Experiments with Tropo-Scatter on 24 GHz By Rex Moncur VK7MO and David Smith VK3HZ While it is possible to readily work up to around 200 km on 24 GHz with line of sight propagation between mountains, those who seek to maximize their grid square score will often find it necessary to use other modes of propagation where mountains are not available. The authors have undertaken tests on 24 GHz tropo-scatter over non-line of sight paths to gauge the value of tropo-scatter where line of sight propagation is not possible. It is found that by using the digital mode JT65c at times of low absorption it is possible to work up to 260 km on tropo-scatter with modest portable stations (1.5 watts and 40 cm dishes). Tropo-Scatter vs Line of Sight Tropo-scatter involves an increase in loss over line of sight at 24 GHz of about 72 db at 200 km and 78 db at 300 km, Mike Willis G0MJW (1). While atmospheric absorption losses are significant at 24 GHz, these are to some extent reduced by using tropo-scatter as some of the path is at altitude where these losses are somewhat lower. Even so tropo-scatter on a 250 km path is likely to be about 60 db worse than line of sight. Equipment All equipment is GPS locked. VK3HZ: Thales module and 38 cm dish, 1.5 watts TX, FT817 IF. VK7MO: DB6NT transverter, pre-amp and 3 watt PA to 47 cm offset dish, IC-910-H IF. Locations and Test Results Day 1 Tests: VK3HZ was located at VK3ES s QTH on Mt Macedon at an elevation of 700 metres with a clear take-off towards all locations. QF13: 136 km. VK3CYs location on a rise of around 100 metres. SSB signals with reports of 5-4 and 5-3 with heavy QSB QF14: 177 km: On flat farmland 16 km south of Quambatook. SSB signals with reports exchanges of 5-2 and 4-1 with heavy QSB. QF15: 268 km: On flat farmland 12 km north of Moulamein in VK2. JT65c reports of -28 and -30 with both short and long term QSB. QSO took around 30 minutes to complete on QSB peaks and required the use of single tones for RRR and 73. Day 2 Tests: VK3HZ was located at Johns Hill lookout at an elevation of 408 metres with a clear take-off towards all locations. QF21: 106 km: On flat farm land 12 km North West of Geelong. SSB reports 5-5 and 5-5 with heavy QSO.
QF11: 134 km: On a rise of about 50 metres above flat farm land. SSB reports 5-2 and 5-1 with heavy QSB making it unreadable at times. Predicted Propagation Losses The traditional formula found in most amateur texts gives the propagation loss increasing at the cube of frequency or 30*LOG10(f). We were surprised that applying this formula showed our results exceeded the predicted values to a significant extent. After looking for an explanation we noted that in his presentation (1) G0MJF uses a modified relationship 25 Log (f) 5 [log(f/2)]^2 which was more in line with our results. Investigation of ITU Recommendation on tropo-scatter indicates that the 30*LOG10(f) (2) relationship is limited to 4 GHz and a separate ITU paper that discusses propagation up to 50 GHz (3) applies the formula that G0MJF uses in his notes. After applying this formula our results are now consistent with low absorption as applied during our tests (Fig 1). Fig 1: Comparison of Test results with those predicted by the method in ITU-R P.2001 Based on the equipment used for these tests Fig 1 gives an estimate of the distance that might be worked on tropo-scatter for SSB, CW and JT65c. These estimates are based on locations with good take-offs close to zero degrees as applied in the case of these tests. Graphs are included for the basic propagation loss and also after allowing for low and high absorption losses based respectively on 30 db and 100 db at 250 km. These values are typical of the range at temperate latitudes such as Southern Australia as apply at the surface.
Tropo-Scatter Spreading Fig 2 shows a Spectrum Lab waterfall (frequency spreading) and time (signal to noise ratio) display when receiving a constant tone of 1270 Hz. The waterfall display shows the signal is spread by up to +/- 10 Hz and the time display shows QSB of up to 20 db with variations within a second or so. The spreading is presumably due to the fact that tropo-scatter can take multiple paths due to the atmospheric discontinuities in the common volume used for scattering. The spreading is somewhat wider than the 10.8 Hz that is used by JT65c but as most of the energy would be within the central part the loss of sensitivity is probably marginal. This compares to 144 MHz where the spreading is typically around +/- 40 mhz on a (600 km) tropo-scatter path. If one proportions the 144 MHz spreading for the increase in frequency it is seen they are of the same order. Fig 2: 136 km path. Top shows tropo-scatter spreading in Hz and bottom shows troposcatter QSB in db when receiving a 1270 Hz tone
Fig 3: Data for 176 km path with a 1270 Hz tone Fig 3 shows somewhat lower spreading but from the signal to noise ratio data the average is around 25 db compared to about 28 db for the 136 km path and this reduction is signal level might explain the apparent reduction in spreading.
Fig 4: 268 Km path. Spreading and Signal to Noise ratio with a 1270 Hz tone Fig 4: At 268 km the 1270 Hz tone is only evident on peaks that occasionally reach a peak signal to average noise ratio of 15 db. While spreading appears to be less this is probably only because the spreading is in the nature of a Bell curve and thus only the central part is above the noise. Conclusions With JT65c and small portable stations (1.5 watts and 40 cm dishes) it is possible to work 24 GHz tropo-scatter out to around 260 km if absorption losses are low essentially this means winter conditions with temperatures close to zero. SSB is useful out to around 170 km if absorption is low. While spreading of the signal due to multi-pathing is up to +/-10 Hz most energy is still retained within the JT65c bandwidth of 10.8 Hz and thus JT65c performance is not significantly reduced by tropo-scatter spreading. Tropo-scatter on 24 GHz exhibits very rapid QSB of up to 20 db.
References (1) Microwave Propagation, propagation you can t rely on, Mike Willis G0MJW www.mike-willis.com/tutorial/rt%20propagation%20lecture.pdf (2) International Telecommunications Union, Recommendation ITU-R P.617-2 (02/2012) Propagation prediction techniques and data required for the design of trans-horizon radio-relay systems. (3) International Telecommunications Union, Recommendation ITU-R P.2001. A general purpose wide-range terrestrial propagation model in the frequency range 30 MHz to 50 GHz