Should each service panel have its own GEC?
I would say no, that one GEC connected to both panels is sufficient so long as its properly grounded at the ground rod or Ufer.
But the local AHJ has the final say.
Each service disconnect would require a GEC connected to the GES unless the connection of the GEC is upstream of the service disconnects. So if the GEC connects in the meter it would be upstream of the service disconnect(s) and would not be required in each if the service disconnect(s).
Thank you, Robert, It does not appear to be connected upstream, only to a driven rod.
@rmeier2 can you point me to the code reference on this?
Thank you again for your insight.
Take a look at 250.64(D) and 250.36(F). There should be a GEC connection in both service disconnects. You can run a single GEC to the vicinity of the panels and then tap off with a bonding jumper to each panel.
Isn’t that what the OP pictures show? Or am I missing something?
To me it looks like there is a GEC connection from GES to one panel and then from that panel to the other panel.
So would the downstream panel be considered a bond as opposed to a ground?
I believe that according to the OP there is no connection of the panel on the left to the GEC or the GES.
Got it, for some reason I thought the top picture, where the GEC is looped, was the connection between the panels, which isn’t visible in the picture with both panels.
It looks like the GES may be at the meter… you say it was connected with a driven rod, that GEC poking out of the siding may go into the bottom of the meter box
It does appear that the top picture with the looped GEC is the left panel. In the second picture you can not clearly see the looped GEC or where it exits the panel, but the right panel has a single GEC attached to the left (bonded to enclosure) bus bar. Without being there I would surmise that the GEC that originates from the grounding electrode loops at the left panel and continues to the right (little other reason to have it looped like that). My conclusion is that both panels share the same grounding electrode.
There is (1) GEC that loops from one service panel to the 2nd service panel then is exiting through the wall to the exterior.
So my question is this acceptable or should this be set up differently?
I don’t normally run into a setup like this with 2 service panels.
Probably depends on the date of construction and the code adopted at the time.
If that’s the case then why would it not be compliant? I have had some local electricians tell me that 2 ground electrodes are required here (MD).
There are a few methods that are code complaint when there multiple service disconnects, the GEC tap method I mentioned (see graphic below), a separate GEC to each service disconnect or connection of the GEC to the neutral upstream of the service disconnect. If the installer looped the GEC through the lug in one panel then into the other that is not code complaint. Also looping a conductor in a terminal is never code complaint unless the terminal was tested and listed for the conductor loop.
How are the “taps” done?
Can be a simple split bolt (the most common method) or some other type of connector like a crimp or mechanical connector. The conductors tapped off of the GEC would be bonding jumper and since you’re not splicing the GEC an irreversible connector is not required.