Central Idaho Amateur
Radio Club
     
CIARC - Operating On Linked Repeater Systems
Operating On Linked Repeater Systems

The following information was presented at the August 2015 Central Idaho Amateur Radio Club general meeting.

This presentation discusses:

A .pdf copy of this article can be downloaded by clicking here.

Why Repeaters

VHF and UHF communications are primarily line of sight. Obstructions can prevent communications between two stations, even at relatively short distances.



Mountainous terrain will prevent station A from being heard by station B.



Similarly, mountainous terrain will prevent station B from being heard by station A.



One way to mitigate obstructions that are blocking communications is that we can send someone to the top of a mountain to act as a relay station. This requires assignment of a third operator, deploying the relay station operator, and relaying each message means that communications will be slow since every transmission made by either station A or station B must be repeated by station C to complete the relaying process.



A repeater, which is installed at the mouton top communications site, provides a permanent and automatic real time relay station.

A repeater, in its simplest view, routes audio from the repeater receiver to the repeater transmitter.



With a repeater, station A transmits to the repeater and the repeater retransmits station A audio, in real time, to station B. No relay station operator is required and there is no significant delay in communications.



Similarly, station B can transmit to the repeater and the repeater will retransmit station B audio to station A.



In a repeater, the repeater receiver squelch circuit creates a carrier operated switch, or COS.

When the repeater receiver detects a signal, or carrier wave, the COS signal tells the repeater controller that a signal is present at the repeater receiver.

The repeater controller responds to the COS activation by keying the transmitter while forwarding audio from the repeater receiver to the repeater transmitter.

When the receiver no longer receives a signal, the COS signal will go inactive, telling the repeater controller that a signal is no longer present. The repeater controller will delay unkeying of the repeater transmitter for a short period of time to accommodate signal drop out of short duration by the transmitting station. The delayed unkey of the repeater transmitter is called a squelch tail. All Central Idaho Amateur Radio Club repeaters have a squelch tail of 1.5 seconds in duration.



Without additional protection, noise sources, interference, or stations in overlapping coverage from a distant repeater operating on the same frequency, may gain access to the repeater transmitter. In order to protect the repeater transmitter from unintended access, a tone decoder is introduced. When a tone decoder is used, the repeater controller must see both the COS and an indication that the proper tone was decoded before the repeater controller will activate the repeater transmitter.

Tone decoders are slow, and can delay the keying of the repeater transmitter. In turn, a delay in keying of the repeater transmitter can cause a loss of information at the beginning of the transmission.




Repeater Control Delays - The Downside Of Protecting The Transmitter

In this timing diagram, the C O S indicates that the repeater receiver is receiving a signal. We can also see that the tone decoder, indicated by the C T C S S signal, will go active after a delay period. The delay period is simply the amount of time required by the tone decoder to detect a tone.

The tone decoder delay time has a relationship to the tone decoder frequency. Lower frequency tones take longer to decode than higher frequency tones, but higher frequency tones may be heard as a buzzing sound at the station receiver. Tone selection is always a compromise between tone decode speed and having a buzzing sound present at the station receiver.

Once the repeater controller sees both an active COS signal, and a valid tone, the repeater controller will activate the repeater transmitter by keying the push-to-talk or PTT signal.



Antenna Radiation Pattern Imposed Limitations

Antennas generally look toward the horizon if nothing is done to modify this behavior. This means that a mountain top antenna will see other mountain tops but not stations located directly below the mountain top antenna.



An antenna with a narrow vertical beam width will look toward the horizon and not support communications near the base of the mountain where the repeater is located.



An antenna with a wide vertical beam width will cover areas in closer to the repeater, but a log of the radiation pattern wastefully sits above the horizon.



Antenna radiation patterns can be mechanically or electrically modified to provide down tilt. This reduces waste of the radiation pattern above the horizon, and provides coverage in areas closer to the repeater.



But something is still wrong...

The areas close to the repeater are not in the radiation pattern and may suffer from degraded communications capabilities.

The areas behind other obstructions, such as mountainous terrain, are shadowed and will also suffer from degraded communications capabilities.



So, how do we mitigate these deficiencies?

By adding repeaters with complementary coverage, the areas below the repeater, and shadowed areas, can be filled in. In this slide, the left repeater sees below the right repeater and the right repeater sees below the left repeater.

But there is still a problem to solve.

Repeaters are on unique channels or frequency pairs in order to avoid interference between repeaters. A station may have to move from one repeater to the next, but if the station they are working doesn't have coverage on the other repeater, and is unable to move to the same repeater as the station Being worked is moving to, then communications will be lost.

Linking repeaters attempts to eliminate both the loss of coverage and the loss of communications. Linking ties the repeaters together. Anyone using one repeater is heard simultaneously on all repeaters in the network. Only the station that is loosing coverage need change to a repeater with coverage, and doing so will not result in a loss of communications.




