Well casing ground wire connection

After many google searches I give up. Is there a special ground strap for the connection of the well casing to the ground wires of the pump wire circuit? My well cap is made of some sort of light alloy so it is not a candidate, nor does it have a grounding screw on it anyway. Drill a hole in the well casing for the bond?

If the casing is metal then it needs to be bonded, I just drill and tap the casing, use a lay in lug, with a 1/4-20 stainless screw.


Typical lay in lug.

I live in NH in I have never seen even 1 artesian well grounded. I would think that they are pretty darn well grounded all on their own. Is there a specific article that states that and artesian well must have a grounding conductor back to the neutral bar??? Maybe I missed something.

I see an issue where if the pump faulted AND the pump grounding conductor failed the breaker may not trip (depending on the severity of the fault of course), but I think that the casing's earth ground is so strong that even if you added a ground terminal the fault current still would not make its way back to the neutral bar.

drick said:

I live in NH in I have never seen even 1 artesian well grounded. I would think that they are pretty darn well grounded all on their own. Is there a specific article that states that and artesian well must have a grounding conductor back to the neutral bar??? Maybe I missed something.

I see an issue where if the pump faulted AND the pump grounding conductor failed the breaker may not trip (depending on the severity of the fault of course), but I think that the casing's earth ground is so strong that even if you added a ground terminal the fault current still would not make its way back to the neutral bar.

Click to expand...

Your completely wrong in your thinking, the EARTH plays NO ROLE in tripping a breaker, so get rid of that idea immeditely.... the article your looking for is 250.112(M).


What your actually doing is bonding the well casing to to the grounding conductor, which eventually ties back to the neutral buss. This provides a low impeadence path for a fault to actally raise the current high enough to trip the breaker suppling the well pump.

Hope this helps, if not ask away.

On the electrical supply to my house, I've got a 6600v single-phase feed buried along my driveway that feeds a transformer on a pad in my front yard, that supplies 120-0-120 volts to the distribution panel. The neutral (0-volt) line is grounded both by a buried copper rod at the transformer and to my copper water lines at the breaker box. There is no separation between neutral and ground--they are connected in the panel. Since most residential well pumps run from 240v, it would seem that either side of the pump electrical feed contacting the ground potential of the casing would trip the breaker whether or not the ground line is bonded to the well casing--they should be at the same potential, but they might not be--because galvanized or bare pipe will corrode in contact with soil and exhibit something much less of a perfect ground.

With dry sandy soil, and PVC water lines to the house I could see that this might be a real safety problem.

I'm not disagreeing with the need to electrically connect the ground lead to the well casing; just trying to offer an explanation of the reasoning behind it.

ChuckG said:

On the electrical supply to my house, I've got a 6600v single-phase feed buried along my driveway that feeds a transformer on a pad in my front yard, that supplies 120-0-120 volts to the distribution panel. The neutral (0-volt) line is grounded both by a buried copper rod at the transformer and to my copper water lines at the breaker box. There is no separation between neutral and ground--they are connected in the panel. Since most residential well pumps run from 240v, it would seem that either side of the pump electrical feed contacting the ground potential of the casing would trip the breaker whether or not the ground line is bonded to the well casing--they should be at the same potential, but they might not be--because galvanized or bare pipe will corrode in contact with soil and exhibit something much less of a perfect ground.

With dry sandy soil, and PVC water lines to the house I could see that this might be a real safety problem.

I'm not disagreeing with the need to electrically connect the ground lead to the well casing; just trying to offer an explanation of the reasoning behind it.

Click to expand...

Your wrong also... You guys have to know the basic rules of electricty.... the return source always, ALWAYS goes back to the transformer supplied by the utility company, it will travel through the EARTH, but with too much resistance to actually trip the over current device. think about this scenario,

lets say we have a fault to earth, the earth has 25 ohms of resistance, 240v / by 25 ohms equals 9.6 amps, now tell me that is going to trip a breaker and clear a fault? and I was being VERY generous on the earth being 25 ohms, it is MUCH, MUCH higher.

Chris75 said:

lets say we have a fault to earth, the earth has 25 ohms of resistance, 240v / by 25 ohms equals 9.6 amps, now tell me that is going to trip a breaker and clear a fault? and I was being VERY generous on the earth being 25 ohms, it is MUCH, MUCH higher.

Click to expand...

Hmm, I just checked NEC; code stipulates a maximum of 25 ohms. But your point is taken.

