Short on fairly large DCC layout - use multimeter to narrow it down?

[juke]

I am having difficulty locating a short. I have read a lot of help topics and to be honest I was already familiar with the steps to go through. So far I'm not seeing it.

 

I think I read a long time ago that a 'better' mm can be used to narrow down the area of the short.

 

I have a Brymen 869s meter that can read the voltage on DCC track and lots of other things so maybe there is a function in there somewhere that could help if I new anything about using this thing<G>

 

What I seem to recall reading is, with the DCC controller and it's PSU removed, set the meter to the ohms setting and check across the the track. This gives reducing or increasing measurment values as it gets nearer the short.

 

Is this correct and which reading is the one high or low?

 

Syd

 

 

[Dagworth]

the nearer you get to the short the lower the value on the meter.

 

The other trick that may work is that you can sometimes hear where the short is. I don't know if it works on all DCC systems but with my Lenz set 100 a short will make a noise when DCC is fed through it. If you can make the layout location quiet enough then pressing the button to restore track power will result in a very brief buzz from the short itself before the DCC system cuts out again. This only works if the short is not absolute, for instance if it is a wrongly set point you can hear the relatively poor contact at the point blade.

 

Andi

[juke]

the nearer you get to the short the lower the value on the meter.

 

The other trick that may work is that you can sometimes hear where the short is. I don't know if it works on all DCC systems but with my Lenz set 100 a short will make a noise when DCC is fed through it. If you can make the layout location quiet enough then pressing the button to restore track power will result in a very brief buzz from the short itself before the DCC system cuts out again. This only works if the short is not absolute, for instance if it is a wrongly set point you can hear the relatively poor contact at the point blade.

 

Andi

Thanks Andi, the lowest value will now be sought!

 

I added to my problems by damaging my Lenz set 100 when checking through all that side of things, a stupid moment's inattention very quickly repaired by A&H models. I'm loath to attach it back to the track until I'm sure everything is OK.

 

I have around 25 Cobalt IP points, two 30'x10' ovals and some connecting points and 4 track sections so a lot to visually fault find. If there was a slight noise from the short at switch-on I probably wouldn't be near enough to hear it, a good idea though.

 

All was working perfectly before I wired in another crossover. As soon as I added power the short occurred, yet it had nothing to do with that work as it is still there even with the two points removed completely.

 

I do wonder if one of the Cobalts have gone faulty in some way as they are the only variable items still attached to the track. All locos and rolling stock is off the tracks.

[shibushe]

Yes shorts can be difficult to find, and frustrating, I know by experience.

I can only suggest you have close and careful look round your tracks and points.

I say this because it could be a track pin lodged somewhere.

My worst short and the time it took to find it was a small wagon screw that fell off, IN A TUNNEL.

[Ian Abel]

I love my Cobalts, but, the one potential failing (ask me how I know!!) is that if you're ham-fisted, as I apparently managed to be, when wiring them, you may leave a stray strand of wire exposed at one of the connections.

I've about 30 motors and after installing the last two everything ran fine, for about 10 minutes.

After I ducked under a section I got a short. Rather a bu@@er to find especially as I suspected the two recently installed motors! Checkwd/disconnected but they weren't the problem.

Turned out one of the motors at the duck-under was brushed as I went by and THAT had a poor connection inducing a short from a stray strand.

I've a separate accessory bus so at least knew it was there. Just had to divide and conquer various sections of that bus.

 

Good luck...

[10000]

I once had to deal with a short on an n gauge layout. It turned out to be insulating rail joins crushed as the temperature went up.

 

The way I found the offending joints was to put a slip of paper between all the switch blades and stock rails. This cleared the short which was then found by removing the pieces of paper one by one.

[smokebox]

If all else fails, divide the layout into 2 halves. See which half of the layout the fault is in and divide it in half. Continue until you find the fault. This halving process can give very quick results with only a small number of divisions. Remember though, if you have a wiring bus, you need to divide the bus AND the track.

[Phil S]

This is one of the main reasons I advocate the use of sub-districts based on a geographical area such as the sidings of a station all fed from their power- district bus... But via a switch conveniently to hand.

Eg my Beechingless Barnstaple loft layout has 4 power districts..

Barnstaple Junction Station, Scenic left, scenic right, and the lower storage level ( exeter and Taunton Storage areas)

Each of these power districts has local switches to isolate eg Barnstaple junction goods yard or the through platforms....

