Spark123
We seem to be on our own - still here is my answer to your points. I don't expect you to see the light and give up this nasty habit (testing r1 + r2 that is
) - but give my responses some consideration.
Q1) What does knowing the resistance of R1 + R2 tell us about the safety of the circuit under test?
A1) The combined resistance of the phase and CPC, that we have continuity of both (but as I said before not necessarily the correct polarity).
Response 1
It does indeed, but why on earth would we want to know that
. I say again, what does that tell us about the safety of the circuit?
So what does it tell us about anything?
That the circuit, which is just part of the Earth Fault Loop Path (EFLP,)is continuous. This is confirmed (NOT
) by the massive test current that is all of 0.2A. The fact that part of the circuit might be connected by no more than a single fine strand of wire is not detected.
Now some, not the contributors to this forum I hasten to add, will insist that this measurement is the be all and end all of safety checks, that an installation is unsafe if it is not taken and recorded - regardless of whether such action is of any use, or even possible (i.e. parallel paths).
Q2) Does an R1 + R2 test prove that a circuit is fit for purpose?
A2) Not on it's own but it gives you an indication when added to Ze if the circuit under test has a low enough efli before you even get to the point of energizing it.
Response 2
Consider the situation where we are using ADS with overcurrent protective devices (OPD) used as the dis-connector. This is generally TN installations. We need to know if the impedance of the EFLP is low enough to operate our chosen OPD in the required disconnection time.
R1+R2 is part of the circuit but not all. The value of R1+R2 may or may not be a major factor - this will depend on whether there are any parallel paths to R2, and if so, how reliable they are. Now in the PVCTWE world that is domestic wiring, parallel paths will either be fortuitous or via an appliance such as a water heater. The latter can be eliminated on small installation for both initial verification and periodic inspection if required. The former, may not be found, but might be suspected - they usually cannot be eliminated.
Now what is the purpose that this EFLP must be fit for?
It is to carry a current limited only by the impedance of the EFLP.
This is:
(i) R1 + R2 and any parallel paths
(ii) Ze
Note that Zs = Ze + R1 + R2 is not a valid equation, the best you can say is that there is a relationship between the parameters.
Ze being the total impedance external to the installation - this includes the impedance of the supply transformer, the supply cables and the bonding of all of your neighbours on a TN system
.
So in the domestic arena a typical fault current might be around 2kA. Still we have tested part of our circuit with 0.2A, so that is alright then
(PS: Don't start me on loop impedance testers, we would be here all day
)
Does R1 + R2 testing prove fitness for purpose - you jest of course.
Q3) Why is a 0.01 resolution specified - this suggests that the site based test method will produce results that are accurate - will it?
A3) The max efli values for an OCPD are given to 2 decimal places so it would make sense that we need to have accuracy to 0.01 ohms.
Response 3
Site based testing can introduce errors as high as 30% - for confirmation look at the standards for the instruments and at the specifications produced by some of the more responsible manufacturers. If you look hard enough you will find that this is the case. Note that when you send your instruments away for calibration they are checked against a resistor. The test may not even use your worn out test leads (croc clips). In an installation you are not testing a resistor - you are testing a load of cable that runs off in various directions picking up noise signals and generating thermal emfs. You connect to this using croc clips - need I say more
.
The results you obtain are then compared against a table in BS 7671. The source of the data is the characteristic curves in appendix 3. These are BS curves, so worst case approximations, not actual manufacturers' data. They are log log making high resolution reading virtually impossible. The figures obtained from the APP3 data are used to calculate an impedance (well resistance really in small installations). They are presented as maxima for a fully loaded cable and you are invited to use a crude temperature adjuster to account for the difference in temperature on load and under test conditions.
Do you still think a two decimal place resolution has any meaning
.
Q4) What is the resolution required of an analogue instrument and should it be acceptable if it is different?
A4) I'd expect you to need one which gives an accurate resolution so the reading is of value, i.e. an ohm meter which only measures in K ohms will not be accurate enough to measure a value required to be added to Ze to give Zs. If you decided to use a wheatstone brigde we could drive amps through it
Response 4
See response 3 - IMO the authors of GN3 have simply not addressed this issue.