info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Hum from Cable
45 views
Skip to first unread message
Bob Simon
Aug 25, 2016, 10:24:18 PM8/25/16
to
I recently moved and connected my new Cox Contour DVR to an audio receiver and TV. (A Blu-ray player is also connected to the receiver and TV; this may or may not have any relevance to my issues.) There was a pretty bad hum so I inserted an isolation transformer between the cable feed and the DVR, which eliminated the hum. I measured the voltage between the outside of the cable connector and safety ground with a cheap VOM and found 0.2VAC and 1-9 mVDC. I have a number of questions that I hope someone can answer for me.
Is this voltage difference the cause of the hum?
Was my measurement approach appropriate?
Do the measured voltages indicate that the cable is improperly grounded?
If the cable were to be bonded to the house neutral with 3/0AWG, would this likely eliminate the need for the isolation transformer?
Whether it would eliminate the hum or not, should I have it done?
Does the cable company have any obligation to do it in order to comply with regulations?
Bob Simon
New Orleans
Is this voltage difference the cause of the hum?
Was my measurement approach appropriate?
Do the measured voltages indicate that the cable is improperly grounded?
If the cable were to be bonded to the house neutral with 3/0AWG, would this likely eliminate the need for the isolation transformer?
Whether it would eliminate the hum or not, should I have it done?
Does the cable company have any obligation to do it in order to comply with regulations?
Bob Simon
New Orleans
stra...@yahoo.com
Aug 26, 2016, 1:01:49 AM8/26/16
to
http://www.rmscommunications.net/mi2120v_n.htm
The Voltage in my system is similar to yours so safety really isn't an issue. The isolator solved the problem and as far as I'm concerned, it's a permanent solution.
G²
whit3rd
Aug 26, 2016, 2:41:10 PM8/26/16
to
On Thursday, August 25, 2016 at 7:24:18 PM UTC-7, Bob Simon wrote:
> I recently moved and connected my new Cox Contour DVR to an audio receiver and TV. (A Blu-ray player is also connected to the receiver and TV; this may or may not have any relevance to my issues.) There was a pretty bad hum so I inserted an isolation transformer between the cable feed and the DVR, which eliminated the hum. I measured the voltage between the outside of the cable connector and safety ground ..
Good; measurement is the right thing to do.
> Is this voltage difference the cause of the hum?
> Was my measurement approach appropriate?
Yes!
> Do the measured voltages indicate that the cable is improperly grounded?
from lightning, and isn't supposed to pass a 'hum' test.
> If the cable were to be bonded to the house neutral with 3/0AWG, would this likely eliminate the need for the isolation transformer?
bonding allowed by code is inside the circuit breaker box. No, confusing
the matter with multiple straps won't help. Any cable with multiple
connections to ground is a ground loop, and those CREATE hum.
If your audio path can be replaced with digital or TOSlink, that would
be another solution (and would simplify the wiring instead of adding
to it). RCA-plug stereo connection is common, and commonly causes
problems, in part because (unless one is clever) it creates ground
loops.
Dave Platt
Aug 26, 2016, 2:50:39 PM8/26/16
to
>I recently moved and connected my new Cox Contour DVR to an audio receiver and TV. (A Blu-ray player is also connected to the receiver and TV; this may
>or may not have any relevance to my issues.) There was a pretty bad hum so I inserted an isolation transformer between the cable feed and the DVR, which
>eliminated the hum. I measured the voltage between the outside of the cable connector and safety ground with a cheap VOM and found 0.2VAC and 1-9 mVDC.
>I have a number of questions that I hope someone can answer for me.
>
>Is this voltage difference the cause of the hum?
millivolts of AC ripple appearing on the audio signal. Since audio
these days is usually 2 volts or so, peak-to-peak, you'd end up with a
serious hum.
Common problem with cable installs, and the solution you've used is a
common one.
>Was my measurement approach appropriate?
A/V equipment, and reconnect it directly to the cable. Then, measure
the AC voltage between its A/V ground (e.g. the shell of the RCA
connectors for the audio-out) and the corresponding AC ground on
something else in your A/V setup.
