had done some scenery etc.  I went down all the track and checked for stray spikes, things across the rails etc, and all looked ok.  I removed all engines and running stock which didn't fix it.

 There might be your clue. Perhaps you got something conductive across one of your gaps, or if you used PC board ties, across the gap sawn in one of them. If not that, check into some of that 'etc', since it WAS working fine before you did the scenery and etc, it has to be something there that now causes a short.

                           --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    October 2006
  • From: Western, MA
  • 8,571 posts
Posted by richg1998 on Thursday, September 24, 2009 5:25 PM

donhalshanks

Rich:  I did not do an ohms mode test as you described and thanks for suggesting it.  Didn't realize there was this kind of a test.

 

When you get time, click on the search engine below and search for:

troubleshooting model railroad with ohm meter

You will get a lot of hits and instructions on using different options in a multimeter. Some sites will be easier to read than others. Some sites assume technical experience.

I find answers to a lot of different subjects on the 'Net. There are enough sites so if a site has some errors, you can cross reference with other sites. Wikipedia is know to have some errors at times as an example.

If you have a library nearby, you can also find material.

Many old timers did not have meters but, used t a buzzer and a battery connected to the track as they worked. A buzz told them they had a short. on the track. Some modelers make the mistake of wiring a without a meter to troubleshoot and have have issues.

Good luck.

Rich

If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.

  • Member since
    November 2002
  • From: Colorado
  • 4,074 posts
Posted by fwright on Friday, September 25, 2009 12:46 PM

MisterBeasley

Contrary to popular belief, things like this seldom break by themselves.  My guess is, you were under the layout for whatever reason, and you unknowingly caught a loose wire and pulled it out, so that it fell back and made contact with something of the opposite polarity.  Check under your layout.  Think about where you might have caught something.  Don't forget your control panels.

Do you have a reversing section?  If so, try disconnecting that and see if you can bring power up.  That's going to be an isolated piece of track, even if you don't think of it as a "block."  A programming track is also an isolated section you can disconnect.

Did you install any bumpers on the ends of sidings?  Some of these are all-metal, and if you connect both rails they will be a short.  You need to gap one or both of the rails with these.

Have you removed all the engines and rolling stock?  Very often, shorts are caused by wheels.  If it's a car or engine causing the problem, it's easy to figure out which one by removal.

One forum member had a problem like this once, and he was kind enough to share his story.  It's one I've remembered, because it's such a simple thing to do, and so easy to overlook.  He had been adjusting couplers, and he left his Kadee coupler gauge on the tracks.  It's all metal, and a dead short.  But, it looks like it belongs there, so as he scanned for misplaced wires and tools, his eyes passed right over it.

That would be me with the Kadee couper height gauge - one of my stupid model railroading tricks.  The sad thing is it took me (supposedly an electrical engineer) about 3 hours to find it, and only after I had cut several feeders in an attempt to find the short.  As Mr Beasley states, sudden and presistent shorts have a cause; they just don't happen.

Unfortunately, DCC lends itself to a monolithic wiring structure with a lack of easy isolation points.  Still, as tedious as it may be, you need to isolate a step at a time until you find the cause - even if it means cutting soldered feeders and/or cutting gaps in the track.  My suggested order would be:

  • remove rolling stock and locomotives 1 piece at a time, starting in the area where you last added wiring.  You can't do resistance measurements (ohms) with locomotives or lighted cars on the track anyway, so this might as well be the first check.  Leave your DCC system connected so that you can tell if/when the short disappears when a piece of rolling stock is removed.  If all rolling stock is removed and you still have a short, you know it's not being caused by derailed or mis-oriented wheels or an internal short in a locomotive.
  • Disconnect your DCC system from the track.  As another poster said, power up the DCC independently of the track.  If the short is gone, the short is in the track and wiring.  If you still have a short in the DCC system, the system has an internal problem.
  • The next step is a coin toss, depending on what you think is more likely.  You can start chasing the wiring, or you can start chasing the track.  But you can't do both together and be certain of your findings.
  • To chase a short in the track:
    • first do a check for anything bridging rails anywhere (Kadee coupler height gauge keeps popping into my mind for some reason)
    • cut gaps around a turnout at a time - all 6 rails - and any feeders attached to the turnout.  Check for a short (0 ohms) on track on the outside of the isolated turnout.  If the short is gone outside the turnout, the problem is/was the turnout and/or its feeders.  With the turnout feeders cut, if you still have short at the turnout, then it's in the turnout itself.  Throw the turnout both ways to make sure you know what is happening.
    • If cutting the turnout feeders got rid of the short in the turnout, you now know where the problem is.  The downside of this precision isolation is that you have to redo the turnout feeders.  For this reason, I strongly recommend starting with turnouts/track that were added since things were working properly.
  • To chase a short in the wiring:
    • start disconnecting feeder pairs, a set at a time from the power bus.  Again, start with feeders that were added since things were working properly.  Also again, you get to redo anything you undid.  Check to see if you still have a short when the feeders are 1st disconnected.

