# Would it be feasible to set up a SSB or AM 2 meter repeater?

Would it be feasible to set up a SSB or AM 2 meter repeater? I want to set up an AllStar full duplex node and like the idea of AM because stations can double without issues. SSB works for weaker signals and it takes up less bandwidth. The circuitry is also way simpler. Just set up a PL tone on the repeater so interference does not get through.

Would it work?

• It's called a "linear translator". – Phil Frost - W8II Mar 10 '15 at 13:24

(This is a purely theoretical answer; I have no experience in repeater building. Sorry.)

The components of a repeater are:

• an antenna,
• a duplexer,
• a transmitter and a receiver,
• and a repeater controller.

The antenna and duplexer are passive devices and aren't affected by what modulation you're using. The repeater controller is on the audio side of the radios, not the RF side. So there is no fundamental reason this can't work.

However, using a CTCSS tone together with SSB modulation would not work. This is because SSB causes an audio frequency shift if the transmitter and receiver are not exactly on frequency (as they never are), so the tone sent by a user of the repeater will not be recognized by the repeater controller unless the tuning is exact to within a few Hz.

(With AM, the carrier provides the frequency reference. With FM, as normal VHF repeaters use, the audio frequency is not related to the RF frequency.)

There are various other coded squelch (selective calling) systems available, some of which might do better than CTCSS, but many of them are only a burst at the beginning of the transmission and thus assume the signal has a carrier which the repeater can use to know when to stop repeating, which is not true of SSB.

Incidentally, there are existing repeaters, of a sort, which can use SSB/AM. There is a device called a linear translator, or linear transponder, which is completely mode-independent: it receives a certain band, frequency shifts it to another band, amplifies, and retransmits it.

These systems can be used with any mode you care to transmit at them. The cost is that since they are not demodulating, they don't have any sort of intelligent handling of the signal (such as tone squelch, timeout, etc.).

Linear transponders are most often found on amateur radio satellites (those that do not carry FM repeaters instead). Because the satellite has very limited power available, it is polite to use SSB, CW, or other power-efficient modes as opposed to AM or FM.

I find the idea of an SSB repeater quite challenging. A repeater has noteworthy the function of allowing communication between points which are mutually inaccessible along the direct line. A repeater has not the function of giving the word (the channel) to one sole participant a time (sic!). I greatly prefer the convivial chat in SSB on the HF bands to the scholarly sequenced conversation in FM on 2m and above! Therefore, the SSB repeater would just be a continuously operating frequency translator, installed at a high place, which takes in a first, lower frequency band, e.g. a 12kHz channel at 434MHz, and gives it out, transposed and amplified, in a higher frequency band, e.g. at 439MHz. For putting this in place, two problems must be solved: 1) to achieve an amplification of 90dB without feed-back, and 2) to achieve steep (rectangular!) channel filtering to avoid crosstalk. The solution to the problem is in the combination of a digital I/Q receiver at the first, lower frequency, e.g. based on the Realtek RTL2832U chip, with digital filtering, amplification and processing to isolate the 12kHz baseband signal, and reemitting the baseband signal at the second, higher frequency with the help of an I/Q switching modulator, followed by a HF amplification stage.

The advantages of passing through the baseband are a) the distribution of the necessary gain stages between three different frequency levels, reducing possible internal feedback problems, and b) the ease of realizing a very steep and phase-linear filter window for the 12kHz channel.

The advantages of digital signal processing are the ease of realizing time-delayed phase-linear FIR filters and the total avoidance of external feedback by an appropriate combination of time-delay between input and output signal, and high-pass filter cutting off all low-frequency signals beyond the time delay of the signal processing chain.

Input and output frequency may be in the same, or preferably in different bands.The operation on the repeater is as ordinary operation, but in duplex.The repeater processes any type of analog or digital input signal and any type of modulation to the frequency-translated output.

Remains the question whether such repeaters are admissible after all, because they work in continuous operation, monopolizing one channel (the emission channel) from the band.However, this might be worth the value of providing a real chatting option in SSB or CW via a repeater. Furthermore, if there is no activity on the input channel, the output is silent, too.

In any case, HAM radio is an experimental discipline, and this would certainly be a field of new experience, which is worth the effort!

• The filtering problems seems likely to me to be a major reason why many amateur satellites have uplinks and downlinks on different bands. No reason you can't do that terrestrially. Or, of course, you could easily achieve the appearance of same-band input/output if you put the two radios some distance apart and link between them on a different band (like you could have site A receive on 144 MHz, link to site B on 1296 MHz, where you retransmit on 144 MHz). – a CVn Dec 14 '15 at 15:22

Rotorua branch 33 of NZART 2m linear repeater is still operational and has been for more than 20 years. It is a solar powered repeater and situated on a high point overlooking Rotorua Town. Output is 144.350MHz with a plus 600 shift.

More details of it can be found here

Look at:

http://www.amalgamate2000.com/radio-hobbies/radio/dunedin%20linear.htm and http://www.amalgamate2000.com/radio-hobbies/radio/Rotorualinear.htm These are the only 2 working examples of terrestrial a linear translator I know of. And I believe they are now both off the air...

