Higher-accuracy low Ze test procedure

[IQ Electrical]

This is the method that we use to ensure a higher accuracy on very low Ze readings.

At first glance, a difference of say 0.09 Ohms does not seem much on a Ze reading but when this value is used (by the instrument or by calculation) to determine PFC then the difference can be massive.

For example: 230/0.10 = 2.3KA 230/0.01 = 23KA

It goes without saying that any connections must be made safely and live testing carried out with equipment in accordance with GS 38.



1 Find a suitable impedance to be used-this should be around about 0.5 Ohms, R2 leads are good for connectivity reasons.

2 Find an outlet on the installation (fused connection unit, socket outlet etc.) that is likely to have a Zs of around 0.20 Ohms.

3 Measure that Zs with the instrument test leads (not the BS 1363 plug lead) and note the value.

4 Measure the Zs again but with the selected impedance (SAFELY) in series with the earth test lead.

5 Subtract the value measured in step 3 from that measured in step 4 and you now have a value for your test impedance (only for that particular installation).

6 Now return to the origin and measure Ze with the test impedance in series with the earth lead.

7 Subtract the value obtained in step 5 and you now have a far more accurate figure for Ze.

Obviously ambient temperature, harmonics, transformer noise etc. all affect readings but unless you have the transformer impedance and the details of the distribution circuit length and CSA etc. then this is about as accurate a reading as you can hope to measure!

 

[Guest123]

Thread Stuck.

 

[Guest 004]

round of applause for IQ

 

[IQ Electrical]

I forgot to mention that the test can also be used to determine Prospective Short Circuit Current (L-N) by using the impedance in series with the neutral test lead.

 

[Guest 004]

I will do a calc for that, as that is all the test instrument does anyhow

 

[T&Earth]

Hi IQ
Thanks for explaining this method, very useful

T&E

 

[IQ Electrical]

Hi IQ
Thanks for explaining this method, very useful

T&E

No problem, it's quicker to do than it looks when it's written out!

 

[Guest 004]

No problem, it's quicker to do than it looks when it's written out!

yes agreed, when I first read it I though 'what the feck', but as I have done R2 readings before the procedure is quite easy to do

 

[mozzo]

Applying basic electronics, I would assume that within the meter very high ohm, low tolerance (accurate) resistors with a high wattage for safety would be used to measure a voltage drop that would be used to calculate the PSCC. The introduction of a low ohm resistor external to the test meter would surely be insignificant in the measuring process. The added external resistor would also have to be super accurate if you then use the external resistance value to subtract from the measured value. Again, for such low measurements, you would also need to measure the value of this external resistor and not juts take the reading on the packet as gospel. However I do see the logic in what you have described as if your meter is only capable of reading to 2 decimal places and the Ze initial Ze measured is near the end of the range then boosting the Ze would bring the reading within a range that is more acceptable and more accurate to your meter. Again this also depends on the meter and the accuracy of the meter in relation to the upper, lower and mid range points of measurement. If your meter is totally accuate throught its entire range then the prescribed technique is not required.

 

[Cammy]

I'm presuming that this method would be ok to use and explain for an (Elecsa) assessment? I have changed a consumer unit and the Ze is coming out at 0.03 ohms; so I have got some interesting PFC results off that. The substation being across the road from site is just great! :furious3:

 

[Mark_Burgess]

I take it this method would work for the PFC/PSCC?

Looks like I will have to try this later, the reading I got yesterday of 50kA was taken 3 times to ensure I was not reading it wrong, would be 100kA between lines!!

 

[IQ Electrical]

I take it this method would work for the PFC/PSCC?

Looks like I will have to try this later, the reading I got yesterday of 50kA was taken 3 times to ensure I was not reading it wrong, would be 100kA between lines!!

Absolutely, where is the transformer sited in relation to the point of PFC measurement?

 

[Mark_Burgess]

Half a mile or more down the road, this factory used to be a joinery and had one hell off power consumption, the incomer fuses must be in the region of 200A, I cant identify them and will need to call SSE out as my customer has decided he really wants to know what they are for the PIR, neither SSE or B,Gas know what is installed. The cable is approx 150mm CSA.

