Voltage on Cable Line

To All:
I just moved in to a townhouse - built in 1985. I've been having cable modem problems since it was installed a few weeks ago - intermittent
connection problems and speed issues. After the cable company replaced everything from the modem to the outside pedestal, they tell me it's a grounding problem with the electrical service in my house.
Apparently, voltage is leaking out of the back of the modem along the coaxial cable. They attributed this to inadequate grounding of the house. It's reproducable on several circuits, leading me to believe that it's the house and not a localized thing inside. It's only 5-10 volts.
I tried running a separate ground wire from one of the outlets we tested to a cold water line where it enters the house but that had no effect.
Anyone have thoughts on how to address this? The cable guys suggested I drive in a new grounding spike 10 feet down and re-ground the main service panel. I have shale a few feet down, so I'd prefer to avoid that if possible. Any other options on how to correct this problem?
Please reply to snipped-for-privacy@gmail.com - I'll get the answer quicker that way, considering I can barely get online from home, but at least I can get e-mail on my cell. Thanks.
Matt
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Cable companies are required to install a grounding interface somewhere on your coax before it enters you house. This is primarily for lightning protection. I've seen them connected to electrical boxes, cold water pipes (assuming metallic, conductive pipes) and driven copper or copper-coated ground rods. Any grounding problems (leaks?) beyond that point are your problem and must be solved at your expense.
If your electrical system ground is open or defective, you could have a dangerous situation that could lead to electrocution or fires. A qualified electrician should be called in to check this, if in fact, you are convinced that this is the problem. There are inexpensive testers that can at least verify an open ground problem at the electrical outlet.
The cable modems i've seen either have separate wall wart power supplies or just two prong (non-polarized) ac plugs. There might be other types (with 3 prong grounded plugs) that I am not aware of. There should not be any current flowing from your cable modem to the shield of your incoming coax. The safety ground (third prong on the AC plug) should not have anything to do with the signal ground (coax shield) on an incoming signal cable.
Beachcomber

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2 Jan 2007 13:19:56 -0800, snipped-for-privacy@gmail.com Gave us:

Check to make sure the connection to your HOUSE ground rod/water pipe is good and tight.
If it is a ground rod, slowly dump a couple gallons of water on it, and see if the problem goes away. Also, the FAULT wire in your electrical outl;et (third wire/green) and the tie point back at your distribution panel. You should have an electrician check those though if you are not electrically inclined or an electrician, etc.

That is the best ground for the structure you can get typically. ALL the connections in the service/distribution panel should be checked for tightness too, and this should ONLY be done by an electrician.

Your current ground rod may be a fudged, shortened version for that reason then. It may be the cause of the problem. An electrician has access to a tool that pipes a sound wave down the rod and can tell the tester how many feet of ground rod there is. It stems from the cable industry installers cutting corners, and making cable companies liable since it does not comply with code.

Check the cable on your house drop, or lock box if it is an apartment. Hard line cable has a 40 Volt DC feed on it, and it should get trapped out by the pole taps, but your voltage could be coming down the coax. Not as likely, but possible.

This is Usenet. You ask in these forums... YOU need to come back to these forums to gather your replies.

So. Many Usenet posters prefer to remain anonymous, particularly as it relates to email.

If you want answers from queries made in this forum, you should expect to get your replies here as well. The information is not merely for you.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Something is "fishy" here.
In most places in the US the neutral is grounded at the transformer. Even if that particular "ground" ain't so hot, the neutral for the primary side provides a connections to other grounds.
If the neutral is bonded to the house ground I just don't see more than, say, 5 volts difference between "grounds" and certainly not even close to 10 volts.
The neutral might actually be broken and the "bonding" is providing the neutral current. If the only L-N loads are relatively light, the unbalance might not be noticed.
If the OP has a VOM, he should measure the L1-N, L2-N, L1-G, L2-G, and N-G voltages while switching on and off a solid 10 amp load like a toaster. The results should give a pretty good idea of where the problem lies.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 7 Jan 2007 21:29:23 -0500, "John Gilmer"

