IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Ze too high
Topic Summary: 200A sub
Created On: 10 January 2017 08:54 AM
Status: Read Only
Linear : Threading : Single : Branch
1 2 Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 10 January 2017 08:54 AM
User is offline View Users Profile Print this message



SKElectrical

Posts: 1124
Joined: 01 February 2009

I'm typing up an EICR I did a few months ago and just finding my mistakes!!
TNCS.
Ze is 0.2
Fused Isolator at cut out fused at 200A with max permissible Zs value of 0.14
So supply to Sub mains exceeds permitted value.
What is the correct way to proceed - not an upfront RCD?? The damn tails are 70mm. Besides I've said everything's okay now!!
 10 January 2017 09:03 AM
User is offline View Users Profile Print this message



daveparry1

Posts: 7749
Joined: 04 July 2007

Why do you think a Ze of 0.20 ohms is too high? 0.35 is the max recommended book value for TNCS.
 10 January 2017 09:05 AM
User is offline View Users Profile Print this message



SKElectrical

Posts: 1124
Joined: 01 February 2009

After the cut out, there is a 200A fused isolator serving a sub mains - this is where the Zs value is exceeded
 10 January 2017 10:02 AM
User is offline View Users Profile Print this message



mapj1

Posts: 9670
Joined: 22 July 2004

How much of the installation is relying solely on that fuse for its earth fault protection?

I imagine most small loads are on their own breakers or fuses, are they happy with the Zs at their point in circuit ?
If so its only the sub main/ tails or whatever you have that we need to consider.
If there was a L-E short, OK the fuse might miss its time target, but look at the makers fusing curve and decide by how much - that tells you how long an exposed voltage greater than 50 may appear on the metalwork of the installation - but also ask yourself is the fault condition you need to have to create that problem plausible in that set-up ? If it needs multiple stupid failures it probably isn't.



Its not as bad as an MCB where the difference between missing the impedance and meeting it is a near instant circuit breaking or not, more of a disconnection within heartbeat and a half instead of within a heartbeart...OK for most of us.

If once you ahve the figures, post back here, but if you think it is really seriously unacceptable, then a smaller fuse or a parellel CPC path are options, but in many cases may not be possible/practical.

you could have a 10 amp earth fault relay or something, but I'd check the fuse curve first, and also what breaking time you think really applies.
Simply saying please only use a fuse from maker XX whose curves are a bit slicker than the euronorm or BS worst case, may be easiest, if that is an option.

PS Whats the L-N loop like - I'd expect the same-ish on TNCS and at 200A, a resistance of 0.2 ohms would equate to 40V drop - is that really the case?

-------------------------
regards Mike
 10 January 2017 10:40 AM
User is offline View Users Profile Print this message



leckie

Posts: 4432
Joined: 21 November 2008

I think this is an option for you

415.2 Additional protection: Supplementary bonding.

415.2.1 and 415.2.2 - so you can provide supplementary bonding to satisfy these conditions so that R = or < than 50V/Ia

You cannot always meet the low Zs values quoted.
 10 January 2017 12:05 PM
User is offline View Users Profile Print this message



OMS

Posts: 22423
Joined: 23 March 2004

How accurate is your Ze value - and please check the limiting impedance for the fuse type - it's most likely to be around 0.2

So, if it is a BS 88 fuse, then you could push 1200A through it (240/0.2) and you will get disconnection in 5 seconds or a bit less (usually less, due to preheating of the fuse under load)

What is the demand profile like - can you put in a 160A fuse

Can you re-measure the Ze - as Mike suggests, at 200A there is going to be one hell of a voltage drop - so that might suggest that the Ze is a bit adrift from what you measured

Regards

OMS

-------------------------
Let the wind blow you, across a big floor.
 10 January 2017 12:06 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 16083
Joined: 13 August 2003

TNCS.
Ze is 0.2
Fused Isolator at cut out fused at 200A with max permissible Zs value of 0.14

Whose permissible? BS 7671 or the DNO's?

DNO values don't always give BS 7671 ADS compliance - just think of a 100A TN-S supply with Ze approaching 0.8 Ohms.

The old fashioned answer for that of course was in effect to have a class II installation up to the next tier of overcurrent protection (e.g. insulated & sheathed tails into an insulating CU in domestic settings). That might be a bit harder with larger supplies and metal-clad gear of course.

0.2 Ohms for a 200A TN-C-S supply does worry me a little though - just on a voltage drop point of view it implies a 40V v.d. when drawing 200A - which would put it outside of the statutory ESCQR limits (unless the DNO are doing something unusually clever with auto tap changers and there aren't any other consumers on the supply end of the LV line). So either the DNO have cocked up - or perhaps your meter isn't 100% accurate (cue usual comments about the specification of loop testers being so wide to make some results akin to fruit machines). Maybe pop back with a fresh set of batteries and see if it's any better?

