Ground Loop - alarm panel.

[spark master]

The fire alarm company installed their panel in the equipment room. and ran a ground wire directly to the water main.
It's also grounded from the EMT 15 amp feed directly into the alarm panel. There is no ground wire in the EMT, just the EMT itself.

I say they created a ground loop, and the alarm panel is throwing grounding errors.

What is your opinion ?

 

[gadfly56]

The fire alarm company installed their panel in the equipment room. and ran a ground wire directly to the water main.
It's also grounded from the EMT 15 amp feed directly into the alarm panel. There is no ground wire in the EMT, just the EMT itself.

I say they created a ground loop, and the alarm panel is throwing grounding errors.

What is your opinion ?

Is the water pipe bonded to the electrical panel's ground bar? If the water pipe is bonded to the electrical panel, this shouldn't be an issue.

 

[rlundsrud]

Why do you think it is throwing grounding errors? Is there a specific error code being generated? I work on BAS equipment integration and I have seen techs loose site of the real cause because they got hung up by something like this. I am not trying to say you are doing this, I just want to caution you so you might explore all possible causes. One troubleshooting trick you might try is to remove all I/O and only have power connected and see if the problem goes away. You can also temporarily connect a small battery UPS to see if that alleviates the issue. You could also disconnect the ground at the water riser, but turn off the power to the fire alarm first.

Good luck

 

[spark master]

For the same reason you don't bond a neutral in a sub panel, correct?
Why create a ground loop..... ?

 

[don_resqcapt19]

For the same reason you don't bond a neutral in a sub panel, correct?
Why create a ground loop..... ?

Multiple EGC paths are not an issue. It is not the same as having multiple neutral to grounding connections on a system. It is almost impossible to install a system where there are not multiple paths for the EGC.

 

[gadfly56]

Multiple EGC paths are not an issue. It is not the same as having multiple neutral to grounding connections on a system. It is almost impossible to install a system where there are not multiple paths for the EGC.

Exactly. If you ran the ground inside the EMT you'd still have parallel paths.

 

[spark master]

ok.... but why would you want group loop after ground loop ?
I know, if I'm mounting a 1900 box to building steel, it's grounded to the footing. But to ground something directly to the water main..... not to mention the alarm contractor probably used the same ground screw as the panel ground, using the same ground clamp. .....which we know is wrong.

 

[gadfly56]

ok.... but why would you want group loop after ground loop ?
I know, if I'm mounting a 1900 box to building steel, it's grounded to the footing. But to ground something directly to the water main..... not to mention the alarm contractor probably used the same ground screw as the panel ground, using the same ground clamp. .....which we know is wrong.

The water main should be part of the building ground. It is therefore at the same potential as the ground. Therefore no ground loops.

 

[iwire]

ok.... but why would you want group loop after ground loop ?
I know, if I'm mounting a 1900 box to building steel, it's grounded to the footing. But to ground something directly to the water main..... not to mention the alarm contractor probably used the same ground screw as the panel ground, using the same ground clamp. .....which we know is wrong.

Again it is not a problem and is not the cause of ground fault messages.

The ground fault messages are only caused by ground faults on the low voltage side.

 

[infinity]

The fire alarm company installed their panel in the equipment room. and ran a ground wire directly to the water main.
It's also grounded from the EMT 15 amp feed directly into the alarm panel. There is no ground wire in the EMT, just the EMT itself.

I say they created a ground loop, and the alarm panel is throwing grounding errors.

What is your opinion ?

As Bob stated ground errors have nothing to do with the installation that you have described.

 

[cuba_pete]

The water main should be part of the building ground. It is therefore at the same potential as the ground. Therefore no ground loops.

Even though a system may seemingly be at the same potential from an NEC standpoint, from an RFI/EMI or sensitive electronic equipment standpoint there are ground loops which can cause issues. Running several bonds between grounds rods, water pipes, gas lines, etc. creates many ground loops which may affect many different equipments in countless ways.

We all know that electricity (potential) takes all paths. We also all know that one of the primary purposes of grounding and bonding (a fault path) is to ensure an installation is safe and protects from electric shock. The NEC is supposed to be practical safeguarding of persons and property from hazards.

What grounding and bonding does not do as part of the its primary purpose according to the NEC is ensure that sensitive equipment functions correctly in all situations. This is one reason why the NEC has quite a few exceptions which can be utilized to help out with any issues.

The NEC only refers to loops as drip loops, switch loops, loop wiring, cable loops, etc. Nowhere in the NEC does the term "ground loop" come into play. Electromagnetic interference is barely mentioned, and radio frequency interference is not mentioned at all. None are really a concern for the NFPA.

Some of us have to address these issues, but we must stay compliant with the codes (NEC, CEC) as applicable, AHJ's, etc.

 

[iwire]

Even though a system may seemingly be at the same potential from an NEC standpoint, from an RFI/EMI or sensitive electronic equipment standpoint there are ground loops which can cause issues. Running several bonds between grounds rods, water pipes, gas lines, etc. creates many ground loops which may affect many different equipments in countless ways.

We would be glad to learn what equipment that is and the many ways it can be affected.

