Overview
The Skunkworks team at the Gloucester County Amateur Radio Club is always looking for new projects to dig into, learn from, and build something new. So when we found that the “Greencube” satellite had been launched and was a very different type of device we started figuring out how we could work it.
The Greencube satellite is described in this link, so we won’t spend a lot of time covering it here. It has two primary differences from other ham satellites – first, it’s in a medium earth orbit (MEO) about 2200 miles above the earth (by contrast, the International Space Station is about 250 miles above the earth). This means several things – its coverage footprint is far larger than any other current amateur satellites and its operating window over a particular point on earth will be measured in hours, not minutes.
The second significant difference is that it utilizes a “digipeater”, which is a digital repeater that receives digital signals from earth stations and retransmits them from the satellite. Given that the footprint of the satellite will cover an almost an entire hemisphere of the earth there may be thousands of operators trying to access it simultaneously; therefore only a communications protocol that involved short transmissions would be practical. Greencube digital transmissions last about 1/4 of a second, so many stations can be transmitting over a short period of time and still be heard by the satellite.
Most other amateur radio satellites are “full duplex”, meaning that they transmit and receive on different frequency bands, and that users can hear themselves in the downlink of the satellite. This is important because of the “Doppler shift” present in satellite operations in which frequencies need to be adjusted to compensate for the speed of the satellite as it moves overhead. Greencube differs in that it is “half duplex”, meaning the transmission and reception occur on the same frequency (435.310 MHz ). This meant several adjustments to the station configuration.
GCARC Satellite Station
The satellite station at the GCARC clubhouse is close to being state-of-the-art. For receiving it uses an SDR Play software defined radio coupled with the SDR Console user interface that allows the operator to visualize the entire passband of the satellite and make adjustments where necessary. A Yaesu 847 transceiver is used as a transmitter. The satellite antennas are among the best available from M2 – the 70 cm antenna is a 42 element 436CP42UG crossed yagi and the 2 meter antenna is a 22 element 2MCP22 crossed yagi. Antenna rotation is handled by a venerable Yaesu GS-232 rotator with an AMSAT LNB controller. Satellite mode switching for the antennas is usually handled by an Arduino-controlled relay switch, but because of the renovations at our satellite room it was temporarily replaced by two manual coaxial switches that switch the two antennas between the transmitter and the receiver.
This configuration works well for full duplex satellites but had to be modified for Greencube’s half duplex operation. Initially we manually switched the 70 cm antenna between transmit and receive but later were able to add a MFJ 1708B RF-sensing antenna switch that would disconnect and ground the SDR radio when the transmitter was transmitting. That let both the 847 and the SDR Play connect to the 70 cm antenna.
Two other functions are necessary for satellite operations – the antennas must be rotated to continually point at the satellite as it moves through the sky, and the transmit and receive frequencies must be adjusted for the Doppler shift that occurs when working satellites that are moving thousands of miles an hour. Those functions are both handled by the PST Rotator program, which we’ve found to work extremely well with all satellites.
The Greencube Software
When we initially started looking at the Greencube satellite the online references seemed to point us to a receiver that needed to be constructed from GNU radio, which was beyond our capabilities. This initially dissuaded us from pursuing that satellite until we located the satblog.info site that contained the digipeater and telemetry software. We also tried different variations of the UZ7HO “Soundmodem” software trying to identify the proper version for Greencube until we noticed a download link on their website for “greentnc.zip” that contains the modem software written specifically for the satellite. Soundmodem audio is fed from SDR console through a virtual audio cable and it decodes the audio packets into raw data. That data is fed through a TCP port into the Greencube decoder software that lets the operator view incoming packets, call CQ or respond, and also log the QSO. It’s a really neat set of software.
The final program is the telemetry receiver, which updates about every 45 seconds from packets transmitted by the satellite. This displays current values for various telemetry fields from the satellite, with the new vertical bar appearing in the lower panel with each new set of telemetry. There’s an accompanying program that will upload this data to the SatNOGS database but unfortunately I was not able to overcome a Windows error that occurs when I ran this program.
Initial issues
Initially we were somewhat successful with this configuration, working a number of stations throughout the middle of a pass. Two problems became quickly apparent – although we had a 42 element yagi antenna we weren’t able to decode signals near the beginning and end of the pass. Unfortunately, this is where the more interesting stations appear since they’re also the more distant stations. Most other Greencube stations appeared to have less sophisticated antennas but are also using preamplifiers mounted at the antenna, which we decided we needed.
The other issue was that that the transmitted signal frequently didn’t seem to be heard by the satellite. The FT-847 is only rated at 20 watts output on 70 centimeters and ours seemed to be putting out significantly less power, which doesn’t appear to be enough to create a reliable and readable signal at the satellite. From some online research we found that some other stations appeared to be using Icom IC-9700 transceivers that run 70 watts on 70 centimeters, so we decided we needed to upgrade to a more powerful radio.
A confounding problem also appeared to be finding the correct base frequency for transmitting. Setting PST Rotator to the published frequencies required adjustment to make the received audio frequency center around 1500 Hz. But what about the transmit frequency? We didn’t know exactly where the satellite would be listening, and being off by several hundred Hertz could make us unreadable. Finally, we found a reference telling us to set both frequencies lower by about 800 Hz, which put the received frequency perfectly in line for decoding by Soundmodem and also appeared to create the correct transmit frequency to be decoded by the satellite.
Adding the preamp
Al KB2AYU came to our rescue on the receive issue with a mast-mounted preamplifier for 70 centimeters that he installed on the antenna boom. We installed the power injector at the feedline switch and found that the preamp significantly improved receive performance. That preamp was initially switched out of the line by sensing the RF; however we later were able to hard-switch it using a direct connection to the PTT output of the transceiver.
Replacing the 847 with the 991A
The issue of low power output was solved by replacing the FT-847 transceiver with a newer FT-991a radio from my personal station. Initially we had some COM port issues which we traced down to having defined a virtual COM port at the same location that Windows assigned to the standard COM port of the 991. We also found that the 991 appeared to significantly reduce the high frequency component of the audio signal when set in USB-DATA mode (we found this by using the calibration function in Soundmodem and looking at the output from the low and high tones, finding the high tone output being significantly reduced). We solved this by operating the radio in USB mode, not USB-DATA mode. The complication with this arrangement is that there is apparently no adjustment on the transceiver for the audio level coming in through the USB port, so we had to adjust this in Windows for the maximum level that would not kick in the ALC. Finally we decided that we had everything working as well as we could.
How does it work?
At this point the Greencube station seems to work as well as we could possibly expect. We can decode signals virtually throughout the entire pass and can view our transmitted signals on the downlink most of the time. At this point we’ve worked 38 different countries on five continents, with the longest distance QSO being 11,000 kilometers between us and the station in Japan. We’re hopeful to work satellite DXCC although we’re not confident that there are actually 100 operational Greencube stations within the potential footprint of the satellite from our location.
But so far working this satellite has been an exciting new challenge, allowing us to learn much about how our equipment operates and how to create success on a digital half duplex satellite. We are planning to continue active operation on this satellite, so if you see W2MMD please give us a shout.