Help on Noise Cancellation 1

[gokulkrish2]

Hi All,

We have a coaxial copper cable used to communication and interlocks running parallel to a 13.8KV power cable at a distance of 1/2 ft in between them.

Problem our process engineers are telling is that they have a lot of noise and sometimes that trips the circuit.

My question for you guys is:

1. Would that power cable be able to induce significant noise inside that co axial cable? If it can then what is the minimum distance we should have in between them in order to get rid of the bad signals?

2. Is there a way i can shield the coaxial cable to be insulated from the inductance of the power cable.

Any suggestion on this matter would be appriciated.

I am not very familiar with communications so any literature to read about would also help me.

Thanks all

gokul

 

[pstuckey]

What kind of coax, and is replacing the coax a possibility? Most inexpensive coax has a single shield, which provides ~40 dB isolation. Double and triple shielded coax is available for applications which need more isolation.

Peter

 

[MikeHalloran]

One way is to not let them be parallel and anywhere near each other.

E.g. triple the length of the coax, and route it far away from the power cable, say around three sides of a rectangle, with the power cable being the fourth side.


Or add magnetic shielding to the coax, e.g. by running it inside steel conduit.


Or replace the coax with fiber.



Mike Halloran
Pembroke Pines, FL, USA

 

[gokulkrish2]

Thank you so much for the replies....

It is a BELDEN 3092A RG-6/V Type Controlnet Cable. It is a Quad Shield 1C18 cable....

Is this cable good enough or is there any other suggestions????

Also Moving the cable to a distance might work, but its a overland conveyor and there is hardly room there to move it far apart.

Among the options replacing with fiber optics is the best, but i am not sure if we can do controlnet functions with fibre optics. Can WE?

Please advice

Thanks

gokul

 

[Comcokid]

This is a case where you need to shield the coax as the 13.8kv noise is being coupled either electrostatically (E-field to the shield) or magnetically (inductively like a poorly coupled transformer) to the shield and/or center conductor.

Can you run the coax through a conduit by itself? Conduit is great as it provides electric and magnetic shielding.

You can also try tri-axial cable. Use something like RG11 which provides 75 Ohm impedance between the inner shield and inner conductor (same impedance as your RG6) and ground the outer shield to the frame of the conveyor. However, if your problem arises from inductive coupling this will not help.

 

[VEBill]

Perhaps the receiver circuit isn't designed to allow high noise on the shield. It is easy to design a system that fails to protect the effectivity of the shielding. Lack of attention to grouding paths potentially carry high currents, lack of 360-degree shield termination to a shielded chassis, poor common mode rejection, no transformer coupling, etc.

Shielding only works if the entire system is designed to manage where the common mode noise currents are directed.

For example, if the RG6 was connected to an F-connector installed in a plastic box, so that the common mode noise on the shield had to flow into the box to complete its loop via ground, then the poor receiver circuit wouldn't have a chance.

This is just another consideration.

It might be a combination of several problems.

 

[zappedagain]

It may only be semantics, but you want to be thinking 'noise reduction', not 'noise cancellation'. They are two very different areas.

I still highly recommend Kimmel & Gerke's book for EMC/EMI noise reduction:

http://www.emiguru.com/books.htm#EDN

It is comprised of 14 different articles that were published in EDN magazine in 1994.

John D

 

[gokulkrish2]

Thank you all for those inputs.... zappedagain.... very useful information. Take a star for sharing the resource.

Kudos....

gokul