Have been thinking of upgrading my 6 meter repeater to a linear translator, using a low band TV transmitter amplifier for the output stage. Believe going from FM to SSB will yield a 10 dB improvement, and perhaps a 3 times improvement in range...

And should be even better, using Digital Voice (DV) over SSB !!!

But just may not be worth the effort, based upon lack of 6 meter activity..

• Hello, thanks for the answer, and welcome to ham.stackexchange.com! Please consider taking the tour to get the most from the site. 73! – rclocher3 Mar 12 '17 at 17:06

Yes. If you truly want to allow doubling then you want SSB. You want SSB anyway for the significant SNR advantage over AM.

You need to consider voice-activated squelch so that VoIP knows when to stop/start sending audio to the internet (unless continuous streaming audio is OK). This can be implemented in software, particularly for software defined radio (SDR).

Until a few years ago, one might have done this in hardware. However, it is now feasible to do this with a full-duplex SDR (or separate SDR receiver and transmitter). In fact, I believe it might be possible for about $100 using an RTL2832 USB stick receiver and Raspberry Pi transmitter PLUS a duplexer appropriate for your transmit/receive separation assuming you wish to use a single repeater antenna. Since the duplexer would cost a couple hundred dollars used, you might just start by using two antennas on a "garage repeater" setup or 2m/70cm with an inexpensive diplexer and single antenna. You might not find off-the-shelf code for this, but the appeal is you can do it for$100 of hardware and a weekend of learning instead of a bench full of hardware.

I would start by understanding the GNU Radio examples for ham radio, in particular this example set or the simpler GNU Radio ham radio examples.

• What would the software be doing except moving unmodified bits from one device to another? A wire would accomplish the same objective much more cheaply. In fact if you set the LO to mix to the transmit frequency rather than baseband, you can do it with the parts in just one SDR, except without any of the digital parts. – Phil Frost - W8II Mar 12 '17 at 23:00
• Hi Phil, three key factors are provided by SDR here. 1) The same transmit amplifier would be used for an SDR or analog system. For a garage (short range outdoor) repeater demo, no transmit amp is needed. 2) Flexible frequency translation--crossband or inband repeating as budget and antennas allow. 3) Voice operated squelch in software--while maintaining full AGC for SSB. Many hams already have a Raspberry Pi and RTL2832, so they can make a low power repeater for near zero cost assuming they have an antenna already. – Michael Mar 13 '17 at 4:37
• Factors 1 and 2 are provided entirely by the analog components of the SDR, or any other receiver. Granted, squelch would be implemented in software, though this is not complicated or expensive to realize without software. Point being, when you said cheap I think you meant easy (for someone familiar with a Raspberry Pi and RTL dongles). The analog solution is demonstrably cheaper since you could make it by stripping all the digital components from an RTL dongle and adding a comparator and simple filter. There are people for which this is easier, too. – Phil Frost - W8II Mar 13 '17 at 14:58
• I agree Phil, especially since tiny programmable oscillators are so inexpensive and Minicircuits, Analog Devices, etc. make little inexpensive modules or just use the individual semiconductors. So there are at least two pathways: one for the electrical/RF engineering oriented and another for the DSP/software engineering oriented. Great! – Michael Mar 13 '17 at 15:13

Tone decoding is a bit more problematic on SSB. So PL tones may not be detected if the SSB receiver's BFO is off. AM would be better at tone detection. FM is superior due to the FM capture effect for multiple keying transmitters. AM tends to allow noisy hetrodyning with multiple transmitters at the same time (i.e. double-keying). I personally suggest FreeDV over SSB with tone decoding. http://www.freedv.org/

• FreeDV would seem to defeat the objective of allowing doubling. – Phil Frost - W8II Sep 18 '17 at 14:07
• When you say "doubling", are you referring to the way SSB "triples" the ERP via the SSB method (i.e. PEP)? That would be a good reason to use SSB over AM, however, the tone decoding is still impacted by accidental BFO detuning or drift. Using FreeDV is encouraged by the developers to use SSB mode. The audio quality is enhanced with FreeDV, as well as readability over noise. It is a DIGITAL signal too not a analog one. However, if you were referring to "double-keying", I am not familiar with how FreeDV handles that. I'd think it also has a capture effect like FM. I am not sure... – SpookySr Sep 19 '17 at 14:26
• I mean two people can talk at once while remaining somewhat intelligible to the receiver. And indeed, FreeDV has something of a "capture" effect, though it's more like a mutual-obliterate effect. Unless one station is much stronger than the other, neither will decode. – Phil Frost - W8II Sep 19 '17 at 16:07
• Also, "FreeDV over SSB" isn't really a thing. It's just FreeDV. A USB transceiver doesn't add any modulation: it just shifts up/down in frequency. – Phil Frost - W8II Sep 19 '17 at 16:08
• Correct. You can use FreeDV over (via) any modulation scheme. It is just 16 QPSK streams over audio. However, SSB has better range due to its inherent PEP scheme than lets say AM at same input power. So I think that's why the inventors chose SSB over AM or FM methods. Not that SSB added anything to the FreeDV modulation scheme. I do have one question though, does FreeDV get impacted by SSB receiver BFO drift or can it compensate somehow? – SpookySr Sep 20 '17 at 16:34