Connect with us

Building Ground (long-...sorry)

Discussion in 'Electronic Repair' started by jakdedert, May 9, 2006.

Scroll to continue with content
  1. jakdedert

    jakdedert Guest

    I've read a good deal about the above. Ideally, the grounds for various
    (telephone, cable, power) electrical utilities should all be common; and
    the service entries for those sundry services should be located within
    15' of each other.

    If I ever build a new house, I will take that into account.

    However, it's not practical at this time to realize the ideal. What I
    have currently is as follows:

    The phone drop (two lines) comes in on one side of the house and is
    grounded to the water supply line which comes in the front of the house
    (about of 20' of wire clamped to the pipe where it comes through the
    foundation into the unfinished basement).

    The power drop (100 amp 220v single-phase) enters at the back of the
    house, about 30' diagonally from the phone service. A ground wire
    snakes from the panel (inside a utility porch) around a couple of
    corners and through the floor to a ground post of unknown length/depth
    in the crawl space beneath the adjacent kitchen...approximately 20 feet
    of wire with at least two 90 degree bends.

    The cable drop is around 15' feet from the power service entrance, and
    grounded to an adjacent outside faucet a couple of feet away (all
    plumbing in the house is copper).

    Over the years, we've had a good deal of surge and lightning related
    damage to devices in the house...most recently a DSL modem.

    Would I derive any advantage by driving a new ground post outside,
    adjacent to the power drop and to run all the various service grounds to
    it (around 15' for cable, 25' or so for phone)?

    Alternatively, I could move the telephone ground wire to the existing
    power drop ground post (probably using the same 20' wire), and also
    extend cable ground to this point. That would give me a 'star'
    configuration with all utilities having around 20'-30' of wire from each
    drop to ground.

    Moving the phone service drop at this time (the ideal) is not practical.

  2. Could anyone explain why a US telephone cable needs a local ground? Aren't
    they balanced?

    The UK system only used a local ground for shared local lines which went
    out years ago.
  3. Eric Law

    Eric Law Guest

    System is balanced and isolated in US too. The ground is just for surge / lightning protection.

    Eric Law
  4. It is used for lightning protection.. Believe me you really need if you
    happen to live in Florida.

    - Mike
  5. Right. Few telephone cables in the UK are overground, so that explains it.
    The surge arrestor here is just wired between the incoming pair.
  6. hmm.. I can't say I know of many overground telephone wires in my area,
    except in older houses that have overhead power service, they used to string
    the telephone wire under the power wire, but the main phone lines are all

    I found out the hard way that burrying the wire doesn't help with lightning
    protection when I ran a cat5 ethernet wire from my house to a friends last
    year. That thing got zapped evey time we had a bad storm..

    - Mike
  7. Warren Weber

    Warren Weber Guest

  8. Warren Weber

    Warren Weber Guest

  9. The problem wasn't caused by the ethernet wiring; it was the difference
    in ground potential between the two homes.

    You should have optically isolated the two ends to avoid the problem.
  10. No I'm not talking a small spark when you plug it in. I'm talking about
    lightning hitting the wire during storms. It would arc off the wire when it
    wasn't plugged into the arrestor.

    - Mike
  11. w_tom

    w_tom Guest

    Described by jakdedert is a building all but begging for lightning
    damage. For example, a cow stands in an open field when lightning
    strikes a nearby tree. The cow is killed. Killed by electromagnetic
    fields? Of course not. Killed because cow was part of a path from
    cloud, through tree, into earth to earth borne charges maybe miles
    distant. The electrically shortest path was not under the cow. It was
    up cow's hind legs and down fore legs. Cow was part of a direct
    lightning strike from cloud to distant earthborne charges.