 

[Mark_Burgess]

Phew, thats better, 4.9kA!!

 

[Engineer54]

50ka is High!! What you'd expect from a 2MVA TX, even then you would need to be dammed close to that TX ...lol!!
Think you must have been missing the decimal point there...lol

 

[Mark_Burgess]

[QUOTE Think you must have been missing the decimal point there...lol[/QUOTE]

I did think that first time, so repeated the test 3 more times, even put the back light on the display to make sure!!

 

[telectrix]

never had these problems when the most advanced test equipment we had was a AVO 8. LOL.

 

[Mark_Burgess]

Managed to get it powered down today and main isolator switch opened, 250A fuses and found a cable marking of 185mm, so I was close before but not quite there.

 

[VoltzElectrical]

This is the method that we use to ensure a higher accuracy on very low Ze readings.

At first glance, a difference of say 0.09 Ohms does not seem much on a Ze reading but when this value is used (by the instrument or by calculation) to determine PFC then the difference can be massive.

For example: 230/0.10 = 2.3KA 230/0.01 = 23KA

It goes without saying that any connections must be made safely and live testing carried out with equipment in accordance with GS 38.



1 Find a suitable impedance to be used-this should be around about 0.5 Ohms, R2 leads are good for connectivity reasons.

2 Find an outlet on the installation (fused connection unit, socket outlet etc.) that is likely to have a Zs of around 0.20 Ohms.

3 Measure that Zs with the instrument test leads (not the BS 1363 plug lead) and note the value.

4 Measure the Zs again but with the selected impedance (SAFELY) in series with the earth test lead.

5 Subtract the value measured in step 3 from that measured in step 4 and you now have a value for your test impedance (only for that particular installation).

6 Now return to the origin and measure Ze with the test impedance in series with the earth lead.

7 Subtract the value obtained in step 5 and you now have a far more accurate figure for Ze.

Obviously ambient temperature, harmonics, transformer noise etc. all affect readings but unless you have the transformer impedance and the details of the distribution circuit length and CSA etc. then this is about as accurate a reading as you can hope to measure!

I feel stupid for not understanding this. Perhaps there are too many words? How do I do this with my Megger 1553? I know how to do Loop testing with it, but i don't really get what is meant here by 'safely in series with the earth test lead'.

Can anyone clear this up, it sounds interesting.

Thanks

 

[adamh]

Isnt this a live test, with the main protective bonding conductors disconnected? (so no good for an assessment) What is the extra impedance for?

 

[ElectriCol]

Pardon my ignorance, but...! Isn't this method merely describing an alternative to 'nulling' the meter with the test leads attached? Or are you saying that the apparent resistance of the (R2) test lead would vary with different installations, depending on the Ze????

 

[Sparky Ninja]

It's to do with how the accuracy of the meter becomes more inaccurate, or inconsistent, when the impedances are very low. And as IQ has stated the resulting calculated/measured fault currents vary greatly. By applying a controlled resistance into the circuit, the meter can perform more accurately/consistently.

 

[pebsham]

If you have a calibration check box could you use one of the resistors there? They are meant to be accurate & consistent and have good connection points

 

[IQ Electrical]

I don't see why not-as long as the connection is sound and safe.

 

[David Marshall]

This is the method that we use to ensure a higher accuracy on very low Ze readings.

At first glance, a difference of say 0.09 Ohms does not seem much on a Ze reading but when this value is used (by the instrument or by calculation) to determine PFC then the difference can be massive.

For example: 230/0.10 = 2.3KA 230/0.01 = 23KA

It goes without saying that any connections must be made safely and live testing carried out with equipment in accordance with GS 38.



1 Find a suitable impedance to be used-this should be around about 0.5 Ohms, R2 leads are good for connectivity reasons.

2 Find an outlet on the installation (fused connection unit, socket outlet etc.) that is likely to have a Zs of around 0.20 Ohms.