- Andy.
 10 January 2017 03:56 PM
User is offline View Users Profile Print this message



leckie

Posts: 4432
Joined: 21 November 2008

So I assume that we are talking about a 200A switch fuse with BS88 gG fuses. BS7671 says 0.18 max Zs, so 80% is 0.144 ohms, so that where the 0.14 ohms figure comes from I would think.

As already said it probably is just inaccuracy in the measurement, I've got three EFLI testers that all pass calibration tests and all give different measurements once they are used to measure low values of impedance.

There are 200A MCCB's that have a max Zs of 0.12 ohms - so 80% is approx. 0.1 ohms which might be harder to achieve.

So if that were the case, cant my suggestion of supplementary bonding as additional protection be used to achieve compliance? We only need to keep R low enough between extraneous and exposed conductive parts enough then don't we, and that is often quite easy to achieve.
 10 January 2017 04:03 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 16083
Joined: 13 August 2003

Although strictly speaking the 80% figure need only apply to conductors inside the installation (see appendix 14) - the DNO lines should already be "warm" so no de-rating needed for that bit - hence raw BS 7671 figures should be OK for the origin.
- Andy.
 10 January 2017 04:19 PM
User is offline View Users Profile Print this message



SKElectrical

Posts: 1124
Joined: 01 February 2009

Originally posted by: mapj1
If so its only the sub main/ tails or whatever you have that we need to consider


you could have a 10 amp earth fault relay or something, but I'd check the fuse curve first, and also what breaking time you think really applies.

Whats the L-N loop like - I'd expect the same-ish on TNCS and at 200A, a resistance of 0.2 ohms would equate to 40V drop - is that really the case?


Ze was 0.19 I think. 3 metres of 70mm singles to 200A BS88 gG (max Zs 0.18 , and 0.14 @ 80%) fused isolator. 10m x SWA 4 core 50mm with 2 x 16mm separate earths to TP DB pulling about 35A / phase *currently, so for time being 100A main fuses is the simplest option.
Yes only the cleated SWA supply to DB is effected.
The earth fault relay is a clever idea but over engineered for this installation.
I don't follow you on the volt drop side of things. It's something I tend to ignore. Surely vd is only relevant from isolator to DB Sub??

Low battery on Megger is a possibility for erroneous results but two meggers used on this job.
 10 January 2017 04:21 PM
User is offline View Users Profile Print this message



leckie

Posts: 4432
Joined: 21 November 2008

But isn't SK saying the 200A switch fuse is on a set of consumers meter tails to the cut-out, so the measurement is Zs not Ze? I'll have another read of appendix 14 - I seem to recall it gives an alternative method of calculation.

Anyway, even at 100%, if you had an MCCB with a max Zs of 0.12 ohms for 5s for example, and this could not be achieved I think supplementary bonding achieves compliance unless I am misunderstanding this. So a few bit of green/yellow and everything is OK?
 10 January 2017 04:52 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 8792
Joined: 23 April 2005

SK

What were the model numbers of the 2No. Meggers you were using.

Time to break out JPs"passing" meter I think!

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 10 January 2017 05:00 PM
User is offline View Users Profile Print this message



SKElectrical

Posts: 1124
Joined: 01 February 2009

All singing MFT 1730 v2 - they even have Bluetooth although I have no idea how this could possibly help.

Calibrated mid Sept 2016
 10 January 2017 05:37 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9670
Joined: 22 July 2004

I don't follow you on the volt drop side of things. It's something I tend to ignore. Surely vd is only relevant from isolator to DB Sub??


OK - back up on the voltage drop. In TNCS, the loop impednace is either set by something indoors you can parallel strap an earth to, or if the ohms are on the DNO side, then as its a PEN, the L_N loop impednace and the L_E loop impedance at the origin are the same thing.
Now between you me and the meter, a few feet of 70mmsq is not going to alter that much, so I reckon that if we trust the readings, unless something is cooking, then most is outside.
But if there really are nearly 200 millioms distributed outside the building, round the DNO transformer windings a few times and then back again then if ever you were to try and draw the 200A load it seems to be designed for, then on any one phase, some 40 volts will be lost.
You may not mind that, but it corresponds to the lights dimming quite markedly, and the fuse being not much cop for L_N faults either - i.e. if it is true, then its not really an adequate supply for 200A loads anyway, so the fuse can be lightened up. It also means the voltage drop allowance for the rest of the installation is negative, as the supply has taken it all.
Be aware however that measuring low impedances with battery powered test meters especially 'no-trip' ones that use very small test currents can have a large variation - after all the meter is tying to add and remove a few mA of extra load, and look at how much the voltage droops in response.
- If as a noddy example the test current were 10mA and the loop impedance was 200milliohms, then we are looking for the nominal 230V twitching by 2000 microvolts.
hat is the sort of measurement I can easily make in the comfort of the lab here, but on a supply shared with all sorts of loads that click on and off, it is easy for the meter to be fooled. A good meter does in effect make the test many times in rapid succession and average over a carefully designed pseudorandom waveform to avoid the effect of big clicks mid test, but even so, to believe the last few digits is asking a lot.
Sometimes "Fruit machine readings" indeed.

