Noise suppression in cables

[Ben Pope]

Hi everyone,

What's the principle behind those blocks at the end of signal cables (and
what are they called)? I think they're made out of iron or something
similar - you find them on the end of monitor cables, that sort of thing.

Cheers,
Ben.

 

[Don Pearce]

Hi everyone,

What's the principle behind those blocks at the end of signal cables (and
what are they called)? I think they're made out of iron or something
similar - you find them on the end of monitor cables, that sort of thing.

Cheers,
Ben.

They make the cable much more inductive locally, and this tends to
stop the signals spreading into the rest of the cable. Because the
block (ferrite) is round the whole cable, the signal path inside the
cable had a component in each direction, and its field cancels - so it
doesn't interact with the block,, and passes unhindered.

d

_____________________________

http://www.pearce.uk.com

 

[Ben Pope]

They make the cable much more inductive locally, and this tends to
stop the signals spreading into the rest of the cable. Because the
block (ferrite) is round the whole cable, the signal path inside the
cable had a component in each direction, and its field cancels - so it
doesn't interact with the block,, and passes unhindered.

Thanks for the quick reply Don.

If I have a monitor cable without them, is that likely to be the cause of
around 6 visible "echoes" on high contrast vertical edges?

I have just purchased this cable:

http://catalog.belkin.com/IWCatProductPage.process?Product_Id=65410

And it's crap.

I thought this one might be better
http://catalog.belkin.com/IWCatProductPage.process?Product_Id=83865

"Constructed of 3 coaxial and 5 twisted pair cable to minimize cross talk,
noise, and other interference. 3 coax conductors for the RGB signal maximize
color and imaging"

So if the other one is a 15 wire cable with "straight through" configuration
and no attempt at noise suppression, they have to be asking for trouble with
frequencies around 100MHz. How can they even sell such a thing.

Ben

 

[Don Pearce]

Thanks for the quick reply Don.

If I have a monitor cable without them, is that likely to be the cause of
around 6 visible "echoes" on high contrast vertical edges?

No, these device have no effect at all on signal passing through a
cable. If a cable is causing these kind of problems, then it is poorly
constructed from the wrong impedance internals.

I have just purchased this cable:

http://catalog.belkin.com/IWCatProductPage.process?Product_Id=65410

And it's crap.

I thought this one might be better
http://catalog.belkin.com/IWCatProductPage.process?Product_Id=83865

"Constructed of 3 coaxial and 5 twisted pair cable to minimize cross talk,
noise, and other interference. 3 coax conductors for the RGB signal maximize
color and imaging"

So if the other one is a 15 wire cable with "straight through" configuration
and no attempt at noise suppression, they have to be asking for trouble with
frequencies around 100MHz. How can they even sell such a thing.

Ben

These cables should always be coax, for reasons both of performance
and interference suppression. I've never taken one apart, so I have no
idea what variation in construction quality you are likely to find. It
isn't cheap assembling a proper set of coax and normal conductors,
though. My current monitor uses coax leads terminated with BNC
connectors.
d

_____________________________

http://www.pearce.uk.com

 

[Ben Pope]

These cables should always be coax, for reasons both of performance
and interference suppression. I've never taken one apart, so I have no
idea what variation in construction quality you are likely to find. It
isn't cheap assembling a proper set of coax and normal conductors,
though. My current monitor uses coax leads terminated with BNC
connectors.

Thanks. The cable attatched to my monitor is huge, and the image quality is
excellent. This new cable is about half the diameter.

Looks like I'll have to spend considerably more on a cable of reasonable
quality.

Again, thanks.

Ben

 

[John Larkin]

These cables should always be coax, for reasons both of performance
and interference suppression. I've never taken one apart, so I have no
idea what variation in construction quality you are likely to find. It
isn't cheap assembling a proper set of coax and normal conductors,
though. My current monitor uses coax leads terminated with BNC
connectors.
d

_____________________________

http://www.pearce.uk.com

The cheap VGA cables are just a bundle of wires; video is terrible.

John

 

[Ben Pope]

The cheap VGA cables are just a bundle of wires; video is terrible.

Tell me about it.

It's a good way of making £700 worth of monitor and video card look like
crap.

Ben

 

[Walter Harley]

They make the cable much more inductive locally, and this tends to
stop the signals spreading into the rest of the cable. Because the
block (ferrite) is round the whole cable, the signal path inside the
cable had a component in each direction, and its field cancels - so it
doesn't interact with the block,, and passes unhindered.

I think ferrite beads like these work because the core material is lossy,
i.e., resistive. High-frequency common-mode signals get inductively coupled
into the core, and dissipated as heat.

 

[Don Pearce]

I think ferrite beads like these work because the core material is lossy,
i.e., resistive. High-frequency common-mode signals get inductively coupled
into the core, and dissipated as heat.

The core is non-conductive so you can't couple into it inductively. It
does work by virtue of its high relative permeability which increases
the common mode inductance of the cable hugely. It is beneficial that
it is so lossy though, as this gives the energy somewhere to go (which
it must eventually) other than back into the computer.

Personally I always found the yellow coded ring cores from Phillips
the best for this kind of application.

d

_____________________________

http://www.pearce.uk.com

 

[Walter Harley]

The core is non-conductive so you can't couple into it inductively. It
does work by virtue of its high relative permeability which increases
the common mode inductance of the cable hugely. It is beneficial that
it is so lossy though, as this gives the energy somewhere to go (which
it must eventually) other than back into the computer.

Can you say more? What does it mean to be lossy if there's no resistance?

 

[Don Pearce]

Can you say more? What does it mean to be lossy if there's no resistance?

What I mean is that if you measure that core with a meter you will
find that it doesn't conduct. But the rattling around of the domains
is pretty lossy, and that makes it absorb to a fair extent. But that
is only a small part of the mechanism that makes it reduce common mode
signal, which is primarily a huge increase in local inductance.

d

_____________________________

http://www.pearce.uk.com

 

[Animesh Maurya]

They make the cable much more inductive locally, and this tends to
stop the signals spreading into the rest of the cable. Because the
block (ferrite) is round the whole cable, the signal path inside the
cable had a component in each direction, and its field cancels - so it
doesn't interact with the block,, and passes unhindered.

d

_____________________________

http://www.pearce.uk.com

Iam also having a question in the same regard, it may be very silly.

Why are those ferrite blocks placed at the extreme end of the cable,
why not at the center or some where else.

Thanks

Animesh Maurya

 

[Don Pearce]

Iam also having a question in the same regard, it may be very silly.

Why are those ferrite blocks placed at the extreme end of the cable,
why not at the center or some where else.

Thanks

Animesh Maurya

It doesn't actually make a great deal of difference where you put
them. Convention has it, though, that you put them near the end of the
cable which is the source of possible interference.

d

_____________________________

http://www.pearce.uk.com

 

[Fred Abse]

Why are those ferrite blocks placed at the extreme end of the cable, why
not at the center or some where else.

If you put them at the *source* end of the cable, less noise will be
radiated by the cable.

Conversely, if you put the suppressor at the "load" end, interference
picked up by the cable will be reduced.

Think of the cable as an antenna.