Max Zs to BS7671 or Manufacturers Spec

[ktech]

I've had differing opinions on this.

Say a final circuit Zs falls outside that required by BS7671 but taking the manufacturers data it is within, ie. the manufacturers device exceeds the requirements of BS7671 for disconnection and therefore allows a greater Zs.

An example of this is would be a 100A 60898 Type C with BS7671 max Zs of 0.23. However max Zs of a Moeller PLHT Type C 100A 60898 would be 0.51, taken from manufacturers data in Amtech.

I have come across this on periodics when the circuit would look not to comply on BS data but taking into account the manufacturers specs it is ok.

I asked NICEIC about this and they said it does not comply with BS7671 and is therefore a code 2 - requires improvement. But it is technically safe so what is the problem? Does the regs state anywhere that you can use manufacturers data when available?

 

[Spudnik]

Must admit this is a very grey area with regards to manufacturers specs and using them.

Your Zs, regardless of specs, should be within the max stated in BS7671 for the type of supply.

If this cannot be achieved then a 30mA RCD may be used, although the suspect circuit should be investigated.

Hope that helps!

 

[ktech]

I agree it is a grey area as i can't find anywhere in the regs that suggests more accurate manufacturer data can be used. NICEIC say it has to be to 7671 values but such as Amtech say it's common place to use manufacturers data when available.

Common sense says to me that manufacturers products may quite often exceed the requirements of BS EN 60898 and therefore provide a greater tollerance on Zs whilst providing disconnection in the required time as per 7671.

It's a real pain not been able to get a consistant answer as half of the site i'm looking at would be classed as suspect and require additional RCD protection.

 

[Guest123]

Hi Ktech.

Tough call this one, but I think I would have to go with BS 7671.

At the end of the day that is what we all have to work to and should the worst happen then that is what will be used against us.

I'm not dis-respecting moeller in any way and if they say their device will operate at that value then who are we to argue, but surely the bods that write BS 7671 would take into account data from all manufacturers before publishing the values.

Personally I would be a lot happier going by the BS values.

As a remedy, do the breakers absolutely need to be 100A, what are they supplying?? how much out is your Zs?? Could a larger CPC be installed. Not sure how easy it would be to RCD protect these circuits.

 

[Mr Mark Sparks]

on the inside cover of an niceic minor works pad it says

QUOTE
the impedance values are based on worst case limits allowed by 7671 and in certain cases where the manufacturer of the protective device claims closer limits necessary for operation of the device than allowed for by the standard the values maybe modified accordingly

QUOTE

where measured eli exeeds relevant tabulated value further investigation will be needed to confirm compliance with 7671 has been achived

my head hurts

 

[Guest123]

on the inside cover of an niceic minor works pad it says

QUOTE
the impedance values are based on worst case limits allowed by 7671

I've always taken this to mean that these values are the absolute maximum allowed full stop, and anything over and above should be investigated.

This is only my take on it though.

Cheers

 

[ktech]

markspark666:
My head hurts as well. I think you could look at these two quotes and say they contradict each other, OR that if you measured the eli and it was outside the tables then further investigation would be required. Further investigation and in this case looking at manufacturers data would make you happy it complies. Although I assume on a periodic you would state the tabulated values in the max column but make some note of manufacturers data been looked up.

lennytheloon:
The 100A relates to a planned new circuit, but let me give you an example that already exists and hopefully this demonstrates the issue:

The case is as follows:
Ze at origin (TN-C-S) 0.17.
Distribution circuit to DB1 in 25mm 5 core using integrated cpc and armour. Protection at origin is via 80A Type C (Moeller PLHT).
Zs DB1 feed, or Ze at DB1 is 0.30.

Board powers two appliances required to be connected on 63A type C. Applying diversity a 80A breaker was used at the origin. We look to have some issues with discrepancy with this but that's another story.

BS tables state max Zs of 0.22 (80% value).
So the eli is out of tollerance according to the regs. However using Moeller data from Amtech software the max Zs is 0.51 (80% value) so well within tollerance for the distribution circuit.

 

[Spudnik]

So where would we stand if this was a TT system with an EFLI of, say, 20 ohms?

This would be way over the manufacturers max of .51.

Or am i missing something??

 

[ktech]

My understanding would be that TT should have a Ze of less than 200 ohms but more than 100 ohms requires further investigation. In addition TT requires additional RCD protection and with a maximum touch voltage of 50v the maximum Zs would become 166 ohms. e.g. 50v / .3ma = 166, for 100ma rcd 50v / 0.1ma = max Zs of 500 ohms?

 

[Spudnik]

Indeed but the RCBO or MCB is, as you said, rated at .51 so this would be way over?

 

[ktech]

The Zs would still be way over but because you have the RCD this is not an issue. One way of negating the problem of high Zs could often be to install an RCD, as generally the case in a TT system.

I've quite often seen 1667 been written in the 'max Zs allowable by BS7671' written on periodics for circuits protected by an RCD. 30ma rcd works out as 50v / 0.03 = 16667 ohms. I will try find this part in the regs.

Interesting point about TT systems though. The installation I am talking about has around 25 boards with a few of these fed from a standby generator. Checking with the NICEIC they advised that two periodics would need to be completed, one when connected to the grid and one on the generator. Makes sense as the earthing arrangements are different. The generator is TT. So in the end these dual fed boards will probably need an RCD on the incommer to compensate for the higher Ze when on standby power and limiting touch voltage. Interestingly on initial investigation the generator does not have a earth electrode and is earthed to the building fabric so effectively connected to the grid earth.

 

[Spudnik]

Ah yes of course im with you now i think.

Having a middle age moment there

 

[johnboy6083]

Sorry for resurrecting an old thread ( I saw it at the bottom of the page and I piqued my interest), but I disagree with the opinions above.

Somewhere in BS7671, im sure it states that manufacturers data can be used if available, with BS7671 tables as a worst case scenario. Personally unless told otherwise, I'll design to the nominal values as described in 7671. In the case of a periodic, then if the Zs values are close to/ above 7671 limits, then I'd go with manufacturers data.
I recently did an EICR on a 3ph DB full of 3871 type 4s. I couldn't find a table in 7671, so I looked at the 60898 standard, which mentioned that the tripping value of type 4's are 10-50 x In. So on a 30A type 4 the max Zs would be 230/(50x30)= 0.153 ohms.
Looking at manufacturers data meant that the max Zs was actually 0.6ohms.