RF Column 35 - September 1994
Copyright (c) 1994,1995 H. Douglas Lung ALL RIGHTS RESERVED

TOPICS:   Telemundo's switch to digital video for distribution
          LNB selection for satellite digital signal reception
          Common problems with digital satellite reception
          Competition improves LPTV transmitter options
          Information from the Internet - USENET newsgroups

-----------------------------------------
It's been a busy month. Like last month, I'm completing this column a
few days past my deadline (if you noticed an issue missing my column,
now you know the reason). Last month I mentioned I was in the process of
testing Scientific Atlanta's digital video compression system on our
auxiliary transponder. Last week we put it on the air on our main
transponder with programming. I'll describe how it went and give you
some additional tips on how to plan for the time when you need to pull
down digital signals from satellites. I've been promising to cover some
of the new developments in LPTV transmitters and this month I've got the
space. Finally, you've read my comments on CompuServe in past columns
and my exploration of the Internet in more recent ones. This month I'll
describe another facet of the Internet - USENET News Groups.

In my last column I described what "bit error rate" is and gave some
target numbers for different performance levels using the S.A.
compression system. One thing that became obvious after the first day of
digital broadcasting was that it is absolutely essential to get the
lowest error rate possible! Most of the cable companies or stations that
had difficulty that first day had settled for error rates that appeared
to give a perfect, solid picture. Unfortunately for them, any
interruption or interference will cause a much more noticeable
disturbance to a digital signal than to an analog signal. Our stations
using large dishes in quiet locations saw few problems (except when
another uplink apparently didn't notice we were transmitting on Galaxy 4
transponder 8 and put a carrier on top of us the second day!). Other
stations and cable companies with smaller dishes or older LNB's (or
misaligned dishes) saw several instances where the video dropped to
black for a second or two.

I questioned the need for "Digital Ready" LNB's last month. Based on my
experiences last week, I now say they are worth the extra investment.
Why? Because anything which disturbs the digital data stream to the
point the error correction can't handle it will result in a loss of
picture. With an analog system, a few sparkles now and then or a line
flashing across the screen wasn't considered serious. Many people see
similar disturbances on their TV sets from near by radio transmitters or
cars noisy ignition systems passing by. On the other hand, even a short
dip to black is noticeable on an otherwise perfect picture.

There are couple of specifications to look for when choosing an LNB for
digital service. John Joslin, Sales Manager at Dawn Satellite in
Michigan (810-969-0377) related some of the problems he'd seen with
conventional LNB's pushed into digital service. Excessive phase noise
was one problem - phase noise generated in the LNB caused jitter on the
phase modulated digital data, increasing the chances of an error when
demodulating it. Another, which I hadn't heard of before but which makes
sense, was frequency (and phase) modulation of the LNB's local
oscillator by vibration. This vibration may be caused by heavy rain,
hail or even wind shaking the LNB. Absolute frequency accuracy or long
term frequency accuracy isn't as important as short term stability when
receiving full transponder digital signals. Dawn Satellite sells
non-phase locked loop LNB's that are designed for superior short term
stability (including resistance to vibration) and low phase noise
without the cost and complexity of phase locked loops. I haven't had a
chance to try these, so I can't say how well they work.

Getting back to my own experiences, I found digital satellite systems
working at a decent bit error ratio under normal conditions may
experience an occasional video hit. At one station the hits appeared to
be caused by nearby lightning. Lightning generates RF (remember reading
about "spark gap transmitters" used in the first part of this century?)
over a wide frequency range. At another of our stations on the east
coast all the usual suspects were been ruled out. The latest suspect,
courtesy of Howard Fine in Los Angeles, is radar from foreign carrier
aircraft with spurs in the 4 GHz. C band range. Howard said he
experienced this problem during a downlink under the flight path for Los
Angeles International Airport. We're not talking about a lot of hits
here - maybe three or four a day. However, in a broadcast operation,
even one hit will be noticed! Therefore, when installing a system that
may be used for digital signals, I recommend taking every step possible
to insure the most re liable signal -- frequency/phase stable LNB's
included. Follow ALL the steps I mentioned last month to improve
performance.

What were the most commons problems experienced during our digital
transition? As far as equipment was concerned, LNB's topped the list
followed by cable and splitters. We even had some bad out of the box
IRD's. Operationally, the biggest problem was dish alignment. Under
LNB's, we found many systems, particularly in Latin America, that were
still using LNA's or the older 270 MHz. output LNB's. Even replacing
LNA's with new LNB's didn't work unless the cable was also replaced,
since most LNA's used 50 ohm cable like RG-213 or RG-214. You'll
remember from last month that reflections or mismatch in the cable can
cause reflections that distort the phase and frequency response of the
system. Since LNB's and all digital IRD's are designed for 75 ohm input,
the mismatch from 50 ohm cable significantly increased the number of
errors. A lot of LNB's were not stable enough. My comment last month to
replace LNB's over two years old stands. Underline it.

