Remote Autopatch Link Ideas and Techniques

Up one level (Tech Index)
Back to Home How do I get a phone line linked into my repeater site?
by Kevin Custer W3KKC
Maintained by Mike Morris WA6ILQ

The Question: (from the repeater-builder mailing list...)

>I want to provide autopatch to my repeater but I cannot
>run a phone line up to the site. The site is within a
>few miles of my house and line of sight. I want to RF
>link the phone connection to a second phone line at my
>house. Does anyone know of a stand alone RF link that
>can replace the phone line input to the controller? This
>problem has to have come up before and someone must have
>an inexpensive solution.

My Reply:

A while back I had a site that couldn't be fed easily from the local phone company, so here is the solution I came up with:

My controller is an ACC RC-96 which, at the time, the factory said it could not be modified for a radio phone link. I opened the controller and with the aid of the schematic discovered it was going to be easy to separate incoming and outgoing voice signals because it already existed that way before it was turned into duplex audio. This was the easy part.

There were concerns as to how I was going to complete signaling and audio to and from the site to the remote location. One concern was the remote site (my home) only had one phone line and if it were busy, I did not want the autopatch to go through. Since I was using my own phone line I couldn't use the reverse patch or call myself on the autopatch, because you can't call yourself on your own phone line. If I would have paid for a separate line at my house (the preferred way) this situation would not happen.

To make a long story a little shorter, here is what I did to make the patch work:
1) I converted the autopatch section of the controller from duplex audio to separate receive and transmit audio paths. Most controllers that have an autopatch option are similar.
2) I built two UHF links from Hamtronics transmitters and Motorola Motrac receivers. Other brands would work just as well.
The uplink and downlink frequencies were separated in frequency by between 6 and 7 MHz. I chose (and coordinated) an odd split to help confuse people wanting to hack it. The increased split also provided a little more isolation because I did not want to have to use duplexers. My 3 element homemade beam antennas were separated by about 10 feet vertically, which with the tight front ends of the Motrac receivers was enough.

At the house, I used a Heathkit hybrid phone patch to connect the uplink transmitter and downlink receiver to the phone line. At the repeater site, I connected the downlink transmitter and uplink receiver to the controller's modified telephone port.

I programmed the ACC 96 to regenerate the touch tones into the phone line, and also to wait till the phone line is stable before dialing. Most controllers that have an autopatch option have this ability. This pause before touchtone transmission made it nice because I did not want the duplex link in service except when the autopatch was actually in use. I connected the coil of an additional relay across the autopatch offhook relay in the controller. Contacts of this added relay supplied power to the downlink transmitter at the repeater site when the autopatch became active (like grounding a PTT line). For this method to work, the controller must be able to regenerate the touch tones and the remote phone line must have touch tone dial service.

At home, I controlled a relay from the link receivers COS line (you can use the PL decode line if your link receiver has a decoder or can have one added - the ComSpec TS-32 can drive a relay). When the link receiver at home became active, this relay connected the phone line to the Heath Hybrid Patch, and keyed the uplink transmitter from my home back to the repeater site.

When you used the autopatch, here is what happened:
At the repeater site, the controller closed the relay I added and the downlink transmitter became active. As stated earlier, the COS (or PL decode) on the link receiver at home sensed the repeater site downlink transmitter signal and closed the relay connecting my phone line to the hybrid phone patch, and also activated the uplink transmitter back to the repeater site. The controllers pause allowed the system to become stable before the dialing tones were dumped into the downlink transmitter and ultimately into my home phone line. Now, since the uplink transmitter was activated at my home, the audio from my phone line was available back up at the repeater site, completing the duplex audio path.... remotely. You could actually hear the dial tone from the house phone line before the controller started dialing. The receivers at both ends were on all the time, ready to hear something from the other end.

One other situation.....
As I had said, the remote location was my home, and I didn't want the autopatch to dial over an existing conversation (my wife????) so here is what I did. The ACC RC-96 has a function called "Shared Phone Line". This is a logic input to allow the controller to sense that the phone line (at the site, if there was one) was busy. I went to Radio Shack and bought a telephone recording device. This device sampled my home phone line and if the line was offhook, a relay would activate (normally intended to control a tape recorder). I used this device to control the uplink transmitter at my home. If my phone line was in use the uplink transmitter at my home would become active along with a PL tone. The PL tone was decoded at the repeater site and this logic was used to tell the controller that the phone line (at my home) was busy and not allow dialing. The controller would say "BZ" in Morse. Because I used my own phone line I never wanted the reverse patch to work, so I never went through the effort of sending a ringing signal through the link to tell the controller someone was calling. Again, if I had chosen to pay for a second phone line at my house this situation would not happen. Reverse patch ringing could have been indicated simply by using a different, second PL tone on the uplink transmitter (and a second PL decoder on the uplink receiver) to simulate a ring signal into the original autopatch ring detect circuitry. Multiple tones are easy using a ComSpec TS-32, just jumper an open DIP switch with a relay contact. They are just as easy with a reed based encoder, just use DPDT relay contacts to select a second reed.

