 |
|
|
|
|
|
|
|
|
Noise and Filtering, but let’s just define noise first. - Which One Should I Use - Part IX
by Phil Kingery
Hello!?! Hello?!? Captain Coupling is having a hard time hearing
you…there’s too much "noise" on the line….
|
Which One Should I Use, Part IX (Preamble)
The voting was hot and heavy and the overwhelming winner was
"Noise and Filtering". (…yea, and the crowd goes wild!!!) For those of you who have no
idea what the heck I am talking about, I will give you a one-breath synopsis of what you have
missed: (Ah, hum, clear throat, deep breath, here goes…)
I started writing articles for Bob Hetherington (the benevolent, all
powerful master of this website), way back in February 1997 with the first one, called "Which
One Should I Use, which gave a very basic overview of X-10 receivers and tried to explain the
differences between 2-wire units and 3-wire units, then continued that in part 2, in April 1997, on
controlling motors and transformers, which led nicely into part 3 for the June ’97 issue of HTI,
where we discussed simply residential coupling, which of course, led directly into the August 1997
episode on more complex coupling, which also led into the next one, part 5, where I tried to
explain the complexities of coupling dim/bright commands, but then in the next one, for last
December’s issue, we got into 3-way circuits, which turned out to be a very long one and so it was
continued in part 7, last February, and then even more, just this last April where I finished up
3-ways, 4-ways and troubleshooting them. (AHhhh-catch-my-breath. That has got to be the longest
run-on sentence I have ever written.
Editor says ... Yes and the MSWord grammar checker underlined the whole thing
... way to go Bill :-)
)
Which One Should I Use, Part IX
(Noise and Filtering, but let’s just define noise first.)
Yup, this one looks like it’s going to be a long one. As I begin
writing it, I already suspect that it’s going to be a 3-parter. In this first part, I think we will
begin with noise and what it looks like on an oscilloscope.
 Figure
1 is just to get you interested. It shows an o-scope display with a little bit of noise.
We will try to explore noise and its effect on the original X-10
receivers, the AGC receivers and possibly some new stuff on the horizon. The next part (in the
August issue of HTI) will talk about what filters are available and how they differ in design and
purpose. Then, in the third section, we will try to tie it all together and actually see how
filters work on reducing electrical pollution. If possible, I will also squeeze in whole house
filtering, but if not, that may be pushed into its own episode. We’ll just have to see how it
goes.
Second only to 3-way circuits (which was the subject of the last three
chapters in this series), "noise" is probably the most misunderstood problem in the X-10
world. Let’s begin with the term, "noise". Personally, I rarely use that term. The only
reason it appears in the title of this section is that it seems to be the generic and
preferred term by most of the rest of the world. I suppose the word interference is
acceptable but I prefer using the term "Electrical Pollution".
I used to use those three terms interchangeably but decided to favor
the more accurate electrical pollution after a customer once said to me, "What do you
mean, noise? I don’t hear nuthin’!".
(…long pause here….just for effect…)
My handy-dandy desk dictionary defines noise as, "a sound of any
kind especially when loud or disagreeable". Obviously this is too restrictive a definition for
our purposes (although I like the "disagreeable" part) and does little to give the
necessary weight and blame to the problem. The word "pollute", however, has a definition
that seems more appropriate: "to make impure, defile". Your local electric utility
generates power (at 60Hz here in North America, 50Hz in most of the rest of the planet) that is
relatively pure. Unfortunately, many companies manufacture devices (mostly electronic in nature)
that generate an unnecessary byproduct of high frequencies and dump them back onto the line from
which they are powered. They pollute the electrical environment.
I really get on my soapbox when it comes to the problem of electrical pollution.
I think we would all agree that everyone should be allowed to use the water in the river but not
one should be allowed to pollute the water in the river. Why should an electrical environment be
any different. Everyone should be allowed to use the electricity but no one should be allowed to
pollute the electricity. Apparently this concept is lost on many people. A lot of companies believe
that the problem is not the pollution coming from their products, but the inability of ‘my’
products to work in spite of it. Sorry, I don’t buy that. When the Exxon Valdeze ran aground
several years ago in Alaska, no one would dare hold up a dead fish and say, "What you
need are stronger fish".
Why am I so adamant about this? It’s because it’s not just an X-10
problem. Electrical pollution adversely effects everything that uses electricity. Have you ever had
a corrupted file on your computer? …have you every had your TV suddenly die? …lost a fax? …had your
radio quit? …had to lie still on that cold table, writhing in pain, in the emergency room while
they take another X-ray because the first one was fuzzy? All of these problems, and many more, are
all attributable to electrical pollution. Many hospitals and research facilities have long rejected
the use of electronically ballasted fluorescent lights in their buildings because they adversely
effect sensitive equipment. (There are now many brands of clean electronic ballasts but we
will get into that later.)
