Insulation resistance testing

[williamglossop]

This is more of an academic question rather than a specific problem.

Can anyone explain how insulation resistance tests work? I understand the test procedure and can interpret the results of the tests, but I don't really understand the physics behind how the tests work. Is it that 500v are slung down the conductor and a measurement is made at the other end to see if there has been any voltage leakage along the way.... im puzzled. Purely interested from a physics perspective rather than a real world problem I'm trying resolve.

Merry

 

[333rocky333]

No measurement is taken at the other end, it tests from one end only and measures resistance between the twoleads and whatever they are connected to.
Not sure how it does it though.

 

[crystal ball]

Usually on a single phase system IR checks are mad between L-N, L-E. N-E all from the same end

 

[ericmark]

E/IxR so E which stands for volts so is 500/I which is current = R the resistance, so it measures current but displays resistance.

Impedance and resistance are simular, but not quite, so although ohms law may show 5 mA should flow, it may well be that 10 mA flows because of inductive and capacitive links.

 

[sparxxxx]

Its a test between conductors, and its measuring any leakage current generated by the tester between eg L and N, the tester computes that leakage current and gives a reading in ohms, the lowest acceptable reading on a 230v circuit is 1MΩ or 1 Megohm (1,000,000 ohms). A reading of less than 1MΩ could be made up of one area of leakage current or many different points. It can be used to monitor deterioration in a circuit by comparing historic readings.

Although the minimum allowed is 1MΩ, this would be unacceptable on a new installation as its actually an indication of a fault, expected readings should be 100MΩ+ for all new circuits

 

[plugwash]

The tester applies a test voltage, measures the voltage and current and calculates resistance.

I'd imagine the hardest part in designing such a tester would be designing the very sensitive and yet robust (sooner or later someone is going to test a short-circuit) current measurement.

The tester I used had a maximum reading of 4GΩ At 500V that translates to 125 nanoamps.

 

[bernardgreen]

designing the very sensitive and yet robust

It is possible the the circuit generating the 500 Volt test voltage is designed with current limiting to prevent the current being too high for the current measuring circuit. If the current limiting did become effective then the device would have to indicate a failure in some way as the applied voltage would be less than the required test voltage.

 

[SimonH2]

The voltage generation would inherently be current limited anyway - simply by stepping up battery voltage to 500V (or whatever is selected). Mains powered equipment could be a different matter though !
As for sensitive electronics, consider that your typical DMM on auto-ranging will go down to it's most sensitive range looking for something to display. Then you go and connect it to anything up to 1000V which it has to detect and auto-range to a higher scale. The techniques have existed for decades, though I don't know enough about the electronics involved to know how it's done.

 

[JohnW2]

It is possible the the circuit generating the 500 Volt test voltage is designed with current limiting to prevent the current being too high for the current measuring circuit. If the current limiting did become effective then the device would have to indicate a failure in some way as the applied voltage would be less than the required test voltage.

My Fluke MFT reduces the test voltage as/when necessary to reduce the current to some limit (which I think is not much over 1 mA) AND then also displays what voltage it has had to reduce to, as well as displaying "<xxxx" in its display of the IR. With low resistances, the test voltage essentially falls to zero.

Kind Regards, John

 

[Risteard]

No measurement is taken at the other end, it tests from one end only and measures resistance between the twoleads and whatever they are connected to.
Not sure how it does it though.

It measures the voltage drop to determine the impedance.

 

[JohnW2]

It measures the voltage drop to determine the impedance.

What? !!

 

[Risteard]

What? !!

That's what it does. It puts a load on the circuit and measures the voltage drop to calculate what the impedance is.

 

[bernardgreen]

It puts a load on the circuit and measures the voltage drop to calculate what the impedance is.

That is one method that can be used to measure the impedance ( or resistance ) of an intentional circuit ( such as the end to end impedance of a ring final conductor ).

It does not work well if at all for measuring the impedance of an unintentional circuit ( such as where current is flowing through damp insulation )

 

[JohnW2]

That's what it does. It puts a load on the circuit and measures the voltage drop to calculate what the impedance is.

A load on what circuit? One only measures IR on a de-energised cable or circuit. The tester applies a voltage, but that only results in any current flowing if the insulation has a sub-infinite resistance and, even then, there is no 'load put on the circuit' by the tester.

Also (the clue is in the 'R' of 'IR') what one measures is resistance, not impedance, since the measurement is undertaken using DC.

 

[Risteard]

A load on what circuit? One only measures IR on a de-energised cable or circuit. The tester applies a voltage, but that only results in any current flowing if the insulation has a sub-infinite resistance and, even then, there is no 'load put on the circuit' by the tester.

Also (the clue is in the 'R' of 'IR') what one measures is resistance, not impedance, since the measurement is undertaken using DC.

Sorry. I thought we were discussing loop testing for some reason. Didn't realise we were discussing insulation resistance testing.