Grounding Electric Actuators on a dam

[jaker]

Planning for installation of electric actuators on a dam that will be connected to a SCADA system. The actuators have 2 possibilities for grounding, an M5 screw to frame internal to the terminal compartment where incoming power is connected (120VAC/30A), and an external case grounding lug. Here's the problem. Incoming electric service (w/ ground rod) will be approx. 150 ft. from site of these actuators which will be on top of a concrete dam. The dam is also built upon granite ledge with very little possibility of driving a ground rod except for possibly in the water upstream. But actuator manufacturer recommends using separate ground rod to outside ground lug on actuators at that distance (150 ft.) from incoming service, which makes sense. My questions:

1. Would a ground rod fully submerged in water (potentially at high flows) driven partially into streambed be a sufficient ground for actuators?
2. Would there be issues with incoming electric service & PLC/SCADA being on a separate ground from the actuators?
3. Using both grounds (M5 screw a back to service ground rod and external ground lug to local ground rod) with separate rods would create ground loop issues no?

I'm just the instrumentation guy looking to get a plan and wire schedule in the hands of an electrician, but I'm always looking to learn more about this stuff.

 

[texie]

You must have a properly sized equipment grounding conductor run with the branch circuit conductors. Any ground rod(s) you install in addition to the required grounding electrode system at the service are optional and not required by the NEC. You can not have just a ground rod serving as the ground for anything in the NEC as to do so would not have the required fault return path.

 

[jaker]

You must have a properly sized equipment grounding conductor run with the branch circuit conductors. Any ground rod(s) you install in addition to the required grounding electrode system at the service are optional and not required by the NEC. You can not have just a ground rod serving as the ground for anything in the NEC as to do so would not have the required fault return path.

Thanks. I thought it seemed crazy to not connect a ground with incoming service. For 150ft. run would grounding conductor need to be oversized or is it sufficient if grounding conductor is same size as hot/neutral of branch circuit? I.e. are there distance/impedance specifications for the grounding conductor?

Also, do you know if you would you ever connect both the grounding conductor run with branch circuit as well as the outside chassis ground lug to a separate "local" ground rod? I would think this would cause ground loop issues for instrumentation. But perhaps this outside chassis ground is sufficiently isolated from the internal service ground.

 

[tom baker]

Your actuator mfg is incorrect in recommending a ground rod at the actuator, in the event of lightning, that allows another path into the equipment. We often see this recommendation at CAD/CAM equipment, probably as the equipment is not designed well and its an attempt to make you responsible. I once found a research paper from the EPRI on ground rods at equpment, they said sometimes it helps, makes no difference and often makes the issue worse. I sent the paper to Mike Holt and he made a great graphic. I will post it later.
By the way I spent 38 years as a instrument tech and still am a master electrician. I have wired and installed a lot of actuators, including two at our dam. I would be curious as to who makes your actuators. I used rotork and they did not call for an external ground rod. The worst actuators I worked on were German.

 

[tom baker]

Its not grounding conductor its equipment grounding conductor. in grounding and bonding its very important to use the correct terms. Your electrician probably just says run a ground and I would like you to look up the definition of ground in Art 100

 

[jaker]

Your actuator mfg is incorrect in recommending a ground rod at the actuator, in the event of lightning, that allows another path into the equipment. We often see this recommendation at CAD/CAM equipment, probably as the equipment is not designed well and its an attempt to make you responsible. I once found a research paper from the EPRI on ground rods at equpment, they said sometimes it helps, makes no difference and often makes the issue worse. I sent the paper to Mike Holt and he made a great graphic. I will post it later.
By the way I spent 38 years as a instrument tech and still am a master electrician. I have wired and installed a lot of actuators, including two at our dam. I would be curious as to who makes your actuators. I used rotork and they did not call for an external ground rod. The worst actuators I worked on were German.

Thanks. The actuator is a Limitorque MX, but when I tried to call with tech questions I got routed to a regional sales manager. He wrote back to my email indicating I should use the outside ground lug to a "new ground rod in close proximity to the actuator." I wrote back for clarification and I suspect I may get a different answer from someone in tech support. So I figured I'd pose the question here in the meantime.

 

[jaker]

Its not grounding conductor its equipment grounding conductor. in grounding and bonding its very important to use the correct terms. Your electrician probably just says run a ground and I would like you to look up the definition of ground in Art 100

The earth

So would the EGC be what you would connect to outside chassis ground lug on actuator? Would this ever be necessary for safety reasons as an additional ground-fault current path? Or is an equipment grounding conductor run with branch circuit conductors always considered sufficient (no matter length of run)?
Grounding Conductor, Equipment (EGC). The conductive
path(s) that provides a ground-fault current path and connects
normally non–current-carrying metal parts of equipment
N together and to the system grounded conductor or to the
grounding electrode conductor, or both. (CMP-5)

 

[tom baker]

