Let me start with a Happy Easter to everyone! Having a few days off for a long weekend really lets me get back to what I like doing – which includes writing blogs. This post continues in the series of satellite hunting posts which I started with the hunt of Optus 10 at 164 degrees East earlier in the week. It’s probably a post that very few would appreciate, but as I use this blog as a bit of a reference and an “engineer’s logbook”, I decided I should write it up anyway.
For this upcoming series of posts, I will be focusing on capturing the spectrum and beacons of less-popular or even unused satellites which are in orbit and beaming to Sydney, Australia in the Ku band. Catching them will give me a chance to find out what is on the satellites through blind-scanning and spectrum analysis, and verify their operational status.
The well-known satellites might not get covered in their own post, or will get covered last. The well-known Ku satellites in Sydney include:
- Optus D2 at 152 degrees East
- Optus C1 and D3 at 156 degrees East
- Optus D1 at 160 degrees East
- Intelsat 19 at 166 degrees East
Chances are, if you have a dish, or you see a dish in the air, they will be pointed at one of these, as they carry the vast majority of direct to home broadcasting for Australia. But what’s on the other satellites that are in the sky that very few people are pointed at?
Intelsat 18 at 180 degrees East
We start this posting with hunting for Intelsat 18 at 180 degrees East. This particular satellite is the easternmost Ku band satellite targeting Sydney, and is a relatively new satellite, launched 5th October 2011 to replace Intelsat-701 which was previously operating in this orbital slot.
The Lyngsat transponder listing shows that there is only a small group of services in Ku which can be received in Australia, and these are the CanalSat Caledonie multiplexes. However, there is a complication – these transponders are operating in the “extended” Ku-Low portion of the band below 11700Mhz, and cannot be received with the regular Wideband 10700Mhz “Foxtel standard” or regular 11300Mhz LNBs! Additionally, the multiplexes are in DVB-S2 mode, meaning your “cheap” set-top-box-in-a-portable-unit satellite finder (which I use quite regularly) won’t be able to lock or blind-scan these.
As a result, I had to change the LNB on the dish over to a Universal “dual-LO” type (9750 and 10600Mhz) to be able to receive it. Further to this, I would have to go “blind” with a needle-based satellite finder which responds to total power. It’s critically important to make sure the LNB is being powered for Horizontal low-band (i.e. no 22khz tone) otherwise the finder will see only noise.
The response was weak on the needle based finder but was sufficient to align the dish. A scan with Crazyscan was enough to confirm the pointing was correct.
As expected, the New Caledonia transponders came up, and each of them had about 22Mbit/s of null packets, so they could do away with a whole multiplex altogether. Aside from that. some narrower data only services were found as well – all on Horizontal polarity.
I had to confirm whether it was expected to see signals in these locations and find the beacon frequencies, so frequencyplansatellites.altervista.org came to the rescue with a downlink plan, excerpted below:
The scanned transponders seem to align with the layout, thus validating the findings. The data services seemed to span several generations – for example, the “regular” DVB-S style constant coding and modulation (CCM), transport-stream based type:
There is a newer continuous-multi-stream type utilizing DVB-S2 adaptive coding and modulation with short frame-length:
There is also one utilizing a packetized-single input stream option, which I haven’t seen on the air before:
One of the signals was quite strong, and 32APSK constellation, intermixed with QPSK pilots, can be seen, which is also not very common on the air.
An image of the spectrum, including 50Mhz either side of the band, was taken with the RSA306. As will be a common theme, you will notice there are several spikes towards the top end of the Ku High band scans which are a result of the analyzer itself.
Ku-Low, Horizontal Polarity
Ku-Low, Vertical Polarity
Ku-High, Horizontal Polarity
Ku-High, Vertical Polarity
A large array of beacons were listed, with four 6 telemetry beacon listings in the C-band, and one regular beacon in the C-band. There were three listings for beacons in the Ku band, at 11198R, 11452R and 12502R. As a result, I expect there to be carriers but no modulation on both linear polarities at one or more of these frequencies. In fact, some are already clearly showing in the scans above.
It seems clear that the lower beacon has been found, unmodulated, on both polarities as expected. It does not appear to be a result of internally generated spurs in the RSA306.
The middle beacon was also found, with no modulation and expected characteristics of being on both linear polarities.
Here, we can see that the spike on the right is internally generated, but the spike near the middle is definitely the beacon. All three beacons in the Ku were received with expected results, which gives us additional reassurance that we are indeed listening to Intelsat 18.
There seemed to be quite a few signals visible by energy but not lockable by DVB-S2 blind-scan. They did seem to have a decent QPSK constellation at times, which was puzzling. A closer look revealed a TDMA-style transmission mode, carriers of varying width (uplink vs downlink) and discontinuous transmission. These images were taken on the Ku-Low H-polarity setting.
