A curious short

[RobinofLoxley]

I am currently building a new layout that is wired for DCC but I'm using a DC controller for testing, as I have only one converted loco and I want to test the build quality using a variety of locos. Ive built a fair amount and now Im working back wiring point motors and such like. I hadnt tested the track systematically but had just put a loco on in a few places to check power was flowing OK. So today i started more systematic testing and I found the test loco wouldnt run at all and all the indications were for a short. Everything is on droppers and colour coded, no obvious errors so I started disconnecting feeds board by board. I got back to the first board and the loco started to work on all tracks. I introduced the feeds on the second board one by one and found a single straight piece of track, isolated from everything else by IRJ's, that caused the short to appear. I say short, the Guagemaster D100 power indicator remains red when the power dial is turned up, either direction, which I take to be a short. All I could find on testing was a dry joint on one side of the isolated track. I cant understand why that could give the appearance of a short, can anyone explain? Have I overlooked something? I should add that I had run the test loco over this track section earlier with no problems.

 

BTW I have yet to repair the joint, a job for tomorrow.

Edited May 17, 2021 by RobinofLoxley

[kevinlms]

2 minutes ago, RobinofLoxley said:

I am currently building a new layout that is wired for DCC but I'm using a DC controller for testing, as I have only one converted loco and I want to test the build quality using a variety of locos. Ive built a fair amount and now Im working back wiring point motors and such like. I hadnt tested the track systematically but had just put a loco on in a few places to check power was flowing OK. So today i started more systematic testing and I found the test loco wouldnt run at all and all the indications were for a short. Everything is on droppers and colour coded, no obvious errors so I started disconnecting feeds board by board. I got back to the first board and the loco started to work on all tracks. I introduced the feeds on the second board one by one and found a single straight piece of track, isolated from everything else by IRJ's, that caused the short to appear. I say short, the Guagemaster D100 power indicator remains red when the power dial is turned up, either direction, which I take to be a short. All I could find on testing was a dry joint on one side of the isolated track. I cant understand why that could give the appearance of a short, can anyone explain? Have I overlooked something?

 

BTW I have yet to repair the joint, a job for tomorrow.

The easiest way to find shorts, is to not create them in the first place!

 

Try connecting a multimeter with a continuity tester (or buzzer) across the tracks and leaving it on. If by attempting to connect a wire the beeping starts, there is your short. Much easier than soldering up everything, then finding that there is a short!

[RobinofLoxley]

All my droppers are connected to bus power through quick release connectors. Trouble shooting isnt difficult. In any case the section would have passed your test first time, the problem appeared later.

[AndyID]

From your description there must be a short on that section. Disconnect both feeds and measure resistance between the rails and across all the IRJs. They should all be open-circuit (infinite resistance).

 

[DCB]

You need section switches so you can isolate chunks of layout rapidly to narrow down the fault location.   Shorts don't just appear when a layout is new, they can appear after 30 years or more if you displease your personal gremlins. Sooner or later you are going to get a fault and have no idea where to start. Metal wheels across irjs is a favorite of theirs and the wheels don't short initially but wait for a suitable moment such as when your mates come round for a beer and a running session.

[kevinlms]

8 hours ago, RobinofLoxley said:

All my droppers are connected to bus power through quick release connectors. Trouble shooting isnt difficult. In any case the section would have passed your test first time, the problem appeared later.

Your original posted mentioned 'a dry joint' which indicates soldering. Many people solder the whole job up and then wonder why there is a short, which they have accidentally caused. Which can be a pain to find. Already you said you 'started disconnecting feeds, board by board'.

So obviously you have a problem somewhere.

 

I can't see how quick release connectors, can suddenly develop a short. More likely is a poorly terminated screw in block, with loose strands or similar.

 

[Harlequin]

You may have accidentally introduced a new problem while disconnecting and reconnecting feeds to find the original one.

I'm sure that a bit of methodical detective work will sort it out.

 

[Pete the Elaner]

10 hours ago, kevinlms said:

 

Try connecting a multimeter with a continuity tester (or buzzer) across the tracks and leaving it on. If by attempting to connect a wire the beeping starts, there is your short. Much easier than soldering up everything, then finding that there is a short!

 

That is surely a much better method than trying to activate some sort of overload protection?

Get the layout back to a fault-free condition & add connections 1 by 1. The advantage over testing with a cutout is that the buzzer will sound immediately when problem occurs rather than having to manually run a test after each connection, which not only takes more time but can too easily become testing after 2,3,4 or more connections, then you don't know which one has caused the fault.

[RobinofLoxley]

I am using the testing process to check how trains run over crossings, points etc, as well as electrical continuity. I am doing that for every section (block) so max 4 droppers are connected at once. But my own process, once I had got over the business of this ran OK last time I tried it, whats going on, I located the fault in about 2 minutes. What i cant see is why the dry/poor joint caused problems elsewhere, so i guess I have a second problem to find, which I will, as I re-connect section by section.

[Platy787]

13 hours ago, RobinofLoxley said:

 I introduced the feeds on the second board one by one and found a single straight piece of track, isolated from everything else by IRJ's, that caused the short to appear. I say short, the Guagemaster D100 power indicator remains red when the power dial is turned up, either direction, which I take to be a short.

I have a Gaugemaster Combi (which may operate differently to your D100) which has a red led as Track indicator. It increases in brightness from off to bright as the control is moved from 0 to 100. It comes on at about 10 on the dial when it is not connected to anything. When I short circuit the output of the controller the red led still comes on, but at 20 rather than 10.