Back in the early 1900’s, coal miners used to take canaries down into
the mines with them. If the birds fainted, they knew that there was dangerous gas in the mine. If
they died, they knew they had to get out now. Think of your X-10 system like that. If your
X-10 system is having troubles because of electrical noise, everything in your house that uses
electricity is also being effected, it just takes longer. Your X-10 system may experience a variety
of problems raging from receivers randomly going on and off, to an occasional problem to a
completely dead system.
All right, I’ve made my point. Anyone who manufactures or sells devices
that pollute the electrical environment needs to take responsibility for their actions and made to
correct their error. (And then they should be hung by their thumbs!)
(I will now step down from my soapbox…)
When a user installs a new X-10 based receiver and it doesn’t work, the
only question that can be asked is "why". If the receiver is a normally operating unit
(meaning it is not an "out of box" failure) then there can be only three reasons:
-
Signal at the wrong place on the sine wave (but this is so rare in
residential installations that we will disregard this possibility for this article).
-
Insufficient signal (can be corrected with proper coupling and
sometimes filtering).
-
Electrical pollution (noise, interference, whatever you want to call
it).
Is this electrical pollution a new phenomenon? No, but it seems to be a
bigger problem then ever before. That is because of 2 main things. First, there are more electronic
devices, more compact fluorescent lights, more computer power supplies, more big screen TV’s and
more high-tech doodads than ever before. Second, there are more X-10 systems out there to be
effected by all those new high-tech doodads. Electronically "noisy" devices are allowed
to exist because there are no clear regulations against them. The FCC is mostly concerned with
radiated noise above 450kHz (the low end of the AM radio band), the utilities are only concerned
with harmonics of the 60Hz (because it burns up their transformers) and UL is only concerned with
the device being "safe". The band of frequencies that most effect X-10 systems are in the
80kHz-to-150kHz range but unfortunately, there just aren’t many regulatory policies that deal with
that area.
My opinions should be obvious by now. I feel very strongly that the
responsibility for correcting this problem is on the shoulders of those manufacturers who have
caused it. However, having said that, I must admit that I am not going to hold my breath waiting
for that to happen. We know that electrical pollution out there, so we are going to have to try to
co-exist with it, at least, as much as we can.
The Pico engineers recognized the need to get their signal through
moderate amounts of noise even in 1978, when they decided to transmit and receive at the 0° point
on the sine wave. Although, not as much today as in years past, that location is usually the
cleanest part of the sine wave.
Another great idea was to use a "complementary pair" system
of sending the digital data. (One of the choices offered at the end of the last section was
"X-10 Binary Codes Including the New Extended Code ", but I only received a few
votes for that. Consequently, this part will refer to the X-10 binary bit pattern but you will not
have the luxury of referring back to an earlier article to help you understand it. Maybe after we
get done with this one, we will try to do a piece on the X-10 Binary Codes.)
In standard TTL (transistor to transistor logic) protocol, sending a
stream of zeros simply means sending "nothing" for a predetermined amount of time.
 Since
most of the time, the powerline consists of hours of nothing (as least as far as X-10 is
concerned), it would be inappropriate to interpret that as a nearly endless string of zeros.
Therefore, the engineers at Pico (the guys who invented this stuff) decided to define a
"Zero" as 0-1 and a "One" as 1-0 (figure 2). This is a far more reliable method
of sending binary information than just sending "something" as a one and
"nothing" as a zero. Figure 3 shows a typical X-10 data segment.
In 1989, the X-10 engineers also improved the way the binary data was
interpreted by the receiver circuitry, which again improved reliability. Finally, we have the new
generation of interference-resistant AGC and Gated-AGC receivers. Before we get into what AGC is
and does, let’s look as some noise.
Figure 4 shows an oscilloscope displaying a fairly clean sine wave.
 If
you stick your nose right up to your computer screen, you might see that there is just a hint of
interference on the sine wave, but for the most part it’s pretty good. Most residences in N.
America have about 10-25mv of background interference all the time. It may rise and fall in
amplitude as appliances and electronic devices are switched on and off, but in most houses, the
overall interference level is good and more than tolerable to X-10 systems.
I need to let you know a few things about the graphics you will see in
this and the next couple of installments. First, they will be greatly exaggerated. Trust me, this
will all make sense in the end but for the time being, I need to be able to show you the sine wave
(which is very high in amplitude) along with the X-10 signal and/or electrical pollution (which is
usually very small by comparison). If I displayed them both using the same reference, we would not
be able to see the signal nor the interference, just the big sine wave. I will explain the how and
why of this later when we get into using an oscilloscope. Second, I will be showing you a
simplified version of the X-10 signal. Again, I am just asking that you trust me on this. All will
become clear at the end of the series. (Gee, sounds like one of those complicated Myst
-like computer games, doesn’t it.)
 Figure
5 shows the o-scope display with an X-10 signal on the line. (Remember that it is greatly
exaggerated in amplitude.) You notice in this graphic that the signal is clear and precise and
there is no danger of confusing it with any background noise. The original X-10 receivers saw just
about what you are seeing. They easily picked out the bit strings and decoded them and reacted
accordingly.
Then as we got in the 1980’s, our technologically advanced society
demanded that we all buy more and more high-tech gizmos and doodads to put in our homes and impress
our neighbors. Unknown to us, the electrical pollution level began to rise. The X-10 engineers
responded by improving the method by which the binary pulses are decoded in the receivers. It’s too
complicated to go into here, but suffice it to say that the new method actually counts the
half-sine wave "bumps" of the 120kHz pulses. This helps narrow the frequencies to which
the receivers will respond.
The 1990’s saw us with even more computers, fax machines, big screen
TV’s, electronic lighting and more electronic gadgets than ever. X-10 responded by developing
receivers with "AGC". Automatic Gain Control
may be a misnomer. For many of us who were originally trained in the radio telemetry end of the
wide electronics field, know AGC as a description denoting an output variation. Put simply,
if a device had AGC, its output was no longer constant. Instead, it was variable based on the needs
of the system and controlled by a feedback algorithm. It’s like some of those new car radios that
turn up their own volume based on the road noise. If you were driving on the highway with your
windows down, the volume is automatically turned up. When you stopped at a traffic light, it turns
itself down.
AGC takes on a somewhat new meaning with X-10 receivers as they are
capable of varying their input receive threshold depending on the amount of background
interference. I suppose it should have been called "ATC" for Automatic Threshold Control.
(If they had only consulted with me before hand…) AGC (or even the newer "Gated-AGC")
receivers are a great improvement but they still have problems in high interference environments.
The AGC algorithm requires that the X-10 signal be at least twice the amplitude of the background
interference.
 Figure
6 shows an instance where even the newest X-10 AGC receivers would not operate reliably. It shows
an instance where the X-10 signal is higher in amplitude but not twice as high as the background
interference. We here at ACT have some new X-10 "compatible" receivers that far
exceed the performance of X-10’s own receivers, but we are using our new technology only for our
industrial-commercial customers. Sorry, guys. Soon we hope to have some new stuff for you HA guys.
I think that is about as much as I should get into in this installment.
Just to give Mr. Hetherington a headache, I am including one little mpeg file. This little 11
second video started out as a huge 6.5MB .avi file that would have been unbearably long to
download. Even with my 56k modem, it takes over 20 minutes and so I figured you guys out there with
14.4's, 28.8's or even 33's, might give up before you got all of it. I did a quick web search and
found a demo program that did a painless conversion to a more manageable mpeg format. (However,
since it was a demo version, I will need to pay for it if I want to continue using it.) The good
news is that now it's only 201KB. Since this is my first attempt at putting video into one of my
articles, you should know a little about this 12 second masterpiece. Even though it is recorded at
30 frames per second, the oscilloscope displays an image at 60 "events" (sine waves) per
second, so you can see that this is not going to be nearly as good as seeing it with your own eyes.
All in all, I think it is pretty neat. If we get a lot of good feedback, we will try to do a lot
more mpegs. (I may even consider buying that software instead of just using the demo
version.)
As usual, I am writing this at the last second. I am listening to the
Indianapolis 500 as I write. By the time you read this, you will know who won the race, last week.
The next chapter (August 1998) will continue this same subject of electrical pollution and what you
can do about it. Even though I am not asking for votes, I would still like your emails. Do you like
the .avi files? Do you have any questions on this first part? Are you a rich relative that I don’t
know about?
Well, children, until next time....
What will become of Captain Coupling?
Is he really going deaf?
Stay tuned, children! Same Bat Time, same Bat Channel!
Phillip Kingery is the representative of Advanced Control Technologies,
Inc. and teaches X-10 related classes around the country. Email him at
pkingery@act-solutions.com
|
|
|
|
|
|
|
|
|
|
|
|
The Smartest DIY Home Improvements