I make the most of dcc simplicity by parallel wiring the platforms, and similarly the goods sidings.... But each is then a simple observable area in which to look for the fault.

 

An early zero1 example I used to quote on engineering courses... To illustrate why an area is subdivided to aid fault finding:

My original zero1 Beechingless was 1 power district with no sub districts.... It had 2 power busses.... 1 for track and 1 for accessories.

It also had an ac ammeter and an oscilloscope connected too.

A short occurred ... Somewhere ,.... Thankfully only 1 train had been running... But in a tunnel. On plain track. I lifted the track at either end to isolate it, and the short was still there. I felt along the track bed in the tunnel.. Awkward in a loft....and found a single strand of wire in the ballast which had been moved by the passing train.

 

More recently... On the present layout, when a friend was visiting... A derailment occurred, and a short circuit was a natural consequence... I put the train back on, but still there was a short.... I turned the track power off ( it wouldn't stay on anyway, of course) and measured with the ohms setting. ... Not simple continuity beep.... And. Followed the path of least resistance.... To find a metal clad 9v battery laying across the track.... It had been in my hand when the derailment occurred, and I simply put it down... On the track.

 

now I have PSX intelligent breakers on my 4 areas to cope with sound locos etc, but also because they try to reset every 2 seconds while I am some distance away locating the fault. I also have ramp meters on each district ... Normal current 0.3- 1.3 A depending on coach lighting etc. I can see the currents change as a train goes round the sections. In early dcc days I had a continuous current of 3.5 -amps flowing and showing on my ztc511 which had a built in meter.... A faulty point switch was shorting and melted the point internally.

 

My sub-district / power district central switch panel also has LEDs built in on the track side of the switch.... They will light if an adjacent section is sending power even if the switch is off... Due to a metal wheel bridging a gap, for example.

 

Yes, I used to teach maintenance so I design diagnostics in to make fault finding easier.

Buffer stop lights indicate track has power. Street lights or platform lights etc show either Ec or ac or accessory busses etc.

On a current portable layout, the veissmann emotion working child's seesaw confirms that 16v has passed through all boards to reach the last one....

[juke]

Thanks for all the replies, they're really appreciated.

 

I thought I had located the area that was faulty using the meter, but so far (couldn't continue today) I haven't spotted it.

 

I have noticed the 'problem' with stray wires touching the adjacent wire on Cobalts and have checked them all with glasses on and a torch, without success.

 

I've been around the entire length of the track with a bright LED wander light at least three times. Above and below.

 

It will no doubt end up being something simple but I have resolved to split the layout into sections with switches inserted into the bus. I have a problem in that the ballasting is well advanced so inserting insulated joiners will be tricky. I'm wondering if I just cut the track with a fine rotary diamond cutter is each line likely to stay put? I think maybe the ballast will hold it and it would make things easier.

 

I think I'll add an accessory bus, and rewire all of the Cobalts to that as an extra precaution.

 

Phil's longer advice post will be reread tomorrow, rather a lot to take in first time after a frustrating day!

 

Syd

[juke]

I finally found the short!

 

When I built the layout I used two kitchen base units as a starting point for height. It was never my intention to store anything in them but of course boxes started getting piled up in one of them.

 

The wiring has all been done with droppers wrapped round exposed sections of the bus and soldered. I made sure that the two wires did not have any chance of accidentally touching so nothing is insulated.

Of course, at the back of the cupboard the two wires came close togetherwhere they went through the side panel but I couldn't see that one joint every time I rechecked the wiring.. The stack of boxes got high enough that they moved the bus wires together

and the bare wires met. It must have happened after the last operating session as all was well when switched off.

 

I have the layout separated into two sections now, I'm rather tired of rewiring for now because I had removed, trimmed and reinserted a number of the Cobalt wires, more mindful now to ensure the insulation runs right into the connectors.

I'll introduce more sections wnen I get some energy back <G>

The fault wasn't anywhere were I had laid the new points, though of course I was convinced it was around that area that was at fault. The MM homed me in, so that was great.

 

The bad news is that the LZV which had been fitted with refurbished boards has a fault in that it is feeding too high a voltage (20.3) and it fails to change no matter how many times I have carried out the CV7 procedure, I gave up and bought a new unit.

It arrived today and all is now well. Maybe the old unit would be OK for an O gauge layout, as everything else is working fine on it.

 

Syd