Register to reply

Electrical noise

by david90
Tags: electrical, noise
Share this thread:
david90
#1
Jan31-13, 07:16 PM
P: 303
I work with industrial electrical system and I got a question regarding electrical noise caused by electromagnetic radiation. In an industrial environment, cables carrying an analog signal are shielded and its drain wire is earth grounded at one end of the cable. When electromagnetic radiation hits the cable, the induced current drains to the earth via the drain wire.

Does the induced current in the shield prefer to travel back to earth? Would the shield of the cable still be effective at blocking the radiation if the drain wire was connected to a really big piece of metal that is not earth grounded?
Phys.Org News Partner Engineering news on Phys.org
Greater safety and security at Europe's train stations
Fingerprints for freight items
On the way to a safe and secure smart home
berkeman
#2
Jan31-13, 07:32 PM
Mentor
berkeman's Avatar
P: 41,109
Quote Quote by david90 View Post
I work with industrial electrical system and I got a question regarding electrical noise caused by electromagnetic radiation. In an industrial environment, cables carrying an analog signal are shielded and its drain wire is earth grounded at one end of the cable. When electromagnetic radiation hits the cable, the induced current drains to the earth via the drain wire.

Does the induced current in the shield prefer to travel back to earth? Would the shield of the cable still be effective at blocking the radiation if the drain wire was connected to a really big piece of metal that is not earth grounded?
There are lots of considerations for how to terminate the shield of shielded cable. Factors like the frequency of the EM noise, potential sources of ground noise, and how the shielded cable is routed (near a metal wall, or away from any metal). Optimizing the effectiveness of the shield takes some intuition about the electrical environment of the cable installation, and the nature of the noise to be shielded.

But to answer your question, if the frequency of the EM noise is high enough, and the piece of metal is big enough to have significant parasitic capacitance to Earth, then yes, it will be fairly effective to tie the drain wire to it.

Note though, even with an Earth ground connection at one end of the cable, if the EM noise has a frequency so that the length of shielded cable is a quarter wavelength, then the cable shield will go resonant (like a resonant quarter-wave monopole antenna), and you will get a lot of noise coupled to the inner conductor...
jim hardy
#3
Jan31-13, 09:51 PM
Sci Advisor
PF Gold
jim hardy's Avatar
P: 3,686
Many years ago i bought this book. I would say that for anybody working in industy it is quite a good reference to own, in that it explains the hows and whys of shieding.
http://www.amazon.com/Grounding-Shie.../dp/0470097728

Mordred
#4
Jan31-13, 10:25 PM
Mordred's Avatar
P: 1,857
Electrical noise

I have some good grounding pdf's I can post them for you on monday. They were from the Rockwell Knowledge base.
The files are on my flash stick at work. I also have the R56 standard. Which is for radio towers.

The AB article also includes grounding practices and cabinet layout recommendations with regards to drives, soft starters. Which has high varying frequencies due to the PWM
signals. Its a very extensive coverage of a large variety of grounding needs. Included is machinery grounding. Such as conveyers etc.
david90
#5
Feb1-13, 05:19 PM
P: 303
Quote Quote by Mordred View Post
I have some good grounding pdf's I can post them for you on monday. They were from the Rockwell Knowledge base.
The files are on my flash stick at work. I also have the R56 standard. Which is for radio towers.

The AB article also includes grounding practices and cabinet layout recommendations with regards to drives, soft starters. Which has high varying frequencies due to the PWM
signals. Its a very extensive coverage of a large variety of grounding needs. Included is machinery grounding. Such as conveyers etc.
That would be nice. Thank you!
david90
#6
Feb1-13, 05:26 PM
P: 303
I read a wiki article on electrical noise and it said that two wires running in parallel can cause noise due to capacitive coupling.

"Capacitive coupling - A current through two resistors, or any other type of conductor, close to each other in a circuit can create unwanted capacitive coupling. If this happens an AC signal from one part of the circuit can be accidentally picked up in another part. The two resistors (conductors) act like a capacitor thus transferring AC signals. There may be other reasons for which capacitive coupling is wanted but then it would not be thought of as electronic noise."