    Cow could have lived is a halo ground surrounded the cow. That
    buried conductor would have, instead, routed electricity around (not
    through) cows. The concept is called single point earthing. Cow with
    separated legs has multiple earthing connections - therefore dead. Cow
    centered in a halo ground has a single point ground.

    jakdedert describes here (and previously) utilities (ie mutli-line
    phones) entering and earthed more like the cow. Building is even worse
    because earthing points are farther apart. Destructive charge can
    enter building on telephone line (overhead or underground line) either
    from its grounding connection or via utility wire (from nearby struck
    tree, from other struck building, or entering via ground rod). That
    transient crossed building, destructively through appliances, to obtain
    earth via AC electric.

    Connecting phone line with a 20' plus ground wire or via pipes
    accomplishes little. Wire has impedance. That means earthing from
    each incoming utility to a single point earth ground MUST be less than
    10 feet. Other features such as no splices, no sharp bends, no solder
    joints (on wire or pipe), etc also required to lower impedance. Not
    resistance - impedance.

    A minimal single point ground is a grounding rod. That means even
    incoming cable TV wire must make that 'less than 10 foot' earthing
    connection to earthing electrode. Better earthing is a halo ground
    (what saved the 'dead' cow) or even better, Ufer ground.

    What does a protector do? A protector only connects from AC electric
    or phone lines (that cannot be earthed directly) to an earthing
    electrode. Protector is nothing more than an connection. No earth
    ground means an ineffective protector - which many overpriced, plug-in
    protector manufacturers hope you never learn. Plug-in protectors that
    have no earthing connection, then, connect to what? They hope youj
    never ask that question.

    Cable TV does not need protectors which often degrade cable modem or
    TV signal. Cable is earthed directly - hardwired - to earth without
    any protector for superior protection. Wire does better than a

    An electric utility demonstrates bad, good, and ugly earthing. Ugly
    because the earthing electrode must be 'lengthened' so that all
    utilities make a common earthing point:

    Water pipe typically is not good earthing. Pipes too long, too far
    away, too many sharp bends, solder joints, etc. A major difference
    between earthing for human safety verses earthing for transistor
    safety. A major difference between resistance and impedance means wire
    distance is more critical that a low resistance ground. Worse,
    jakdedert describes grounding to pipes or water faucets. That means
    ineffective and multi point earthing - that also killed the cow.

    Most critical component in a lightning protection system is earthing.
    Earthing defines uality of that protection 'system' and effectiveness
    of protectors. Ineffective plug-in protectors avoid all mention of
    earthing to sell hyped products at higher profits. Such ineffective
    products have no dedicated earthing connection AND avoid earthing
    discussions to keep customers ignorant. Bottom line: a protector is
    only as effective as its earth ground.

    Effective protector manufacturers have names such as GE, Polyphaser,
    Square D, Intermatic, Siemens, Cutler-Hammer, and Leviton. Their
    effective products have that dedicated earthing wire.Notice that names
    such as APC, Tripplite, Belkin, and Monster are specifically not
    mentioned. The telephone company already installs an effective 'whole
    house' protector in their NID (premise interface) box. But again, you
    (the owner) define its effectiveness by providing an earthing system.

    UK residents who suffer so few lightning storms also suffer frequent
    and unnecessary damage. This because UK incoming phone lines don't
    have that necessary earthing. BT does install effective earthed
    protectors on their end. But subscribers are expected to pay for their
    own protection - which is provided free in North America.

    Also is a myth that underground wires are better protected. Does not
    matter as demonstated by the 'dead' cow. Any utility that does not
    first connect to single point ground before entering a building is an
    obvious incoming path for household electronics damage. As the
    'dead' cow demonstrates, single point earthing means even a nearby
    lightning strike can be a direct strike into building electronics - if
    building's earthing is not properly installed and connected to by every
    incoming utility wire.