Repeater Linking 101

A set of interlinked repeaters forms a repeater network.

In a linked repeater network, each repeater usually forwards audio to a central distribution point, or hub. The central distribution point, or hub, then forwards the audio to each satellite node repeater.

Links can be implemented through a number of means, including a dedicated phone line, radio frequencies, or one of several voice over internet protocol (i.e V O I P) methods.



Up linking is the process where satellite node repeaters forward audio to the main distribution point, or hub.



Down linking is a process where audio is distributed from the main distribution point, or hub, to each of the satellite node repeaters.



But why link?

Repeater coverage can be increased by one of three methods:

Increasing receiver sensitivity and transmitter power has limited utility and can cause de-sense behavior.

Increasing antenna gain has limitations due to available gain, and can be expensive.

Altitude increases also can be limited by site availability or by prioritization of antenna position on the tower by the site manager.

By linking repeaters, the propagation predictions of each of the individual repeaters within a repeater network can be composited together, resulting in not only an increased coverage foot print, but also better penetration into marginal repeater coverage areas.


About Repeater De-Sense Behavior

With de-sense behavior, the repeater receiver is less sensitive when the repeater transmitter is active. A weak signal may be able to cause the repeater transmitter to go active, but once the repeater transmitter goes active, the repeater receiver becomes less sensitive, and the weak signal may become noisy, or, in a worse case scenario, become too weak to hold the repeater transmitter in the active state. If the signal drops out due to de-sense that is introduced by the repeater transmitter being active, the repeater transmitter will go inactive. Once the repeater transmitter goes inactive, the repeater receiver can, once again, hear the weak signal and then reactivate the repeater transmitter. This causes an endless cycle of the repeater transmitter going active, then inactive, then active, etc.

Such behavior can be extremely difficult to avoid where there is insufficient isolation between the repeater transmitter and repeater receiver. Some repeater bands are more difficult to achieve sufficient isolation than others. In cases where sufficient isolation cannot be achieved, it is advisable to reduce the repeater transmitter power level, if necessary, to avoid de-sense behavior.

De-sense behavior should always be avoided. This is one reason why the CIARC repeaters do not run with high power transmitters. If you cannot communicate, due to de-sense behavior, or have to move off of the repeater channel to a simplex channel to communicate, the repeater has lost all utility. It is far more useful to have a repeater that can hear at least as well as it is heard than have a repeater that can be heard over a vast area and not hear a signal that is well within the repeater coverage footprint area.


Repeater Coverage Prediction

About Repeater Coverage Prediction

In each of the repeater coverage prediction images, red indicates strong coverage while yellow indicates weak coverage. Please note that the software used to generate the coverage predictions was authored in Canada, and uses the metric system to describe coverage prediction area.

Please also note that the predictions smooth the terrain by the prediction resolution. Any obstructions within the resolution area are artificially smoothed and will result in errors in the predicted coverage. The resolution is 500 square meters.

An example of how this smoothing produces errors in the prediction coverage occurs where the No Business Mountain repeater antenna is seen as being at the same elevation as Lookout Peak while the antenna is actually located on the north downward slope, relative to the peak and approximately 30 feet lower than the peak. The coverage prediction does not show the realized degraded performance to the south that the obstruction of the peak creates. As such, the predictions can be, and are, overly ambitious, and the realized coverage footprint is often smaller than the prediction. The predictions only approach an accurate value when there are no obstructions within the area of prediction resolution.


Brundage Mountain - KC7MCC Repeater - 146.900 MHz


STRONG SIGNAL COVERAGE 21,584 km2 8,334 sq. mi.
WEAK SIGNAL COVERAGE 41,845 km2 16,156 sq. mi.


McCall - N7IBC Repeater - 444.125 MHz


STRONG SIGNAL COVERAGE 2,019 km2 779 sq. mi.
WEAK SIGNAL COVERAGE 3,540 km2 1,367 sq. mi.


Cascade - W7CIA Repeater - 441.925 MHz


STRONG SIGNAL COVERAGE 1,098 km2 424 sq. mi.
WEAK SIGNAL COVERAGE 1,910 km2 737 sq. mi.


No Business Mountain - KC7MCC Repeater - 147.020 MHz


STRONG SIGNAL COVERAGE 20,287 km2 7,833 sq. mi.
WEAK SIGNAL COVERAGE 38,363 km2 14,812 sq. mi.



Please take special not on the transition from previous image to the next image as this best illustrates what has been done to improve repeater network coverage when comparing the stand alone 2-meter repeater on No Business Mt. and the network of repeaters that is now available for use.


