Bonding question

[Evans Electric]

Existing workshop with no bonding .    

Supply is an SWA     possibly 50mm  sq   or 70mm  X 4 core    TPN   to a 200A  Sw/fuse     I can only guess the size by looking .  

According to Table 54.8  the bonding  needs to be  25mm    .............BUT  ..........   the origin of the supply is a  BBC  40 mtrs  away  with the metering and  3 X 100A   service fuses  with 25mm tails . 

My question is ,  should I size the bonding to the 25mm tails  or the  50 / 70mm   SWA  sub - main  which is for the volt drop I imagine.    The 200A Sw/Fuse  must be  for the physical size   :C Can't switch off to look at the fuses , not that it matters much as the service is 100A .  

 

[binky]

any links between buildings other than the cable?

 

[Evans Electric]

No mate ,  two totally different buildings 

 

[Ardet R]

For PME supplies the bonding size is based on the incoming supply cable to the property; with a 100A service fuse the incoming cable is likely to be 35mm² (just a guess) and so the bonding, even at the remote building, would only need to be 10mm², assuming you have a PME supply.  The size of the distribution circuit should not be relevant to the size of the bonding.

Use the distribution cable as bonding back to origin and the switch fuse as your earth marshalling terminal for the bonding connections at the remote building.

 

[Evans Electric]

For PME supplies the bonding size is based on the incoming supply cable to the property; with a 100A service fuse the incoming cable is likely to be 35mm² (just a guess) and so the bonding, even at the remote building, would only need to be 10mm², assuming you have a PME supply.  The size of the distribution circuit should not be relevant to the size of the bonding.

Use the distribution cable as bonding back to origin and the switch fuse as your earth marshalling terminal for the bonding connections at the remote building.

The supply is probably 70mm   I can't switch anything off to look at the cores  , can only go by the outer sheath .      Theres a cell phone transmitter fed from this workshop  ...I'm assured it WILL NOT  be turned off .  

Can't use the switch fuse , it will have to be on  the side of the  Dist. Bd .    

 

[Ardet R]

But you seem to say that the distribution cable to the workshop is 50-70mm², this is not the cable size you are meant to be referring to to get the size of the bonding.

The bonding should be sized by the incoming cable from the DNO (like that is ever available to test!) for a 100A supply fuse the DNO would usually have a 35 or 25mm² supply cable though if this is an industrial site then the supply cable may be sized for a more substantial supply.

If the incoming DNO cable is also 70mm² then the bonding would need to be 25mm², but your distribution cable (SWA using the armour for earthing) would then not meet the size required for bonding and you would need a 25mm² copper conductor back to the origin.

If you have your EMT after at the dist board then the switch fuse to dist board earthing/bonding cable also needs to be of sufficient csa. The cable from the switch fuse to the dist board in the workshop is less than the bonding size required then you would have to get to the switch fuse (or go back to the supplying building!)

It would be much easier if the supply were TNS!

 

[Sidney]

Theres a cell phone transmitter fed from this workshop  ...I'm assured it WILL NOT  be turned off .  

Depending on the nature of the cell site and its location in relation to the building you may well find there's also a requirement for SPDs, there's a very high chance of lightning protection on the site.

On a side note, turning it off isn't a problem, it's the restart when the power comes back on that causes issues. All technologies will try to come on air at the same time which causes a large current surge, combine that with anything in the workshop then OCPDs would start to operate.

As for your bonding conductor size, Ardet R has pretty much summed it up.

 

[Sidewinder]

It's interesting that it is being suggested that the Steel Wire Armour of the Steel Wire Armoured (SWA) is not suitable as a bonding conductor, yet it is also being suggested that the bonding only needs to be 10mm2.

It is correct to state that the bonding for a TN-C-S/PME supply must be sized in accordance with the incoming supply cable size.