Traditional meters with a higher test current fare rather better.
I have been known to do it by hand, and switch a known load of many tens of amps, and measure the change in volt drop - its a far more realistic test.

-------------------------
regards Mike
 10 January 2017 06:59 PM
User is offline View Users Profile Print this message



SKElectrical

Posts: 1124
Joined: 01 February 2009

Yes the earthing supply is unusual...
Seemingly half TNS / half TNCS.
(We must have had xray eyesight that day...)
90mm Ali straight concentric into Large DNO Fused box. The metal enclosure was bonded to the Neutral.
*I think the Neutral concentric conductor was Cu.
A separate earth 25mm also enters building with Supply cable and bonded to the metal enclosure.
In addition another Cut out TNCS adjacent 25LN / 16mm earth.
The supplies are bonded in 25mm.
 10 January 2017 07:20 PM
User is offline View Users Profile Print this message



OMS

Posts: 22423
Joined: 23 March 2004

Originally posted by: SKElectrical

Yes the earthing supply is unusual...

Seemingly half TNS / half TNCS.

Not likely in my experience

(We must have had xray eyesight that day...)

Sure, don't all electricians have it

90mm Ali straight concentric into Large DNO Fused box. The metal enclosure was bonded to the Neutral.

*I think the Neutral concentric conductor was Cu.

OK - typical CNE supply

A separate earth 25mm also enters building with Supply cable and bonded to the metal enclosure.

It exits to an electrode somewhere along the service cable run (probably just outside the door) - this is required usually because of the ESQCR requirements - your building has a particular disposition on the distributor out on the road

In addition another Cut out TNCS adjacent 25LN / 16mm earth.

OK - perfectly normal - the buildinh has experienced an upgrade at some point


The supplies are bonded in 25mm.

Sounds reasonable - as the supply neutral doesn't exceed 95mm, then 25mm is fine



Given the current load, I'd downfuse the 200A fuses to 160A or even 100A

Regards

OMS

-------------------------
Let the wind blow you, across a big floor.
 10 January 2017 07:59 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 8792
Joined: 23 April 2005

SK

I don't rate Multi function testers for industrial work. The best MFT. I have found to be the Kewtech KT65DL which I trialed on a large installation.

As stated above by Mike the higher the test current the higher the accuracy. Most loop testers on the so called "high current" range limit the test current to 6A or less to avoid tripping 6A circuit breakers. The lower limit of the calibration range for most loop testers is 0.2 ohms, below that you are operating outside the calibration range. Loop testers may have a resolution of 0.01 ohms but resolution is not accuracy. Remember 2 places of decimals means you are looking to measure 1/100s of an ohm which is a big ask for a meter costing less than £1K. Some meters will indicate 3 places of decimals but that 3rd figure should be treated with great suspicion.

If you want really good and much more accurate loop tester Megger do one for circa £5000.

For low impedance supplies my Megger LT 300 high current loop tester out performs any other meter.It has a test current of 12A. Whatever test kit you use bin the leads that come with the meter and get a set of Kyritsu leads and probes and crocs.from Kewtech. They are expensive at around £40 a set but outperform others and last for ages.

The best all round loop tester I think is the Megger LTW range. I told my friends from Megger this at an Elex show and got ear to ear smiles all round and then rapidly brought them back to reality when I told them their leads were cr&p.

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 10 January 2017 08:04 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 8792
Joined: 23 April 2005

SK

The supply you have sounds like a Wavecon cable so the supply is TN-C-S. The DNOs often bring in an addditional earthing conductor in parallel with the Wavecon but the supply is still TN-C-S as the neutral is combined with the Earth on the supply side.

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 10 January 2017 08:10 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 8792
Joined: 23 April 2005

PS

I wish to claim copy right to the expression "fruit machine readings".

Also " door step competency" in relation to Part P which I am proud to say has been used by MPs.

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 10 January 2017 10:07 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9670
Joined: 22 July 2004

how far back for your precedence JP?

some nutcase using the fruit machine phrase on here in 2008
Think he brought the expression with him from his place of work.
Along with other metalanguage 'feezling' 'Ground bounce' etc.

-------------------------
regards Mike
IET » Wiring and the regulations » Ze too high

1 2 Next Last unread
Topic Tools Topic Tools
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..