Advice on meter readings

[J_T]

Hi
I wired the first fix upstairs in a new house 2 years ago and am now returning to carry out the second fix.
On the lighting radial I have been using a fluke multi meter to check the resistance from ceiling rose to celing rose as I go along to check that all is good with the connecting 1.5mm T&E.

All seemed fine fine until the 3rd rose along, 1 live was 0.7 ohm (obs from the previous rose) but the next line (which shouldn't be connected to anything) gave 4Mohm with ref to the previous rose.

I tried a seperate test from the next un connected rose to the next rose after that (just the cable, not terminated) and the resistance showed 40Mohm spuriously.
It would be 40 M one sec the OL then I would touch the probe with my hand and it would jump back to 40M or 50M.

Has anyone come across this? The house is dry and I'm pretty sure no cables were squashed etc since the 1st fix but I ready don't want to start cutting plaster to try and find a fault if there isn't one there.

Thanks

JT

 

[marconi]

By Marconi I have enjoyed pondering this problem but you no doubt might be frustrated with it. It is not completely clear to me whether you are simply taking continuity measurements on unconnected lengths of cable runs or end-to end on a growing length of cable run as you wire in the CRs. So I have assumed the former but what I am about to write applies to the latter as well.

A digital multi-meter, when set to measure resistance, does so by injecting a constant current through the circuit under test and then measuring the voltage across the circuit. Using Ohm's Law it then calculates the resistance of the current's path. As the resistance of the circuit under test increases, the injected constant current is reduced. This is because the DMM is both battery powered and for safety the applied voltage is limited to around 10 Volts. With the meter you are using the resistance scale is automatically changing up or down, so in turn is the injected constant current. By choosing the inject current to be 1, 0.1, 0.001 mA etcetera for example a straight voltage reading is the same numerical value as resistance for an applied emf of 10Volt - 10k, 100k, 1000k Ohms etcetera

Now let us think about your test arrangement. You are using one short test lead and one long wandering lead. These are connected to the part of the lighting radial circuit you want to test for continuity. These long leads form an aerial and which will pick up electro-magnetic radiation. A small alternating current will flow - micro-Amps - but comparable to the injected current on its higher resistance scales.

There are inevitable capacitance effects which albeit small need charging and discharging. Since the injected current is small on the higher range scales there is a settling time before reliable measurements may be taken.

The connections you make for a test - using your crocodile clips, test lead plugs, bared conductors , dirt- may be covered in grease, dirt, dust and oxidation. Not only do these act as resistive layers, they can also act in a non-linear way akin to a diode. (Railway signalling systems using the track and train wheels suffer from this especially in autumn). This is another interference mechanism to the injection current especially at higher resistance scales.

Before a current - a flow of electrons - can take place between metallic surfaces the so-called wetting current must be exceeded. For switch contacts carrying small currents a capacitor is sometimes connected across them which is discharged when they close to boost the wetting current.

There are then a number of electrical effects which make the DMM unreliable in your test rig when operating on the higher resistance scales. I suggest you only use the DMM for continuity and select an instrument with a higher applied voltage and current to measure higher resistance ; I do not know if the MFT does but the earlier poster recommends its use.

I hope this helps.

 

[DPG]

Very in depth reply from Marconi as usual there I would do as he and Hellmooth suggested, and use the low ohms range on an MFT. The MFT will use at least 200mA drive current which will help to avoid spurious readings. Unlike your multimeter, which will have tens of Megohms input impedance, and therefore be too sensitive for the type of test you are carrying out. Daz

 

[J_T]

Great answers! Thanks for the input.

marconi, you are correct. I have disconnected any terminations I had and am measuring individual lengths of cables.

I continue to to see what I was seeing while retesting this morning with a fresh head. I am planning to use a megger next week to test with higher voltage but am stating to suspect capacitance, inducted voltages from parallel runs or antenna interference as being a factor.

Will post after I have carried out the megger test for future reference.

Thanks again

JT

 

[Hellmooth]

Can you elaborate on your post Marconi...........

 

[Dave OCD]

:lol::lol::lol::lol::lol:

 

[PEG]

Poor connection? He's getting a reading between two points which are not supposed to be connected at all.

...Aaah,and there's the rub,an assumption is made that the conductors are separate...when,in the previous 2 years lul,someone could have "joined" them...but "poorly" :evil:

These questions always get us cluckin'...when really,it's like someone asking the way from the bus station to the opticians...in a town in Poland...but not naming the town...:winkiss: