Topband: WATCH OUT !!Did you get a new cable modem from Comcast? Arris Modem equal lots of QRM

[Tom W8JI]

All of this centers around one thing, at lower frequencies devices that 
generate noise need "antennas" to cause problems. The cures will all prevent 
the devices from effective "antennas".

The original post contained information about an unusual thing we should 
keep in mind, that some devices have battery power. When the mains were 
turned off remotely, the battery took over. Since the device was not 
disconnected from the "antenna", the radiation was not altered. Had it been 
unplugged, even with the battery, the noise would have either greatly 
dropped or disappeared.

This should serve as a warning to NOT just depend on flipping breakers off 
to find noise. There is no assurance removing power will alter noise, 
because some devices run without line power. Radiating devices all change 
significantly when the radiating "antenna" or  "loop path" is opened (or 
shorted), but not always when mains power is removed.

Outside of new equipment I design or special situation, I virtually never 
use chokes and beads. From an engineering standpoint a series choke system 
is actually an unpredictable solution. I tend to avoid unplanned "throw 
something at it" cures.

My first step is **always** to close the loop. I buy a cheap TV/power outlet 
surge protector outlet, and I run every cable and wire in a local equipment 
cluster through that point. I make sure the line strip has suitable bypass 
capacitors that are across the line, and from neutral to ground. I make sure 
the coaxial line shield is bypassed to the common ground point.

This does several very important things, including things beads cannot do:

1.) It closes the loop for lightning and surge ingress into the system. Most 
lightning damage is common mode stuff that loops through gear, like into the 
cable port via the shield, through equipment, and out the power line. The 
external common point, even without a ground, keeps most of the unwanted 
current out of the equipment.

2.) RF from our transmitters follows the same path. Instead of going through 
the gear, the RF is "shorted" and bypassed around the gear. The sensitive 
equipment is not in the loop.

3.) RF inside any device cannot drive the power lines and other cables in 
"push pull". This is exactly what a bead does, but better. Instead of adding 
a series impedance via a choke and depending on the ratio of the newly added 
series impedance being very high compared to the outside world impedance, 
this method "shorts out" the outside world path for RF.

Human focus tends to mindlessly follow the herd of sheep. Since soft iron 
cores began, and the first TV deflection yokes were split apart to wind 
power cords around, we have focused on throwing beads at systems. We have 
not considered the path is two modes, and we have not considered how a bead 
actually works.

The primary problem paths are either differential between conductors in one 
cord or cable, differential between two cables, or a mixture of the two.

A bead or choke does nothing for differential on a given cable or cord. It 
does very little for lightning. Its effectiveness is also highly dependent 
on the differential impedance between inlet and outlet at the insertion 
point.

A bypassing arrangement greatly reduces differential excitation of the power 
feed, something the choke does not do at all. A bypassing arrangement also 
greatly reduces differential excitation of the mains against other cables. 
It is generally far more effective than a choke for suppression, because 
external leads almost always have modest to high differential impedance.

I can often do, with just a simple piece of hookup wire, a better job than a 
choke.  I cured an apartment complex from severe RFI that Buckeye Cable had 
given up on mostly with wire. Buckeye used chokes, filters, double and 
triple shielded cables, and were generally just marginally successful. I 
fixed almost all of it with simple jumpers. Just a few apartments out of 
hundreds needed anything more. I'd guess they invested many thousands of 
dollars, and the nearly perfect fix really just cost a few dollars per 
building.  :)

73 Tom