Strange IR readings......DOH!!

[davelx]

So there I am doing IR tests on an existing lighting circuit prior to moving it from a BS3036 CU to a new RCBO CU. All lamps, tubes, dimmers etc. removed, checked for caps in the fluorescent fittings etc. All light switches on.

L-E checks OK, N-Es check OK, L-N - Zero Ohms! WTF? Did a continuity check - L-N 140 Ohms!

Turn off all light switches - no change. Re-connect to the old CU and check every light works OK. Measure current with all lights off - about 10mA. Hmm, 230/140 is a lot more than 10mA!

Disconnect, test again - no change. Rack my brains and realise the under cabinet tube in the kitchen is still connected - but switched off. Disconnect that and test again. No change. Made a cup of tea.

An then it came to me - DOH! The shaver socket in the bathroom is connected to the lighting circuit and the 140 Ohms is the DC resistance of the transformer primary. Disconnect and all is well.

Moral of this story is ---- A nice cup of tea makes everything OK

 

[ricicle]

Easily done. Good job it wasn't anything electronic.

 

[davelx]

Easily done. Good job it wasn't anything electronic.

I always start at 250V just in case! In this case I knew there was an X10 LD11 dimmer under lounge floor, so removed that first. It's going into the new CU which is a much better place for it.

 

[Taylortwocities]

The TV amp hidden in the loft is often the first one to die courtesy of 500VDC up its chuff!

 

[JohnW2]

[I always start at 250V just in case!

I have to say that, for the sake of a few seconds, I always do 'ELV' resistance (aka 'continuity') measurements before attempting IR measurements, to detect things like your overlooked shaver socket. I'm probably overcautious, but I'm not convinced that all electronic things woud necessarily be happy with 250V DC, and a continuity test might detect at least some of them (as well as well as passive things, like the shaver socket).

Kind Regards, John

 

[ricicle]

A friend of mine who I sometimes help out at weekends always insists on continuity tests between L/N before we move 'up voltage' on IR.
This is especially as he tends to do PIRs in very large, multi-million pound call centres !

 

[JohnW2]

A friend of mine who I sometimes help out at weekends always insists on continuity tests between L/N before we move 'up voltage' on IR. This is especially as he tends to do PIRs in very large, multi-million pound call centres !

As I said, that's what I do - but not because I deal with multi-milion pound call centres

As I implied, it's probably a case of being unnecessarily cautious - since almost anything that is used to having 230V AC between its L & N is likely to tolerate 250V DC for a very brief period (I'm not sure how long IR test pulses are - the manual of mine doesn't say). Let's face it, if one starts at zero-crossing, for the first 10mS of 50Hz AC, the load doesn't know whether it's connected to AC or DC (with a peak voltage of ~325V in the case of 230V AC). However, for the sake of those few seconds, it 'feels' like a reasonable precaution to start with a very low voltage test!

Kind Regards, John

 

[EFLImpudence]

The Fluke 1652 will not carry out the IR test if it detects very low ohms continuity in the circuit.

It will show 0.00MΩ @ 0V.

At higher ohm readings it will stop at (I presume) whatever voltage has been necessary to determine the reading.

 

[JohnW2]

The Fluke 1652 will not carry out the IR test if it detects very low ohms continuity in the circuit. It will show 0.00MΩ @ 0V. At higher ohm readings it will stop at (I presume) whatever voltage has been necessary to determine the reading.

Mine's a 1652. The manual is not clear as to exactly how it works, the only specified situation in which it will not do the test being if it detects >30V before startingthe test (I would have thought it should inhibit the test if it detects any significant voltage!). So, I've just done some experiments. It never "does not carry out the test", but I think I've worked out what it does do. The following lists actual resistances tested and the test voltage as displaced by the 1652 (500V range):

200 Ω 0V
330 Ω 1V
560 Ω 1V
1 kΩ 2V
5.6 kΩ 11V
10 kΩ 17V
39 kΩ 66V
120 kΩ 193V
220 kΩ 342V
470 kΩ 515V
1 MΩ 520V

What is clearly going on here is that the meter is using a constant current source of something in the range 1.5-2.0mA, fed from ~520V (thereby establishing a ceiling). It will always put (or try to put) ~1.5-2.0mA through the resistance being measured (up to the ceiling), and the resultant voltage across whatever is being measured (as displayed on the meter), which is used to calculate the displayed resistance, will be proportionate to that resistance.

It is therefore clear that it cannot doi any harm to low resistance thingys, since it will only put a very low voltage across them, and can never put more than about 2mA through them.

Kind Regards, John

 

[JohnW2]

I should perhaps have added that the 1652 manual says that the IR test current is 1mA at all resistances - although, per the figures I posted, my 1652 is clearly using a bit more than that. Over the range of resistances from 5.6kΩ to 220kΩ (avoiding the inaccurate bits at the bottom, and the 'ceiling' at the top), calculated currents (using nominal values of the test resistors) ranged from 1.55 mA to 1.96 mA.

Kind Regards, John

 

[Spark123]

Looks like Fluke have re-written ohms law then

 

[EFLImpudence]

How can you determine what the meter is doing by quoting the display results?

It will display any resistance between 1MΩ and 500MΩ and still display 500ishV or

between 1MΩ and 1000MΩ and still display 1000ishV

 

[JohnW2]

How can you determine what the meter is doing by quoting the display results? It will display any resistance between 1MΩ and 500MΩ and still display 500ishV or between 1MΩ and 1000MΩ and still display 1000ishV

Of course, but that's because it's reached the 'ceiling' voltage that the constant current source can deliver (i.e. roughly the voltage that is powering the constant-current generator).

If, say, the 'constant current' being generated is 1.5mA and the maximum voltage the constant current generator can provide is 520V, then the voltage across the resistance being measured will be ~520V for any resistance above about 347kΩ. Similarly for the 1000V tests. If one wants to examine what the meter is actually doing, one has to look at resistances which are below this 'ceiling' (i.e. the saturation point of the constant current source) - in the above case resistances below about 347kΩ - and if one does that, one finds that E/R is roughly constant at 1.5mA or so, regardless of the value of the resistance being measured.

Is that clearer?

Kind Regards, John

 

[JohnW2]

Looks like Fluke have re-written ohms law then

More likely they changed the test current they use after the manual I've got was written!

I foolishly did not note down the actual resistance readings being displayed when I did my tests (so calculated test current only on the basis of the nominal values of the resistors), but they were all pretty close. I'll do it again (without that oversight!) so I can determine precisely what current it's using.

Kind Regards, John

 

[Adam_151]

A friend of mine who I sometimes help out at weekends always insists on continuity tests between L/N before we move 'up voltage' on IR.
This is especially as he tends to do PIRs in very large, multi-million pound call centres !

I'm supprised he is doing any IR testing between L-N in that kind of installation at all