Satellite Hunting: AsiaSat 4 (122 E), Thaicom 4 (119 E) & AsiaSat 3S (118 E)

The hunt continues with this fourth installment, where we take a look at a few satellites towards the west.

AsiaSat 4 at 122 degrees East

AsiaSat 4 was launched April 2003, on a high power variant of the Boeing BSS-601 space bus platform with XIPS propulsion. It has a nominal lifetime of 15 years, due to expire in 2018, but is going to be replaced in 2016 by AsiaSat 9.

This particular satellite is an easier satellite to aim for, as AsiaSat 4 is carrying one TV service from the Taiwanese TAS provider which is targeted at Australia. The signal is, however, a DVB-S2 carrier, so users of “set top box in a portable unit” satfinders are out of luck and will have to do it the “old fashioned” way with a needle-based IF-power finder.

The downlink plan for the satellite (excerpted from shows that AsiaSat 4 services the Ku High band, with only Vertical polarity for Australia. This means I can revert to my preferred single-LO 10700Mhz “wideband” LNB.


There are two C-band beacons, and three Ku band beacons, with only the 12252V beacon coming down the Australian beam (and 12253V/12254H going down the East Asia beam).

Crazyscan Results

Alignment of the dish was pretty straightforward, using only the needle based finder. The Crazyscan spectrum and the locked carriers are shown in the below plots.



Some areas showed signal power unit no locking, indicating non DVB-S/S2 systems. The wide carrier is the DVB-S2 television service carrier by TAS in Taiwan. This carrier features a lot of null packets – room for more channels.

Checking for the Beacon

Even though we are confident based on the services that we have hit the target satellite, I would still like to record reception of the beacon as proof. Luckily for me, AsiaSat 4’s beacon is strong, and received at a frequency which would imply ~12252V as expected. No signal was received in the Horizontal polarity.


Spectrum Analysis


Horizontal Polarity


Vertical Polarity

The spikes in the upper part of the horizontal polarity scan is due to spurs in the spectrum analyzer. The beacon line is clearly visible in the vertical trace, as is the pattern of services.

Some closer analysis of the signals on the IF shows some strange behaviour – near the beacon, if you listen long enough, there seems to be occasional “blips” of carrier, which seem regularly spaced. This implies there is something occupying what appears to be silent space, or using it in an unusual way I haven’t seen before. The carrier to carrier spacing is about 3.375Mhz and continues on for a bit.


This thing is also evident when looking near some TDMA-based data carriers – the left two blips seem to have semi-periodic pulses in pairs – and this extends to inside the data carriers as well (click for full size). This doesn’t seem to be an artifact of my measurement equipment!


Some of the wider data carriers, which you might expect to be DVB-based are in fact not, despite showing up semi-decent constellations in Crazyscan. As it turns out, it seems to be some sort of TDMA system, with discontinuous transmissions, noted as hairline breaks in the “bands”.


Some sections of the band feature narrower carriers, as if to imply the use of an FDM-style uplink return path for two-way satellite.


By looking at the modulation pattern, we can speculate about the purpose of some of the carriers – for example, this set of data carriers sees a very strong CW-tone on one side – why might that be?


My best guess is that the strong carrier is used for frequency reference purposes. Remember, in two-way satellite data systems, the remote terminal has to do uplinking to the satellite – with narrow carriers so closely spaced, the oscillator used must be extremely accurate. It might not be practical to build modems with oscillators accurate enough for this purpose, and the modems themselves might not have the ability to use other sources for oscillator discipline (e.g. GPS), so instead, the head-end may be equipped with a very accurate “clock” which is sent out over the satellite so that all the terminals can use this as a reference.

Some other systems don’t seem to use a continuous tone, but instead, this one seems to have a “on-off” patterning to it. Maybe it’s for slot-alignment determination, although it seems unlikely due to the satellite delay causing such a system to be inaccurate unless the signal time-of-flight is taken into account.


Thaicom 4 at 119 degrees East

Thaicom 4, also known as IPSTAR 1, was launched in August 2005 and was a highly sophisticated satellite featuring 87 individual Ku spot and regional beams intended for use in satellite internet services and broadcasting, along with Ka band transponders for high speed satellite internet. The satellite has a service lifetime of 12 years, expiring 2017.

The satellite seems to be built with individual spot beams as “bent pipe” repeaters, involving no processing logic on the satellite’s behalf. This allows for redeployment of virtually any satellite internet modems, taking advantage of the smaller “cellular” coverage patterns which allow for frequency re-use, alleviating congestion and providing improvements in security (since the data is not sent across a wide area).

This particular satellite turned out to be a chance discovery, as I was aligning up for AsiaSat 4 above, I noticed a strange “double blip” of the meter while lining up. It had to be another satellite, but which one? I scratched my head as Lyngsat only mentions Indonesian and Cambodian service for Thaicom 4.

As it turns out, it was IPSTAR 1. A check of the map (excerpted from Thaicom’s website) shows it beams to Australia with an array of spot beams – and I know for a fact that farmers in country NSW and elsewhere have been using IPSTAR for their satellite broadband internet services.


But how could I be sure, with no beacon information or transponder map available? Well, I made a deduction based on what I found on the satellite and the position (i.e. a slight twitch to the west from AsiaSat 4).

Crazyscan Results

Not knowing the transponder layout, I decided to play it safe and go with an Universal Ku LNB. I would suspect that such a capacious satellite would seek to use as much spectrum as possible, despite the frequency re-use due to spot beams. As it turns out, that guess proved to be a good one, as we had results on both Ku-Low and Ku-High bands.

