Here’s the deal:
I recently purchased a 1907 bungalow with galvanized piping. I plan on replacing most of it, but have a couple questions:
- The main service coming from the street is galv. It is grounded back to the panel/ground rod. Assuming I keep the ground on the galv, is there anything else I need to do by code, or as a precaution?
- Where I make the union between the copper & galv (right after the meter & shut off) I am using a dielectric union. Do I need a ground between the different pipes? I would think that it would defeat the dielectric union. Right now, there is a ground across the meter from galv to galv.
- There is a grounding wire connecting the cold (in) and hot (out) pipes of the gas water heater, is there a need to do that in the case of copper?
- There is one other location in the basement (exposed joists) where there is a grounding wire connected to a galv supply line, leading into a wall. I assume it is for the old ungrounded wiring (there is a mix of new romex and old 2-wire ungrounded, I plan on straightening out/replacing in the near future). What do I do in the mean time to ensure I don’t screw up the apparent grounding of the old wiring?
- Is there anything else related to the grounding issue with pipes I should be aware of?
Thanks for the help guys!!!
Replies
"Assuming I keep the ground on the galv, is there anything else I
need to do by code, or as a precaution?"
What you are doing is "bonding" the piping. The purpose of this is to ensure that you do not electrify any of the large conductors (pipes) in your house, except those that you intend to (wires). If by some manner, a hot wire contacts a bonded pipe, the breaker should trip. If the pipe is not bonded, it is a good shock waiting to happen. Bond each pipe system. hot water, cold water and gas piping, if applicable.
You will not defeat the dielectric union by bonding across it. The galvanic corrosion that it prevents is caused at the point of physical contact between two dissimilar metals.
"What do I do in the mean time to ensure I don't screw up the apparent grounding of the old wiring?" - leave it alone.
"Is there anything else related to the grounding issue with pipes I should be aware of?"
If you or anyone else has added a filter or a softener or any other device that breaks the continuous metal path, bridge across that device to ensure that no piping in the house can be electrified,
is there anything else I need to do by code,
As a general rule, newer codes don't apply to older work, with various exceptions.
Newer codes require a ground rod or UFER ground (not applicable here) as well as the bonding of the water pies (which can also act as a ground.)
It wouldn't hurt to check with an electrician re upgrading the grounding system.
The other issue is that that it was allowed to pick up "local grounds" for outlets from a near by water pipe.
That is no longer allowed. The ground wire has to go back to the ground bus in the pannel. But you are allowed to run a separate wire that is not part of a cable (romex). Also you can run it to a ground from an other circuit that does run back to the pannel.
'You will not defeat the dielectric union by bonding across it. The galvanic corrosion that it prevents is caused at the point of physical contact between two dissimilar metals."
Correct me if I am wrong, but at some point when bonding across the union, there will be physical contact between the copper bonding wire and the galv pipe. What type of joint should be used to reduce/eliminate the galvanic corrosion at that point (brass or copper saddle, etc)? Is it worth worrying about?
Thanks for the help!
There are clamps made specifically for bonding pipes. I believe they are made from brass. You are correct in that ANY dissimilar metal to metal contact has a galvanic potential, however, when not regularly exposed to water, the potential for corrosion is mimimzed. You will see some minor corrosion develop over time, due to moisture in the air and pipes "sweating" on occasion. In 20 years, you might have to spend another $2 for a new clamp. In my book, not a significant concern.
Concerning the "advice" not to bond your piping if you have a grounding rod; ignore it. The NEC requires bonding of pipes. The purpose, as stated previously, is not to ground the service, but to provide the safety feature of not potentially having electrified pipes. As a secondary benefit, it does improve the overall grounding of the service.
I offer a recent repair as an example of what can happen when bonding to pipes has not been done: Homeowner went to replace the vinyl vent hose on his dryer with a metal one. The metal hose was in contact with the water heater (gas) and when put in contact with the dryer (electric) he noticed sparks. Dryer checked out OK, but found 120V to ground at the gas water heater. Checking the crawlspace, found melted wires to the AC/furnace (gas) creating a short between one hot leg and furnace frame and gas line. Why didn't the breaker trip? Inside the panel the ground wire had burned in two, probably due to a nick in the wire and a high amperage arc. My guess is the breaker tripped initially, but was able to be reset. Had the gas line been bonded (water lines are CPVC), the breaker should have continued to trip. After seeing the scorch marks on the floor joists and thinking of how easy it would have been for someone to contact the water heater and dryer at the same time (ZAP), I told the homeowner he was very lucky.
If it's metal. BOND IT!
It would seem to me that if you have a ground rod you shouldn't need any bonding at any of the waterlines. I think that if you were to check for ground at the water service when its disconnected from the house you would find that it is definitely an Earth ground. The water in the line is a pretty good conductor so bonding accross a Dielectic union would have little or no purpose in my minds eye.
The only thing that comes to mind is that maybe the receptacles and fixtures were changed at some point but the original wiring did not include a ground wire. This would mean that the grounds would have to be established some other way and maybe the least expensive method was to get the ground from a nearby water line.