    Damage could have been from voltage potential between different
    buildings OR from buried wire carrying transient from a nearby struck
    tree. Multiple sources of damage - all due to a building owner who did
    not install the most critical component in a protection 'system':
    single point earth ground. Why does a telco Central Office, connected
    to every other building in town by copper wires, not suffer damage?
    The solution has been standard for so many generations - proven
    multiple generations before transistors were created. Protection is
    and is defined by earthing. Even protectors are only as effective as
    their earthing.
  12. jakdedert

    jakdedert Guest

    Yes, I've had problems which I have detailed here before. Still, the
    above (and snipped portions) still beg the question: It's gonna be at
    least 20 feet of copper between the service and ground. Still better to
    single point? That's the 'hit' I'm getting....

    Anything would be better than what we have, right? Upgrade the ground
    conductor? I've read (here?) that 1/8" copper tubing is superior to the
    (looks like) 12 ga. wire currently used in the phone and power grounds.

    From the above, I'd assume that 'anything' I did to lower (and
    equalize?) the impedance to ground would be--even though not ideal--at
    least an improvement. How about multiple ground rods, one at each
    service drop--connected together with the aforementioned tubing?

    I know that if I could get the telco to drop the lines in between the
    cable and power drops, the job would be significantly simplified...and
    yes, I have a 100'+ oak tree within 20' of the house.....

  13. w_tom

    w_tom Guest

    'Tree' does not mean it must be a tree. Lightning could strike a
    rock, a neighbor's house, a transcontinential pipe line, or even a
    water box to create same effect. Earthing is the one solution always
    required even if lightning strikes something distant or if lightning
    strikes street utility wires.
    Did you view 'bad, good, and ugly' figures from That
    earthing (in this case a 'right' solution) must conform to two masters.
    One, for earthing transients (ie lightning). Two, to meet electrical
    code defined in NEC Article 250 Section III (Artcile 250.50 through
    250.70. Section III defines seven types of grounding electrodes and
    numbers that apply including wire sizes.

    For example, install a ground ring terminated by rod electrodes (8+'
    copper clad steel rod). Since that ground ring has a ground rod where
    each utility enters, then each utility can make a 'less than 10 foot'
    connection to top of ground rod. Each rod is the same, large, single
    point ground.

    Code demands a ground ring be 2 AWG bare copper wire buried at least
    30 inches. From your perspective, this so that ground wire is below
    frost and remains in soil of constant humidity. Suggested is to obtain
    of copy those five pages from an National Electrical Code book (maybe
    in the library) to better appreciate what is required of each (of seven
    type) electrodes.

    Above to meet code. However grounding wires (to attach to that
    single point ground) also must not have splices, no sharp bends, not
    inside metallic conduit, routed away from all other non-earthing wires,
    and must be firmly attached with proper connectors - for conditions
    beyond what code requires. Best that all earthing wires remain
    separate until all meet at the single point earth ground. Don't worry
    about exceeding wire diameter. Worry more about wire length. Every
    foot shorter than 10 feet means less electricity from lightning will
    seek earth ground via household electronics.

    Next part of that system would connect every wire from every incoming
    cable to that earth ground. Telco has a protector from each (of two)
    wires in cable to the ground wire. AC electric has three wires - only
    one connects to earth directly. Therefore 'whole house' protector (see
    manufacturer list that includes GE, Square D, etc) in AC mains box
    connects other AC wires to that earthing wire.

    Coax for cable TV and satellite dish use a ground block ($2 available
    in Home Depot, Radio Shack, and Lowes) and 10 AWG wire for earthing.
  14. jakdedert

    jakdedert Guest


    Thank you, Tom...I think. I'll have to reread your post several times
    to get the gist of what you're saying. I'll also do some (more)
    research and get the relevant parts of the Code.

    'Tree' in this case does mean a tree...a big one, the highest single
    point on my entire street...even though there are houses which sit
    considerably higher than mine. This is one big tree...probably at least
    part of the reason I seem to be plagued.

  15. default

    default Guest

    I'm assuming the DSL modem and computer are plugged into the same
    outlet . . .