Eutelsat 172A at 172 degrees East
The next satellite across on the Clarke belt that beams into Sydney would be Eutelsat 172A. This satellite started life as AMC 23, launched in December 2005, which was then transferred to GE, and then bought from GE by Eutelsat. This satellite has never really been known to carry anything consistently into Sydney, with the Lyngsat transponder listings being mostly empty. This is an interesting position, as the satellite is due to be replaced in the coming years with Eutelsat 172B in 2017.
Unfortunately, the lack of consistent signals means a challenge for satellite finding. While I could slew the dish slightly towards the north and increase its elevation, do a scan, and then refine, I decided to try a better method.
In order to explain this, I will need to explain the system configuration.
The system at the moment consists of a 75cm Ku band offset dish on a tripod which I use to aim satellites. Two LNBs, one universal (9750/10600Mhz) and one wideband (10700Mhz) were used. The LNB is cabled up to an 8×1 DiSEqC switch which allows for a long feed to be shared amongst other dishes pointed at “regular use” satellites. The cable goes through a window using a flat window cable (lossy), and is powered and selected by the TBS 6925 satellite card using Crazyscan. The loop-out port is connected up to the Tektronix RSA306 Real-Time Spectrum Analyzer which is running on a second laptop.
This combination actually makes for powerful satellite hunting, as I realized that I can select the DiSEqC port, horizontal/vertical, high/low band using Crazyscan on the main PC, and then see the LNB IF signal on the laptop. Utilizing VNC remote networking and my smartphone, it was possible to be outside near the dish and be able to make adjustments based on seeing the screen of the spectrum analyzer software. This way, I could optimize for a beacon power, and or see the whole band change as I moved the dish. Because I know the “shapes” of the spectrum corresponding to our “regular” satellites, I could use them as reference points in the sky for quick and easy alignment.
A phone is much lighter than any cabled analyzer, and I could slip the cover into the dish or hold it onto the dish with a hard drive magnet. This arrangement made my life a lot easier when it came to hunting “quiet” satellites.
Aiming at Eutelsat 172A’s Beacons
Based on the information provided at frequencyplansatellites.altervista.org‘s AMC-23 entry (excerpted below), it was expected to find beacons at 11199R and 12749L in the Ku band. It was also known that this satellite requires the use of Universal LNB to ensure reception of all transponders, just like Intelsat 18.
Swinging the dish around based on what I know of its “neighbours”, I was fairly confident I had found a signal consistent with the expected position. The beacons were a little different to expectations though.
11199R (?) Beacon
A signal was found in the vicinity of the expected frequency in the horizontal polarity, but not the vertical polarity, which suggests the beacon listed as R is actually sending as H.
12749L (?) Beacon
Vertical Polarity, Zoomed In
Also, like the previous beacon, it was not received on both polarities and was received on vertical polarity only. The signal is confirmed as likely from the satellite when zoomed in – the frequency drift is due to the LNB’s local oscillator drifting, whereas spurs from the spectrum analyzer generally remain “still”.
Given the R was received as H, and L was received as V, I suspect the beacon information is just incorrect, and had been noted as circular polarity when it isn’t. We can further verify this when we look at the active multiplexes.
From Sydney, it seems a regular 11300Mhz LNB would have been enough (missing out on the lower beacon) as there wasn’t anything particular showing up in the Ku-Low band. The only services were in the upper end of the Ku-High band, horizontal polarity, consistent with the downlink map.
The services are mostly data services, it seems, but not all of them are DVB-S/DVB-S2 type.
Ku-Low Horizontal Polarity
Ku-Low Vertical Polarity
Ku-High Horizontal Polarity
Ku-High Vertical Polarity
The spectrum analysis results seem to mirror that of Crazyscan closely, with the beacons resolvable in the scan. A closer-up look can be had of the Ku-high-horizontal area.
From 12200 to 12800Mhz
Of course, we can get even closer.
The wider carriers at the left aren’t that interesting, so lets take a look at the data carriers on the right.
Again, it seems they belong to specialized satellite data modems of some sort, with more narrow-carriers, and TDMA/DTX operation.
In this satellite hunting post, we’ve managed to confirm reception of Intelsat 18 at 180 degrees East with all beacons, and Eutelsat 172A with what appears to be incorrect beacon information. On Intelsat 18, the Canalsat Caledonie transponders were well visible, and are the only television services on the satellite. The rest of the active transponders carried data, just like those on Eutelsat 172A.
In general, satellite bandwidth is expensive, so you wouldn’t expect an active station-kept satellite to be “quiet”. However, due to the beam and transponder configuration, it doesn’t seem that these satellites offer that much bandwidth to Australia, and may seem quiet for that reason. They are probably quite active on their other transponders, but not receivable from Sydney due to the shaped antennas on the satellites.
Now that I’ve seen the spectra of the satellite, revisiting it would be much quicker.