Layout Short - Update

[sschnabl]

Hello All

I went down to the basement on Saturday to continue installing my roundhouse tracks in my Elroy yard.  After connecting the two feeder wires, I turned on track power to test, and I got the wonderful Digitrax beeping sound of a short.  Awesome  .  I spent the better part of the afternoon trying to trace the short down, but so far have been unsuccessful.  I have it isolated to my Elroy power district, but unfortunately I have almost 30 turnouts in this area.  I was able to last run trains earlier last week with no issues.  Here is what I have done so far:

1.  Disconnected the last two feeder wires installed
2.  Disconnected all of the right rail feeder wires of the roundhouse
3.  Removed all tools that might be on the track
4.  Removed all locos and cars on the track
5.  Vacuumed up all the track in this area
6.  Started disconnecting feeders from the bus (But I have a feeder attached to every rail, unless it is soldered to another rail.  So I might have 75 - 100 feeders dropped)

Any suggestions as to an efficient way to troubleshoot?  Since I was running trains earlier in the week, I don't think I have any wires crossed.  I'm thinking something is bridging a gap somewhere, but I just don't know where to start.  I was thinking of disconnecting my Accu-Lite SNAPS from all of my turnouts to see if that clears it, then plugging them back in one-by-one to see if I have a turnout frog that is the culprit.  I would like to avoid disconnecting all of my feeders if possible.

Thanks in advance for any suggestions.

Scott

[GaryHinshaw]

You mentioned Accu-Lite SNAPS; are your turnouts thrown with Tortoises by any chance?  If so, I suspect that one of them threw a switch contact arm and it lodged in a shorted position.  This happened to me recently at a most inopportune time and place:

https://www.therailwire.net/forum/index.php?topic=24108.msg826112#msg826112

If one of yours suffered the same fate, you can try throwing all the machines in that district a few times to see if you can dislodge the offending arm.  That will clear the short, but it will also mean that your frog is no longer juiced.  But try this one machine at a time so you know which one to repair/replace -- if it is indeed the problem.

Good luck!  Model Railroading is Fun!

[rodsup9000]

 I've had rail expand so much that one rail pushed against a frog on a Atlas #10 and #7 causing a short (if the points wasn't routed for that side)

[garethashenden]

As it was working, I think disconnecting wires will only cause more problems. Similarly, a bad frog should only show up when a piece of equipment enters onto it.
It’s going to be slow going, but follow one rail at a time and check the gaps. I don’t know how you have made then, but I can put a thin blade through each of mine. If you’re using plastic insulating joiners it may be harder. Somewhere, a gap will probably have closed up. That’s the most likely cause. Start with the front most rail and slowly move backwards. That’s the best suggestion I can offer without seeing what you’re working with.

[sschnabl]

All of my turnouts are hand laid using Fast Tracks jigs.  The gaps are small between the rails and the frog, but I have not as of yet put any styrene in those gaps.  The rails on each side of the gap are soldered to a PCB tie, so I don't think they are moving.  I am using Tortoise machines to throw them.

Thanks for the feedback so far.  I will update when I get a chance.

Scott

[John]

Start your trouble shooting by dividing the problem in half .. you said that the problem started after installing new track.  Isolate the track that was there before, and take it out of the equation .. good luck

[sschnabl]

Well, after looking at all of my turnouts and not finding any issues with the gaps, I decided to remove the Accu-Lite SNAPS from the Tortoise machines to see if that would help.  Once I had disconnected all 30 and turned the power back on - VOILA, no short!  So I plugged them back in one at a time, and found the offending turnout.  I pulled up the turnout to better examine what was going on, and I found out that a shortcut I took when wiring it up was most likely the culprit.

Because N scale rail is small, even 18ga. wire can look out of place when soldered to the side of the rail.  For that reason, I opted to solder my feeders to the bottom of the rail to hide the wires.  As I mentioned earlier, I've been using Fast Tracks jigs and PCB ties to construct my turnouts.  I've also used the QuickSticks ties to speed up the assembly time.  Anyone who has used these ties knows that the wood ties are held together by the same material that runs underneath the rail.  There are gaps left in the ties for where the PCB ties go, so it's an easy drop in with the soldered turnout.

So my shortcut on some of these turnouts was to solder my feeders to the bottom of the rail right between a QuickSticks tie and a PCB tie where the rail was exposed.  Much faster than cutting out a chunk of tie material between two wooden ties, right?  Well, I think what happened was that some solder had creeped up the side of the PCB tie and ended up on the bottom of that PCB tie, where it is of course ungapped.  So I basically had a red wire and a black wire both touching the bottom of the same PBC tie and thus causing a short.  I ended up filing a gap on these ties, just like you are supposed to do on the top side.  And the problem went away.

So my advice to anyone reading this would be:  If you are soldering feeders to the bottom of the rail, do NOT put those feeders next to a PCB tie.  Find a spot between two wooden ties to avoid this issue.  I'm just glad I was able to find this before the track was ballasted.  Before moving on, I will inspect all of my turnouts to see if I took this shortcut anywhere else and rectify those.

Thanks to all who provided suggestions, it is appreciated.

Scott

[peteski]

Good troubleshooting!
I'm curious why 18 AWG wire feeders? In N scale this seems like a serious overkill, and such thick wire is not easy to work with.  Even if the feeder is few couple feet long, a 24 or 26 AWG wire would  still have low enough resistance to not cause any appreciable voltage drop, at even few Amperes draw (which is unlikely in N scale).

[robert3985]

I hand-lay all of my turnouts, and I learned early on that I should also gap the bottoms of the PCB ties.  This will prevent a lot of hard-to-find phantom shorts.