Every telephone pole I ever saw had the big #10 Ga minimum bare solid copper strand in a groove cut down the face of the pole, Some were insulated wire, but ALL were tied to a grounding rod sank right at the base of the pole.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 7 Jan 2007 21:29:23 -0500, "John Gilmer"

Good diagnostic procedure.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
- I'll get the answer quicker that

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Do you have any communications experience at all?
Either the cable installation has a poor ground back at the ground block, or a bad fitting was cut, or the house wiring has grounding issues.
Learn to snip too, twit.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Ok, thanks to everyone for posting so quickly.
The thing is - the voltage isn't coming IN through the coaxial cable. If I hook up the modem to house AC - via non-polarized 2 prong plug, incidentally - and measure the center pin of the coaxial in the BACK of the modem, not connected to anything else except house power, to ground, I get between 7 and 13 volts. There's no way the problem is coming IN through the cable because the power is coming out of the back of an unconnected modem.
That being said, the ground rod is connected solidly - I checked that. I have no house drop - the electrical in my neighborhood is below ground. But that doesn't matter because the voltage cannot be coming in through the coax - see above.
Tonight I tried connecting the modem to a UPS system, but the voltage coming out of it is only 85 VAC, so I can't get a reliable answer from that. I also connected to several other outlets, including a surge protector on my TV, and I get the same results.
However, I ran the following test - I connected an old surge protector I had with coaxial protection on it. I plugged the cable modem into house power. The cable in the back was NOT connected to the incoming cable feed. But I fed it through the surge protector and measured voltage on the other side and it was negligible. AND that seems to have made my internet connection better.
I can't see how a house ground problem can be causing these problems. But I'm not an electrician. I'd appreciate any insight you all have on this. I don't doubt that it can still be the equipment the cable company is providing, but it can't be the incoming cable feed.
And by the way, it was a courtesy request to CC a copy to my e-mail. It certainly doesn't take any more time or effort to CC an e-mail address on a posting. No need to get snippy about it.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2 Jan 2007 18:14:46 -0800, snipped-for-privacy@gmail.com Gave us:

I meant the house drop for the cable. Where it attaches to the house should be a "P hook" (typically). There should be an "expansion loop" of additional cable, and at some point between the attachment and the entry into the structure should be a grounding block, and ground wire. Sometimes it is in the attic or the inside the house side of the entry point, but not usually.
THAT ground point should be your house ground, OR its own ground rod OR the cold water service.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2 Jan 2007 18:14:46 -0800, snipped-for-privacy@gmail.com Gave us:

Sounds faulty.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2 Jan 2007 18:14:46 -0800, snipped-for-privacy@gmail.com Gave us:

The word for today is baseline noise. You could also have an open shield in the coax run at SOME point in your house or as far as the taps on the splitter outside.
The cable co here used to place a trap on the cable modem segment of the lines, I assume to cut any digital noise from getting back into the cable channels... I dunno.
They don't use it any more though.
They did have to come place a specific splitter in our building's lock box that "passes DC" so I could get digital phone service. My Net hooks have been screwed up ever since. I get in streams fast, but a lot of retries on outgoing packets. Not like that before.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2 Jan 2007 18:14:46 -0800, snipped-for-privacy@gmail.com Gave us:

As I said, I like to keep my email to myself, and sending you one means revealing mine, even if only to you.
It isn't snippyness so much as it is an obligation for respect to the forum. You asked here. It isn't asking too much to expect you to complete the dialog here.
Dude... I have been calm the last few days. You should see me when I am snippy...
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@gmail.com wrote:

Where "ground" is what.
If measured to power system ground, the modem is not connected to the power system ground and will float. If measured with a digital meter, the meter impedance is very high and the meter can show a phantom voltage produced by a very low leakage current.
If measured to the coaxial connection around the center conductor there shouldn't be 7-13V.

A meter that is not "true RMS" will read wrong on non-sinusoidal waveforms, but if I am thinking right should read high. If you just connected the modem, the UPS may need to have a larger load to produce the right voltage (a guess).