"The only thing that comes to mind is that maybe the receptacles and fixtures were changed at some point but the original wiring did not include a ground wire. This would mean that the grounds would have to be established some other way and maybe the least expensive method was to get the ground from a nearby water line."
That is all more the reason for the bonding and jumpers. One problem is that you never know when a section of pipe in the middle might be replaced with plastic.
And water, as a conductor, is not reliable. For a quick test I just stuck two probe, about 1" apart in a glass of water.
It was about 250k ohms. While that will vary with the type of water and area of contact that is not near low enough by many orders of magnitude to be a good ground.
"That is all more the reason for the bonding and jumpers. One problem is that you never know when a section of pipe in the middle might be replaced with plastic. "
Good point, one I had not considered.
Here's a thought to ponder - What might occur if at the moment I was washing my hands, my garbage disposal shorted out to the frame. Would I get shocked severly before the breaker could react to the overloaded situation. Or perhaps someone had broken the continuity with a plastic pipe fix and the disposer ground was on the ungrounded side of the pipe.
These are some of the things I think about as good reasons for running the grounds back to panel with the grounding rod and eliminating the many plumbing grounds by simply grounding the panel to both the grounding rod and plumbing at the water service point of entry.
Edited 11/6/2002 11:52:47 AM ET by no one
"These are some of the things I think about as good reasons for running the grounds back to panel with the grounding rod and eliminating the many plumbing grounds by simply grounding the panel to both the grounding rod and plumbing at the water service point of entry."
Exactly.
But to clarify something you should both have the grounds from the recptacles and devices like the disposal back to the pannel ground. Plus bonding across any dilectrics or plastic pipes that are in the middle of the plumbing.
Bill, how about placing the ground on the street side of the Dielectric union and then using GFI receptacles or breakers for appliances that require both plumbing and electrical connections. I'm no code expert but I think that's what is required now. Its the way my 5 year old home is done. Everything with both types of connections are on GFI circuits and all of the receptacles within 6 feet of a water source are GFI. The ground to the plumbing is coming from the panel and is located on the street side of the water service below the union. Directly below the panel I also have a grounding rod. I looked under the house and there are no ground saddles/wires on any of the plumbing, copper or galvanized. At least none that are visible assuming they should be. Of course that doesn't mean that my house is correct. It could be that the contractor got away with something.
Edited 11/6/2002 3:27:00 PM ET by no one
"Bill, how about placing the ground on the street side of the Dielectric union and then using GFI receptacles or breakers for appliances that require both plumbing and electrical connections. I'm no code expert but I think that's what is required now."
First you need to keep clear the difference between a ground and bonding. The current code requires a connection to a GROUNDING SYSTEM. The grounding system is usually one or more ground rods driven into the ground, but there are other options.
The connection to the incoming water pipe is to "bond" to the electrical ground. That is to make sure that every thing is at the same voltage.
A couple of things that goes into this. 1st alot of water distrubution systems are now plastic.
Also "ground" might not be ground. There different reasons why there might be a small voltage between the "ground" from the water pipe and the "ground" from the ground rods. This is a real problem for cows which are very sensitive to small voltages. Also for swiming pools where you are wet. The real purpose is to make sure EVERYTHING is at the same voltage level.
The best way to do this is to make sure all metal is grounded by connecting back to a common point, the pannel.
GFCI's are really designed to protect in cases where the equipment can't be grounded, for example portable appliances.
The code requires all countertop recptacles to be grounded, but only those on countertops. Because they will be used with portable appliacnes and near water and grounded surfaces (sinks, stoves, refigerators).
The best protect from a fault in a fixed appliance is the grounding system. There is no need for a GFCI on the dishwasher or disposal. And especially a GFCI should NOT be used on a refigerator. Not only is not needed as it is connected to the grounding system it is too easy to have a false trip turn off the refigerator.
For the same reason while unfinished basements and garages require GFCI's for convinence outlets there are exceptions for fixed equipment such as freezers, refigerators, and sump pumps.
OK. that all makes sense, thanks.
Also "ground" might not be ground. There different reasons why there might be a small voltage between the "ground" from the water pipe and the "ground" from the ground rods. This is a real problem for cows which are very sensitive to small voltages. Also for swiming pools where you are wet. The real purpose is to make sure EVERYTHING is at the same voltage level.
Bill, what are some of those reasons? Voltage on the grounding electrodes has been mentioned at an HI site, but I've never seena good explanation for it.
Bob
I am not an electrican, but electrical engineer working in electronics and software. So I don't have any personal experience. But I have have pick bits of info over the years.
The earth is not a very good conductor. So anytime that you have current flowing through the earth you will have a voltage drop.
It is usually caused by the failure of the neutral feed. Either on the utitlity side or a feed to an outbuilding.
The neutral is grounded at the transformer and again in your pannel. If the neutral fails then the two grounds complete the circuit through the earth.