 

In a Combi the Track indicator is not directly across the output terminals, so it will still glow, at a higher dial setting than normal, even when there is a short circuit on the output, but the Track indicator is not an indicator of a short circuit.

 

If you put a 12V 18W or 12V 10W automotive globe in series with one of the ac supply lines, or one of the track feeds, you will get a very obvious indication (and current limit) of any short circuit.

 

Mark

[RobinofLoxley]

2 hours ago, Platy787 said:

I have a Gaugemaster Combi (which may operate differently to your D100) which has a red led as Track indicator. It increases in brightness from off to bright as the control is moved from 0 to 100. It comes on at about 10 on the dial when it is not connected to anything. When I short circuit the output of the controller the red led still comes on, but at 20 rather than 10.

 

In a Combi the Track indicator is not directly across the output terminals, so it will still glow, at a higher dial setting than normal, even when there is a short circuit on the output, but the Track indicator is not an indicator of a short circuit.

 

If you put a 12V 18W or 12V 10W automotive globe in series with one of the ac supply lines, or one of the track feeds, you will get a very obvious indication (and current limit) of any short circuit.

 

Mark

The same I think, I was trying to avoid writing war and Peace on the original thread. In normal operations the led goes from red to orange to yellow more or less; in this case with the questionable piece of track in place it goes to a rather dull orange. makes no difference if a loco is on that section of track or not.

Edited May 18, 2021 by RobinofLoxley

[AndyID]

4 hours ago, RobinofLoxley said:

The same I think, I was trying to avoid writing war and Peace on the original thread. In normal operations the led goes from red to orange to yellow more or less; in this case with the questionable piece of track in place it goes to a rather dull orange. makes no difference if a loco is on that section of track or not.

 

Would I be correct in thinking you don't actually have a multimeter? IMHO it's a false economy not to have one. They are dirt cheap these days and you'll find the problem in ten minutes rather than messing around with all this overload stuff.

 

 

 

Edited May 18, 2021 by AndyID
Predictive nonsense

[RobinofLoxley]

No I have one but I dont have any confidence in it. It was a random ebay purchase, and appears to read values at random. I need to junk it and buy another but I just havnt got round to it. If I had a friend locally who was very good at electrics I could probably get the meter sorted out as it could be a user problem...

 

The whole thing is proving to be useful learning for me anyway in respect of DCC wiring (individual droppers) and what can happen with it. I havnt found the source of the problem exactly but I have got the power back in areas where I had lost it, which happened because when I said I had disconnected everything inter-board I had overlooked possible power connections through the track! So next time I am working on it I will find the fault for sure. Actually with so many droppers hanging down under the boards, a short would be easy to make without trying. 

[RobinofLoxley]

The coda to this is, I isolated the fault to a R8078 long point. No faults on inspection, but the multimeter revealed a short across the stock rails. The frog connections on this kind of point are buried inside the sleeper mouldings so there is nothing to see. The point operated normally for 3 years on the previous layout, the only thing I have done is solder droppers onto it as best practice for DCC for the new layout. So I theorise there must have been some thermal expansion during soldering, enough to cause the problem. I said it was curious...

Edited May 19, 2021 by RobinofLoxley

[kevinlms]

17 hours ago, RobinofLoxley said:

The coda to this is, I isolated the fault to a R8078 long point. No faults on inspection, but the multimeter revealed a short across the stock rails. The frog connections on this kind of point are buried inside the sleeper mouldings so there is nothing to see. The point operated normally for 3 years on the previous layout, the only thing I have done is solder droppers onto it as best practice for DCC for the new layout. So I theorise there must have been some thermal expansion during soldering, enough to cause the problem. I said it was curious...

A short across the 2 stock rails? That sounds odd.

It seems like you have made a mistake wiring the frog. Do you have a switch, because it sounds like the switch and stock rails are in conflict.

[RobinofLoxley]

2 hours ago, kevinlms said:

A short across the 2 stock rails? That sounds odd.

It seems like you have made a mistake wiring the frog. Do you have a switch, because it sounds like the switch and stock rails are in conflict.

Absolutely not. The Hornby points are wired so that the frog rails are connected to the switch rails via a metal strip that passes through the moulded area where the apex of the frog is. They must be separated by a thin strip of plastic inside the moulding and somehow that plastic strip has been broached. I have replaced the point with another one the same and everything is normal now.

[Right Away]

On 17/05/2021 at 21:22, RobinofLoxley said:

........ I introduced the feeds on the second board one by one and found a single straight piece of track, isolated from everything else by IRJ's, that caused the short to appear.......

Not certain if I’ve grasped exactly your track work, but your reference of an IRJ isolated straight section - could the polarity of the feeds to this section be reversed?

 

[RobinofLoxley]

Just now, Right Away said:

Not certain if I’ve grasped exactly your track work, but your reference of an IRJ isolated straight section - could the polarity of the feeds to this section be reversed?

 

What happened was this. My layout has a series of boards with a continuous power bus that has terminals on every board. I take a spur from the terminals, and connect droppers to the spur using a daisy chain of Waygo connectors. Sometimes it was more convenient to connect a dropper to a spur on an adjacent board. So when I was troubleshooting, I disconnected the spur feeds one by one but inadvertantly left a dropper connection in place, which meant that through a combination of track joins and droppers part of the layout that I thought was disconnected was still live. Once I correctly reduced the power distribution to a single baseboard I was able to connect droppers one by one and quickly identified where the fault was. All my droppers are colour coded.