The original English text can be read below (not all pictures appeared in the article in OSCAR NEWS). Please be informed that the status of the satellites mentioned in the article is not updated since the article was written! If you are interested in working satellites as a portable station, Google can give you more webpages about this subject, for example a similar setup of KB5WIA.
The current fleet of operational AMSAT satellites consists solely of low-earth orbit (LEO) satellites. Although high-earth orbit (HEO) satellites like AO-40 (and the upcoming P3E) delivers worldwide communication capabilities and interesting technical challenges, using LEO satellites is also interesting and fun, and has its own challenges to deal with.
This article describes a portable satellite setup for working LEO voice satellites. The setup was developed and optimized over several years, and the article mentions some rationale behind design decisions and practical experiences with the setup. Excellent results were achieved in terms of making QSOs via LEO satellites as a portable station.
The setup described here has the following main features:
To be able to work both FM and SSB satellites with a truly portable setup, transceivers like the IC-910 are too heavy. For the setup described here, two Yaesu FT-817ND multi-mode portable transceivers are used. This small all-round transceiver has sufficient functionality and is small enough to form the base of the portable setup. The FT-817ND (further mentioned as '817') covers the amateur radio bands between 1.8 and 430 MHz; for this setup only the 2m and 70cm bands are used. A single 817 does not have full duplex capabilities, but two of them make an excellent full duplex combination. A setup with similar second-hand equipment, for example an FT-290R2 and an FT-790R2, is also possible, which will reduce the overall price of the setup.
Both 817 transceivers are externally powered from a single 2.2 Ah rechargeable sealed lead-acid battery , which is carried in a hip bag. One fully charged battery could be used for a couple of passes. On the 'KA7OEI FT-817 pages'  several useful tips about saving energy with the 817 can be found. The two most important tips are the use of the front antenna connectors and, when possible, turning the display backlight off. Another big energy saver is to leave the transceiver for TX switched off while you are waiting for the satellite to become audible at the start of the pass.
A custom-made box to hold both transceivers was designed to make the setup as rugged and portable as possible and fits inside the carrying bag (designed for photo equipment). Figure 3 shows four pictures of the equipment mounted into the box. The bare box consists of four main parts, made of 2 mm thick aluminium sheet:
Both 817s are powered by one power cable from the battery (figure 3(c)). Fuses are present in the + and – lines, and each transceiver has an anti-parallel diode mounted to prevent damage from reverse polarity. The factory supplied ferrite clamps are mounted just before the power cables enter the 817s. Details of these components mounted in the box can be seen in figure 3(b). A final cardboard protection is installed at the back of the box to protect the cabling when the box is inserted into the carrying bag.
At the back of the 817s there is space left to access the ACC connector, which can be used for Doppler correction in possible future projects. Also the data connector of the upper 817 is still accessible, so one 817 can be used with BPSK-31 on HF during holidays. At the front of the 817s angled BNC connectors are used to feed the antenna cables to the side so they are not in the way while operating the 817s.
As a backup antenna, a portable dual-band HB9CV antenna  for 2m/70cm can also be used (figure 4). It is heavier than the ultra-light Arrow, and it needs an extra handle to point the antenna towards the satellite. The extra handle can easily be attached with the available screw thread in the boom of the HB9CV.
Another commercially available antenna specially designed for portable satellite operations is the dual-band Elk antenna , but the author does not own this antenna. The Elk is a log-periodic antenna with only one connector for both 2m and 70cm. For the setup described here, this would require an extra diplexer to be inserted (which is fortunately is not necessary for the Arrow and the HB9CV).
To compare the performance of these three antennas, table 1 shows the approximate values for forward gain for both bands for all three antennas.
|Forward gain||Arrow ||HB9CV ||Elk |
|2m||5.9 dBd||5.5 dBd||6.6 dBd|
|70cm||8.2 dBd||5.5 dBd||7.0 dBd|
For 2m the gains are almost equal. In practice, there is no noticeable difference between the Arrow and HB9CV. The almost 3 dB difference in gain between these two antennas for 70cm is clearly noticeable as the downlink signals become undetectable at low elevations of the satellite. The best antenna to choose will actually depend on gain, price, weight, size, purpose, availability and, personal taste. Some people also build similar antennas themselves.
In some portable satellite setups, the Arrow antenna is mounted on a tripod, which will work fine. To have optimal and fast control of the antenna polarisation and pointing direction, I prefer to hold the antenna by hand.
The most common solution to this problem, but not effective here, is to add a so-called 'mode J' (old designation for mode V/U) filter in front of the 70cm receiver. This mode J filter is a notch filter that prevents the strong 2m signal from entering the 70cm receiver (or preamp in many cases for a fixed station).
The desensing is caused by a strong third harmonic from the 2m-uplink signal. The addition of a low-pass filter (LPF, in figure 2(a)) that filters out the third harmonic on 70cm solves this problem, which is shown in the basic calculations in table 3. These calculations are performed for the worst-case scenario where the frequency of the third harmonic of the transmitted signal is close to the frequency of the signal received from the satellite. In order to determine what amount of unwanted signal is entering the receiver, the signal separation between the 2m and 70cm ports of the antenna is required and must be measured. A transmission measurement was performed (figure 2(d)) with a handheld spectrum analyzer (FSH3, Rohde & Schwarz, Germany) between the two connectors of the antenna, for both the Arrow as well as the HB9CV. Table 2 shows the results of these measurements. Although only the 70cm numbers are relevant here, the 2m results were also measured. It is remarkable that, although for the HB9CV the 2m and 70 cm antennas are mounted in the same physical plane, the separation on 2m is still higher than for the Arrow.