I'd expect a well casing full of water to have lower resistance to earth than the standard grounding rod connected to the distro transformer center tap, what the setup amounts to is that things get reversed a bit. The system ground will end up being the well casing not the grounding rod.

In any case, relying on the breakers to detect a ground fault seems like risky business. I'm surprised that code doesn't require a GFCI for well pumps.

Gfci?

quote; I'm surprised that code doesn't require a GFCI for well pumps.

Now wouldn't that be a recommendation from a rocket scientist, given the tendency for GFCI's to phantom trip and turn off whatever is connected to it.

ChuckG said:

Hmm, I just checked NEC; code stipulates a maximum of 25 ohms. But your point is taken.

Click to expand...

Thats not what it says....

ChuckG said:

I'd expect a well casing full of water to have lower resistance to earth than the standard grounding rod connected to the distro transformer center tap, what the setup amounts to is that things get reversed a bit. The system ground will end up being the well casing not the grounding rod.

Click to expand...

Your misinformed, ground rods are for lightning, line surges, or unintentional contact with higher-voltage lines. Ground rods are NOT for clearing faults. Cant stress this enough, an electrical system will operate just fine without any type of grounding electrode installed.

ChuckG said:

In any case, relying on the breakers to detect a ground fault seems like risky business. I'm surprised that code doesn't require a GFCI for well pumps.

Click to expand...

Just the opposite, A egc (equipment grounding conductor) trumps GFCI protection any day.

hj said:

quote; I'm surprised that code doesn't require a GFCI for well pumps.

Now wouldn't that be a recommendation from a rocket scientist, given the tendency for GFCI's to phantom trip and turn off whatever is connected to it.

Click to expand...

Thats not true either... no such thing as a nuisance trip... The only reason a GFCI will trip is that its doing its job.

Chris75 said:

lets say we have a fault to earth, the earth has 25 ohms of resistance, 240v / by 25 ohms equals 9.6 amps, now tell me that is going to trip a breaker and clear a fault? and I was being VERY generous on the earth being 25 ohms, it is MUCH, MUCH higher.

Click to expand...

It's worse than you state. The voltage between the third wire and either side of the line is only 120v. That would require less than a 4 ohm resistance to trip a 30 amp breaker--and it could likely trip only one side. I'm not comfortable with the numbers on this for shock or fire protection considering that the pump is probably out a couple hundred feet from the panel. It might be worthwhile for lightning protection of the pump.

Chris75 said:

Thats not what it says....

Click to expand...

I read NEC 2008 section 250.56 saying that when grounding is performed by a single rod (which is the case that I mentioned), that the measured resistance be 25 ohms or less. If this value is exceeded, a second rod must be used not less than 6 feet away from the first. Unfortunately, there is no specification for the resistance of the combination. (I doubt that any inspector actually checks this, however).

Do you read 250.56 differently?

hj said:

Now wouldn't that be a recommendation from a rocket scientist, given the tendency for GFCI's to phantom trip and turn off whatever is connected to it.

Click to expand...

Thanks for the compliment, hj--I've never worked on rockets, but have done work for the space program (NASA MDSF) and have been an IEEE member since 1974.

The reason a GFCI trips is because current is finding a path other than through the supply lines to whatever it protects. If you get consistent false trips, you either have a faulty GFCI or there's leakage to ground in the equipment that you're using.

I'll take a false trip any day to getting fried.

ChuckG said:

It's worse than you state. The voltage between the third wire and either side of the line is only 120v. That would require less than a 4 ohm resistance to trip a 30 amp breaker--and it could likely trip only one side. I'm not comfortable with the numbers on this for shock or fire protection considering that the pump is probably out a couple hundred feet from the panel. It might be worthwhile for lightning protection of the pump.

Click to expand...

It would be supplied by a 2pole breaker.

ChuckG said:

I read NEC 2008 section 250.56 saying that when grounding is performed by a single rod (which is the case that I mentioned), that the measured resistance be 25 ohms or less. If this value is exceeded, a second rod must be used not less than 6 feet away from the first. Unfortunately, there is no specification for the resistance of the combination. (I doubt that any inspector actually checks this, however).

Click to expand...

Exactly my point, no one performs a test to actually test the resistance, so we drive two rods and call it a day.

ChuckG said:

The reason a GFCI trips is because current is finding a path other than through the supply lines to whatever it protects. If you get consistent false trips, you either have a faulty GFCI or there's leakage to ground in the equipment that you're using.

