The discussion about GPS sync requirements for HD systems brought a few technical questions to mind"

What is the "analog diversity delay"--does this refer to the delay that is introduced to the analog audio so that at the receiver end, the analog and digital audio match in time? I haven't heard this term before.

Does an upgrade to HD require that the studio to transmitter link be of a specific type?

When a station does an HD conversion, what equipment changes need to be made at the studio?

How is audio processing handled? In other words, are HD exciters all-in-one boxes that include processing algorithms, or are external processing devices used?

Wow. I have to invite you to a party sometime....

> Wow. I have to invite you to a party sometime....

Ha ha ha ha ha ha ha ha ha ha ha ha ha!



Ha ha ha ha ha!

Some good nerdy stuff here:

http://www.orban.com/about/press/release/2007_0807.php

I think hd exciters can provide delayed output for the analog transmitter, early models were prone to crashing fairly often, taking the analog off the air...ouch!

can they boost up the bit rate for more high speed data,to increase theaudio defination by 2x or 3x?
if we get lucky, maybe some component they used in the generator will be prone to failure when it gets hot!!?????!!!hopefully causing HDshutdownw/analogdead air!!???

Alfredo, I don't know if the HD transmitters use the GPS for the same principal as we do in the paging business but here is how we use it and it seems somewhat similar.

We use GPS time sync in all our paging transmitters. It is used for two things. It is the reference frequency so the transmitter stays exactly on frequency and we use it for timing also.

In a market the size of Portland, for example, we have about 20 transmitters that all come up and transmit the same data at the same time, almost.

Since RF travels at 5.4 uS / mile, we can calculate where the overlap or simulcast zones are using propagation programs that take into consideration; terrain, antenna gain, antenna height, transmitter power, free space loss and several other variables.

We introduce a very small frequency offset of maybe 10 Hz to 60 Hz into each transmitter as well as a timing delay of anywhere from 5 uS to 70 uS.

The combination of the frequency offset and simulcast delay minimize the interference that the transmitters would cause each other.

I would have to assume that since the HD has an analog transmitter and a digital transmitter that maybe they are doing something similar?

It's only a guess, but it's all I've got!!!

Edit add: Here is an interesting article on simulcasting written by a gentleman I used to work with. He is known in the paging industry as the "Father of Simulcasting"!

http://www.braddye.com/simulcasting.html

Did you see the link that Chrisweiss posted on the HD ramblings thread:

http://www.npr.org/euonline/kcsn_report_1.pdf

Apparently IBOC transmitters do not (yet) have the "dial-a-delay" function, that would reduce timing issues from multiple transmitters . When timing differences from two transmitters are greater than 75us (and signals are within 10-20dB of each other) HD reception fails. They need to take a good look at what the paging industry is doing.

The discussion about GPS sync requirements for HD systems brought a few technical questions to mind"

What is the "analog diversity delay"--does this refer to the delay that is introduced to the analog audio so that at the receiver end, the analog and digital audio match in time? I haven't heard this term before.

The difference in time between the analog audio and the HD-1 audio. Ibiquity requires them to be "in sync." It's about 8 seconds.

Another problem is sync'ing the 44.1 kHz sampling rate to GPS at both the studio and transmitter ends. If they drift apart, the audio buffers in the Importer and/or Exporter can empty or overflow which will mute the audio.

Does an upgrade to HD require that the studio to transmitter link be of a specific type?

It depends on the location of the HD equipment (Importer, Exporter, etc.) and if HD-2+ is being transmitted. With just HD-1 you just have to make sure that your STL has 20 kHz of audio bandwidth.

For HD-2+ you need another 20 kHz stereo path, or about 300 kbps of data bandwidth if the "Importer" is to be at the studio end.

When a station does an HD conversion, what equipment changes need to be made at the studio?

None, as long as the programming has 20 kHz of audio bandwidth. A software update might need to be made to display artist & title for HD PAD/PSD.

How is audio processing handled? In other words, are HD exciters all-in-one boxes that include processing algorithms, or are external processing devices used?

External processing devices need to be used. Most folks went with new boxes that do the analog and HD-1 together, complete with the delay. Each HD2+ source usually needs its own processor. Since pre-emphasis and clipping are not needed for HD Radio, the processing can be a lot simpler, except for dealing with the effects of extreme data reduction.

I did a Google search on the terms "HD Radio Importer," and I came up with the following link:

http://radiomagonline.com/eyeoniboc/radio_hd_radio_system/

I will attempt to read through and digest this tomorrow. It looks like pretty good reading!

For the five Entercom stations at least, the Importer is at the studio. It's a PC with a really good soundcard. We put AES audio in and through the network it feeds the Exporter at the transmitter.

The Exporter does all the heavy lifting and feeds the exciter.

If you look on http://www.bdcast.com, the Importer will be an iDI-20 and the Exporter will be an FSi-10.

Then there's Exgine, but we don't use that configuration, yet.

IBOCNin band on channelnoise IBACNin band adjacent channelif station is usin a reason not to be AMStereo.doesn't the same reason exist to not do HD??be consistant please!!!

Interestingly enough, the STL has been a major stumbling block for solid HD installs. Previous to HD it was easy enough to get by with a 15 kHz audio link and a phone line to the transmitter for remote control data. Now you've got to be able to deliver very large packets of data to the transmitter site, and sometimes 20 kHz, stereo audio (linear preferred). Most microwave STL systems are just barely able to handle the data capacity.

As we started the HD rollout at the Clear Channel FM site, we switched over to a dark fiber for our primary STL link. The current mux on that fiber provides a 100 mbps WAN extension, as well as 4 T1 lines. The primary audio paths for the 3 FM signals (analog and digital) is provided by the T1 gear at 44.1 kHz. The AES signal is split for each station at the site into separate processors (Omnia and Orban) for analog and digital transmission (they don't come together again until they are a few feet in front of the radiating element on the panel antenna). The AES signal from the HD processor goes into the Exporter, and then via UDP data link to the exciter (Harris Flexstar today).

The HD2 signals all originate in audio servers of our NexGen system at the transmitter site. The low-latency WAN extension allows us an excellent connection to our database servers at the studios. The HD2 NexGen audio servers are running RCS Importer software. This links via the LAN to the Exporter for the SPS stream of the HD signal. The Importer tells the Exporter how to divide the data path before it is sent to the exciter via the Exgine card.

All told, our primary STL is running 106 mbps (with a bit of headroom). Previous to the HD rollout we were running 4.5 mbps. As Kent pointed out, these systems can be run on much lower bandwidth, but the stability can be compromised as you run close to the line. It will be interesting to see what kind of bandwidth will be required to move to the service modes in the extended hybrid operation (MP3-MP11). We are currently running service mode MP1. White papers at the Ibiquity web site provide a better description. http://ibiquity.com/broadcasters/quality_implementation/iboc_white_papers