CIARC LINKED REPEATER NETWORK - CONPOSITED COVERAGE PREDICTION

COLOR DESCRIPTION
RED This color represents the area of the repeater coverage footprint that presents strong signal coverage.
YELLOW This color represents the area of the repeater coverage footprint that presents weak signal coverage.
ORANGE This color results from the mixing of red and yellow, due to overlapping coverage from at least one repeater with weak coverage and one repeater with strong coverage, and represents the area of the repeater coverage footprint that presents strong signal coverage from at least one of the repeaters.



Accumulating Repeater Control Delays When Keying Up

In order to best take advantage of a linked repeater system, it is necessary to both understand the impact of control delays and how to best interact with repeater network system behavior.

This section will explain the impact of control delays, and discuss both a hardware solution and practical operating procedures that will mitigate the impact of control delays.

This set of slides shows what happens to the repeater network when you initiate a transmission. We'll be walking through this sequence step by step while showing the progression on a time line, and with a goal of understanding how control delays accumulate. In our time line pictorial, earlier events are on the left of the time line while later events are on the right.

The thing to keep in mind is that we're only applying what we've already learned about repeater control delays in the earlier discussion of repeater control and tone decoder delays, but, instead of looking at a single repeater, we are now going to look at repeaters that repeat to repeaters, that is, the repeater network as a system.

Let's consider longest path through the repeater network, which is when a station accesses a satellite node repeater to communicate with another station that is accessing a different satellite node repeater. This situation presents the worse case control delay through the repeater network.

Lets start with station A beginning a transmission through the node 1 satellite repeater to station B, who is using the node 3 satellite repeater. This will cause the COS at the node 1 satellite repeater to go active.



After the tone decoder delay, which is approximately 200 milliseconds, the tone decoder will signal to the repeater controller that a proper tone is being detected.



Once both the COS and the tone decoder indicate that the received signal should be retransmitted, the node 1 satellite repeater controller will key the repeater transmitters (i.e. both the repeater transmitter and the uplink transmitter).



When the uplink transmitter at the node 1 satellite repeater goes active, this will cause the COS signal at the hub uplink receiver to go active.



After the tone decoder delay, which is approximately 200 milliseconds, the tone decoder will signal to the repair controller that a proper tone is being detected.



The hub repeater controller will now activate the hub repeater transmitters (i.e. both the repeater transmitter and the downlink transmitter).



After the tone decoder delay, which is approximately 200 milliseconds, the satellite node repeater tone decoder will signal to the repair controller that a proper tone is being detected.



After both the satellite downlink COS and tone decoders indicate a valid signal, only then will the satellite node repeater transmitter be activated. This occurs 600 milliseconds after station A initiated transmission.

If the audio from station A is not delayed by 600 milliseconds, then station B will not hear any of the audio that was transmitted during the cumulative control delay of 600 milliseconds.

In order to avoid loss of information, or requests to retransmit, the audio must be delayed through the network. The only question is how to implement the audio delay.



But wait, it gets worse...

Other repeaters are able to connect to the Central Idaho Amateur Radio Club repeater network by using a remote base. For example, the Idaho ARES District 3 NET establishes a remote base connection from the K7ZZL repeater, located on Snowbank Mountain, to the Central Idaho Amateur Radio Club repeater network. Further, the K7ZZL repeater also uses a link to the K7BSE repeater that is located in the Treasure Valley. These two connections have the potential to add two additional tone decoder control delays when communicating between Treasure Valley stations and stations on the Central Idaho Amateur Radio Club repeater network.

There are two options to mitigate the effects of tone decoder control delays.

The first option is to install audio delay boards to delay audio from the receiver to the repeater controller, with the delay period equal to or exceeding the tone decoder delay.

For this hardware solution, a delay board is required for each receiver. That is one for the repeater receiver and one for the link receiver. With two receivers per site, and four sites, that works out to eight audio delay boards, at an expense of $100 per board, or $800 system wide.

The second option for mitigating tone decoder control delays uses an operating procedure that will result in no loss of information, a reduced number of requests for repeated transmission, and best of all, it is free.

The operating procedure has the transmitting station key the transmitter, wait for approximately one half second, and then begin speaking. In essence, the station operator is acting as a hardware audio delay board by delaying their speech relative to when they key their transmitter.


A common method of counting seconds is to count one-thousand, one, one-thousand, two etc. This can be applied to radio operating procedure by:

  • Key the transmitter
  • Count off one-thousand silently and to yourself
  • Begin speaking

The operating procedure is a good option, and only fails when participating station operators fail to implement the procedure as described. Adherence to the operating procedure is a simple matter of training and developing the proper operating habit.

For Idaho ARES District 3 NET operations, the additional tone decoder control delay that is introduced by connections to additional repeaters requires that participating stations increase the delay between keying the transmitter and speaking. Although it is known that the K7ZZL repeater does implement audio delay boards, it is not known if the K7BSE repeater does so. Counting off one second between keying your transmitter and beginning to speak will ensure that all transmitters in the network have activated prior to your speech being conveyed across the network.