    It sounds like you have two unknown grounds that are far from the
    ideal. You don't say what the failure mode of the equipment is and
    how severe - I'm guessing it is probably common-mode developing
    between the two ground systems. If you get something like bridge
    rectifiers in power supplies shorting - that could indicate a
    differential voltage spike.

    Or, better yet, start at the beginning . . .Check the power
    transformer. There should be a lightning arrestor for the transformer
    - a kind of insulator that sits off to the side of the can that is
    clamped to the can and has a small 1/2"-1" air gap between it and the
    HV terminal. (designs vary quite a bit but it should be there in some
    form - a collection of broken porcelain around the base of the pole
    and some carbon blocks means it is done its job once too often)

    The transformer pole and transformer must be grounded. There should
    be a thick soft copper wire running from the transformer down the
    length of the pole and into the ground. Usually the wire is wrapped
    in a spiral and nailed to the bottom of the pole.

    Without the transformer grounded - anything you do may be wasted
    effort. I had no problem getting a neighbor's lightening arrestor
    replaced with just a call to the power company and, in another
    instance, a pole ground wire replaced - they are subject to being
    stolen by people trying to salvage the copper. Two different power
    companies and no arguments - they just went out and fixed it.

    Your best option is probably to get the grounds on the same circuit.
    I'd sink a ground and wire to it, so I knew what its condition is
    like. (ten foot length of half inch dia copper water pipe washed into
    the soil is better than anything you're likely to drive into the
    ground) Braze or solder the heavy wire on. Put it at the power
    entrance as directly under the power meter as you can with no bends in
    the heavy wire. Five or six feet to ground if you can manage it with
    no bends. Use a large radius bend if you go around an obstacle. It
    may not be practical, but do the best that the conditions allow.

    What to do about the telephone line and power line separation? Sink
    a separate ground for the phone line and just use it for the
    lightening arrestor on the line and make sure it is bonded to the main
    power ground. There was some good stuff in an ARRL article some years
    ago that mentioned a similar problem. He solved it by adding a ground
    for the phone line directly under the phone's entrance box to the
    company supplied arrester then added a set of gas filled spark gaps in
    addition to the arrester.

    Cold water pipes are iffy - they can be excellent grounds or very poor
    grounds. Too many variables - you'd have to know the material and
    condition of the pipe, how straight the run to ground is, and what the
    joints are like electrically. Some old systems used cast iron pipe
    and the joints are not connected electrically. Some really old cities
    still pull up a wooden water main from time to time . . . Galvanized
    threaded joints are usually pretty good if the pipe was clean when the
    pipes where joined and someone didn't get carried away with Teflon

    Beyond that you can still get fancy with isolation transformers or
    optical links.
  16. Bud--

    Bud-- Guest

    An excellent paper from the IEEE on surge and lightning protection
    (which came from a w_tom post) is at;

    Contrary to what w_tom says plug-in point-of-use surge protectors do
    provide protection and are recommended in the paper above. All the
    electrically interconnected apparatus, like tuner, amp, has to be
    connected to the same surge protector. If there are external lines, like
    cable TV, the apparatus can still be protected using a multiport plug-in
    surge protector that, in addition to the power protection, has through
    ports for the cable connection (and/or phone line, LAN connected to
    devices not on the same surge protector, ...). Multiport surge
    protectors, and how they protect, are described in the IEEE paper.

    Another paper is from the NIST
    It also recommends point-of-use plug-in surge protectors.

    I agree with w_tom that single point grounding for wires entering a
    building (also dish antenna) is a very good idea.

    (A multiport surge protector provides a local single point ground at
    protected equipment.(

    Why don't you get objectionable drop through the ground ring? (Not to
    say that the ring isn't a good idea.)

    When you talk about a "halo ground", as for your cow, I presume you are
    talking about a ground ring. The only use of "halo ground" I have seen
    is as in PolyPhaser papers - a conductive ring around a room
    ceiling-wall edge, that may or may not be earthed, to counter the field
    effects from the down current from a lightning strike on an adjacent
    antenna tower.

