LOW ZE READINGS

[Farmelectrics]

carrying out an eicr on a building recently we found the external ze reading to be extremely low ie 0.02ohms. so by measurement and testing are fault current read 20ka. now most of the mcbs rcbos ets have an icn of 10kohms. this poses a problem as obvisoluy the mcbs etc would self destruct under fault conditions. what would you code this and more importantly how would we rectify this problem, Interested to here your thoughts.

 

[Wilko]

Hi - Some testers can be inaccurate at very low values, so I’d be redoing the measurement with a known length of cable added in series and seeing if it stacked up. I have a Long Lead that’s about 1 Ohm that I’d use for this.

 

[James]

2 options available,

1 extend the supply cable by running a length of swa from incomer to fuseboard, 50m would probably bring the reading down enough.

2 change the breakers to a higher rating of fault current.

 

[Farmelectrics]

Wilco it is pretty close to the sub station however I've always been a great believer the meter never lies even checked the meter on my cal card all readings very accurate.

 

[Strima]

3. Test using three lead no trip and get a higher Ze reading therefore lower PFC... ??

 

[James]

I had a case where I was measuring about 3m away from transformer and the meter was giving readings that were far higher than I expected.
changed the values on cert. to what the transformer manufacturer stated.

 

[Farmelectrics]

are you suggesting I cheat and higher the ze reading. ha ha

 

[Farmelectrics]

all the mcbs are 10k icn would be extremely expensive to change them what code would you give to this

 

[Julie.]

The reading may be perfectly correct, and the mcbs ok as well.

If the installation is protected by fuses, the let through may be less than the rating of the downstream mcbs.

Just check the fuse let through first.

 

[James]

are you suggesting I cheat and higher the ze reading. ha ha

not cheat,
figure can be found by measurement or investigation.

if the transformer manufacturer says max pfc is 20Ka and the meter reads 92Ka then the manufacturer wins on my paperwork.

all the meters suggest that they may be inaccurate when close to transformer but not 1 of them says in the instructions that below x meters away you should ignore the reading on meter!!

 

[Julie.]

all the mcbs are 10k icn would be extremely expensive to change them what code would you give to this

If the let through is greater than 10kA, its c2

If its less, then no code

 

[Farmelectrics]

Julie were would I find the let through value and are you talking about the main incomer fuses

 

[Julie.]

Julie were would I find the let through value and are you talking about the main incomer fuses

Yes, so if they are say bs88 industrial fuses, anything below about 125A would be less than 10kA let through for 20kA perspective.

You need the manufacturer's data, there will be a curve showing this data, select 20kA on the prospective axis and follow it until it reaches the appropriate fuse line/curve then on the other axis you can read the actual let through current

 

[Julie.]

Here's an example if you select 20kA on the lhs and go across to the 100A fuse line, then follow it down - or up, it gives a let through of 9kA

 

[Farmelectrics]

ok I understand that thank you. so the breakers in the dbs that we have tested are all Schneider type, and there all 10ka breaking capacity and with a fault current of 20ka. so if I look into the let through value and that's above 10ka im happy to code 2 it. But if the client asks how we resolve it were do I go with it . thanks in advance





Like Quote Reply

Report

 

[Wilko]

But if the client asks how we resolve it were do I go with it

I can’t focus on the graph just now, but perhaps pick a fuse with less let through?

 

[Marvo]

At 0.02 ohms you're way outside the range of accuracy for any normal tester.

Make a 3 separate Zs tests at a socket at least 10 or 15 meters away from the point of supply. Note the average result.

Remeasure the Zs 3 times with a wander lead in series with the tester earth lead and take the average reading. Hopefully it's at least 0.5 ohms, if it's much less try find a socket outlet that's further away from the supply and use that one instead.

Take the first Zs reading from the second Zs reading to get an accurate measurement for the wanderlead.

Use the wanderlead in series with your tester earth probe and take a Ze measurement.

Take your test result and subtract the known wanderlead impedance. This should give you a far more accurate Ze result

Even if the new and hopefully more accurate result is 0.01 or 0.02 ohms higher than your original result it will have a profound effect on the fault current by calculation.

 

[Farmelectrics]

marvo thank you take your thoughts on board. might of been easy just putting max value on saves me a headache.

 

[Dartlec]

Ran into a very similar issue today in a nasty Barratt flat. Ze of 0.08, but L-N 0.03 according to my Megger, though it gave me a calculated pfc of 6.3k which suggests it was calculating to more digits than it was showing me.

Anyway, sub main to the flat 16mm T&E protected by a 63A MCB with 6kA rating. Also a main fuse somewhere, but not in the same building it appears (red fuselink).

Given that the source feed is an MEM enclosure with MEM breaker, and the flat board is LeGrand throughout, I'm assuming I can rely on the type testing for 16kA so no need to consider coding in my case.

As pointed out, the extra 0.01 can make quite a difference.

 

[pc1966]

Sadly MFT don't have the option for 4-lead Kelvin style measurements, and the ones that do cost something like x10 more so not an option for most folk

GN3 points out the impact that the odd 0.01 ohm error can make at high PFC value.

As Julie already mentioned, if there is the typical incomer BS88 fuse of 100A or so it is going to limit the PFC let-through to a value that will be safe. Remember, fuses are your friend!