The cable company demarc coax was burned and melted. New cable installed. Said the problem was bad house wiring. I checked all the plugs and all the lighting in the house. Found no real problem. Then disconnected the house coax cable and tested voltage to ground and found 52 volts. Then I turned off each breaker in the dist box one at a time to see if the voltage went away. Found that the back bedroom breaker did remove the voltage, from 52 to 3. SO i then disconnected every single load on any outlet. Voltage with nothing plugged in was 3 volts. (Cable shield to ground) Knowing that the problem was in the back bedroom I focused on that. I first plugged in the cable company modem, voltage went to 29 volts. I plugged in the cable company router and voltage immediately went to 52 volts.

QUESTION: Has anyone else experienced this problem and what did they do to fix it? The power supplies only have two prongs connected to AC, no ground, so where is the ground potential coming from if not the internal solid state circuits?

  • 2 things come to mind. The cable lines coming into the house are supposed to bonded for lightning strikes and the cable guys often do it wrong. And I have heard that it's possible for induction to data lines from running parallel to mains but have never actually witnessed it. Otherwise it could be shorted somewhere and not grounded and holding static potential. Be careful to not get zapped. 24ish volts is considered to be the upper limit for contact before resistance of your skin breaks down.
    – Joe Fala
    Feb 20, 2019 at 23:54
  • Are you renting that equipment from them? My experience is they'll let you bring that into an office and swap it at no cost. However, first I'd check the output voltage of any wall-wart power supplies in use. Feb 21, 2019 at 0:23
  • Can you run a wire from the cable company demarc point to your main breaker box or grounding electrode conductor? Feb 21, 2019 at 0:30
  • @Harper I thought about that, but I thought, for both devices too be adding voltage to the CoAx would be like winning the lottery. If it's free to swap them out then why not try.
    – Joe Fala
    Feb 21, 2019 at 4:52

2 Answers 2


The voltage measured on the coax doesn't tell the whole story, especially if it's measured with a high-impedance digital multimeter. The two cable boxes will have isolated power supplies inside, meaning their low voltage circuitry is isolated from the high voltage mains. But there can be capacitive effects that would cause a voltage to appear between the mains and the low voltage circuits (ie the coax connector) even though there's no path for a meaningful amount of current to flow.

To check whether that voltage really means anything you'd need some kind of load. I'd grab a resistor in the range of a few hundred to a couple kilo ohms from my parts box, put it between the cable shield and ground, and measure the voltage again. It should drop to near zero.

I'm trying to think of an alternative that a "normal" person who lacks an electronics parts box might try instead.. If the meter supports current measurement, one could measure how much current will flow between the cable shield and ground. Use an ordinary incandescent or halogen light bulb (not LED or compact fluorescent) in series to limit the current in case there actually is a fault in the cable boxes. A regular household bulb, a large C7 or C9 bulb from old-style Christmas lights or a night light, etc will do. Using your third and fourth hands (or an assistant) put the meter into its highest-range AC current mode, hold the coax cable shield to the shell of the bulb, and touch the meter leads to the button terminal of the bulb and the electrical ground. Repeat with the meter in DC current mode. (Don't forget that meters usually require the red lead be moved to a different socket for measuring current, and definitely remember to move the red lead back to the voltage socket immediately when finished!)

If the voltage didn't drop to near zero in the load resistor test, or if the meter measures more than just a few mA in the current test, that could be a problem and replacement of the cable boxes could be advisable. If these tests check out ok it would seem that the energy that melted and burned the demarc probably didn't come from those two cable boxes.

  • that's really interesting and I hope Tom Stinson tries your suggestion and reports back.
    – Joe Fala
    Feb 21, 2019 at 4:56

Greg Hill has an excellent answer, but before I would go to that length I would like to say this. I have worked with customers who have had similar problems and I seem to want to say that the problem is usually not your wiring. The problem is usually the cable installer who says "the problem is bad house wiring". I have to consider the source.

Cable installers are usually given some tooling and about 2 days of training and sent out to install systems. I will say that there are some who might be highly skilled but I have not found this to be the norm. Most are not skilled electricians or skilled electronic technicians. Most are not even employees of the cable company but contract labor or contractors (the Gig Economy). I know this because most electricians are paid better since they have more skill. I wouldn't be surprised if they were trained to give bad wiring answer to get you off of there back since they would be paid per installation and not warranty repair.

What I do know is that it takes a lot of power, not just voltage to melt insulation, and I would suspect a large surge or lightening the utility side or some poorly installed equipment. You were right to contact stack exchange or some other professional source and seek a second opinion. Now that that's done and if the cable problem hasn't been fixed. I would contact the cable company direct and tell them what you have done and asked that they send a qualified technician who is actually an employee out there to repair their problem.

Good Luck.

  • 1
    Today the cable company replaced the modem and router and the voltage was about the same on the new equipment. I agree that this voltage cannot support a load because there is no shock or arc when connecting it or touching it. I am going to try a resistor in my multimeter connection later today. The cable guy seems to think that this reading will always be in this range for this equipment so i am going to measure other houses in this neighborhood to be more informed. Feb 21, 2019 at 18:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.