3 Measure that Zs with the instrument test leads (not the BS 1363 plug lead) and note the value.

4 Measure the Zs again but with the selected impedance (SAFELY) in series with the earth test lead.

5 Subtract the value measured in step 3 from that measured in step 4 and you now have a value for your test impedance (only for that particular installation).

6 Now return to the origin and measure Ze with the test impedance in series with the earth lead.

7 Subtract the value obtained in step 5 and you now have a far more accurate figure for Ze.

Obviously ambient temperature, harmonics, transformer noise etc. all affect readings but unless you have the transformer impedance and the details of the distribution circuit length and CSA etc. then this is about as accurate a reading as you can hope to measure!

Why dont you just test it normally like everybody else I don't understand the logic behind this most readings on TNS supplies are about 0.2 to 0.3 ohms and PME is a little lower, Tower blocks tend to be a lower if you test them in the flat because of the bonding and you don't get access to the main intake but in general it's an easy test why complicate it

 

[Des 56]

Why dont you just test it normally like everybody else I don't understand the logic behind this most readings on TNS supplies are about 0.2 to 0.3 ohms and PME is a little lower, Tower blocks tend to be a lower if you test them in the flat because of the bonding and you don't get access to the main intake but in general it's an easy test why complicate it

The reason
Widdler quote
It's to do with how the accuracy of the meter becomes more inaccurate, or inconsistent, when the impedances are very low. And as IQ has stated the resulting calculated/measured fault currents vary greatly. By applying a controlled resistance into the circuit, the meter can perform more accurately/consistently.

The outcome
The level of Pfc measured could mean the difference between a rated device passing or failing,with readings that are low he demonstrates the large variation that can occur
I thought it was an excellent and informative post by a guy who definitely knows his onions

 

[Silly Sausage]

The trouble with this method, at such low readings (instrument resolution of 0.01 Ohm, looking a measurement of 0.0x Ohms), is that it doesn't eliminate the instrument's 'number of digits or counts' error.
So, it's essentially a futile exercise!

 

[David Marshall]

The reason
Widdler quote
It's to do with how the accuracy of the meter becomes more inaccurate, or inconsistent, when the impedances are very low. And as IQ has stated the resulting calculated/measured fault currents vary greatly. By applying a controlled resistance into the circuit, the meter can perform more accurately/consistently.

The outcome
The level of Pfc measured could mean the difference between a rated device passing or failing,with readings that are low he demonstrates the large variation that can occur
I thought it was an excellent and informative post by a guy who definitely knows his onions

There is only oe approved method of measuring Ze my advice is stick to the industry guidelines in GN3

 

[Engineer54]

There is only oe approved method of measuring Ze my advice is stick to the industry guidelines in GN3

Yes, but if the DNO's TX is basically in the garden, your MFT Ze/PFC values obtained will be next to useless, so unless you know the TX size and it's impedance along with other details to calculate these values your snookered. Then again, you could always use a very expensive high resolution (0.00X) ELI tester!! IQ's method will give you a higher level of accuracy applying the use of a known/controlled resistance into the circuit being measured....

Oh but i forgot, the picture book GN3 doesn't actually mention anything about thinking outside of the box when needs arise does it!!

 

[GEOFF S]

I dont see this technique helping.

Lets say the Impedance is 0.02 ohms, at 230 V this equates to 11.5kA.

Now if we introduce an impedance of 0.5 ohms, our circuit is now 0.502 ohms.

At 0.502 ohms standard meters will have a 10% +/- and a 5 digit error +/- around the 0.5 area.

So this equates approx to a +/- of 0.1 ohm, so even with the 0.5 ohm added the reading could be 0.6 ohms or 0.4, a differnace of 0.2 ohm's.

Cheers

 

[Engineer54]

I dont see this technique helping.

Lets say the Impedance is 0.02 ohms, at 230 V this equates to 11.5kA.

Now if we introduce an impedance of 0.5 ohms, our circuit is now 0.502 ohms.

At 0.502 ohms standard meters will have a 10% +/- and a 5 digit error +/- around the 0.5 area.