|2m||45 dB||50 dB|
|70cm||23 dB||8 dB|
After these measurements, the signal strength was calculated with and without the extra LPF added. The results are shown in Table 3. The values at the bottom of the table (in bold) show the minimal signal intensity that can be received on 70cm while transmitting on 2m. All values are approximate values, but this does not alter the principle.
|Value||Without LPF||With LPF|
|2.5 W output power FT-817ND (2m)||34 dBm||34 dBm||34 dBm|
|Third harmonic suppression FT-817ND (70cm)||60 dB||-26 dBm||-26 dBm|
|Extra attenuation LPF (70cm)||50 dB||-76 dBm|
|Separation Arrow (70cm signals)||23 dB||-49 dBm||-99 dBm|
|BDR FT-817ND receiver||60 dB||-109 dBm||-159 dBm|
These calculations show that in some cases it is impossible to receive weak signals without the extra LPF, such as signals from AO-51, which are between –105 and –117 dBm (from 'Calculating Link Budget for AMSAT-OSCAR 51', AMSAT-NA). It also shows that no desensing will appear when the extra LPF is added, which was verified in practice.
The added LPF is based on a 2m/70cm diplexer design by HB9ABX, where only the 2m part is used for the LPF. The schematic is shown in figure 5(a). Adjustment was done with a dummy load and a VSWR meter. The LPF is home built and is shown in figures 5(b) and (c). A clip made from thick copper wire makes it possible to clip the LPF onto the carrying bag (figure 3(d)). Alternatively, the 2m part of commercial 2m/70cm diplexers can be used as a LPF. When operating in mode U/V, the LPF can be mounted in the receiving path to limit the transmitted 70cm signals flowing into the 2m receiver, as shown in figure 2(b).
A creative method was used to mount the S-band reception parts all together onto the Arrow antenna, as can be seen in figures 6(a) and (b). First an angled N-N connector is used to mount the patch antenna to the converter, and a strait N-N connector is used to mount the bias-T to the converter. From the bias-T to the receiving 817, regular RG58 coaxial cable with BNC connectors is used. It is recommended to add an extra signal attenuator (around 20 dB) between the converter and the 817, because of the high gain of the down converter (or to reduce the gain inside the down converter box itself). After all S-band reception parts are mounted together, all this is mounted onto the boom of the Arrow antenna with the 70 cm part removed. A small piece of aluminium was used to mount the converter to the boom, where the existing mounting holes in both the converter and Arrow were used.
The Heil Traveler dual headset works satisfactorily in this setup (although this headset is discontinued by Heil, Wimo has it still on stock). No RFI is detected during transmissions, which is sometimes the case with cheaper computer headsets. The audio delivered by the microphone element sounds clear and powerful, a warm welcome when working QRP. PTT can be operated from the push button in the cable to the headset, and is used for FM repeater operation.
One expensive Heil adapter cable (HSTA-YM) is required to connect the headset to the 817. On the 817, the microphone/PTT are connected via a different plug than the headphone audio, so the two plugs of the HSTA-YM adapter cable can be plugged into corresponding connectors on the two transceivers.
A portable setup like this is ideal to show other people both the fun and technical achievements of AMSAT satellites. Full-duplex operation is highly recommended for successfully making QSOs, but it also prevents people from listening in to a QSO because of audio feedback. In an attempt to overcome this problem, the audio was also fed to a car radio FM transmitter (Tunecast II, Belkin) to let other people listen to the satellite signal on their iPod or similar devices. However, this was unsuccessful as the transmitter generates so much RF that all satellite reception is blocked completely. A more rigid way of letting people listen in is the use of a simple audio splitter (Rockstar, Belkin, figure 7) and separate headphones for the extra listeners. A safe option is to use an extra audio splitter together with an audio extension cable (figure 7) so people do not have to stay very close to you, thus no harm is done to the equipment if they move one step away from you. When multiple headsets are connected to the audio output of the 817, the SP-PH switch can be set to SP to provide more audio power.
To be able to log the QSOs made, a small audio recorder is used (Zen Nano Plus, Creative, figure 7). This can be inserted in series with the headphone cable, or more conveniently connected in parallel with the audio splitter.
For single channel FM operation, is it convenient to use the multi-band memory capability of the 817, to have the Doppler correction pre-programmed for the entire pass. For SSB transponder QSOs, manual Doppler correction is required instead of using pre-programmed memories (as a starting point for manual tuning, a memory channel can be used to store the frequencies at the start of the pass). Memory programming can be performed with the 'FT817 Commander' software and a CT-62(U) interface cable.
S-band reception while making a QSO with this portable setup is very challenging, due to the weight of the antenna and the Doppler shift. Almost continuous manual Doppler correction is required, especially near the time of closest approach (TCA) of the satellite pass. While both the uplink and the output of the converter were on 2m, no interference between both was observed. The amount of noise in the 2.4 GHz band is enormous, with all wireless networks in the neighbourhood in the same band as the downlink signals from AO-51. Despite all this QRM, the signals from AO-51 were clearly readable.
Items are marked and labelled as much as possible, to make life easier when the setup is used outdoors, where small mistakes are more likely to be made than at home. The elements of the Arrow antenna are marked with the element sequence for fast mounting, all coax cables are marked with their frequency band to prevent wrong connection at the antenna or transceiver, and the transceivers are marked with 'RX' and 'TX' to indicate their function, together with different background colours of the displays.
Although the transceivers are marked with their function to be a receiver or transmitter, all memories are programmed both for reception (RX) and transmission (TX). In the case when a transceiver breaks down on RX or TX only, both transceivers can be exchanged to get the setup working again. In the case when one transceiver breaks down completely, the other one can still be used for working the FM satellites in half-duplex mode, which was also done successfully in the time when I owned only one FT-817ND.