I used the controls on the hybrid patch to null and adjust audio levels to get things sounding right from the downlink receiver into the phone line and from the line to the uplink transmitter.

The path was 16 miles one way but almost line of sight. The little two watt Hamtronics UHF transmitters were plenty of power for DFQ both ways, 32 miles round trip. This was a cheap, easy way of doing a phone patch remotely, and worked for over 8 years with little difficulty. The phone link is not in service now due to me moving (my home) to another location and I just have not had time to get it back up yet......but I will...... even though I'm now a little farther from the repeater site, I'm still line of sight!

Hope this helps,
Kevin Custer W3KKC
Last updated March 2003

PS - If you can't find one of the Radio Shack tape recorder controllers to sense the phone line status you can use the circuit described here: A Phone Line Busy Sensing Circuit.

The below was added Febuary 2005:
It was pointed out in an email that the description above implies that the remote autopatch / telephone point-to-point link system was operating illegally for multiple reasons:

1) As described the link transmitters appear to not be identified. Well, the the Motrac receiver audio circuits have low-cutoff audio filters (around 210 Hz) in them from the factory to keep PL tones out of the received audio (the highest PL tone at the time they were designed was 203.5 Hz). The link transmitters PL tone frequency was in the 70 Hz to 95 Hz region, and both had identifiers in the 170-180 Hz region. The PL decoder was not affected by the ID tone, and the ID was not heard in the autopatch audio since the low ID frequency was masked by the factory audio filters while being quite audible in an unfiltered receiver. Past case history shows that the FCC considers such "masked" IDs perfectly legal as long as they are plainly audible on a receiver that has no audio filtering.

2) As the system is described, the uplink transmitter is keyed when someone in the house lifts a phone offhook, and a dead carrier is transmitted for the duration of that offhook period. This technical situation happened due to the number of available relay contacts on the relays that were in the junkbox at the time the circuit was constructed. I was not worried since my wife at that time was licensed, and there were no children in the house. If this situation had changed, the fact is that the uplink PL encoder is the true indication of offhook status, and the system design could have been modified to key the uplink PTT from the downlink COR through the phone line monitor relay contacts (i.e. the uplink PTT would not be active unless there was a downlink signal AND the phone was onhook / not in use), thereby having a licensed ham (the one initiating the autopatch phone call) in charge of all PTTs. The design change would simply have involved replacing a DPDT relay with a 3PDT or a 4PDT and moving a few wires around. And besides, twice in the above article I clearly state that having a separate phone line dedicated to the patch is the proper and recommended technique.

Neither of the above points was mentioned in the original article simply because both are legal aspects as opposed to hardware design aspects, and the original article was intended as a technical dissertation intended to "jump-start" the reader into developing his own solution to his own situation.

And besides, the system described in the above article has not been in operation for some time (years). Cellphone proliferation has killed most autopatches.

Note from the editor / page maintainer Mike WA6ILQ:
The USA amateur radio UHF band is 420-450 MHz, and in most areas the top 8 to 10 MHz is used for FM repeaters. In every case the local band plan determines the use of the rest of the band, but in most areas there is space allocated for point-to-point FM links. In the USA the 406-420 MHz range is 95% used by government agencies, hence is dubbed the "government" range. Radios in this range are available in surplus and work quite well as 420-425 MHz range links. Many 450 MHz radios can be crystaled up and tuned to as low as 436-437 MHz without problems. You could put your autopatch link at 420.5 MHz or so in one direction and 439.5 MHz or so in the other direction. Just ask your local coordinator for the link band plan and coordinate a pair that works with your available hardware. Even if you are in an area where the band plan has reserved 440-442 MHz for links I'd be willing to bet that the coordinator will have no problems with assigning you a 420.xxx / 439.xxx MHz pair. And 900 MHz or 1200 MHz are other viable alternatives.

Update 2024: A 2.4 GHz point-to-point link is under $175 per end and 20+ mile links are common (the record is 140 miles, from Bertha Peak in California to Potosi Mountain in Nevada). With one of those you could bring the internet to your site and have not only an autopatch but also allow an Allstar link and an internet-connected laptop for remote repeater controller managememt / programming. I'm doing this at 7 mountaintop sites.

Back to the top of the page
Up one level (Tech Index)
Back to Home

This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.