Bonding requirements for ground rods

[jucesanc]

I have a 50,000 BBL tank that is grounded via two ground rods, that are accessible through two ground wells. The tank has instrumentation, lighting, piping, etc. It stores gasoline, so the area is class 1 div 2.

Near the tank (5 feet away from the tank), is an instrumentation terminal box. The box itself is grounded through a separate ground rod, and accessible through its own ground well.

My recommendation is that we bond the ground rod for the terminal box to tank. This is based on my interpretation of NEC 250.50: "All grounding electrodes as described in ..... that are present at each building or structure served shall be bonded together to form the grounding electrode system. Where none ..."

Is my interpretation correct in that the terminal box ground rod should be bonded to the tank?

 

[LarryFine]

Is my interpretation correct in that the terminal box ground rod should be bonded to the tank?

I'd say yes with or without the interpretation. Are you sure they're not already connected underground?

 

[jucesanc]

Are you sure they're not already connected underground?

I'm very certain that they're not. The terminal box installation happened recently and I have all the installation details for it, and for it's ground rod.

 

[LarryFine]

Definitely should be connected. The separate rods could increase hazard.

 

[jaggedben]

Is there not already an equipment grounding conductor effectively bonding the rods between the two? (What kind of wiring is running from one to the other? Line voltage power supply or just signal?)

 

[jucesanc]

Is there not already an equipment grounding conductor effectively bonding the rods between the two? (What kind of wiring is running from one to the other? Line voltage power supply or just signal?)

There are two ground rods connected to the tank, on opposite sides of the tank. And there is a third ground rod connected to the terminal box. This third ground rod is not bonded to the other two. It is in the ground near one of the other ones, but in no way bonded.

Between the tank and it's two ground rods, there is 2/0 cable routed a couple feet. It is clamped to the tank and clamped to the ground rod.

Between the terminal box and its ground rod, there is 2/0 cable routed about 10 feet, It is clamped onto the box and onto the ground rod.

The terminal box just has signal wiring for instrumentation used on the tank (pressure transmitter, radar level transmitter, temperature..).

 

[jaggedben]

I'll preface this by saying I have zero experience with classified areas or practices for that industry, so take this with that big grain of salt.

I think it's debatable whether you have one structure or two, and therefore whether the code section you cited applies. As far as Larry's comment, 5ft is within arms distance, so conceivably bonding the two parts together prevents someone from being shocked by any potential difference between the two, including one from static buildup. If thus treating it as one structure, it probably would have been better to bond the terminal box without giving it its own electrode, but I gather it's too late for that now. My understanding is that having grounding electrodes 'all over the yard' and bonded together increases the danger of a nearby lightning strike causing damage by finding paths along the bonding jumpers. If you bond the two together, I'd run the jumper underground to mitigate that issue. Other than that, however, the bond probably somewhat reduces hazards and doesn't do any harm.

You didn't directly answer the equipment ground question, but I'm gathering there isn't one and that you thus still face a question of whether to bond the terminal box to the tank somehow.

 

[jucesanc]

Is there not already an equipment grounding conductor effectively bonding the rods between the two? (What kind of wiring is running from one to the other? Line voltage power supply or just signal?)

To be honest, I didn't completely understand your question. There is no conductor bonding the rods. The two tank rods are connected through the tank. But this third one is connected to the box, and the box is not connected to anything else.. The instrumentation wiring inside the box has an isolated ground. There is no wiring directly or indirectly connecting the box to the tank.

 

[LarryFine]

My understanding is that having grounding electrodes 'all over the yard' and bonded together increases the danger of a nearby lightning strike causing damage by finding paths along the bonding jumpers.

I believe the opposite us true, that a lightning strike nearby can cause voltage gradients between rods that are not bonded together.

 

[jaggedben]

I believe the opposite us true, that a lightning strike nearby can cause voltage gradients between rods that are not bonded together.

But that's of little consequence if there's no low resistance path between the rods. Lightning creates that gradient regardless of whether the rods are bonded. Bonding the rods together encourages more lightning current to follow the path of the bond. Otherwise the earth will carry the current (relatively) evenly, same as it would if the rods weren't there.

The bonding is fine as long as the bonding jumper is not large enough to be damaged by that, and/or is not run along something that would be damaged by excess heat or flashover following that path. It's a matter of probability whether that is more of a concern than other hazards that the bonding addresses or mitigates.

 

[LarryFine]

I believe that, using this example, a lightning-induced voltage gradient between the unbonded tank and terminal box rods can cause damaging currents between the tank circuitry and the box circuitry.

Think of a properly-sized conductor interconnecting the tank and box rods as equipotential bonding. Otherwise, the voltage difference can damage the rest of the wiring between the two separate parts.