I strongly suspect you'll get roughly the same voltage rating.
>
>Do the measured voltages indicate that the cable is improperly grounded?
>
>If the cable were to be bonded to the house neutral with 3/0AWG, would this likely eliminate the need for the isolation transformer?
current-carrying wire, and it can be pulled several volts away from
ground by voltage-drop in the wiring. The only place this isn't the
case is back at the service panel, where neutral and ground are bonded
together.
The only place you should bond grounds to, is other grounds (and
ideally do so at a single point).
You don't need 3/0 to comply with NEC - 14/0 or heavier is apparently
adequate.
>Whether it would eliminate the hum or not, should I have it done?
fixed your problem, Be Happy!
Grounding the coax at the point of entry (per NEC) would provide you a
bit of additional protection against something like a near-strike by
lightning onto the cable wiring.
>Does the cable company have any obligation to do it in order to comply with regulations?
supposed to have their braids bonded to building ground at the point
of entry. A lot of installers skip this step, as there's often no
"good" ground at the point of entry, and they don't want to run a
heavy ground wire back to the panel and do a proper job of bonding at
both ends.
National Electric Code, article 820, seems to be the relevant one.
I don't have the full text here (my old salvaged copy of the NEC is at
home) but the summary I see on-line does indicate that this is
necessary:
http://ecmweb.com/code-basics/article-820-community-antenna-tv-and-radio-distribution-systems
Sidebar: 820 Tips
Determine point of entrance.
Ground the incoming cable as close as practicable to the point of
entrance.
If you run cables above a suspended ceiling, route and support them to
allow access via ceiling panel removal.
If you use a separate grounding electrode, bond it to the power
grounding system.
Use the correct cable type and raceway for application the general,
plenum, or riser.
jurb...@gmail.com
Aug 26, 2016, 6:55:20 PM8/26/16
to
Unless you are using an indoor antenna, somewhere the cable from wherever is grounded. It is obviously not grounded at the same place as the electrical ground.
While newer electrical installs will have a ground rod stuck ten feet in the ground, I think in most areas cable and phone installers can still use a cold water pipe because those are more for lightning than ground faults in appliances, like if an internal hot wire gets shaved and touches the metal body.
The cable or antenna ground is so that if lightning strikes it does not arc across your house and kill you from ten feet away. It doesn't really protect the equipment much either.
There are ground gradients, I had a similar problem with an electician's house. He has a large house, so large that he decided to use two ground rids. We sold himm a video projectot which at the time was the only piece of his system that had a three prog grounded plug. He comes back and says that as soon as he plugs in the cable he gets a bar running up the screen. This is called a hum bar and had the exact same cause as your audio hum, but when the video gets the hum you get a bar. Our solution was to find an antenna isolator from an old hot chassis type TV and adapt it to be F to F rather than F to modified RCA. Finding such an isolator these days might not be so easy since TVs all when to SMPS type supplies and thus need no isolation.
You can actually build one, in case you would rather have that isolation transformer for other uses. If so, I'll look into the best/easiest way to do it, but if you are happy now and this is all academic you can just leave it as it is. However, your equipment is no longer grounded, which should not be a problem.
While newer electrical installs will have a ground rod stuck ten feet in the ground, I think in most areas cable and phone installers can still use a cold water pipe because those are more for lightning than ground faults in appliances, like if an internal hot wire gets shaved and touches the metal body.
The cable or antenna ground is so that if lightning strikes it does not arc across your house and kill you from ten feet away. It doesn't really protect the equipment much either.
There are ground gradients, I had a similar problem with an electician's house. He has a large house, so large that he decided to use two ground rids. We sold himm a video projectot which at the time was the only piece of his system that had a three prog grounded plug. He comes back and says that as soon as he plugs in the cable he gets a bar running up the screen. This is called a hum bar and had the exact same cause as your audio hum, but when the video gets the hum you get a bar. Our solution was to find an antenna isolator from an old hot chassis type TV and adapt it to be F to F rather than F to modified RCA. Finding such an isolator these days might not be so easy since TVs all when to SMPS type supplies and thus need no isolation.