So as the bonding is required to be sized according to the incoming DNO cable, then if this is 25 or 35mm2, then according to BS 7671, table 54.8, we only need a bonding conductor of 10mm2.

If the DNO head is only 100A, then it is unlikely that it is greater than a 35mm2 incoming cable is it not.

If the sub-main cable is 50 or 70mm2 then the cpc only needs to be half sized to the live conductors in that circuit to be compliant with BS 7671.

So that would mean a circuit protective conductor (cpc) of 25 or 35mm2., depending on the cable live conductor size.

Again, by table 54.8, a 25mm2 bonding conductor is suitable for an incoming supply of up to 95mm2, a 35mm2 bonding conductor is suitable up  to a 150mm2 incoming supply conductor cross sectional area (csa).  Both of which are a lot larger than the likely incoming supply cable size.

The actual incoming supply live conductor size can be obtained from the Distribution Network Operator (DNO), by enquiry, must be provided free of charge and should not require a site visit.  For compliance with ESQCR they must have this information in their records and provide it to those who have reasonable grounds to require it.

According to the IET information in GN8 based on either BS 5467 or BS 6724 cables we have the following:

To meet the k1/k2 x S/2 constraints, we can simplify this equation by calculating the S/2 first, which we know is 25 or 35mm2 to be suitable as a cpc for a 50 or 70mm2 live conductor.

Thus this becomes k1/k2 x 25, or k1/k2 x 35 mm2.

GN8 table B2 gives the result of this calculation for 50 & 70mm2 as 56.37 or 78.92 mm2 of steel being required to be the steel equivalent of copper of 25 or 35mm2 respectively.

Then B2 goes on to give us the steel csa of the armour for the relevant cable sizes/core count.

50mm2  2 core - 68, 3 core - 78, 4 core - 90 mm2

70mm2  2 core - 80, 3 core - 90, 4 core - 130 mm2

From this it can be seen that the steel csa is greater than that required to fulfill the function of a cpc for these sized live conductors.

Note though, that this calculation as described here, limits the cable operating temperature to 70 Deg C.  With a 100A shared incoming supply and a 50 or 70 mm2 sub-main, it is doubtful that the cable would operate above 70 Deg C, and it is also probable that the devices at the terminations of the cable cannot be allowed to exceed 70 Deg C anyway, therefore limiting the cable to 70 Deg C, thus making this a valid calculation.

Ergo, the area of the swa is greater than that required to give the equivalent of 25 or 35mm2 of copper.

Considering that BS 7671 only requires consideration of a single fault.

Where is it a non-compliance with BS7671 to use the armour of the SWA cable to be a bonding conductor for a TN-C-S/PME supply of a maximum of 35 mm2 incoming conductor size, which only requires a bonding conductor of 10mm2 when the copper equivalent of the steel armouring is greater than 25 or 35 mm2?

If a conductor is to be run in parallel with the SWA then it must be a full sized cpc for the circuit, you cannot run a bonding conductor of 10mm2 in parallel with a cable with 50 or 70mm2 live conductors, unless you can prove by calculation that it is capable of carrying the full potential fault current under all fault scenarios by calculation, then by measurement of installed values.

Please note that this is based on the statements made in the earlier posts of this thread only.

Please discuss?...

 

[apprentice87]

Theres a cell phone transmitter fed from this workshop  ...I'm assured it WILL NOT  be turned off

Have to laugh.. If it needs to be turned off, i would just phone the HSE and turn it off in front of them. Lets see who is going to turn up and argue then..

Or you could do like this chap..





john..

 

[Phoenix]

Sounds good to me, not sure if I missed it or you skimmed over it, but following:

So as the bonding is required to be sized according to the incoming DNO cable, then if this is 25 or 35mm2, then according to BS 7671, table 54.8, we only need a bonding conductor of 10mm2.


 


If the DNO head is only 100A, then it is unlikely that it is greater than a 35mm2 incoming cable is it not.

