Basic installation testing with minimal disturbance

Most accurate measurements are made using Wheatstone Bridge and similar zero affect methods where one is looking for zero voltage and zero current in the indicating equipment when the bridge is balanced. The value being measured, the unknown of the four bridge elements, can then be calculated from the three known elements.
That's perfectly true in theory, and quite probably still used for some precision purposes, but the practicalities of a bridge to measure the extremely high or extremely low resistances we are discussing are far from trivial.

Kind Regards, John.
 
Sponsored Links
For crying out loud. Up to page 3, the answers related to the original question. Does a discussion about testing sewage plants or what types of antique test meters we have in our cupboards really contribute th this?

BAS advised SimonH to get a copy of GN3 and a decent text book.
SparkyBird advised to do some I&T and get a qualified electrician in and then compare results.
 
Interesting though that the closer to the ideal you can get the better your instrument but if you could actually reach the ideal it would be no use at all...
Electro-philosophy is it? Fair enough! Although what you say is, I think, inevitably true of any electromagnetic device (since some energy is clearly needed to make it work), I'm not sure that there is any theoretical limit to how close one can get to perfection with electronic devices. Extending the concept of, say, an IGFET, I would imagine that one could approach the situation in which an applied voltage (the voltage being measured) was able to control some other current flow without any current at all flowing in the 'input' - although there would then probably still be some niggling physicist tapping on the inside of my skull and muttering about 'getting something for nothing'! At a particle level, something would presumably have to happen at 'input' level in order for the output to be affected? I think one needs a glass in one's hand, preferably on a Friday evening, to discuss this one!

Kind Regards, John.
 
For crying out loud. Up to page 3, the answers related to the original question. Does a discussion about testing sewage plants or what types of antique test meters we have in our cupboards really contribute th this?
Nope - and you obviously are not obliged to read any of it.

Kind Regards, John.
 
Sponsored Links
For crying out loud. Up to page 3, the answers related to the original question. Does a discussion about testing sewage plants or what types of antique test meters we have in our cupboards really contribute th this?
Not really, but that's the thing about discussions (once you get past the trench warfare) - they tend to drift. In this case, the thread has drifted into a discussion about measurement of very large and very small resistances - via a real world example of insulation resistance reading varying with measurement voltage. It may not be of relevance to the original question, but it's of interest to those still taking part.

No-one is making you read it. If you want you can just click the link below that says "Stop watching this topic" and you'll get no further notifications about it.
 
Sorry that I'm only responding now John, but yes I have come across faults which do not show at all at lower voltages but at higher voltages have read an unacceptable value.

I have come across faults which show up when insulation testing at a higher voltage which do not present themselves when insulation testing at a lower voltage so it's certainly not impossible.
I have conceded that it's not impossible (voltage-dependent insulation breakdown does occur) but, as I said, I suspect that it's extremely rare, particularly with PVC cables and modern accessories/equipment.

As a matter of interest, have you seen such cases in which the measured IR was very high with 'low' (say 250V) voltage testing, but low (low enough to fail the test) with higher voltage testing? FWIW, from very limited experience, on some of the occasions I've measured very low IRs at 500V (<1M &#937;) I have, out of interest, tried with a DVM, and have got virtually the same answer at &#8804;9V.

Kind Regards, John.
 
Sorry that I'm only responding now John, but yes I have come across faults which do not show at all at lower voltages but at higher voltages have read an unacceptable value.
Interesting. When you speak of 'lower' and 'higher' voltages, are you talking about the difference between, say, 250V AND 500V (which is what essentially started this discussion), or voltages differing by appreciably more than that?

Kind Regards, John.
 
For crying out loud. Up to page 3, the answers related to the original question. Does a discussion about testing sewage plants or what types of antique test meters we have in our cupboards really contribute th this?
Not really, but that's the thing about discussions (once you get past the trench warfare) - they tend to drift. In this case, the thread has drifted into a discussion about measurement of very large and very small resistances - via a real world example of insulation resistance reading varying with measurement voltage. It may not be of relevance to the original question, but it's of interest to those still taking part.

No-one is making you read it. If you want you can just click the link below that says "Stop watching this topic" and you'll get no further notifications about it.

My post was for your benefit, as the thread is well and trully hi-jacked.
In the middle of this is some good advice, but it is buried.
 
OK, an update.

As it happened, I had reason to have a socket off so was able to measure r1, rN, and r2 for the RFC, and added a L-E link in the CU to measure R1+R2 (and repeated L-N for R1+RN) for the rest of the circuits.

I did find one fault - rN was a bit high (about 1R where r1 was about 0R36).

So, if you have a socket off, and one of the measurements is a bit high, how do you go about narrowing down where it is ? I'm curious to know how others do it, and I'll say how I did it shortly.

I also went round each socket in turn measuring r1+r2. Mostly out of curiosity to see how it varied around the ring. I was slightly surprised how much variation there was - at first I found two sockets on one double place with different values, then I found flicking the switch a couple of times brought it down. It was much the same for most of the sockets.
 
how do you go about narrowing down where it is ? I'm curious to know how others do it, and I'll say how I did it shortly.
I have used a constant current source drive 4 amps round the CPC conductor being tested and then measure voltage differences between the earth pins of sockets using a wander lead. These voltages give reasonably accurate indications of the resistances of the CPC between the sockets. With a high impedance meter taking no significant current from the test points any resistance in the wander lead and/or between the socket's CPC terminal and earth pin does not affect the reading.

This method also assists in mapping out the route of the cable.
 
Hmm, I did think of something along those lines.
Alternatively, a long lead for a return and just go round measuring the resistance from where I've split the ring. As it is, without popping home I didn't have much with me.
 
Measuring reistance of cables through the socket using the resistance function of a multi-meter can result in un-reliable measurements due to resistance inside the socket and between probes and and socket. Not a real problem for general test and certify but possibly a significant error when looking for a defective conductor.
 
Yes indeed.
I was using a plug in adapter - and took to several insertions, and flicking the switch a few times, to wipe the contacts clean. Didn't take much to get a consistent reading.
 
Yes indeed. I was using a plug in adapter - and took to several insertions, and flicking the switch a few times, to wipe the contacts clean. Didn't take much to get a consistent reading.
Indeed. I think we tend to forget how iffy is the measurement of low (say <1&#937;) resistances. The measurement technology itself is fine down to milliohms and below, but the quality of the connection is obviously crucial. You've only got to try to get a zero measurement by touching together the probes of a (recently 'zeroed') test meter to see the degree of variation which is possible. With a 'resistance meter', mutiple attempts at a measurement and manoeuvures such as you describe are all one can do to attempt to get reasonably accurate results - but Bernard's approach of measuring voltages when a known (substantial) current is flowing is theoretically better. That approach doesn't solve the problem of small amounts of ('variable') resistance at, say, switch contacts, but it does (with a very high resistance voltmeter) at least more-or-less eliminate problems due to the quality of contact between meter and circuit being tested.

Kind Regards, John
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top