Dish alignment also posed a problem for many stations. Many fixed dish
mounts do not make it easy to do fine adjustments. My experience working
with 3 meter dishes at our LPTV sites was that you could not obtain a
satisfactory bit error ratio tweaking for analog signal strength and
analog performance. With small (5 meter or less) dishes the final
tweaking had to be done by minimizing errors, not merely peaking signal.
This was confirmed time and time again as the tech's in Telemundo's TOC
(who, by the way, did an excellent job becoming digital satellite
experts in a matter of hours) talked to cable companies trying to reduce
dropouts in the digital signal. Polarity is critical, since interference
from adjacent transponders will cause significant errors. The General
Instruments "Digicipher I" system uses a narrower bandwidth signal and
probably has less problem with this than the Scientific Atlanta system.

Perhaps the most frustrating part of the whole transition was trying to
explain the differences between analog and digital to engineers at cable
companies and stations that had never dealt with digital signals before.
Cable engineers that had successfully launched digital radio or
Digicipher video knew what to expect an had few problems. I got the
feeling some tech's, once they saw the "perfect" digital picture, said
"good enough" and went home. It was very difficult to explain that a
perfect picture wasn't "good enough". The very nature of the digital
transmission guaranteed a perfect picture, provided the data got
through. You know the problems with "good enough" thinking and you've
got a jump on these people, especially if you read last month's column.

I don't want to give you the impression that digital video compression
is difficult to get working. It isn't, especially if you recognize any
weaknesses in your existing equipment and fix them first. Most of our
stations and cable companies switched without difficulty.

Comments on video quality have been very complementary. At one
commercial downlink site (with huge dishes) I heard the operator
compared our digital feed to Fox's analog feed and declared ours looked
noticeably better. So far I haven't had any reports of compression
artifacts like the ones I mentioned last month. For some reason,
however, we have gotten reports of an occasional (perhaps once a day)
glitch where the picture breaks up with a mosaic like block pattern for
a few seconds. As its very intermittent, I haven't been able to
determine if it happens at all sites at the same time, at some sites at
the same time or at random sites at various times.

Overall impressions? Quality is excellent. At sites with well aligned
dishes and no interference reliability is very good. For the rest of the
sites, I'd like to see the system handle brief hits (whether caused by
interference, poor dish alignment or lightning) better than it does.
Perhaps a freeze frame while the data stream gets re-established would
look better. Perhaps the length of each block of data needs to be
reduced (see last month's column) so that the error correction code can
recover from hits faster. Of course, this would reduce the data rate and
thus increase the change of introducing compression artifacts. I
understand Scientific Atlanta is working on a solution to this
occasional "dip to black". I have another suggestion that might make it
easier to align the dish for best performance. Add an easy to use analog
output that reflected bit error rate. This would allow an operator to
use a familiar tool (like a meter) to align the dish instead of
wondering why a 9E-5 was b etter than a 2E-4 but not as good as a 1E-6
error reading!

Competition Improves LPTV transmitter options...

A few months ago in my NAB column I mentioned an LPTV transmitter that
caught my eye because of its spec's and construction. The unit I was
alluding to is the I.T.S. Corporation Model ITS-830 1000 watt UHF solid
state transmitter. I.T.S. sent me a copy of the FCC Type Acceptance
Report for the unit, including schematics. What impressed me is the unit
looked like it was designed for high end broadcast applications. Full
correction circuitry, including ICPM correction and features like an
output bandpass filter (a must at many sites), circulators on the output
of each of the two 600 W driver amplifiers and full remote control and
metering belie the fact this is a budget transmitter. Reading over the
Type Acceptance Report showed the ITS-830 easily met FCC requirements
for Part 74 (LPTV) service. The results were better than many FCC Part
73 (full power) transmitters I've seen.

When reviewing a proof of performance or type acceptance report I like
to see values that are well inside the FCC requirements. If the
transmitter just squeaks by, chances are you'll have trouble making it
look as good on site. The ITS-830 met all the spec's easily, although I
did notice that for spurious output measurements an aural carrier level
of 5% rather than the traditional 10% aural was used. The lower aural
carrier level isn't unusual in Europe and probably doesn't have any
significant effect on coverage.

Acrodyne Industries was a pioneer in building solid state LPTV
transmitters. I know -- I bought several of them. (I wrote a review on
them in this magazine a few years ago.) The first ones had some
problems, which Acrodyne corrected. Later, Acrodyne surprised the LPTV
market by selling a solid state 1000 Watt LPTV transmitter - the
TU-1000/KSCE - for around $45,000. To meet this price point, they had to
take some short cuts. Linearity correction was minimal. Early units
didn't have ICPM correction until Bill Barrow at Acrodyne found a simple
way to improve ICPM. No bandpass filter was included. A circulator was
extra cost, as was a dual power supply configuration. However, even
after adding all these items it was still a bargain. The Acrodyne KSCE
series, in its bare bones configuration, remains the most economical.
Once the accessories are added, the I.T.S. unit becomes more than
competitive.