Perhaps need to point out that that 10mm of requried copper translates to 10x8.9 = 89mm² of steel for equivelent conductance, which shouldn't be a problem if the cable is 50/70 mm (This is a different calaculation than the conversion of K1/K2 which is relevant for CPC sizing). It does however start to bite if the nature of your supply requires a 35mm or 50mm bonding conductor though, it ends up being quite a lot of steel csa required...

 

[Ardet R]

I think sidewinder has confirmed my posts and added the calculations.

The SWA armour to function as a cpc is fine for all cases in the above described situation.

If the bonding needed to be 25mm² then the armour of the SWA would not be sufficient csa and a separate conductor would be needed.  

As we all say if the incoming supply cable is even 35mm² then only 10mm² bonding is required and the armour of the SWA (for 50 or 70mm² 4 core cable) is sufficient to provide this service.

Phoenix has correctly identified the variation in calculations for bonding vs cpc  when using a non copper conductor.

The cpc size can be calculated using the adiabatic equation and would allow a 10mm² cpc but I might be hesitant on using such a size as an independent conductor, but for no readily apparent reason, it just does not sound right!

Question answered, I think!

 

[Evans Electric]

This is getting complicated .  

Surely I don't size the bonding to the incoming supply , which is ,  I think , 200A   X TPN  .

This would mean the small units on the site  , some with a 30A    SP  supply  would  need  70mm bonding .

This supplies :-

7 small units 

1 car sales

2  repair garages 

1  car body repair

1  valet unit 

1  spray booth

1  Heavy plant hire     ( my bonding job ) 

Plant hire is 40 mtrs  away from the incoming supply ,  with a   buried  SWA  .   Theres no way  a bonding cable will be going back there TBH .  

The service to Plant Hire is fused at  100A  X TPN  ...  in my humble opinion ,  and I stand  corrected  ,  I ignore the size of the sub main    ( If the meter room was next door it would be a 25mm methinks)

The transmitter is not my concern ,   they pay big bucks I believe to have it there  and no one messes with it ,  including me .   

I see my obligation  as  bonding the gas , water & steelwork  back to the incoming feed of  the remote building  itself. 

However some very interesting points have been raised.

Sounds good to me, not sure if I missed it or you skimmed over it, but following:

Perhaps need to point out that that 10mm of requried copper translates to 10x8.9 = 89mm² of steel for equivelent conductance, which shouldn't be a problem if the cable is 50/70 mm (This is a different calaculation than the conversion of K1/K2 which is relevant for CPC sizing). It does however start to bite if the nature of your supply requires a 35mm or 50mm bonding conductor though, it ends up being quite a lot of steel csa required...

When I first looked at it I thought ...35mm  green/yell  ....  based on my guess at the size of the incomer .        Some time later , having access to meter cupboard I note the supply of the sub-main is fused at 100A   

 

[Ardet R]

This is getting complicated .  

Surely I don't size the bonding to the incoming supply , which is ,  I think , 200A   X TPN  .

This would mean the small units on the site  , some with a 30A    SP  supply  would  need  70mm bonding .

The joys of bonding, you have got it in one.  The incoming supply for the whole installation is the defining factor as the bonding is all connected back to the MET at origin.

Unless the earthing system is divorced from the incoming supply, by making the building a TT system, the main bonding is the same size for the whole installation.

I attach a  diagram from GN8 showing 35mm² bonding on a remote building (B3) where the sub main to the building is only 25mm² but the incoming supply conductor is 120mm².



With the incomer as a 200A supply (possibly 70/95mm²) the main bonding csa may well exceed the csa of the armour of the sub main to the workshop.

 

[Phoenix]

Deke, You say 200A supply, but in the first post you say 100A w/ 25mm tails.... I'm confused :blink:

And building on what Ardet has said, consider an industrial building which is no longer used for what it once was (and such does not have the demands it once did), says its used for warehousing after being dormant for a few years with the meter removed. 100A TP is all thats requried now, so CT chamber comes off, 100A whole current meter goes in with 25mm tails, 100A fuses fitted in the J type carriers in the existing cable head, but 54.8 references the supply side neutral conductor, not customers tails, so the bonding might have to be bigger than one might think at first glance....