I am not an electrican, but electrical engineer working in electronics and software
So you're just kind of guessing, right {VBG} (very big grin)
Thanks for the info.
Reminded me of another VBG.
Was at a meeting discussing harmonics where a plant facilities manager related that he alway told his guys to check the "brown" wire first in 3 phase systems. (e.g. if the white wire has turned brown, there is too much neutral current - usually triplett harmonics)
Earth ground is different in many places. In my neck of the woods, Wisconsin, the potential varies in farms and factories due to moisture where the ground rod is placed and other issues. The stray voltage is more prevelant in older structures or multiple electrical service facilities.
Theorize that the average old line factory has expanded like a hunting shack. First there is a 20 by 40 building then they add on, and on, and on, until it covers an acre or two for making some food product or similar manufactured good. The original building had an electrical service with a ground rod. As the building expands the surface of the earth is covered by concrete and is generally well drained by foundation tile. The bigger the area the more coverage from concrete and roof. This affects the quality of ground. They also add additional electrical service drops which have different potentials to whatever "ground" is.
Theorize this too, that the distance of 90 foot on a wire provides an easily measured differential to ground. This also factors into the problem of stray voltage.
"Where you are is where you are at" is strangely applicable to this stray voltage. In application on a variable frequency drive that is controled by 4-20 ma signal you need to separate the controled end from the controlling end. This translates into using a twisted shielded wire for the control potentiometer that is only grounded at the end where the VFD is located. I've had systems that were in food plants where the system would run at a set speed until someone leaned on the control cabinet. As soon as they touched the control cabinet the speed of the motor would change. Control has to be grounded where the controlling device resides. (in an on-off control this is not as necessary)
But, for a cow, the drinking trough and the vacuum milker better be grounded where they are or they will get a milivolt trickle charge in their teets. The analogy is like chewing tinfoil for a person with fillings.
Final thought 1907? Galvanized? more than likely lead pipe supply line. Check it with a magnet. 1910, for my place, and the feed for the community is lead not just the drop to the house. Sounds awful but it oxidizes and forms a highly resistant crust. Checked the water many times and no trace in our tap H2O. I am loath to change the galvanized pipes for the fear I'd disrupt the crust.
Sorry for the long explaination but Stray voltage is spooky."The time has come," the Walrus said, "To talk of many things: Of shoes--and ships-- and sealing wax--Of cabbages-- and kings, and why the sea is boiling hot-- and whether pigs have wings"
Bill,
"GFCI's are really designed to protect in cases where the equipment can't be grounded, for example portable appliances."
This is partially wrong. Ground Fault Circuit Interrupters are designed to protect any electrical device in the event of a ground fault. Specifically, in wet areas where a normal breaker would not trip. If the device senses a difference between the current in the hot leg vs the neutral leg, using groung as a reference, the circuit is tripped. A grounded portable appliance can still give you a serious jolt if you put yourself between the line voltage and anything resembling a "semi-ground". As long as the current is not a very high short circuit current, the breaker will take some time to trip. The GFCI protects on a small imbalance, not total current or theraml overload (they do this as well, if the leg currents are equal).
"protect any electrical device" - e.g your heart
BTW, heard that the GFCIs (very fancy version) on the space station were disabled due to nuisance trips, apparently had been made to respond to every possible "old mother hubbard" safety concern, couldn't deliver power reliably?
I take exception to your exceptions <G>.
"Ground Fault Circuit Interrupters are designed to protect any electrical device in the event of a ground fault."
No it is designed to protect the user, not the equipment. Human responses to current are what determine trip point in the GFCI. Here is some interesting info on this.
http://www.elec-toolbox.com/Safety/safety.htm
"A grounded portable appliance can still give you a serious jolt if you put yourself between the line voltage and anything resembling a "semi-ground" (or real ground, Bill)."
That is right. It is interesting that also this same discussion just went on in another forum, but for different reasons. I think that explained this part in the other forum. With a portable equipment you also often have abused and mis used tools. How often have you seen a tool will a frayed cord, or the ground pin cut off. So there are ways to get connected to the hot lead.
Edited 11/7/2002 10:07:51 AM ET by Bill Hartmann
Bill,
"...it is designed to protect the user, not the equipment."
I agree.
"How often have you seen a tool will a frayed cord, or the ground pin cut off."
Stupid people will always find ways of hurting themselves. The whole concept of the GFCI came about to protect idiots from killing themselve in a tub full of water using an electric hair dryer. You ever see some of the warnings on things like lawnmowers.
Personally, I believe that stupid people should not be completely protected from self-inflicted death or injury. A little chlorine in the shallow end of the gene pool is not a bad idea. Ever read the "Bill of No Rights"? Pretty good, basic stuff.
No disagreement with you.
Some accidents should be considered "Darwinian" and a benefit to society as a whole.
The water in the line is a pretty good conductor so bonding accross a Dielectic union would have little or no purpose in my minds eye.
I used to think that but have been soundly corrected by a couple of electrical engineers.