So this equates approx to a +/- of 0.1 ohm, so even with the 0.5 ohm added the reading could be 0.6 ohms or 0.4, a differnace of 0.2 ohm's.

Cheers

So without going into another pointless in depth discussion so you can prove a point, when was the last time you applied the +/- 10% and 5 digit error or so, to any value that you measured with you're MFT/EFI tester, and recorded that amended value on a report sheet??

The above method is just a means of enabling a relatively standard MFT/ELI meter to more accurately measure values that are very close to it's ultimate limit, nothing more, nothing less!!

 

[GEOFF S]

So without going into another pointless in depth discussion so you can prove a point, when was the last time you applied the +/- 10% and 5 digit error or so, to any value that you measured with you're MFT/EFI tester, and recorded that amended value on a report sheet??

The above method is just a means of enabling a relatively standard MFT/ELI meter to more accurately measure values that are very close to it's ultimate limit, nothing more, nothing less!!

I Dont, but then i wouldnt add a 0.5 ohm impedance to the circuit either, chocolate fireguard!

Well is not going to give you a more accurate result, how can it, when even with the added 0.5 ohm impedance you could get a negative result!

The example i gave was 0.02, now the meter error could be 0.2 ohms, so far from accurate, misleading infact!

Cheers

 

[Engineer54]

I Dont, but then i wouldnt add a 0.5 ohm impedance to the circuit either, chocolate fireguard!

Well is not going to give you a more accurate result, how can it, when even with the added 0.5 ohm impedance you could get a negative result!

The example i gave was 0.02, now the meter error could be 0.2 ohms, so far from accurate, misleading infact!

Cheers

I'm not going to even bother going there....

 

[mrloy99]

Lets say the Impedance is 0.02 ohms,

''Now if we introduce an impedance of 0.5 ohms, our circuit is now 0.502 ohms.''

Hmm is it? 0.5 plus 0.02 is 0.52

sorry old post I know....I personally prefer to use a resistor with value of 1 ohm as its a lot easier to subtract at the end

Cheers[/QUOTE]

 

[davesparks]

I Dont, but then i wouldnt add a 0.5 ohm impedance to the circuit either, chocolate fireguard!

Well is not going to give you a more accurate result, how can it, when even with the added 0.5 ohm impedance you could get a negative result!

The example i gave was 0.02, now the meter error could be 0.2 ohms, so far from accurate, misleading infact!

Cheers

If you look at the specification of a loop tester such as the megger offerings you'll see that their accuracy is stated as being higher within a certain band of resistances so adding a dummy resistor in can put the test results into the more accurate part of the range.

 

[David Marshall]

Lets say the Impedance is 0.02 ohms,

''Now if we introduce an impedance of 0.5 ohms, our circuit is now 0.502 ohms.''

Hmm is it? 0.5 plus 0.02 is 0.52

sorry old post I know....I personally prefer to use a resistor with value of 1 ohm as its a lot easier to subtract at the end

Cheers

[/QUOTE]
I can't see the point in this method, it's a total waste of time. Just measure ze and move on adding a resistance to the figure only to subtract it will only give you what you would get anyway. If you get 0.02 just record it and recommend suitable mcbs that's all you need to do.

 

[davesparks]

I can't see the point in this method, it's a total waste of time. Just measure ze and move on adding a resistance to the figure only to subtract it will only give you what you would get anyway. If you get 0.02 just record it and recommend suitable mcbs that's all you need to do.[/QUOTE]

The point is to make use of the most accurate part of the test meter's range.

 

[David Marshall]

I can't see the point in this method, it's a total waste of time. Just measure ze and move on adding a resistance to the figure only to subtract it will only give you what you would get anyway. If you get 0.02 just record it and recommend suitable mcbs that's all you need to do.

The point is to make use of the most accurate part of the test meter's range.[/QUOTE]

How accurate do we need to be, I'm happy with the approved method described in GN3 which in all fairness is the only method recommended, why complicate this test. My advise is to stick to the approved method because at the end of the day nobody can fault you for complying with your training and non statutory regulations,
This is my final post on this subject, please don't reply as I will not respond.
Thanks David