Contestia on weak signal VHF/UHF SSB
KH6TY, Howard S Teller Jr
Why Contestia on VHF/UHF SSB?
VHF and UHF, and particularly UHF, is very subject to distortion by Doppler spreading and, to some extent, Doppler frequency shift of the SSB phone signal. On UHF circuits, this occurs almost daily, and signals, normally Q5, might be S3 and it is not possible to understand the voice.
On UHF, many transceivers used for SSB phone do not have temperature-compensated crystal oscillators, and can drift off frequency over the course of a QSO - sometimes as much as 200 Hz. Narrow modes, like PSK31, or PSK63, and even MFSK16 (which requires tuning within about +/- 4 Hz to maintain decoding, cannot handle this degree of drift, even with fairly strong AFC.
Contestia 64/1000, during daily schedules on weak-signal UHF paths, with QSB fading, multipath reflections, and Doppler distortions, often will produce 95% to 100% print when SSB phone is unreadable.
Bandwidth makes a significant difference. The wider Contestia modes perform much better than the more narrow ones in the presence of Doppler shift and Doppler spreading, such that a minium emitted Contestia bandwidth of at least 1000 Hz had proven to be necessary. There is plenty of room on UHF and VHF, so the emitted bandwidth as great as a SSB phone signal is not problem, but only if the drift is kept to 100 Hz or less over the QSO transmission. Tuning is a problem when signals are very near or under the noise threshold, because it is not possible to tell which is the signal and which is the noise for accurate tuning. Therefore, coupling RSID with Contestia, and searching the entire IF bandpass will automatically tune accurately for the signal, since the RSID signal has a very limited character set, with greater sensitivity than Contestia.
If each station's transmission is kept down to a length in which the frequency drift (usually from receive to transmit) is less than 100 Hz, RSID will correct the tuning, with the effect of the two stations "walking together" down the band a little bit. After the QSO has gone on for a while, temperature generally stablilize, and the total drift from the initial frequency is generally less than 200 Hz. The Contestia 128/2000 mode is 2000 Hz wide, so if the lower tone frequency starts out at 500 Hz, the accumulated drift is usually no more than 200 Hz, resulting in a low tone frequency of 300 Hz, which is still within the IF bandpass, and there is no problem with decoding.
So, while Contestia 8/250 is a good mode for HF (where spectrum is limited compared to VHF and UHF), the best Contestia modes for VHF/UHF turn out to be Contestia 64/1000 and 128/2000 because of the need to handle transceiver drift and Doppler distortions. For transceivers that drift excessively, Contestia 64/1000 is preferable to Contestia 128/2000, even though the minimum S/N is not as good, because there is more room to fit in the IF bandpass during successive periods of drift correction.
Contestia 64/1000 has often produced better copy than CW (the mode of last resort in weak-signal communications), because the CW note is sometimes distorted by the Doppler effects so much that it is difficult to understand the CW note. The advantage of Contestia 64/1000 over CW is not extremely great, but the 30 wpm typing speed is a comfortable speed for a QSO and under very weak signal, noisy, conditions, maintaining copy at 30 wpm with CW is often difficult, as too many characters may not be
Although slower versions of Contestia have slightly better minimum S/N requirements, 30 wpm seems to be the slowest comfortable typing speed for a "ragchew" QSO, and the faster speed seems to get more characters through in conditions of fast QSB fades than the slower versions.