Online assistance for electrical trade people in New Zealand and Australia Login  |  Register  |   Forgot Password
Assistance for electrical trade people




Click here to send Forum Admin a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: Bonding

Jul 23 2019 18:38

We have some 4m stainless steel workbenches installed in some Chemistry labs. On these will be various bits of lab equipment (centrifuges, heating mantles, stirrers, vacuum systems etc etc) There are gas and water outlets between the benches mounted in wood

Do these benches have to be bonded ?

Jul 23 2019 21:49

5.6.2 Are the benches in contact with the earth, independent of the installation earthing system? I doubt it.

Jul 25 2019 09:43

connecting the benches to earth could create a "substantially conductive situation", triggering application of ESR 89.

Even without that, earthing them may reduce electrical safety because it increases the chance that someone is touching something earthed when an incident occurs.

Jul 25 2019 12:43

My comment would be , that you have asked if they should be bonded, ( thats not earthing )
I would suggest that if the benches ( you have indicated there is more than one ) are within arms reach of each other then I think I would be bonding them together . NOT earthing , for the reasons that Alec explained.
Although the risk may be small it would be possible if there was a fault with something on one bench and and earthed piece of equipment on another then someone could get a shock if they touched both benches at the same time. If you bond them together this can't happen.


Jul 25 2019 13:34

If you want to guard against that sort of thing, an RCD does it better.

Jul 25 2019 16:00

The benches are fixed. They are on metal fames, and these are solidly bolted to a reenforced concrete floor.

One bench that has no earthing has just a thin coating of silicon sealant between that and an earthed sink bench that is earthed.

My view of the risk includes water, glass shards, chemicals, burnt cables are common as they contact heated surfaces. People have to lean over the benches to turn off gas, water,power, so my concern is that a 4m long stainless bench could become live.

Fitting RCDs is not currently an option.

Jul 25 2019 18:51

Whether they have to be bonded and whether they should be bonded are two very different questions.

What are your thoughts on how the wiring rules apply in this situation?

Jul 25 2019 20:09

If it’s above earth potential keep it that way. Eg PVC water/ gas pipes then no bond. Metal pipes touching/ connect to the bench then bond it.

Jul 26 2019 08:43

That's the complete opposite of what equipotential bonding is about.

The point of bonding is to avoid having items at different potentials. Different potentials can arise because the earthing system of the installation is tied to the distribution PEN conductor (for an MEN system, ie to a remote bit of ground at the source and also to a nearer bit of ground at the installation (via the MEC & electrode.
Where structures or services are conductive, and are in electrical contact with other bits of ground, they can be at different potential to the earthing system.
Bonding connects all such parts together, thus keeping them at same potential.
If something isn't in electrical contact with any ground; it can't be at different potential and bonding will serves no purpose.

If the benches are in contact with metallic services that are in contact with ground, the answer is NOT to bond the benches but to bond the service - as close as practicable to the point of ground contact.

And since the OP's description says these services are between the benches, mounted in wood; it's clear that connecting the benches to the earthing system can't help if the services are at different potential.

To answer the OP's question: NO the benches do not have to be bonded. And if the services need to be bonded, that should already have been done.

What he and others are trying to deal with is absolutely nothing to do with bonding. The issues raised relate to dealing with faults arising from damage to equipment being used in the labs. Specifically, damage to cords leading to possibility of live parts coming into contact with the benches and so livening them. The question then becomes - what else is simultaneously accessible that would be at (for example) earth potential; and how likely is it that someone would be touching both ?

Earthing the benches cannot be an effective answer, because you cannot be sure that the damaged part will actually contact the bench. The risk of someone touching a damaged item directly stays exactly the same, but by earthing the benches you pretty much guarantee that they will get a shock. Earthing them will create at least as many issues as it may solve.

RCDs remain the best answer, in that regardless of the actual fault current path operation of the RCD should enshure the shock is non-lethal. Use of RCDs is always an option; whether switchboard-mounted, SORCDs, or plug-in types.

And if burnt / damaged cords are "common", then a revision of work practices is in order, plus a system of routine safety checking.


Jul 26 2019 17:33

Why are RCDs "not an option"?
Seems at odds with the safety trends over the last 30+ years.

Jul 29 2019 16:06


Thanks for you feedback. A question I have for you in regards to a statement you made.

“If the benches are in contact with metallic services that are in contact with ground, the answer is NOT to bond the benches but to bond the service - as close as practicable to the point of ground contact.”

So are you saying that the bench needs to be bonded by the service pipe it’s in contact with? And if so what if there is a high resistance between the bench and metal service pipe( grater than 0.5 ohms) when tested?

Jul 29 2019 17:07

No; that's not the intent at all.
The point is that a metallic pipe - eg water pipe that is in electrical contact with ground is supposed to be bonded, Clause 5.6.2 sets the requirements, and fogs 5.4 & 5.5 illustrate them.

If the pipe is bonded, then it can no longer be at a different potential from that of the installation's earthing system.
As per the clause - the bonding should be done as close as practicable to where the pipe is in contact with the ground,(not at the farthest end).
Not because it makes any difference to the effectiveness of the bonding, but simply to avoid a section of the pipe being replaced at some later date with non-conductive, rendering the bonding ineffective.


