Strange Reading

[Ray H]

Before connecting any form of controller to the freshly laid track on my new OO layout/trackwork yesterday I thought I'd check out continuity with an analogue test meter - I should perhaps add that the layout has been wired as an entity, with no section switches (ready for DCC operation). At this time the controller was still in the box in another room in the house.

 

The layout uses Peco live frog points with the crossings thereof switched by microswitches that are operated by the manual point control mechanisms.

 

There are two 4'x2' boards plus a 1'x2' board at the end of the station. The majority of the track is located on one of the larger boards, the other large board holding two stub sidings and the connection to the single 36" long cassette that I've so far built.

 

The electrical connections between boards are via 5 pin DIN plugs/sockets with only the two extreme pins connected in each plug/socket. The interboard connecting leads are about 1ft long using low current three core mains cable - the earth conductor being connected to the middle of the five pins on the DIN plug but otherwise unused.

 

The under board wiring is via lightweight droppers to adhesive copper strip. The boards are braced across the layout only at 1ft spacing. Each 1ft section has its own length of tape connected to the neighbouring section by the same (7/0.2) lightweight wire that is used for the droppers. The end sections on each board are connected to the DIN socket by the same dropper wire.

 

Apart from the unintentionally incorrectly wired crossings on the three way point the only other problem that I found was an apparent very high resistance reading when the meter's probes were connected to the running rails - the meter was on the x10 scale and the needle made a miniscule movement whilst the probes were on the rails, dropping back when the probes were disconnected. There was no such needle movement noticed when the meter was switched to the x1k setting and the probes connected to the track.

 

Can anyone explain what may be causing the needle movement and whether I need to be concerned about it?

 

I should add that when I eventually connected the Dynamis controller to the track and put a straight out of the box loco on the track, the loco would move for a short distance, stop and then repeat the cycle after a few seconds. Cleaning the track and loco wheels seemed to generally correct this problem and the analogue meter which by then had been switched to read volts stayed steady all the time.

[Edwin_m]

If you have too much resistance in circuit to any part of the layout then there is a risk that a DCC command station will not shut down on short circuit and several tens of watts of heat could be generated until someone notices and cuts the power. The amount of resistance to cause a problem depends on your command station and track voltage but it is probably about two ohms. This is the total of the resistance "out" from the command station to the track and "back" again, and can be measured by removing the command station, connecting its two rail connections together and checking the resistance between the rails around the layout. However meter probes can be a bit hit-and-miss especially on dirty rail.

 

I've not used copper tape but I think it would be quite easy to end up with the main bus tape too narrow a poor eletrical connection at one of your joints. You could and should do a coin test with the Dynamis connected, to be sure whether the cutout works, but it would also be a good idea to do some more probing with the meter and see if you can locate any particular problem areas.

[Ray H]

The problem that I found with the meter was discovered before I took the Dynamis out of the box, let alone connect it to the layout. All the meter was seeing at the time was the rails, dropper wire, copper tape, micro switches and the plugs/sockets. That's what is puzzling me. I've never noticed the phenomenon before on any circuit.

 

The short circuit protection on the Dynamisdoes appear to work as it tripped the system several times with good reason after it was connected to the layout. However, I've noted your comment (for which I thank you) and will keep an eye on it. Luckily (?) I don't expect to be moving more than one loco at a time so that should help minimise the risk.

[bertiedog]

I think you should disregard the reading, you do not state the resistance exactly, but a 10x range suggests that you are measuring mega ohms, the resistance across the rails, and any sensitive meter will register the overall resistance of the layout.

 

Any figure from 100,000 ohms upwards would work, and not affect the DCC signal, as it is not in series, but in parallel with the load, and so much relatively higher, it can be safely disregarded.

 

The insulation is lowered by ballast, glue, paint, damp, conductivity of wood and even air, and may range from 100,000 ohms or so to several mega ohms. If it reads under 50,000 ohms it may mean trouble, and a few thousand then it is serious trouble.

 

The other false signal on meters, is that even when measuring resistance, is that the meter, (any type), may read spurious AC electromagnetic residue from the whole layout acting as an aerial, to signals from the household wiring, local radio and TV transmitters, power lines, and phone masts etc.,

 

The whole layout may also act as a capacitor if the insulation is high at over a mega ohm, and actually store a minute amount of electrical power despite not being connected to anything!

 

None of these affect DCC, the signal would only be affected by massive faults, very low insulation, or high resistance feeds.

 

Hope this helps,

Stephen.

[Crosland]

If you have too much resistance in circuit to any part of the layout then there is a risk that a DCC command station will not shut down on short circuit

The OP is measuring the *open circuit* resistance between the rails and getting *less* than expected.

 

Andrew Crosland

[Edwin_m]

The OP is measuring the *open circuit* resistance between the rails and getting *less* than expected.

 

Andrew Crosland

 

OK sorry folks, I'll go back to sleep...

[Dagworth]

The under board wiring is via lightweight droppers to adhesive copper strip. The boards are braced across the layout only at 1ft spacing. Each 1ft section has its own length of tape connected to the neighbouring section by the same (7/0.2) lightweight wire that is used for the droppers. The end sections on each board are connected to the DIN socket by the same dropper wire.

 

Two things here, firstly if your tape is stuck directly to the board then you probably have a small element of leakage through the residual moisture in the timber. As others have said you can ignore it, but it may be where some of that leakage is coming from.

My bigger concern is the use of 7/0.2 wire to join the sections of copper tape, I'd be wanting something much heavier, probably 32/0.2 minimum as these links form part of the bus lines.

I would also think about the DIN links, if I was using these connectors then I would be looking at a heavy gauge two-core cable, and using two pins of the five paralleled up for each connection.

 

Andi

[Ray H]

Update:

 

Thanks for all the comments and advice.

 

I couldn't recreate the problem today - the meter needle never moved.

 

I'll take on board the comments about the wire gauge and upgrade the inter section links.

 

The interboard/DIN links are a much heavier gauge wire whch I've managed to solder the excess conductors around the plug/socket pins.

[Donw]

If you hands are touching the metal ends of the probes that could be the reading you got. Again depends on the humidity of your fingers.

Don

[Follower]

It might also be worth considering that the rails do form a (low value) air-gap capacitor and this will affect a meter's readings while it charges; particularly when the meter is on a high resistance range.

This effect is more noticable on the old tinplate base HD thre-rail track!

 

Simon.