If you use a plug-in suppressor all interconnected equipment (like computer and printer) should be connected to the same suppressor, or interconnecting wires, like LAN should go through the suppressor. Other external wires like phone, CATV, ... also should to go thorough the suppressor. A plug-in suppressor works by clamping the voltage on all wires to the common ground at the suppressor.

Verify that the ground block at the cable entry is connected to the grounding system for the power. It can't (in the US) just be connected to its own ground rod.
If you have underground metal water service, that is a better ground than a rod. If at least 10ft of underground metal pipe it has to be included as a grounding electrode in the US.
-- bud--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Gave us:

Not true. We were REQUIRED to drive OUR OWN 8 ft rod whenever we did a post wire on an apartment complex.
This was Time Warner (Warner Amex CUBE system at the time)
Single home drops MAY be attached at the house tie point, or a ground rod CAN be driven. It is for lightning protection, and the problems being discussed in this thread are not related to that.
That alone points toward the issue being with HIS power wiring, and NOT the cable system.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
JoeBloe wrote:

Earth ground serves many functions. It is for human safety. National Electrical Code is concerned with human safety. NEC demands that cable and AC electric share a common earthing electrode. It even states a maximum connection length to that earthing electrode.
Then earthing also addresses lightning protection. For lightning, the installation must also do things that exceed what NEC demands. For example no sharp bends in the earthing wire, no splices, no bundled or adjacent to other earthing wires, not through metallic conduit, etc. And also for lightning, all earth utilities (AC electric, cable, satellite dish, and telephone) must be earthed to the same electrode.
Cinergy even demonstrates how to do this when somebody screws up - brings all utilities in at separate locations: http://www.cinergy.com/surge/ttip08.htm
Good news is that cable companies are finally telling installers to earth - also for lightning. Previously, cable installations were some of the worst examples of professionalism ever seen. But the earthing must include each and every incoming utility. The earth wire must be short - 10 feet or less. And that earthing is necessary for both transistor protection (lightning) and for human protection (NEC).
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

For power line wiring and distribution.
The cable installation is concerned with human safety from a lightning strike point of view.
Note that most cable modems as well as most set top cable converters are only two conductor fed devices, and no safety wire/fault return exists. for them.
He likely has an issue with his neutral lines.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
In the neighborhood, phones started ringing constantly. Fire and utilities were called. One house suffered a neutral wire failure. Utilities were not properly earthed. So the house sought other grounding paths. One house had cable not properly earthed at its service entrance. Electric company emergency responder borrowed IR vision from the fire department. He could see coax cable inside walls red hot. Why? Had the cable been earthed at service entrance, then cable inside house would not be used to ground the first house. Missing earth ground almost meant a fire in the second house. Fortunately, that same missing ground caused phones to ring constantly. Example is earthing for human safety.
NEC only defines lightning protection in human threat terms. NEC does not care if lightning protection works or does not. NEC does not address transistor safety. But your cable must be earthed so that humans are not electrocuted AND so that transistors are protected. Those are two separate functions - earthing to protect both humans and transistors from many different electrical sources.
Many installers don't understand the concept. Earthing sufficient for human safety would not be sufficient for transistors safety. Therefore we earth to meet post 1990 NEC requirements. Then we enhance that earthing also for lighthning protection. No earth ground means no lightning protection. But earthing does more than lightning protection.
Cable box, etc must not be safety grounded by the cable. A cable box safety grounded by its cable is a threat to human safety. TV and cable box should even contain galvanic isolation as part of what some call 'double insulated'. Just another reason why cable cannot 'bond' the TV or cable box.
If OP had a neutral wire problem, then he has light bulbs clearly changing intensity as major appliances power on and off. But again I am posting what was apparently ignored. There is nothing in that OP's posts to suggest a neutral wire problem. There is plenty to suggest the earthing is not properly installed and that other internal wire problems exist. Provided were measurements to find that failure. The OP never posted back.
. Meanwhile, learn various reasons for grounding all utilities to a common earthing point. Human safety. Transistor safety. Even eliminates ground loops and other noise in electronics. Multiple functions using the same wire.
Scary is that you believe cable wire is acceptable for safety ground to a cable box. Absolutely not. Reasons why should be bluntly obvious to you and your peers - a human safety issue that should be obvious to all installers.
JoeBloe wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Why I even respond to a retarded top posting Usenet twit, I'll never know. If you know so much about convention, learn one!