Isn't noise caused by two wires running in parallel is due to induction and not capacitive coupling? When AC current flows in a wire, it develops a fluctuating electric field that surrounds the wire. If an analog signal wire runs parallel to the ac wire, then noise would enter the analog signal wire via induction because the electric field from the AC wire cuts across the wire.
gnurf
#7
Feb1-13, 05:57 PM
P: 330
Quote Quote by david90 View Post
Isn't noise caused by two wires running in parallel is due to induction and not capacitive coupling?
Both. The electric (capacitive) coupling generates a noise current that can be modeled as a current source in parallel to your load. This noise current is dependent on the capacitive coupling (here: distance) and voltage between two conductors and the signal frequency. The magnetic (inductive) coupling generates a noise voltage that can be modeled as a voltage source in series with your load. This noise voltage is proportional to frequency, mutual inductance (here: loop area), and the current amplitude of the aggressor.

With a little effort you should be able to relate these concepts to Maxwell's Equations.
jim hardy
#8
Feb1-13, 11:16 PM
Sci Advisor
PF Gold
jim hardy's Avatar
P: 3,686
Anecdote - hope it's not boring....

Shielding is basically trying to wrap a Faraday shield around your measurement system and its signal input wires.
That way capacitive coupled current flows in your shield wire instead of your signal wire, and that's the objective.

It is important that signal wire and shield wire be at approximately same voltage, that way there's not much capacitive coupling inside the cable between shield and signal conductors.

That leads to an interesting corollary - if your signal isn't grounded but your shield is, you will have voltage between shield and signal wires hence capacitive coupling between them.
So some books say that proper way is to connect shield to low side of signal at the point of measurement.
Hardly anybody does that. But most signals aren't very much elevated from ground so most folks get away with simply grounding the shield someplace.

For extreme cases, one can drive the shield to same voltage as the signal wire by using another amplifier. We had one instrument that went to that trouble and for good reason.
It was a "Magnetic Flowmeter", a huge electromagnet on an insulated section of pipe with two electrodes contacting the fluid. The electromagnet was excited by several amps of 60hz AC and induced a tiny 60hz voltage in the fluid, proportional to fluid's velocity. The signal produced at the electrodes was something like zero to 1.65 millivolts ac for zero to 10 gpm through the pipe.
That feeble signal travelled 250 feet to control room through a power plant stuffed with huge motors and electrical equipment, all operating at 60 hz.. "Can't possibly work", i said.

Each of the two signal wires contacting the fluid was wrapped with its own foil shield, and each shield was driven from receiving end to same voltage as the signal wire it enclosed. The pair of shielded signal wires was itself wrapped with third shield, that one tied to case of the electromagnet/pipe. So that third outer shield was tied to low side of signal at point of measurement.
So - no capacitive current was coupled into the signal wires . Capacitive current was coupled into the driven shields, but that current flowed only in the shields , meeting the objective.
Were those shields not driven but grounded, there'd have been zero to 1.65 millivolts between the signal wire and its shield, ergo capacitive coupling.

The system worked better than one could believe. It is a testament to the perseverance of 1960's circuit designers.

old jim
gnurf
#9
Feb2-13, 12:57 PM
P: 330
Quote Quote by jim hardy View Post
Anecdote - hope it's not boring....
Quite the opposite, Jim. More war stories please!
Mordred
#10
Feb4-13, 09:08 AM
Mordred's Avatar
P: 1,857
http://literature.rockwellautomation...m001_-en-p.pdf here is the link to one of the better references I found. Tried loading the R56 manual but it runs into errors on loading
Mordred
#11
Feb4-13, 10:14 AM
Mordred's Avatar
P: 1,857
Here is another manual that has useful information

http://literature.rockwellautomation...n041_-en-p.pdf

this one deals with PWM signals

http://literature.rockwellautomation...n001_-en-p.pdf

hope those help


Register to reply

Related Discussions
Converting noise into electrical energy General Physics 115
Electrical dark, and spectrum noise. General Physics 1
Definition of dynamic noise and observational noise in finance Set Theory, Logic, Probability, Statistics 1