I'll take a false trip any day to getting fried.

Click to expand...

I agree, but I would never call them false trips...

Thanks for the info on the method for the bonding lug. You guys brought up some pretty interesting points. I would think the fault current would take all available paths, not just the one of least resistance, proportionally over all the resistances it sees right? Coincidentally I am installing a mid-lot branch panel right next to that well casing, I wonder if it would serve as one of the grounding electrodes saving myself the cost of a ground rod?

BTW just for info on that last question, the local inspector told me I could call it a service entrance, with no meter or grounding conductor run to it from my main breaker panel. I have to bond the bars and drive ground rods though. It was a MAJOR pain finding a bottom feed outdoor rainproof panel though as you can't invert the outdoor versions of them. I ended up buying another RV style pedestal version from Midwest to accomplish it.

MarkHash said:

BTW just for info on that last question, the local inspector told me I could call it a service entrance, with no meter or grounding conductor run to it from my main breaker panel. I have to bond the bars and drive ground rods though. It was a MAJOR pain finding a bottom feed outdoor rainproof panel though as you can't invert the outdoor versions of them. I ended up buying another RV style pedestal version from Midwest to accomplish it.

Click to expand...

I would not have run 3 wires to it, but thats just me, its a no-go with the 2008 code anyway. so might just as well get used to running 4 wires, and a grounding electrode system is always required with remote sub-panels.

MarkHash said:

Thanks for the info on the method for the bonding lug. You guys brought up some pretty interesting points. I would think the fault current would take all available paths, not just the one of least resistance, proportionally over all the resistances it sees right?

Click to expand...

The fault will take all paths, the problem is the earth will not trip the breaker, thats why we pull EGC's (equipment grounding conductors) to begin with, to provide a low impedance path back to the grounded conductor of the transformer.

MarkHash said:

Coincidentally I am installing a mid-lot branch panel right next to that well casing, I wonder if it would serve as one of the grounding electrodes saving myself the cost of a ground rod?

Click to expand...

If it meets the requirements of 250.50(A)(5) then yes.

You guys seem to have slightly different ideas of what actually happens when a short occurs somewhere in the distribution system. I do a lot of troubleshooting of Water Pumps and their electrical supplies. And even though I'm very good at figuring out what's wrong with a system with the help of an ohm and amp meter, I am still amazed with the fact that I can walk into a room, flip a switch and the room lights up. It's facinating. Since we can't see electricity, it's hard for some of us (well at least me) to figure what actually happens with electricity and why it does what it does.

This may be the reason for the differing opinions, and I'm still not sure any of us really knows exactly what goes on inside those copper wires.

bob...

speedbump said:

You guys seem to have slightly different ideas of what actually happens when a short occurs somewhere in the distribution system. I do a lot of troubleshooting of Water Pumps and their electrical supplies. And even though I'm very good at figuring out what's wrong with a system with the help of an ohm and amp meter, I am still amazed with the fact that I can walk into a room, flip a switch and the room lights up. It's facinating. Since we can't see electricity, it's hard for some of us (well at least me) to figure what actually happens with electricity and why it does what it does.

This may be the reason for the differing opinions, and I'm still not sure any of us really knows exactly what goes on inside those copper wires.

bob...

Click to expand...

I do, its not a mystery.... and I'm not offering opionions, just the facts. sorry to be cocky.

Yer not being cocky. But being in awe of electrons moving through wires, in and out 60 times a second plus doing so at just below the speed of light is still amazing to me. Imagine how fast they could go if they moved in a straight line.

bob...

Chris75 said:

If the casing is metal then it needs to be bonded, I just drill and tap the casing, use a lay in lug, with a 1/4-20 stainless screw.

Typical lay in lug.

Click to expand...

Is that lug approved for outdoor/external in the weather type use?

And how are you supposed to keep the mild steel casing for rusting under it.

Also, the building ground is supposed to be 25 ohms or less or it is not acceptable. Years ago when I first heard of this "bond" or "grounding of the metal casing", it was explained as a building ground electrode because of changes in foundation construction and non metallic water lines that had historically been used in the past.

Now you are saying it is for well worker protection although for decades there was no ground to a 2 wire submersible pump. You also say this and ground rods etc. are for lightening strike protection. I hear that many people still suffer damage from lightening strikes, so how well do these bonds and/or grounds actually work for lightening protection or am I missing something?