JULY 1998
Roy Trumbull - Editor
Bill Dempster - Artist
Advertisers for this month's newsletter are:
Zack Electronics - (800) 998-3947
Advanced Marketing - Frank A. Santucci - (650) 365-3944
Riggins Electronics Sales - George Riggins - (662) 598-7007
Orban - Rick Sawyer - (360) 715-1913
MARCOM - Martin Jackson - (408) 768-8668
William F. Ruck, Jr. Broadcast Engineer - (415) 995-6969
Communications Law Center - Philip M. Kane - (650) 369-7373
RF Specialties of California - Bill Newbrough - (888) 737-7321
Hammett & Edison, Inc. - Dane E. Ericksen, P.E. - (707) 996-5200
LeBLANC - David A. Hill - (650) 574-4600
Pacific Research & Engineering Corp. - William Hopkins - (760) 438-3911 Keith Davidson & Company - Keith Davidson, CSBE - (707) 648-0412
Improbable Missions Fource - Mike Schweizer - (888) 4-ISDN4U
Svetlana Electron Devices - George Badger - (800) 578-3852
TFT Inc. - Jesse J. Piatte, Jr. - (408) 727-7272 x504
Audio Accessories - Rosie Alexander - (510) 787-3335
Brill Electronics - Marcie Mearns / Field Sales - (510) 308-1248
Brill Electronics - Larry Shore / Inside Sales - (510) 834-5888
Scala - Michael Wm. Bach - (541) 779-6500
Scala - Everett E. Helm, CPBE - (541) 779-6500
Harris - Ed Longcrier - (800) 315-7285
Anixter - Judy Conner - (510) 489-7430

Our speaker this month will be Al Jason of CMBE, Inc. They do both studio and RF projects.

As usual our luncheon will be at Sinbad's. Sinbad's is just south of the Ferry Building on the Embarcadero near the foot of Mission Street. Please RSVP to Karen Prasek at Zack's: 408-324-0551 x126 as we've been running out of tables and chairs. We meet at 11:30 and are seated at 12:30.

FM Subcarrier Available
Soliciting bids to lease an FM subcarrier channel in the San Francisco Market. Bids are due 8/21/98. For more information contact Pat May, San Francisco Unified School District, (415) 695-5937 Refer to Bid #522.

Senior Electronics Technician
Office of Media Services
University of California at Berkeley
$3179 - 3840 (full-time rate)
Hours: to be arranged
Job Number: 05-241-20/MO
Closing date: Until filled
Job Duties: One full-time or two part-time positions, ending 3 - 6 months from date of hire. Under general supervision, provide technical support for audio-visual systems, video data display units and equipment, both portable and installed. Install, fabricate, modify, repair, and perform preventive maintenance on a variety of audio-visual equipment.

For complete details contact:

NDS says they have a complete package which will be tested at the model station before the end of summer. GI (General Instruments) is showing an interesting packaging concept for an encoding system at their San Diego facility but they have yet to complete all the microcode. (I'm going down to look at their encoder.)

Essentially the image is divided up into strips each of which is handled by a separate encoder. The extent to which the strips overlap and motion vectors are able to be communicated across the boundary is where all the art is at the moment. Approaches to the problem vary.

I got a note via Ray Smith of KCSM to the effect that the Harris DTV roadshow was fully equipped with an ATSC encoder and decoder that were used for their DTV demo of over-the air simulations.

In a matter closer to home, the KRON-TV July 4th DTV demo uplinked a QPSK satellite format that was decoded and delivered to the monitors as video. No ATSC RF tuner/decoders were used in the demo units. The video was 1080i.

On the morning of June 16th KNTV's traveling wave antenna fell off the tower. Our web page has a link to a site with pictures. It appears there was a break above where the mounting plate bolted to the tower. There seems to be rust around about 300 degrees of the circle but I don't know if there's been a chemical analysis to confirm that.

The weather wasn't a factor. Regular visitors to the Loma Prieta site reported hearing a noise present only at low wind speeds which has led to speculation about resonance. No one can say for sure that the noise came from the antenna. I'm not aware of any official analysis.

Ray Smith of KCSM has some interesting comments and insights into the puzzles of ATSC. It is one thing to have a system and quite another to see how it actually winds up being implemented. Because we're out of room again, I'm posting Ray's comments at our web site as part of the web version of this newsletter.

If you have an email address, please contact Warren Reese at Warren is in charge of our email list, which we use when we need to notify members when something has happened between newsletters.

Webmeister Tim Pozer has the current newsletter plus newsletters all the way back to 1996 at:

DTV implementation issues that are causing some confusion
By Ray Smith

There are some fundamental DTV delivery concepts that have been producing misunderstandings amongst some engineering circles including the ones I belong to.

Please understand that these are a snap shot at this time of my own opinions and are offered only as topic points for further discussion. No offense is meant to be issued to any one. I reserve the right to be wrong and at the same time welcome being corrected where I have gone astray. I do not portend to be an expert on anything, I am just an engineer in the trenches that has made some observations at one time or another.

Concept A: according to the great plan, the ATSC data stream does not care nor does the ATSC compliant receiver care what (frame rate) profile is being sent to it. What is important is what happens during the approximately 70 nanosecond group of packets and the approximately 50 microsecond frame data rate.

Concept B: You do not need to convert 24 frame material to 60 frame just for the sake of last mile broadcasting. (frame rate phobia) If you have a choice between pretty pictures with some artifacts or pretty pictures with less artifacts which would you choose? Then broadcast that.

Concept C: Opportunistic data encapsulated with the video is not the only data space available. There is plenty of "holes" in the ATSC stream that can be filled with data. There is space for data in the audio packets - without bothering the 5.1 main channels or even the second audio channel(s), there is program related (time sensitive) data space, there is non time sensitive data space, there is as yet unallocated data space and there is even enough room for ATSC reserved (as yet to be announced what they need the space for) space.

Concept D: If you broadcast 1080 or 480 you are doing so within a 50 microsecond group of packets that also have plenty of room for other information to be carried with it. In other words, you are not "filling up" all the bandwidth only when you are broadcasting 1080, but rather you are "filling up" all the band width if you are broadcasting any compliant format.

You cannot send any more or any less information than what will fit into the 70 nanosecond by 50 microsecond group of packets. Also recall that in digital, the quantity nothing is data. In fact that "no data" data is going to be randomized before broadcasting in such a way that it will cost you transmitter power in order to send "no data" information.

Concept E: Data does not have to be delivered in maximum chunks and highest throughput rates in order to be useful. One bit every 70 nanoseconds over the course of a day is a fairly sizeable, over the course of a year it is quite massive. Just think what you could do with two bits...

Concept F: We cannot engineer our way through the digital transition as we have in times past. At this stage we will have to post our requests and suggestions to the manufactures and just hope for the best as the tools develop. It is in our own best interests then as engineers to keep the communications between us open and not to fall prey to marketing hype, vaporware and market share segmentation.

We've been approached by a firm which makes a device to detect null packets in an existing stream and substitute a properly formatted data packet.