You can actually build one, in case you would rather have that isolation transformer for other uses. If so, I'll look into the best/easiest way to do it, but if you are happy now and this is all academic you can just leave it as it is. However, your equipment is no longer grounded, which should not be a problem.
Dave Platt
Aug 26, 2016, 7:08:10 PM8/26/16
to
>Unless you are using an indoor antenna, somewhere the cable from
>wherever is grounded. It is obviously not grounded at the same place
>as the electrical ground.
In cases like this, chances are that the cable's nearest ground is
>wherever is grounded. It is obviously not grounded at the same place
>as the electrical ground.
either out at a street-side service pole, or a curb-side vault, or a
neighbor's house. There's all kinds of opportunity for current loops
between the cable ground, house ground, and neutral.
>You can actually build one, in case you would rather have that isolation transformer for other uses. If so, I'll look into the best/easiest way to do
>it, but if you are happy now and this is all academic you can just leave it as it is. However, your equipment is no longer grounded, which should not be
>a problem.
rebuild a standard cheap 75-to-300-ohm balun, to turn it into an
isolated 75-to-75-ohm unun. The result is probably rather similar to
what you'd find in a commercial 75-ohm coaxial groundbreaker.
(This very problem was the source of a discussion on the FMTuners
mailing list last week, and that's where I cribbed the reference to
this particular DIY project.)
Ron D.
Aug 27, 2016, 1:53:36 PM8/27/16
to
In a really perfect world, every ground should go back to the single point ground.
The first part is to have Telco, electrical and cable tie together at only one place.
The telco protector is a potential source of problems, but it's on the telco side of the NID and connected to the ground rod. Hum usually shows up in the land line.
Another issue is the cable. I should also be home run if from many locations. A splitter here and a splitter there isn't very good because of potential loops.
Each 120 V branch circuit is already a loop, because it has multiple outlet grounds in parallel separated by some distance. If each outlet went all the way back to the panel, the potential loop would not be there.
Some device along a circuit can potentially put noise on ground from say a switching power supply filter that's grounded. A fault can raise the potential of some device say near the middle of the loop from a lightning strike or ground at the end of the loop.
Ground is supposed to be low impedance and in theory nothing should happen.
Everything works really well when you have one phone, one device connected to the TV or the devices are clustered and you have a single point ground.
USB also uses ground as a common to the power supply. Switching power suppllied that are grounded also put some amount of noise on the ground. The RCA phono jack is the consumer (single-ended) version of what's done commercially using differential signals and DIN connectors. there is a in phase and out of phase signal relative to ground. The wires are twisted to reduce EMI and shielded to reduce RFI.
Shields should actually only be connected at one end, but sometimes that isn't possible.
The first part is to have Telco, electrical and cable tie together at only one place.
The telco protector is a potential source of problems, but it's on the telco side of the NID and connected to the ground rod. Hum usually shows up in the land line.
Another issue is the cable. I should also be home run if from many locations. A splitter here and a splitter there isn't very good because of potential loops.
Each 120 V branch circuit is already a loop, because it has multiple outlet grounds in parallel separated by some distance. If each outlet went all the way back to the panel, the potential loop would not be there.
Some device along a circuit can potentially put noise on ground from say a switching power supply filter that's grounded. A fault can raise the potential of some device say near the middle of the loop from a lightning strike or ground at the end of the loop.
Ground is supposed to be low impedance and in theory nothing should happen.
Everything works really well when you have one phone, one device connected to the TV or the devices are clustered and you have a single point ground.
USB also uses ground as a common to the power supply. Switching power suppllied that are grounded also put some amount of noise on the ground. The RCA phono jack is the consumer (single-ended) version of what's done commercially using differential signals and DIN connectors. there is a in phase and out of phase signal relative to ground. The wires are twisted to reduce EMI and shielded to reduce RFI.
Shields should actually only be connected at one end, but sometimes that isn't possible.
0 new messages