Another PV Grounding Question

[DanS26]

Just have to ask the brain trust here to analysis another grounding question.

Take a look at this diagram....does anyone have any problem with this schematic? I know it has been posted before, but I'd like the experts to opine on grounding system. I'd ask Mike Holt directly if I could...specifically two ground rods bonded underground but could be much distance apart. Is this dangerous to equipment due to the possibility of voltage gradient between the two rods in the event of a high voltage lightning strike?

 

[Carultch]

Just have to ask the brain trust here to analysis another grounding question.

Take a look at this diagram....does anyone have any problem with this schematic? I know it has been posted before, but I'd like the experts to opine on grounding system. I'd ask Mike Holt directly if I could...specifically two ground rods bonded underground but could be much distance apart. Is this dangerous to equipment due to the possibility of voltage gradient between the two rods in the event of a high voltage lightning strike?

NEC690.47(D) is a contentious section of the code, and I do not understand why it has to even exist. Mike Holt says that it should be removed from the code immediately, and says that it makes a system less safe. That is the case when you have mulitple ground rods, that are not electrically continuous with one another underground.

My understanding is that you want everything above ground bonded together, everything below ground bonded together, and you want to make exclusively one location where you connect between them.

Since your two ground rods are bonded together, underground and irreversibly, I don't have a problem with the fact that there are multiple of them. This bonding helps mitigate the ground voltage gradient effects. It is best if these two rods can be as close to one another as practical.

 

[jaggedben]

I remember that thread, sort of, but I'm not going to dig it up. :roll:

I'm no expert on lightning, and my primary comment is that if one is serious about lightning protection then that is way beyond the scope of the diagram, and also way beyond the scope of making sure that the PV system is safeguarded against other dangers (i.e. shocking someone or starting a fire). Also, from the point of view of those non-lightning dangers, I don't have a problem with the diagram.

With that said, it seems to me that in the diagram, the low resistance path for lighting is the ground ring, and any earth gradient will primarily take that path as well as the surrounding earth. Along with the presence of the SPDs, it seems to me that the danger that a gradient will result in destructive currents traveling through the above ground parts of the building would be lessened by this design, rather than made worse. Again though, I don't know much about lighting protection. Real lightning protection, to my best estimation, is not in the diagram.

 

[DanS26]

....... and you want to make exclusively one location where you connect between them.

Well you have touched on the essence of my question. In the diagram there are multiple locations where the above ground system connects with the below ground system. That is the issue!

 

[jaggedben]

Well you have touched on the essence of my question. In the diagram there are multiple locations where the above ground system connects with the below ground system. That is the issue!

If I understood Mike Holt's criticism of 690.47(D) correctly, his beef is as follows:
- An auxiliary electrode in Article 250 is one that is permitted but not required. (Permitted, say, because the manufacturer of your piece of machinery says you should have one, even though they are very likely wrong that it does any good.)
- Because an auxiliary electrode is not required to be bonded to other building electrodes, such an electrode should never be required by the code. It's just permitted, e.g. in case machinery manufacturers are jerks about their warranties.
- In the case of solar PV it's particularly bad to require an auxiliary electrode, because the only bonding to other electrodes is likely to be over the roof, far from the ground.

The diagram isn't as bad as that, because of the presence of the 'ground ring.'

 

[DanS26]

Ben,

Thanks, but you did not directly address my question. Should there be only ONE point of interconnection of above ground grounding system and the below ground grounding system?

IMHO from an EE point of view, the multiple points are a better grounding setup IF the rods are bonded underground. I totally agree that un-bonded rods are a danger......for which Mike Holt has been expounding upon for many years.

And yes, if required by code to drive a rod at the array (if more than x feet from the EC) then can I bond the electrodes underground to make a safer system?

 

[GoldDigger]

Ben,

Thanks, but you did not directly address my question. Should there be only ONE point of interconnection of above ground grounding system and the below ground grounding system?

IMHO from an EE point of view, the multiple points are a better grounding setup IF the rods are bonded underground. I totally agree that un-bonded rods are a danger......for which Mike Holt has been expounding upon for many years.

And yes, if required by code to drive a rod at the array (if more than x feet from the EC) then can I bond the electrodes underground to make a safer system?

Based purely on theory and a limited set of fault/stress conditions, there is a good argument for designing for a single interconnecting point. That has the greatest advantage when you expect a persistent or transient voltage gradient within the earth at the site.
But a well designed multiple-contact setup can do almost as well in theory.
The code does not mandate a single point of connection.
What you do not want is to have separated electrodes which are only interconnected by EGC (above ground in the original dichotomy) wiring. And some but not all of these configurations are prohibited by the NEC

 

[DanS26]

If the code and the AHJ requires me to sink a second rod, I will bond it underground to the main EC if possible. That is the only conclusion I can take from all this.

Thank you everyone for your help and advice.

 

[jaggedben]

Ben,
Thanks, but you did not directly address my question. Should there be only ONE point of interconnection of above ground grounding system and the below ground grounding system?

I'll stand behind Golddigger's response. And just add that sometimes between AHJ requirements and customer preference/economics I am essentially forced to violate the one-point principle. Fortunately I don't work in a high lightning area.

And yes, if required by code to drive a rod at the array (if more than x feet from the EC) then can I bond the electrodes underground to make a safer system?

You are certainly allowed to, but I work on a lot of urban houses where it would be prohibitively expensive. I hope that section disappears again in 2017. Mike Holt's explanations make sense to me, and the justification for the rod is totally nebulous. I also think it's unfair to solar; there are similar large pieces of metal electrical equipment on many buildings which are not required to have electrodes. (Air conditioners and their duct systems, for example.)