I'm pleased to see that Acrodyne has responded to the challenge by
creating a new transmitter designed to compete with the ITS-830. I hope
to see it next week. If I'm allowed, I'll let you know how it compares.
Actually, however the comparison turns out, I'm happy to see two
companies competing by improving quality as well as price. We've gotten
used to that in the computer field, but too often in the broadcast arena
price rather than quality seems to be the point of comparison. Both ITS
and Acrodyne desire credit for improving products while reducing costs.
Please don't take these comments as a slam on other LPTV manufacturers.
I've seen incremental improvements from them as well, but none as
dramatic at the new product introduced by ITS and the new product
undergoing testing at Acrodyne. If you are considering purchasing a
1,000 watt transmitter, check out all the options, compare the system
price once you've added the necessary accessories and finally compare
the specifications. G et a copy of FCC type acceptance data if possible.
It will show you how well the transmitter really can perform.

The last month or so I've mentioned some exploring I'm doing on what
some have called "The Information SuperHighway". I'm referring to the
Internet. Lately I've been exploring an aspect of the Internet called
USENET, which is a collection of electronic mail messages into news
groups organized by various topics. There are close to 4,000 different
topic groups in existence now and I usually get an opportunity to
subscribe to half a dozen more each time I log on. Some are truly
bizarre - there is a group titled "talk.bizarre" and another titled
"alt.pave.the.earth". Some are too strange to mention here. I won't deal
with those. If you cruise the Internet, you'll run into them soon
enough.

There are some groups which should be of interest to my reader(s). A new
group popped up a month ago called "sci.electronics.repair". In this
news group you'll find articles on repairing TV's, VCR's, CD players
among other things. Interesting reading. Another group,
"sci.engr.advanced-tv", seems to promise high tech info on HDTV, but so
far most of the articles have been on MPEG compression. A must-read
group for satellite users is "rec.video.satellite". This news group is
populated by a mixture of consultants (some "high powered"), an engineer
whose organization is listed as "Dreaded Evil Communication Company in
San Diego" who has many insightful comments on direct to home TV systems
and a large number of home dish owners who spend a lot of time scanning
the skies for interesting stuff. Chances are, if something happens on
satellites, you'll read about it here first. The most comprehensive
guide to satellite services I've ever seen is published regularly on
this news group. Yes, it s even better than WestSat, which is saying a
lot. This guide is clearly a labor of love as it doesn't cost anything
for news group participants. Spend some time on "rec.video.satellite",
contribute when you can, and enjoy the benefits. A final news group of
interest is "alt.satellite.tv.europe". If your vision of satellite TV is
limited to the big dish, (called "BUD" for Big Ugly Dish on
rec.video.satellite) this group will give you an idea what satellite
reception is like with multiple DBS options. A number of discussions
deal with programming, but there are also some interesting articles on
international satellite communications, changes on DBS birds throughout
Europe and scrambling (including defeating it).

If this has piqued your interest, the best over-all reference I've found
for getting on the Internet is published by Osbourne - McGraw - Hill,
authored by Harley Hahn and Rick Stout and called "The Internet Complete
Reference". If your method of access to the Internet is via a university
computer or dial up terminal, you need this book. If you decide to get
started with a direct connection to the Internet, I've found the
"...Tour Guide..." series to be a good starting point. Look for Tour
Guides including software for MAC, DOS and Windows computers. If you
don't mind heavy books, Sam's Publishing's "The Internet Unleashed" by
various authors covers all aspects of the Internet, right down to
details of network protocols. Software for the P.C. is included with
this book. If you get tired of reading it, the weight should make it
useful for killing large bugs or small rodents.

That's it for this month. I didn't have room to tell you about one of my
favorite USENET groups - rec.foods.drink.coffee - and probably won't
next month, but I do plan to answer some of the questions you've been
writing me with and cover some of the interesting things manufacturers
have been sending, along with any other interesting news that should pop
up. I've gotten seriously behind in answering letters due to all the
projects at work but hopefully by combining answering letters and
writing this column I'll help not only the original correspondent, but
others as well. I promise to try to do better in the future. The best
way to reach me is via EMAIL. I've several addresses - use
dlung@gate.net for Internet messages or my CompuServe address 70255,460
(70255.460@compuserve.com for Internet). You may also fax me notes at
305-884-9661 or phone me after 6:30 PM eastern time at my work number,
305-884-9664. If you must use mail, send it to my mail service at 2265
Westwood Blvd., Suite 5 53, Los Angeles, CA 90064. I'm trying to get
caught up, but for now expect a 6-8 week or more delay for mail
responses.


Copyright (c) 1994,1995 H. Douglas Lung ALL RIGHTS RESERVED