Bonding of Extraneous pergola frame

[Deleted member 9648]

Would you bond it or not. Getting mixed opinions on it.

No.
If you connect this metal frame to the earthing/bonding arrangements of the entire installation a fault anywhere on the installation may introduce a dangerous potential onto all metal work connected to the system,including the pergola. Within a building this is not an issue because a person is effectively insulated from the general mass of earth by the normally electrically insulating floor.The purpose of bonding is to prevent a PD between conductive and extraneous conductive parts within the building.
Externally a PD on a metal frame would be very dangerous to a person in contact with the frame and the ground,hence IMO bonding the frame increases the shock risk considerably. Neither is it considered a conductive part as it is not part of the electrical installation so earthing it should not be required. It is up to the installer to make sure cables and accessories are installed in a manner which prevents any possibility of a fault to the frame.

 

[SS-2015]

I had a similar thing when having to instal some 110v sockets on some steel work benches. (Bolted to floor)
What decided it for me was the fact that when I tested between known earth and benches I got a reading of 0.9M ohm, if they had been totally insulated (i.e meter maxed out) then I wouldn't have bothered but in this instance I decided that if a fault did occur I would want the fuses to trip asap so decided to bond.

 

[Dave OCD]

Main bonding should be sized according to the requirements of the incoming service, not reduced at any point in the installation

Yes I realise that but took the view that as electrical equipment was being attached to a metal structure then the structure should be earthed, 6mm would be more than adequate in such circumstances as the maximum earth fault current would be limited and of very short duration as the aforementioned equipment was RCD protected.

 

[rattlehead85]

It's clearly evident from the responses to this thread that there is a clear difference of opinion wether the structure should be bonded or not.

 

[Richard Burns]

I had a similar thing when having to instal some 110v sockets on some steel work benches. (Bolted to floor)
What decided it for me was the fact that when I tested between known earth and benches I got a reading of 0.9M ohm, if they had been totally insulated (i.e meter maxed out) then I wouldn't have bothered but in this instance I decided that if a fault did occur I would want the fuses to trip asap so decided to bond.

This is an interesting response.
When you measured the resistance to earth you got a value that would limit any current attempting to flow through the bench to well below the perceptible limit of 5mA, so this would indicate that the bench was not an extraneous part even by the most rigorous standard.
You also state that you would want the fuses to blow as fast as possible and this can only be achieved by reducing the resistance of the earth fault loop path so as to maximise the fault current.
However you decided to bond the bench because your measured resistance was too low and to cause the fuses to blow even though bonding would have no significant effect on the fault current.
Please help me to clarify the reasons for your decisions.

 

[SS-2015]

This is an interesting response.
When you measured the resistance to earth you got a value that would limit any current attempting to flow through the bench to well below the perceptible limit of 5mA, so this would indicate that the bench was not an extraneous part even by the most rigorous standard.
You also state that you would want the fuses to blow as fast as possible and this can only be achieved by reducing the resistance of the earth fault loop path so as to maximise the fault current.
However you decided to bond the bench because your measured resistance was too low and to cause the fuses to blow even though bonding would have no significant effect on the fault current.
Please help me to clarify the reasons for your decisions.

Firstly I'll apologise - should have said 0.09M.

But just interested in you comment that bonding doesn't have any significant effects on fault current?

 

[Richard Burns]

I was just starting to draw a diagram to explain when I found this one I prepared earlier and modified slightly to fit your situation.
Hopefully this demonstrates the limited current that flows though bonding conductors in the case of a fault, and the minimal change bonding would make to the EFLP.

 

[SS-2015]

I get this and would almost certainly apply in the OP's situation but in my situation, this was a workshop with no Gas and water present, therefor there would be no parallel paths, it would be relying purely on main earth and subsequent CPC's and due to benches being bolted to floor and not 100% insulated from ground I felt it necessary to bond.

 

[rattlehead85]

I get this and would almost certainly apply in the OP's situation but in my situation, this was a workshop with no Gas and water present, therefor there would be no parallel paths, it would be relying purely on main earth and subsequent CPC's and due to benches being bolted to floor and not 100% insulated from ground I felt it necessary to bond.

The difference between my situation and this one mainly being that my Pergola is outside bolted to the ground whilst yours is internal where the floor is most probably insulated from the general mass of earth.

 

[SS-2015]

The difference between my situation and this one mainly being that my Pergola is outside bolted to the ground whilst yours is internal where the floor is most probably insulated from the general mass of earth.

Normally you'd be right, but as this was reinforced concrete industrial floor, I'm guessing that the bolts holding bench are touching rebar somewhere....

anyway wasn't my intention to have your thread hijacked.