Jul 29 2019 19:18

Yes I do understand the part about bonding the water pipe etc.
What I’m getting at is Bonding a metal bench. When should it be done etc.

What mowgli as started is a good question
“Whether they have to be bonded and whether they should be bonded are two very different questions.
What are your thoughts on how the wiring rules apply in this situation?”

I have heard a lot of different interpretations in regards to bonding over the years from both sparkys and inspectors.
I remember a few years ago when I was on a refresher course. We had a discussion about bonding. The conversation went around in circles.


Jul 29 2019 20:28

There are some things to ask yourself,
1) does the metal bench have electrical fixings or accessories fixed to it - if the answer is yes then bond it
2) does the metal bench have contact with the ground - via a metal water pipe for example - if yes then bond it.

both of these examples describe how the metal bench forms part of the electrical installation, if it doesn't form part of the installation then don't bond it.


Jul 29 2019 21:41

1. That would be earthing, not bonding, and it's only required where there's a risk of contact with live parts or single insulated conductors. There are simple ways to prevent that, as per

2. Even if the bench was in contact with a conductive pipe, and that pipe is also in contact with the ground, it's only the pipe that needs to be bonded, as close as possible to where the pipe enters the building - not at the bench.

Jul 30 2019 09:20

"What are your thoughts on how the wiring rules apply in this situation?”

There is no requirement in AS/NZS 3000 that applies to your situation. Which means that the rule-makers consider that nothing needs to be done WRT earthing the benches. If you want to go further than the minimum requirements set by Wiring rules, that's allowed.

much of the confusion in this area comes from changes in earthing & bonding requirements over time. We've moved from having a list of items that had to be earthed regardless of circumstances; to having the requirements set by the particular circumstances. This requires us to think, instead of blindly applying a rule.

There's also been a failure to grasp the fundamental difference between the functions of protective earthing (carrying - typically very high - fault current in order to operate a protective device; and equipotential bonding (keeping simultaneously accessible items at closely similar potential).

That failure of understanding was not limited only to practitioner at time the rule-makers suffered from it as well.
Example: when I was an apprentice, we "bonded" meter boxes using the 2.5 mm2 PEC of the 2C &E cable used for the ripple control of the w/h circuit. However the function needed wasn't "bonding" it was "earthing"; and a 2.5 mm2 simply wasn't big enough for a 16 mm2 mains with 63 A supply fuse.

We also "bonded" hot & cold water pipes, plus any metal waste pipes, regardless of whether or not there was any ground contact. Most of which was a complete waste of time & materials; as where there's no ground contact, and the conductive items are left floating, there can be no difference in potential.

We did the same for sink benches, which were required [R 151 of 1976 Regs]to be "earthed" in every case. One of the old "reasons" bandied about, by Inspectors and others, for earthing metal benches was - as some have suggested here - in case a faulty appliance livened the bench. But that "reason" was not given in the Regs, it was invented by the inspectors. In fact there was no official reason, it was just whoever wrote that reg thought it was a good idea.

1993 Regs moved on; the requirement simply disappeared. 1995 brought us ECP 25 in which sink bench tops (but not other bench tops) were "required" - along with shower trays & urinals - to be equipotentially bonded. Noting that the ECP was never mandatory, even after the 1997 Regs cited it as a means of compliance. 2 years later, by amendment to 1997 Regs, the citation changed to NZS 3000: 1997 but the list stayed the same. Then in 2003; a further amendment changed the citation to AS/NZS 3000:2000. That's when the list of "must always bond" items was dropped.
So since 1976 (at least)there hasn't been a requirement to bond bench tops that don't have sinks; and there hasn't been a requirement for sink bench tops for over 16 years.

If the "reason" of faulty appliances had been valid for sink benches, it would have been equally valid for non-sink benches. Looked at that way, it becomes obvious that it was simply an invented "reason", not a real one.

It's attempting to deal with a double-fault scenario; and the safety rules we operate under only require us to deal with single-fault scenarios. It's easy to think that "bonding can't do any harm", but in fact it can. So before going that extra mile, you need to consider not just that scenario, but every other possible scenario; including those for which increasing the amount of earthed metal within easy reach could actually increase the chances of someone being in contact with that metal at the moment they touch something live.
Which is what ESR 89 (1) is about. It specifies the extra precautions necessary for 'substantially conductive" situations - one of them being use of equipment that is BOTH double insulated AND protected by an RCD.


Aug 01 2019 14:52

Thanks for your reply, extremely useful .
Because of differing situations in the labs I will probably end up going 50/50 with one set of benches earthed because of the crappy almost connected situation the others left floating as they are well insulated from anything earth.

Aug 01 2019 15:22

You seem to be missing the points that people have tried to carefully explain above.

Almost earthed, floating, connected to a pipe which is bonded somewhere else...
None of those situations require the bench to be earthed or bonded, and earthing them could possibly make the situation less safe.