 

[cuba_pete]

We would be glad to learn what equipment that is and the many ways it can be affected.

A standard 19" equipment rack in a computer server room, a main fire alarm panel in a high rise building, or a hospital operating suite can be grounded and bonded according to the NEC. This same equipment and many others installed in similar configurations can have demonstrable ground loops.

As one example, close and near field probes can be used to identify sources of EMI and RFI throughout equipment. That energy can be impressed on any part of equipment under the right circumstances. How that circuit is laid out with regard to bonding and shielding can greatly affect the operation of the system. Sometimes the levels of interference are not a problem and in fact are not interference at all, just background noise that may not affect a thing. The FCC sets rules for class A and B devices in most instances.

Even the FCC and NEC do not catch the infinite possibilities for those situations clearly addressed in, for example, the IEEE Green and Emerald books.

 

[tom baker]

For the same reason you don't bond a neutral in a sub panel, correct?
Why create a ground loop..... ?

Two different issues. the neutral is a current carrying conductor, bonding it a feeder panel puts neutral current on any metallic paths. The EGC is not a current carrying conductor, "ground loops" fall into the urban myth area.

 

[cuba_pete]

Two different issues. the neutral is a current carrying conductor, bonding it a feeder panel puts neutral current on any metallic paths. The EGC is not a current carrying conductor, "ground loops" fall into the urban myth area.

The EGC is not an intentional conductor.

I just completed periodic ground vault readings for my facility. In the real world, I measured from zero to 2.88 A of current flow from various connections. That current flow creates noise and is a potential issue for sensitive electronic equipment.

How do you personally explain the IEEE's stance on ground loops?

 

[iwire]

A standard 19" equipment rack in a computer server room, a main fire alarm panel in a high rise building, or a hospital operating suite can be grounded and bonded according to the NEC. This same equipment and many others installed in similar configurations can have demonstrable ground loops.

As one example, close and near field probes can be used to identify sources of EMI and RFI throughout equipment. That energy can be impressed on any part of equipment under the right circumstances. How that circuit is laid out with regard to bonding and shielding can greatly affect the operation of the system. Sometimes the levels of interference are not a problem and in fact are not interference at all, just background noise that may not affect a thing. The FCC sets rules for class A and B devices in most instances.

Even the FCC and NEC do not catch the infinite possibilities for those situations clearly addressed in, for example, the IEEE Green and Emerald books.

That did not say a thing.

Please explain the problem these ground loops can cause or do not clutter this thread with things that have nothing to do with the OPs question.

 

[iwire]

[

I just completed periodic ground vault readings for my facility. In the real world, I measured from zero to 2.88 A of current flow from various connections. That current flow creates noise and is a potential issue for sensitive electronic equipment.

What potential issue?

What sensitive equipment and how does this reflate to the problems the OP is having with groundfault issues with a FACP?

 

[cuba_pete]

What potential issue?

What sensitive equipment and how does this reflate to the problems the OP is having with groundfault issues with a FACP?

I don't understand the word reflate. Perhaps another term?

The sensitive equipment in this case may very well be the fire alarm control panel.

I am merely trying to relate to the OP my experience with ground faults and ground loops and the issues they can cause. These issues may or may not include fire alarm control panels.

The IEEE has put much credence to the effects that grounding and bonding can bring to bear on sensitive electronic equipment.

In my experience and in my workplace the ineffective and casual use of grounding and bonding methods by inexperienced and unqualified personnel may cause nothing but trouble to the actual purpose of an installation.

If the OP thinks I am way off base then I wish they would say so.

 

[don_resqcapt19]

I don't understand the word reflate. Perhaps another term?

The sensitive equipment in this case may very well be the fire alarm control panel.

I am merely trying to relate to the OP my experience with ground faults and ground loops and the issues they can cause. These issues may or may not include fire alarm control panels.

The IEEE has put much credence to the effects that grounding and bonding can bring to bear on sensitive electronic equipment.

In my experience and in my workplace the ineffective and casual use of grounding and bonding methods by inexperienced and unqualified personnel may cause nothing but trouble to the actual purpose of an installation.

If the OP thinks I am way off base then I wish they would say so.

While there is some very sensitive equipment, I really doubt that a commercial fire alarm panel falls into that category.
Based on the information the OP provide, I expect that he will find that there is an actual ground fault on of of the fire alarm initiating or indicating circuits.

In a typical commercial or industrial installation, there are always multiple paths for ground fault current. Without spending a lot of money, there is no way to avoid those multiple paths.
Computer installations and telecommunication installations are more sensitive and they often spend the money to prevent those paths.

 

[cuba_pete]

While there is some very sensitive equipment, I really doubt that a commercial fire alarm panel falls into that category.
Based on the information the OP provide, I expect that he will find that there is an actual ground fault on of of the fire alarm initiating or indicating circuits.

In a typical commercial or industrial installation, there are always multiple paths for ground fault current. Without spending a lot of money, there is no way to avoid those multiple paths.
Computer installations and telecommunication installations are more sensitive and they often spend the money to prevent those paths.

I agree and I hope it's not that sensitive.