This behavior is the very reason why the Idaho ARES District 3 NET control operator specifically asks that all stations key up and wait for the links to come up. The procedure described here is put into place but the radio operators are not given sufficient information to understand why. You now have that information.

Accumulating Repeater Control Delays When Un-Keying

Control Delays also stack up when un-keying, and can also present behavior that is extremely detrimental to communications. Just as it takes time for each tone decoder to detect a tone, it also takes time for each tone decoder to detect the absence of a tone. The detection of the absence of a tone results in a delay in un-keying of the repeater transmitter. And just like the stacking up of delays through the linked repeater network when keying up, the delays also stack up when un-keying.

According to FCC regulation, repeaters must implement a timer to limit the keying of the repeater transmitter to a maximum of 10-minutes. If the repeater does not observe the received signal going inactive within 10-minutes of activating the repeater transmitter, the repeater transmitter is said to time-out and the repeater transmitter is then disabled until the repeater observes the received signal going inactive. Once the repeater observes the received signal going inactive, the repeater transmitter time-out timer is reset and the repeater will then repeat the next received signal.

With control delays stacking up when a station un-keys, all stations must observe not only the stacked up control delays, but also the length of the repeater squelch tail before the next transmission occurs.

The repeater squelch tail is that time that the repeater transmitter remains active after the received signal is determined to have gone inactive. For all repeaters in the Central Idaho Amateur Radio Club linked repeater network, the squelch tail timer is set to a duration of 1.5 seconds.

In order to avoid experiencing the repeater transmitter timing out, and loosing all information that is is being conveyed until the end of a user station transmission, participating stations must delay, after un-keying, for a duration equal to the stacked up control delays plus the length of the repeater squelch tail timer in order to avoid causing the repeater transmitter time-out timers from disabling the repeater transmitters. This is even more critical during NET operations, where vital information could be lost should the repeater transmitter time-out timer expire.

It is advised that participating stations that are using a repeater that is not a full-time participant in the Central Idaho Amateur Radio Club linked repeater network (i.e. K7BSE or K7ZZL), wait 3-seconds after the participating station un-keys their transmitter, before keying to begin a transmission. If you are using one of the repeaters that comprises the Central Idaho Amateur Radio Club linked repeater network, waiting 1 second after the repeater transmitter drops should be sufficient.

Central Idaho Amateur Radio Club Repeater Network Topology

COLOR DESCRIPTION
RED Signal paths that are depicted in red represent the Central Idaho Amateur Radio Club repeater netework, and show link paths that are available full-time.
AQUA Signal paths in aqua depict a link path that are enabled by commanded or by operation of a command scheduler according to a regularly scheduled event, such as the Idaho ARES District 3 Station Readiness NET. When enabled, communications between the Boise K7BSE repeater, operating on 146.940 MHz -Offset 100.0 Hz CTCSS, and the Snowbank Mountain K7ZZL repeater, operating on 443.300 MHz +Offset 110.9 Hz CTCSS, are possible.
VIOLET The signal paths in violet is established via a remote base and is enabled in the same manner as those signal paths depicted in aqua. When enabled, communications between the CIARC Repeater Network (i.e. repeaters linked in red or orange) and the Snowbank Mountain K7ZZL repeater, operating on 443.300 MHz +Offset 110.9 Hz CTCSS, are possible.
ORANGE Signal paths that are depicted in orange are commanded connections that are only available in support of emergency communications services, or in support of public service communications when prior agreement has been obtained. The signal path is enabled only by specified control operators within the CIARC or KA7ERV repeater organizations, and requires either prior agreement (in support of public service communications) or an immediate need in support of emergency communications. This signal path will not be enabled for casual use.
AQUA VIOLET
When both the aqua and violet signal paths are enabled, communications between the Boise K7BSE repeater, operating on 146.940 MHz -Offset 100.0 Hz CTCSS, the Snowbank Mountain K7ZZL repeater, operating on 443.300 MHz +Offset 110.9 Hz CTCSS, and the Central Idaho Amateur Radio Club repeater network are possible.



Idaho ARES District 3 Communications

With the installation of the Central Idaho Amateur Radio Club 2-meter repeater at Brundage Mountain in early July of 2015, this provides an unobstructed signal path between the K7ZZL Snowbank Mountain repeater and the Brundage Mountain repeater. The August 17, 2015 Idaho ARES District 3 NET session saw the first operational use of having the K7ZZL remote base reconfigured to use the Brundage Mt. repeater, and demonstrated full-quieting bidirectional communications between stations located in Adams and Valley counties and stations located in the Treasure Valley.