usually the cause is a bad AMP on the cable company lines on teh poles.

you say that they are grounded to a rod at your house, but are they ground to THE main electrical ground? just because they go to a ground rod does not mean they are on your electrical ground. house we used to live in the cable and telco had a rod right next to teh electrical rod. technically they were grounded, but not PROPERLY. this resulted in similar floating ground issues. i corrected the grounds and all problems went away.

you could also have internal wiring problems with either your cables or your electrical. a bad neutral will give you some wierd voltage issues the same as a bad ground.

Are you using a digital meter? If so try using a analog meter. Digital meters are notorious for measuring false voltages when none are present.

The OP says that the CATV drop is grounded with a "ground block" and verified to be a low impedance connection to (presumably) the service entry ground. I am guessing that the measured V is on the coax shield on the subscriber side of the ground block when same is disconnected from the CATV ground block. This is a fairly common observation if using a high impedance meter. Most TV sets, cable boxes etc. have bypass capacitors inside that couple very lightly from the chassis to the coax shield. If the voltage will not show up on a Simpson 260 or comparable meter (it most likely will not), it isn't going to hurt a thing.

The most common things I have seen that place substantial V on the shield from the subscriber side are defective electric water heater elements (mostly in mobile homes) and an open neutral (once again mostly in mobile homes), TV sets are a very distant third- I saw this once in over 10 years in the CATV biz.

As an aside, if your "cable technician" is unaware of the information above, he/she is either (A) very inexperienced, (B)lacks any appreciable knowledge of electronic theory/practice, or (C)both of the former.

There is a very small chance that any defect in the CATV outside plant would put voltage on the drop, and even less that it would be on the shield.

... A COMMON "problem", although it really isn't one.

For various reasons, the "ground" (shield) of the RF input connector on a TV is connected to the TV's power supply ground (and needs to be, in order for the TV to work).

Many older TVs (and probably any TV manufactured without line-level audio/video inputs, if they still make such a deprived animal) contained power supplies that were/are not in any way isolated from the AC line. Therefore, the "isolation" is provided at the RF input by, typically, a couple small capacitors.

SOOOO you've got a "ground" that isn't one, at an unidentified mystery voltage (quite likely 60V), connected to the F-connector shield via a small capacitor. OF COURSE if there is no "load" you're going to find AC on that connector.

The fact that the Time-Warner guy doesn't already know this disturbs me greatly. There isn't a problem, and there's nothing that needs to be fixed. Unhook those TVs for the duration of the cable guy's presence to shut him up, then hook them back up after he leaves.

Time Warner employees/contractors have IQ that is 1 point above that of a box of rocks. I just replaced two bricks on a rental house that TW broke by drilling through a conventional framed wall from the inside and letting the bit break through outside. Holes down through the hardwood floor inside only 2" apart. They drove big bucket truck over my septic tank- after I told them not to do so! I have some farm property. TW has their cables on the POCO utility poles. In some places the cable goes down (up if you prefer) the pole to transition from aerial to buried. Installers used a plastic "U" shaped cover about 8' long to protect the (big, about 7/8") coax mounted on the pole. Drove lag bolts into the pole to hold the cover. But used a hammer on the bolts instead of a wrench. That broke out the slots for the bolts and the wind blew the cover off. Cover laid there for over a year, requiring dismounting equipment and moving it around for working the field. This is beside one of the main roads where TW trucks pass almost daily. Finally I called TW, told them about the whole situation with the cover and that the cover would be removed and thrown away in one week if not reattached to the pole. The woman on the phone told me that TW does not serve that area! I was very tempted to use the axe that I carry in my truck on that coax and see if it mattered to TW! I discarded the cover a week later.

Thanks for the replies, and sorry for the delay - been away/busy all last weekend and this week. I believe there is a genuine problem with the cable TV cable from my home. It appears the cable tech might be on to something.

This afternoon the cable modem kept loosing signal (flapping) frequently. It almost seemed to happen every time the air conditioning came on, and sometimes when it went off. After disconnecting the cable TV cable from my house (i.e. the one attached to my TV's etc) from the cable company splitter outside there were ZERO problems with the cable modem for the rest of the afternoon. So it seems like the cable tech might be sort of correct. The problem has something to do with that cable connected to the TV's etc.