The EGC is required to be in close proximity to the circuit conductors IE in ditch or in a raceway. If its separated (IE on the outside of the raceway) the magnetic fields don't cancel, increasing the impedance meaning a line to case fault wont clear.
You are required to run an EGC, 150ft is no issue, there is a length limit. You can connect an ground rod to the actuator if you want, its what I call a time and materials rod (does nothing). By the way, the Limitorque MX is a copy of the Rotork actuator that I prefer.
The one at our dam is on the discharge, its a special flow control valve where the flow/position is linear, 700 turns open to close...
When I post the graphic from Mike Holt please send it to Limitorque. I have worked on few of them , old technology lots of limit switches, gears and such

 

[drcampbell]

Concrete is a fairly good conductor, especially if it contains moisture or metal reinforcing.
If the frame of the actuator is fastened to the concrete, I would be inclined to declare it a concrete ("Ufer") grounding electrode.

Ufer ground - Wikipedia

en.wikipedia.org

And if not, I'd be inclined to declare the additional grounding unnecessary, unless a job spec requires it.

 

[Flanative]

Thats sounds like a very typical set up we had at South Florida Water Management District. Even though everything had an equipment grounding conductor run with the Branch conductors to the Limitorques there was also a #4 or #6 Bonding conductor run to everything on gate structure. (I.e. handrails, external lug on the limitorque, grates you stand on, all stainless control boxes, even the unistrut rack it was mounted on. ). All were connected to a ground ring run around the control houses and a lightning protection grid on top.

 

[winnie]

As others have said, the actuators must* have an EGC run along with the supply conductors. The 'Equipment Grounding Conductor' provides a fault current path and also will connect the actuator to the grounded point of the supply as well as the supply 'Grounding Electrode System'.

An additional connection to a 'supplementary' earth electrode is permitted* say if the manufacturer suggests such.

I don't know much about the construction of dams, but I presume concrete with rebar. As others have noted, if the equipment is itself connected to the concrete, and especially if it is attached to the rebar, then the equipment is very well grounded.

Jaker asks about 'ground loops'. IMHO 'ground loop' is an often tossed around term and not necessarily indicative of a problem. Multiple connections to earth will certainly cause multiple current flow paths which include the earth. Due to current flow through the earth (both natural and man made) different points on the earth are often at different potentials, and if you have multiple earth electrodes connected by wires there is a real possibility of current flowing into one electrode, through the wire, and out of a different electrode.

Where 'ground loops' creep into control and measurement circuits is when you have a 'ground referenced' signal (for example a signal on a coaxial cable with a grounded outer conductor, or an RS-232 signal where the message is passed on wires referenced to the equipment ground at both ends). Since all circuits require a complete path, the 'signal wire' _and_ the 'grounded shield' are part of the signal path. When the grounded shield is connected at both ends, non-signal current from all the other stuff connected to ground, including the earth electrodes, gets mixed into the signal path and can cause severe noise in your measurements.

The solution to this is not more and more grounding, but rather making sure that all of your signal circuits are separate from ground, for example by using transformer coupled differential signalling, fiber optic, differential twisted pair, etc.

* And finally, the 'must' and 'permitted' about the egc and supplementary electrodes assumes that this installation is governed by the national electrical code. Given that this is a dam, different codes may apply, and especially with regard to grounding there may be different requirements that go counter to what electricians working under the NEC expect. For example utilities use and _require_ multiple earthing of the neutral conductor, explicitly prohibited under the NEC.

-Jon

In this case a _signal carrying conductor_ is

 

[electrofelon]

Thanks. The actuator is a Limitorque MX, but when I tried to call with tech questions I got routed to a regional sales manager. He wrote back to my email indicating I should use the outside ground lug to a "new ground rod in close proximity to the actuator."

This kind of thing really annoys me, This happens frequently when manufacturers recommend dumb things outside of their scope on things they know nothing about. This company should stick to just making actuators

 

[tom baker]

There is no restriction on have 1, 2 or more ground rods for equipment, the NEC calls them Auxiliary Electrodes. In grounding you want to create a single point ground, having additional electrodes can create a path into the equipment. The EPRI did a study on this, stating a ground rod at equipment can help, or makes no difference, or can make it worse. I sent a copy of the report to Mike Holt and he had this graphic made. If you can send a copy of this to Limitorque, what are their comments?

 

[tom baker]

Couple of other comments, stuck inside from smoke (near Seattle)
1. If the concern is with lightning, have a Lighting Protection System installed, it could help. But if lightning hits the actuator, its toast
2. The old style Limitorques are very reliable, as they are all electromechnical and rugged. But they are a bit harder to set up and set the limits, plus don't have the communications like the newer ones do, such as ethernet-IP
3. the new style limitorque and rotorks are easy to set up, but they are electronic. Make sure you have SPD on the service and any subpanels. When I started installing SPDs, my electronic equipment didn't not have as many failures (normally in the power supply) I used Leviton 42000 and 32000 series, they are about $300-400 each. Typically I had two SPDs at each location. Your equipment is at a dam, maybe overhead power so SPDS are a must.