You're nuts. Phones ring when RING TONES get sent onto their line.
CITE a red hot IR cable story that has photos, jerk. Don't just create a "possible" scenario.

No shit.

Those are not all the reasons, but you are close.

No shit? Personnel... equipment...
Were it not for you, I'd have never known.... Sure, bub.

It had nothing to do with concept. It was our installation policy.

You're not too bright. 40 mA on the skin, across the chest can cause fibrillation. Only 10 mA into puncture wounds (through our very salty conductive blood), and only 2 mA directly applied to the heart in open heart surgery procedures.
With P-N junctions (doesn'y have anything to do with transistors per se) the determining factor is voltage, not current. When the voltage reaches a point where the junction interface gets breached failure occurs, or damage is induced at the site that causes failure in the future far sooner than a non-assaulted junction.

The installations I remarked on were, as stated, back in the late seventies. I also stated that I was quite sure that their policy had likely since changed.

As it relates to power, yes. As it relates to those comm lines, however, the main purpose is the lightning protection, and that protection is for both humans and equipment.

Not in the proper setting, but in a household setting with all the typical associated gear, sure. All hypothetical though since no cable boxes are made with fault lines. My early cable modems may have been, however. I am not talking about the store bought variety after the spec passed either.

Not if he uses fluorescent replacements everywhere.

What is actually suggested is that he may be using a sad meter, making an improperly setup measurement.

He needs an electrician to inspect his service panel, and ideally, load test each branch to identify any anomalies, and more easily pinpoint the source location.

I suspect methodology errors at this point.

I don't need to learn it. I was merely iterating what that company's policy is/was.

A terminology in common use.

A never used term.

I've built HV supplies that have 0.0006% ripple. I know how to ground circuits. From the very smallest pico amp leakage intolerant PMT supplies, to the 180kV X-Ray driver in use at all the airports.

Watch out, you are getting loopy.

Show me where I said that.

I no longer perform cable installations. That was decades ago. I am sure installers today are far better trained as there are more in house personnel than contract installers in some systems.
With the previous post below, all I can do is close by calling you the utter TOFU retard that you are.
Snipped TOFU from the TOFU retard. Get a clue Tom.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
JoeBloe wrote:

The US-NEC 820 covers CATV. 820.100-B covers the earthing electrode to be used. Required connections are all to the power system grounding electrode system including the grounding electrode conductor, or points immediately adjacent.
820.100-A-4 requires the connection from the CATV ground block to the power electrode system above to be 20 feet or less.
An exception to 820.100-A-4 - for 1 and 2 family dwellings only - allows a separate rod to be used if the connection above cannot be made in 20 feet. But it requires bonding the CATV rod to the electrodes in 100-B above with a #6 or larger wire.
The NEC requires all grounding systems to be bonded together. That includes phone and lightning rods. I am not aware of any exceptions.
The IEEE published an excellent guide on surges and surge protection available at: http://www.mikeholt.com/files/PDF/LightningGuide_FINALpublishedversio ... Starting on guide page 31 is an illustration of the problems if the wire from CATV ground block to power system electrode is too long. If the ground block had been connected only to a rod, and not bonded to the power system, the problem would have been much worse. As an approximation about 70% of the voltage drop from a rod to absolute earth potential occurs in the first 3 feet from the rod. If earthing a lightning surge, the CATV rod would be at a far higher potential than the power system earthing point. And if the power system was earthed only at a rod, and the power system earthed a lightning surge, the power system would be at a far higher potential than the CATV earthing point.

Ground rods at different points are not necessarily at the same potential. And with "shale a few feet down" all the earthing is suspect. Bonding the systems puts them at the same potential.

You seem to be the only person convinced it is not a cable problem.
-- bud--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.