Also, when wiring always check for shorts before beginning, then check for a short after each solder joint.  This can save a LOT of time searching for that short, especially if you've completely wired a section without checking.

My father, who was an electrical engineer, taught me this when I was about ten years old and he was teaching me how to solder while I was putting together a Heathkit transistor radio.  Unfortunately, decades later when I ripped out my rat's nest of old DC wiring and then re-wired DCC wiring on my layout, I forgot this rule.  I was very careful  and I had already done one section without any problems whatsoever, so, I proceeded on to the next layout section without even thinking about checking the wiring, but after I was done attaching my track feeders to my sub-busses and turned on the power, I had a dead short.  After two solid days of attempting to find it, I finally found it...a broken point rail across both rails, which because my layout is sectional/modular, I had broken it down and I was working on my modules with their backside clamped to a 6' folding table...so, the broken point rails were towards the wall...invisible with the module on it's edge.  If I had checked for shorts before I began modifying my wiring, I would have caught it immediately...but, I didn't do that.

Have FUN!

Cheerio!
Bob Gilmore

[sschnabl]

@peteski  - it looks like I misspoke.  I am using 22 gauge wire for the feeders.  While much smaller than 18 gauge, I still think they look big against c55 rail.  But I think I will take Bob's advice and gap the bottoms of the PCB ties as well.  Shorts on the bottom of the track, especially after ballasting, would be a real pain to find and then fix.

Scott

[Sumner]

I hand-lay all of my turnouts, and I learned early on that I should also gap the bottoms of the PCB ties.  This will prevent a lot of hard-to-find phantom shorts.....................I finally found it...a broken point rail across both rails,...

Hi Bob still having a hard time figuring out how the broken rail caused the short on the bottom of the ties?

I agree about checking more often as you go along.  I got sloppy about that this week and wired about 40 drops and drops from the frogs and when I went to check yesterday I was reading about 40 ohms between the track rails.  This can happen when I have my test DC loco sitting on the tracks but it wasn't.  I use a Tech II DC throttle with alligator clips to the track to test turnouts and such as I go along with the loco.

I cleaned and vacuumed the track thinking there was something very fine bridging the rails....no luck.  Finally I disconnected the leads fro the Tech II.  That was the problem.  Even though it wasn't on it creates a little resistance.  Still I'll go back to checking with the ohm meter any time I add a drop again (with the Tech II not connected to the tracks.

I like to power the short rails just past the frog isolation cuts in case they don't get good current through the rail joiner that isn't soldered connecting them to the next piece of track.  One way I do it is drill a small hole in the ties at the outer stock rails on both sides of the turn out.  Then with one tie drill a small hole by one or the other stock rails past the frog at the inside of the rail.  I can run a short piece of ESU 36 awg up the hole.....Flatten it and solder it to the tie next to the rail.   Then cut it and flatten it and solder to the bottom of the PCB tie (at this point all the insulation has been removed.  Do the same for the other rail past the frog.  Current goes from the stock rail down to the bottom of the tie......across the tie bottom and up to one of the center stock rails.  Takes maybe 5 minutes to do so not a lot of work.

Sumner

[peteski]

@peteski  - it looks like I misspoke.  I am using 22 gauge wire for the feeders.  While much smaller than 18 gauge, I still think they look big against c55 rail.  But I think I will take Bob's advice and gap the bottoms of the PCB ties as well.  Shorts on the bottom of the track, especially after ballasting, would be a real pain to find and then fix.

Scott

Ah, that makes more sense to me. But for short length feeders in N scale IMO using even smaller wire would still be acceptable (and smaller wire is easier to route or solder).

Getting the technical info  from https://www.powerstream.com/Wire_Size.htm here are some examples.  Assuming the feeder length is 3' (6' round trip), here are the voltage drops in the feed itself:

6' of 22 AWG  wire - total resistance 0.097 ohm - voltage drop @ 0.5A = 0.048V or @2.0A = 0.19V
6' of 24 AWG  wire - total resistance 0.154 ohm - voltage drop @ 0.5A = 0.077V or @2.0A = 0.31V
6' of 26 AWG  wire - total resistance 0.245 ohm - voltage drop @ 0.5A = 0.122V or @2.0A = 0.49V

In N scale, an average sound decoder-equipped locomotive with a conventional permanent-magnet motor will draw less than 0.5A of current (even less with coreless motor). I also gave voltage drop @ 2 Amps for multi-unit lashups.  To me even 24 AWG wire would not result in excessive voltage drop. Of course with shorter feeds, the voltage drop will even be less, and more for drops more than 3' long.

I  think of myself as a king of overbuilding things, and I would still consider using 24 AWG wire for short drops in N scale as having very comfortable current carrying headroom.

[MK]

I was even more brazen when I built my little home DCC spaghetti bowl layout.  I used 26 gauge!     But all my drops were under 6 inches with a handful around 8 inches.

[peteski]

I was even more brazen when I built my little home DCC spaghetti bowl layout.  I used 26 gauge!     But all my drops were under 6 inches with a handful around 8 inches.

No, you were not brazen - you were smart MK.  You made the wiring easy on yourself while still not introducing much of voltage drop.  26AWG 8" drop (16" round trip) would @0.004 Ohm/ft only drop 0.064V.  That is extremely low, even for someone who likes to overbuild things.

[garethashenden]

I don’t mean this as criticism of you, but it is criticism of Fast Tracks. Why are their pcb ties double sided? There’s no benefit and lots of potential downsides. It seems like a poor decision on their part and I’m sure this isn’t the first time something like this has happened. I wonder how many modelers have given up when they can’t find a short?