    If cable and power entrance points are separated, it would seem like the
    cable could be wired from the its entrance ground block to a 2nd ground
    block adjacent to the power service entrance, with a short connection
    from the 2nd ground block to the power service grounding electrode
    conductor. Cable distribution to the building from the 2nd ground block.
    Similarly a secondary phone protector block could be installed adjacent
    to the power service. I have never seen this suggested but it seems like
    a practical way to get a single point ground.

  17. w_tom

    w_tom Guest

    We had been through this before in Those
    who once strongly advocated 'point of use' protectors (ie ex-GE
    employees) have backed off that recommendation. One example is an IEEE
    paper by them about an "Upside Down House". Francois Martzloff and
    Thomas Key in 1994 wrote in "Surging the Upside-Down House: Looking
    into Upsetting Reference Voltages" :
    Why do those who once always advocated only 'point of use'
    protectors now change their tune?

    Effective protection at the appliance is already inside appliance.
    If components inside 'point of use' or plug-in protectors were so
    effective, then those $0.05 parts would already be inside each
    appliance. Once they were installed. But since those parts (currently
    inside plug-in protectors) were not effective inside appliances, then
    appliance manufactures use only other well proven techniques

    This internal appliance protection assumes a transient will be
    earthed before entering a building. That being the purpose of a 'whole
    house' protector that also costs tens of times less money per protected
    appliance. If not earthed, then a transient can overwhelm protection
    inside appliances. As Martzloff, et al noted, excessive voltage can
    occur even "perhaps because, surge protective devices are present at
    the point of connection of appliances". Martzloff was once a major
    promoter of 'point of use' protectors.

    Do we spend $20 to protect every appliance - or do we spend far less
    money to enhance earthing? Per dollar, earthing provides major
    appliance protection. Point of use protectors - if for no other reason
    - cost massive dollars and provides little benefit. Again, if it were
    effective, then those same parts costing so little (and selling at
    exaggerated profits) would already be inside appliances. Shunt mode
    protectors are only as effective as their earth ground. Plug-in
    protectors have what for earthing? So instead, plug-in manufacturers
    forget to mention earthing (since earthing is not provided by plug-in
    protectors) AND forget to mention protection already inside appliances.
    Profits are just too large to be fully honest. Protection is
    earthing.... the most critical component in every protection 'system'.

    How does a shunt mode protector do anything effective when it does
    not shunt to earth? Manufacturer hopes we don't ask that question.
    Plug-in protector manufacturers, instead, cite protection from
    transients that don't typically cause damage - and hope you don't
    notice. They hope you never learn why earthing is so critical -
    profits being too outrageously high to be fully honest. Even those who
    once only recommended 'point of use' protectors are now changing their
    tune in IEEE papers - citing advantages of 'whole house' protectors -
    that also cost tens of times less money per protected appliance.

    Why would an objectionable voltage not exist in a ground ring? Well
    repeatedly cited citation shows a bad, good, and ugly
    solution. The prefered solution puts everything at a single point.
    But the OP does not have every utility approaching a single point.
    Therefore an uglier solution is useful. That solution does make the
    earthing more conductive. It does provide a single point ground. It
    is a major improvement over what he currently has. Others who can plan
    a new house must avoid what the OP has - before footing are even
    poured. What the OP currently has would explain (and may be reason
    for) his many years of electronics damage. Provided is an effective
    and easier solution - since utilities don't like changing services
    without big buck bills.
  18. Bud--

    Bud-- Guest

    This is exactly what a multiport plug-in point-of-use surge protective
    device protects against. These are called Surge Reference Equalizers by
    the IEEE. Another paper specifically about SREs is link. pdf
    This paper is currently available from the NIST in a collection with a
    forward by your good buddy François Martzloff (who was an author of this

    I have provided links to an IEEE paper and and 2 NIST papers, all
    current, that recommend plug in surge protectors. In previous threads
    (and this one) I have not seen any links supporting your view. Its you
    against the IEEE and NIST (and a lot of other people).