Using a digital meter I measure 47.4 ac volts from either the signal (center) wire or the shield on the cable coming from my house to the cable/electrical ground outside at the splitter. Using the same meter I also measure 358 ac micro-Amps from the signal (center) wire and 615 ac micro-Amps from the shield on that cable. This is about 29mW coming from the cable connected to the TVs in the house.

I had an electrician check the electrical panel this week, but they really did nothing more than check all connections were tight and voltages etc were normal. They did not find any issues. The incoming cable splitter is grounded to the electrical ground outside.

Right now I'm away to check other items, including HVAC connections. Any other suggestions to check?

Microamp amp leakage currents are not really of any concern, and any equipment having problems from them is not operating correctly.

What is a "broadband cable modem flapping issue"?
Was the cable modem operating or not?
It should not have any problems with 60 Hz leakage.

Cable modem flapping, as I understand it, is when the cable modem keeps loosing and regaining the signal (or signal sync). When it is in sync it works perfectly. This flapping problem only started in the last few months, and seems to be getting worse. The modem has been replaced once. Right now it is not happening because the cable TV cable from my house is disconnected from the external splitter. So that is why I think there is some linkage between HVAC, cable TV cable and the modem.

According to my meter the voltage on the cable TV cable is not 60Hz but something more like 40 kHz. So it might be more like induced noise from something than AC hum.

ok.

first note, your math is flawed.

You measured the open-circuit voltage, followed by the short-circuit current. During you open-circuit measurement, you've got almost 50VAC, but since it's open-circuit, there is no current, hence 0 mW. During the short-circuit measurement, you've got all those µA of current but (since it's short-circuited) no voltage. Again, 0 mW.

I haven't been around cable modems too long, but speaking from my experience with the digital cable service around here - that extra "leakage" *CAN* piss off the electronics, if the ground isn't really good.

ie, the ground connection to the input side of the first splitter (where the cable modem taps off). A poor ground here means that the 'stray' voltage from the TVs tends to mess with the cable modem (ground loop maybe? I don't know why - it just does).

This does not indicate that the voltage itself is a problem. I can re-word this any way you like, but when it's all boiled down, you're not going to get rid of that voltage unless you care to start throwing away and replacing televisions, and being very picky about the replacements. I'm talking about standing in the electronics department with a multimeter.

There may very well be a grounding issue, or a bad connection, or just a junk cable.

Putting this a different way, the "stray" voltage ISN'T the problem - it's just causing the problem to show itself.

Thanks for that great post. I'll redo the ground connection since I'm not sure I measured that in a manner that would give a good reading.

The grounding of the cable to the earth does nothing for leakage on the cable.
The earth is not an effective ground source at the voltages present.
The earth ground is for lightning protection (and even a buried cable can be affected by lightning strikes and displacement currents as charged clouds move above the earth).
How are you measuring 40 KHz?
The only way you are going to find the source of the problem is by hooking up the cable modem and nothing else.
Verify normal operation.
Then connect other items one at a time until the cable modem stops working.
The problem is liable to be a weak signal as much as anything else.

Speaking from an electronics perspective, and ignoring whatever you might have learned in whatever school electricians go to, having a "zero volts" reference point is important.

At some point, the ground on the cable wire is supposed to be hooked to the ground from the electrical service - granted there may be some rods, pipes, and wires in between, but they're hooked together.

Your AC "leakage" through the tiny capacitor inside your TV gets very effectively shorted to ground through this, and goes away.

Problem solved. Especially since most cable companies (Time Warner included) install a two-way splitter as the very first object in/on the residence, with the cable modem coming off one side of it, and the other side going to the other (bigger) splitter for TVs. The grounding usually takes place very close to this first splitter, and is thus very effective in preventing the cable modem from "seeing" whatever leakage may take place elsewhere in the building.

RESOLVED! I did fix the "voltage" problem, which helped but was not the root cause. All the audio/video equipment uses 2-wire power plugs and was not grounded except through the cable TV cable to the outdoor splitter. Grounding at the receiver caused the voltage at the splitter to be almost zero, which meant less broadband problems but it was still happening. After more cable tech visits they replaced the cable drop from their network box to my house - and the problem was solved. The old drop was 20 years old and was RG5 like cable. The new drop is at least RG6 (or better? it is very thick). The broadband modem now has totally consistent and unchanging signal levels (usually visible on http://192.168.100.1/). Previously the upstream (transmit) power levels would vary multiple times during the day from 42 dBmV to 48, 52, or even 54, which are not good levels. Now it is always 42 dBmV. Hopefully this resolution helps someone else in the future.