 

[Evans Electric]

I know what you mean ,  and strangely I've never considered this particular scenario before .

I'm just of , say ,  hospital complexes I've worked on .... there would be  possibly a 500A   PME  incomer , then loads of separate buildings with various sized sub-mains  to them but ,   as I remember them , the bonding matched the sub-main  supply characteristics  ,   not whatever a 500A would be .  

Phoenix  just to recap ,     the site supply is 200A    PME.....  a BBC  with a load of sw/fuses on it ....... the one to this Heavy Plant  workshop  is via 100A   Henley fuses , a sw/fuse  and  40m    of 50 / 70mm  SWA imperial  from the late 1960's .  

I have to say I've never come across  bonding cables going back to a remote  switch room .   

 

[Sidewinder]

Sounds good to me, not sure if I missed it or you skimmed over it, but following:

Perhaps need to point out that that 10mm of requried copper translates to 10x8.9 = 89mm² of steel for equivelent conductance, which shouldn't be a problem if the cable is 50/70 mm (This is a different calaculation than the conversion of K1/K2 which is relevant for CPC sizing). It does however start to bite if the nature of your supply requires a 35mm or 50mm bonding conductor though, it ends up being quite a lot of steel csa required...

That is the calculation I couldn't find in GN8 yesterday, and had to go and do some work.

Well done Phoenix, thank you, I couldn't think of the word to look for.

It's in GN8 5.2.1, as equation 5.5.

So yes, 

where:

Sm is the minimum csa required for the protective bonding conductor (in a metal other than copper).

Sc is the minimum csa required for a copper protective bonding conductor.

ρm is the resistivity of the metal from which the main bonding conductor is made.

ρc is the resistivity of copper.

For a steel main protective bonding conductor, the ratio ρm/ρc is 8.5. Similarly, for an aluminium conductor, the ratio is 1.68. So for conductors made of these metals, the csa would be required to be 8.5 and 1.68 times that of copper, respectively, in order to afford an equivalent conductance.

So this simplifies when relating steel to copper to  Sm8.5Sc which in this case comes to Sm 85mm2

So 4 core 50mm2 & 3 & 4 core 70 mm2 would meet the requirements as a bonding conductor at that size, I suspect that the cable is 4C?

This may be one of the reasons it has been chosen?

 

[Evans Electric]

This may be one of the reasons it has been chosen?

Won't be that  Sidey ...this was installed before bonding was invented .  

 

[Andy™]

The supply is probably 70mm   I can't switch anything off to look at the cores  , can only go by the outer sheath .      Theres a cell phone transmitter fed from this workshop  ...I'm assured it WILL NOT  be turned off .  

Can't use the switch fuse , it will have to be on  the side of the  Dist. Bd .    

just remind them if its so critical that it has power at all times, then it should have its own UPS/generator to keep it going. in which case, there is no problem turning it off and there is no guarantee that DNO supply will never fail either

 

[Evans Electric]

No Andy ,  I learned many years ago to keep my nose out of such things .  I'll just install the bonding ....  cash the cheque ......and move on . 

 

[Ardet R]

Effectively the only time that the bonding not being up to size between different buildings would be a significant problem is in the case of an open circuit PEN conductor outside the installation that would then mean the bonding would be carrying load current and, if a remote building had very good earthing from the extraneous conductive parts but other buildings didn't, the load current from the whole installation would go that way and could cause the bonding conductors to overheat.

In the case of a general fault to earth the MET would rise above 0V but the smaller earthing and bonding conductor from the remote building would actually ensure that the MET would rise in potential less than if it were full sized.

However do not tell the IET I said this!