  19. w_tom

    w_tom Guest

    Default makes an important point. Defined for a house is a 'whole
    house' protection system - also called secondary protection. Primary
    protection is provided by the utility, as default has described.
    Pictures that demonstrate inspection of a Primary protection system:
  20. w_tom

    w_tom Guest

    In Bud's cited 1993 paper on 'surge reference equalizer' or
    'multiport protector', Martzloff, et al defines a problem common in
    most residences where traditional plug-in (point of use) protectors are
    IOW power strip protector simply creates one of many defined
    transient problems that contribute to electronics damage:
    Suggested by that 1993 paper is a multi-port protector - surge
    reference equalizer- that only uses a principle called equipotential.
    First, $20+ to protect only one appliance; without conductivity to

    Second, defined are six ports that must be part of equipotential.
    A multiport protector must provide equipotential for all ports at a
    point inside the room. But as posted back in April - it does not
    provide equipotential because some ports are not part of that
    equalization technique. Where is concrete floor or linoleum tile
    included as part of multiport protection? Where are baseboard heat,
    air ducts, wall paint, or furniture included? That paper calls them
    'enclosure ports'. Any one port not part of a multiport equalizer
    means equipotential is compromised.

    To have equipotential inside a room means the entire room must be
    constructed to provide equipotential. Therefore we locate
    equipotential where equipotential is easy to achieve.

    And third, protection must provide both equipotential and a
    conductive path to earth. Since neither equipotential nor conductivity
    alone is sufficient, then a protection 'system' must do both. That
    'point of use' protector provides all but no conductivity - no
    effective earthing.

    Not only is equipotential compromised in a room not constructed to
    provide equipotential. Also the 'system' does not provide necessary
    conductivity to earth. All this and $20 or $80 to ineffectively
    protect one appliance? How is that effective?

    Meanwhile his 1993 paper then moves on to describe another protective
    Described is a 'whole house' protector. Note how it is described:
    'Best' protector recommended by Bud's 1993 paper costs about $1 per
    protected electronics. It does provide equipotential to every room (by
    making the equipotential point beneath the entire house rather than a
    point inside one room). And it provides a best conductivity to earth.
    Both requirements - equipotential and conductivity - are necessary for
    a 'best' solution. Surge reference equalizer ... AND more ... that
    is provided for a whole building rather than just for one appliance is
    called 'whole house' protection. As that 1993 paper notes, part of
    that 'whole house' system is already in telephone NID.

    The same author later states in a 1994 paper:
    Curiously, this interest in a 'whole house' solution coincided with
    post 1990 National Electrical Code changes that require earthing an AC
    mains breaker box and all other incoming utilities to a common point.
    A common earthing point that must be adjacent - a short distance.
    Although code is only for human safety, still, those changes make
    'whole house' protection more effective and simpler to install. To
    provide surge reference equalization - AND more.

    What does not change? A protector - the protection 'system' - is
    only as effective as its earth ground. A fundamental demonstrated by
    IEEE papers cited in that previous April discussion.

    The OP (jakdedert) suspects years of electronics damage due to
    transients. His earthing system is defined as defective. Earthing
    that violates principles of single point earth ground and post 1990
    code. Provided were examples from a utility (, further
    details in how to create single point earthing for that building, AND
    how to connect each utility to that earthing. Connections either by
    hard wire or by effective protectors that even cost less - 'whole
    house' protectors. Conductors for earthing and those 'whole house'
    protectors are even sold in Lowes, Home Depot, and electrical supply
    houses. A solution even recommended in Martzloff's 1993 paper and a
    following 1994 IEEE paper. A solution that is routine in virtually
    every telephone switching station, commercial broadcasting, emergency
    response centers, and now in homes constructed to protect household
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day