Accurate earth loop impedance testers or MF tester recommendations for industrial type installations/supplies

CF52 said:

I do not feel my older Megger 1741+ is particularly accurate. When I compare readings to a colleagues modern Metrel touch screen style multifunction tester the impedance readings he obtains are significantly lower.

It does not answer the question, but I assume that both of you have a regular (annual) calibration check. Megger claims an accuracy of ± 5% to the nearest 10 mΩ.

Why do you feel that your MFT is less accurate than your chum's Metrel?

Do you really need to measure such small values? Can max Zs not be calculated (e.g. using DNO data plus impedance of installation cables) and the test used merely to prove a connection rather than deliver accurate values?

   - Andy.

Are you taking the resistance of your test leads into account?

On the Megger MFT1741+ this has to be set manually in the setup menu under 'Loop' and the factory default setting is 0.07 Ω.

Note that although the Megger MFT1741+ will display loop impedance values down to 0.01 Ω the manufactuer's specification only applies down to 0.3 Ω on the high current range and 1.0 Ω on the low current range. This can be seen on the table on the back of the instrument case, or in the manual for the instrument.

I would recommend getting hold of the booklet by Megger 'Earth Electrode & Earth Loop Impedance Testing - Theory & Applications' which explains the perils and pitfalls of trying to measuring low loop impedance values close-up to the distribution transformer.

I have no connection with Megger other than as a user of their products and services.

There is a general problem with measuring low resistances, as you have to use large test currents to generate a decent voltage drop. Ideally this has to be done with a 4 point (Kelvin) connection, so that the test current does not flow down the wires making the voltage measurement. Multi function testers in particular are a design compromise, and therefore make 2 wire measurements, with modest test currents, and the penalty is that the low end of the range is prone to larger errors. 

A reading like this only verifies the earth is not disconnected, but a knowledge of the transformer regulation and the dimensions of the main cables will allow a more accurate calculated estimate.

Mike

EDIT

Very low loop impedance (or very high PSSC readings..)   are a common source of distress when it comes to worrying about things, for example 6000A rated breakers in a CU. In the days of a moving needle instruments, where the high PSSC numbers and infinity were all bunched down at one end of the scale, much like high resistances on the ohms range,




you sort of knew when you were in deep water as it were because a slight wobble of the head got you from 1000 to several thousand,  while the hundreds were well separated -  the problem of 1 divided by something approaching zero.
The problem with the digital instrument is it happily calculates a single value based on a really small voltage drop , and displays lots of significant figures, but it still does so from a measurement where the uncertainty is rising rapidly at one end of the scale, and does not really convey the natural uncertainty.
It is not really a reliable way to make a measurement. cleaning meter lead connectors may reduce the divergence but it is inherently an increasingly imprecise measurement.
(and conventional  meter leads will add a few tens of milliohms per metre ,even when perfectly clean.)

Mike.

Chris Pearson said:

It does not answer the question, but I assume that both of you have a regular (annual) calibration check. Megger claims an accuracy of ± 5% to the nearest 10 mΩ.

Overall service error of ± 10 % ± 2 digits, though, on loop test, I believe ... and not uncommon for a general MFT?

Given the  ± 2 digits, and resolution of 0.01 Ω, any readings that would come in below about 0.05 Ω (equivalent to about 4.6 kA prospective fault current) are effectively indistinguishable with such an instrument.

This is the problem with general MFTs used with high prospective fault current supplies.

An example from the EI Design Guide would suggest the folly of using any of the MFTs. Note the need to determine R and X as separate considerations, something a MFT does not do. 

And from GN3

gkenyon said:

This is the problem with general MFTs used with high prospective fault current supplies.

Quite so. I fear that the original question has not been answered. I would say a specialist EFLI tester, but which manufacturer, and which model?

The description of a '4-Wire Impedance Test' in the Megger booklet I mentioned would suggest something like the Megger NIM1000 Impedance Meter.

• https://www.megger.com/en-gb/products/nim1000 

It uses a test current of up to 1000A, has a measurement resolution of 1 mΩ gives the measurement results as Z Ω / R Ω / X Ω.

However, with a price tag around £8k its purchase and use is probably only economical for DNOs and similar.

I have no connection with Megger other than as a user of their products and services.

- Ross

you do need to ask how often you need that sort of measurement rig. Unless the answer is 'daily' then hiring one in, either with or without the skilled subcontractor to operate it, is probably the way forward.

Much like buying your own telehandler, cherry picker or crane.... which is also rarely owned outright, except perhaps for Bob the builder who seems to own a lot of plant that is mostly underused. But that is a fictional  kids TV programme.

Mike.