119.0E-RFScan 119.0E-RFScan-Carriered

As it turns out, every service that was locked was an ACM/VCM style DVB-S2 service, many of them using Long Frames, for example, the one below. This implies all of the services were data services. Several of the carriers were too weak to lock, and their strengths varied, which seems consistent with “spillover” from the edges of adjacent spot beams. All downlink services found were on the Vertical polarization.


Spectrum Analysis


Ku-Low Horizontal Polarity


Ku-Low Vertical Polarity

At the time I took the spectra, the weather in Sydney wasn’t particularly good, so we had a bit of a rain-fade situation by the looks of it. A similar trend is seen with the Crazyscan results above, with a narrow strong DVB service “tucked in” with a few wider services, clearly poking out.


Ku-High Horizontal Polarity


Ku-High Vertical Polarity

This particular scan seems to come up with an interesting result – the Ku-High Horizontal isn’t quiet like Crazyscan would have you believe. It’s actually got a bunch of narrow carriers – maybe these are the beacons?


I am not sure, but there seems to be a lot of “narrow” tones, spaced about 3.33Mhz. They aren’t constant, and blip from time to time, like those seen on AsiaSat 4 above. So, they are likely not beacons, and are instead a part of a data system of some sort.

I didn’t spot any obvious beacons, but that may be because the beacon is on the Ka band, of which I have no equipment for. Now that I think of it, if I had a proper satellite-internet grade LNB/BUC unit, some of them accept a precise frequency reference up the cable, which would make absolute frequency measurements much more possible.

AsiaSat 3S at 118 degrees East

We come to the last satellite for this part of the hunt, and it’s another one dear to my heart. This particular one is the oldest serving member of the AsiaSat fleet, launched March 1999. This was a replacement for AsiaSat 3, which failed to successfully launch into orbit and was declared a total loss. The satellite has a nominal lifetime of 15 years, which would have expired in 2014, and formerly served in the 105.5 degrees East slot where AsiaSat 7 (launched November 2011, and moved to 105.5 degrees East late 2014) now operates. AsiaSat 3S formerly carried the Telstra Bigpond Satellite broadband service. According to Lyngsat, it currently carries no services.

It is currently parked right next to IPSTAR above, and is currently in a semi-inclined state, with an inclination of 0.86 degrees at this time. A check of the transponder plan confirms that the satellite only uses the Ku-High frequencies, and only the vertical polarization transponders will serve Australia. As a result, I can once again revert to a single-LO “wideband” 10700Mhz LNB.


The beacons were listed as 12250V and 12749H. Having the dish aligned to IPSTAR should allow me to receive it, as it is within the beamwidth, but for better luck, I gave it a slight nudge until the IPSTAR signal fell about 25%. Unfortunately, I could hear no sign of the beacon. This may be because the Ku beacons are not operating or switched off, as AsiaSat 3S also carries two C-band beacons.

Unfortunately, it’s a fail on this one, but there may be a second reason. My view to the west is limited by the roofline of the neighbouring townhouse, which limits my view to elevations above about 36-38 degrees. Unfortunately, without a precise inclinometer, it’s hard to judge, but it could be just behind the roof-line.


In this part, I’ve been able to visit Asiasat 4, and unexpectedly, also Thaicom 4/IPStar 1. Only one TV service was seen, on Asiasat 4, a multiplex from the Taiwanese TAS operator carrying Da Ai 2, Hwazan Satellite TV and Taiwan Macroview. The rest of the services were data related, as usual. Unfortunately, I was not able to positively identify Asiasat 3S’s beacons, possibly due to visibility issues.

There are still further satellites further towards the west which do potentially beam signals to Sydney in the Ku band. These include Asiasat 7 at 105 degrees East, Chinasat 11’s steerable beam at 98 degrees East, NSS6 at 95 degrees East, Measat 3B at 91 degrees East and Yamal 300K’s steerable beam at 90 degrees East. Unfortunately, as I only have sky visibility down to about 36-38 degrees elevation towards the West, I’m not able to see these satellites, and thus the hunt for them is impossible at this time.

As a result, this just about covers all of the Ku band, less popular, satellites beaming to Sydney. I might come back and write something about the “regular” popular Optus series and Intelsat 19 in the future, but for now, that’s the hunt complete!

About lui_gough

I'm a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!
This entry was posted in Radio, Satellite and tagged , , , , . Bookmark the permalink.

2 Responses to Satellite Hunting: AsiaSat 4 (122 E), Thaicom 4 (119 E) & AsiaSat 3S (118 E)

  1. jarofrant says:

    Came across your site looking at faxing stuff.

    There is some interesting History regarding IPSTAR-1, I used to work for an ISP that sold services though the platform and there is some very cool tricks behind the scenes to maximize the available bandwidth. Each of the “Spot” beams can use multiple channels at once if a particular area is over subscribed. Additionally The satellite itself monitors the ES/no of each of the VSAT’s attached, as the TOLL (downlink) signal fades due to rain/moisture on each beam power is used from reserve to ensure good signal for the subscribers. It was the largest commercial satellite up there at the time when I was working there (about the size of a bus) and needed its own airlane rocket to get it into orbit. (normally they take 4)

    Testing of new modulation and modems was starting for the NBN as I left using some of the spare spectrum they have assigned.

    I have plenty of stories to tell, but nothing i’m confident to put online 🙂
    Well, There is one I can put online, During the first few years of its rollout, the ODU assemblies from Parrot had some issues in the joins between the orthogonal splitters and the BUC/LNB’s causing them to fill with water. Turns out the issue was that some of the screws were just a smidge too long, so as the units were assembled they didnt compress the O-rings correctly causing the leaks.

Error: Comment is Missing!