 To prevent this from happening again:

  • test after installation of each feeder pair
  • isolating turnouts with gaps in all six rails is a pretty sound practice, as is keeping all the feeders and wires for the turnout tied to a single pair of wires to your power bus.  The gaps don't have to be at the turnout itself, but need to be before another turnout.  And all turnout feeders should be tie to a single pair of wires to the power bus.
  • test after installation of each turnout

My bottom line is that there's enough pain in wiring under the layout that I only want to do it once.  The thought of having to redo a batch of connections/wires because I made an undetected error a while back drives me to test at every opportunity as I laid out above.  I also like power distribution points (screw terminals are my favorite) for complex wiring areas like turnouts where I can easily disconnect/reconnect as compared to soldering directly to the power bus.

What I gave you above is a generic trouble-shooting process.  In your case, you say the wiring wasn't touched during the scenery application, and all rolling stock was removed.  Chances are it's something across the tracks - a great big uninsulated horizontal power bus set waiting for something to short it out.  It could be something quite small (rail spike size) at a turnout where rails of opposite polarity are in close proximity to each other.  In the old days (and I've done this, too), it was not unheard of for the staples used to fasten screen wire for the scenery to cause short circuits.  Stapling wiring has also caused me grief in the past. 

my thoughts, your choices

Fred W

  • Member since
    March 2007
  • 2,751 posts
Posted by Allegheny2-6-6-6 on Friday, September 25, 2009 10:34 PM

 Take it for what it's worth but having years of automotive electrical trouble shooting behind me I try and apply the same principals to my model railroading.

A: Check the simplest things first, before you go disconnecting wires etc. do a through visual check of ALL your track work thoroughly and slowly check every inch and then go back and check it again.

B: Think back to when the layout was functioning properly and what has changed since then, this is usually where the trouble is, if you installed a turnout or did something to one thoroughly check it, if you added scenery which i believe I read you did go back and check it.

C: I would then start doing my electrical continuity testing first. I personally would disconnect the DCC unit and connect my ohm meter between the tow rail and work your way around the layout. I have a tone option on my multi meter which makes things like this easier. Work your way around the layout and if you come to a spot that has tone or full deflection of your ohm meter thats where your short is. Another method is to hook one ned of the ohm meter to one of the leads that wold be connected to your dcc system lets say the red one and the other lead to the corresponding rail. Work your way around the layout to find the short.

Although I believe what Mister Beasley said in his post it's more then likely you caught some wire or brushed up against something to cause a short.

We had a short on the club layout that seven guys spent four hours trying to find and myself and another guy came in and found it in about 10 minutes. They used real dirt and sand for ground cover and no one bothered to check it with a magnet. well I have ever had it happen but read where as when ever you do such a thing to first run a magnet through it as a lot of dirt has high iron content or particles in it that will cause a short circuit. Guess what they were right.

Just my 2 cents worth, I spent the rest on trains. If you choked a Smurf what color would he turn?
  • Member since
    July 2006
  • 3,312 posts
Posted by locoi1sa on Saturday, September 26, 2009 6:58 AM

 The first rule of trouble shooting. KISS it. Keep It Simple Stupid. 35 years of heavy truck and mining equipment trouble shooting has taught me that. If it a definite short remove power right away. Take a real good look at recent wiring first. Check connectors for solid connections and make sure the wires are in the right slot in the connector. Use your multimeter in the beeping ohms mode across the connections for you DCC system and disconnect sections or newly connected feeders one at a time until the beeping stops.

  One last thing. Is your track wet from the new scenery? Probably dry by now and no short?

      Pete

 I pray every day I break even, Cause I can really use the money!

 I started with nothing and still have most of it left!

  • Member since
    January 2001
  • From: US
  • 406 posts
Posted by donhalshanks on Saturday, September 26, 2009 1:45 PM

Thanks Fred, Pete, Beasley and all you other guys who have kindly responded with methods to use.  I will have time this weekend to work through it, and will let you know how it comes out.  Pete.... you may have hit on it.  I had ballasted and graveled part of the yard leading into the engine house around three lead in tracks.  I had just a little time, and my ohm meter showed zero reading on two of those spurs.  I'm guessing that maybe the track was still wet as Pete suggests because it was not long after I finshed when I got the shorts.  That will be my first look when I get back to it! 

 You have all given me the right methods to go about it.  I'm glad its just not me to realize it is a little like hunting for a needle in a hay stack.

Hal 

  • Member since
    June 2009
  • 36 posts
Posted by Hooty on Wednesday, September 30, 2009 7:47 AM

The idea of wet ground foam is good. I once had the problem of plaster across a track at the base of a mountain when dried

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!