Courtesy of: DIY Space Exploration
A blog about Space, PocketQubes and DIY endeavours.
Tuesday, 7 October 2014
Cool PocketQube Infographic!
The PocketQube Shop and the DIY Space Exploration website have teamed up to create this fantastic PocketQube infographic.
Labels:
PocketQube
Thursday, 4 September 2014
67P Images from Rosatta
In case you've been living under a rock, the ESA spacecraft Rosetta is currently orbiting the comet 67P Churyumov-Gerasimenko.
In an attempt to involve the general public with the mission, ESA have published 4 separate Navcam images with anyone can use to combine together into a mosaic of the comet.
So here's my attempt from 4/9/2014:
Credit: ESA/Rosetta/NAVCAM/Stuart McAndrew
In an attempt to involve the general public with the mission, ESA have published 4 separate Navcam images with anyone can use to combine together into a mosaic of the comet.
So here's my attempt from 4/9/2014:
Credit: ESA/Rosetta/NAVCAM/Stuart McAndrew
Saturday, 19 July 2014
My Hackaday Prize Entry
Many of you out there already know of the website called Hackaday. They happen to be running a competition called "The Hackaday Prize", where people are invited to submit a project that meets a very lose criteria:
So seeing as I'd just bought some of the NiceRF radio modules for testing, I thought I'd give the test board a name and enter it in the competition!
So the test board is called QubeCast Max.
(Pocket) Qube
(Broad) Cast
Max - because it's high power! And leaves room for other models in case I make more of them.
Please support PocketQube's and my entry!
http://hackaday.io/project/1726-QubeCast-Max
Give me a skull if you like the project!
- You must actually build something
- It must transmit data to and/or from another device (computer, phone, duplicate/variation of your device, etc). This could be over the Internet, or using any other method of your choosing.
- Our main requirements have to do with documentation. This includes lists of parts, schematics, images, and videos. Remember, Openness is a Virtue.
So seeing as I'd just bought some of the NiceRF radio modules for testing, I thought I'd give the test board a name and enter it in the competition!
So the test board is called QubeCast Max.
(Pocket) Qube
(Broad) Cast
Max - because it's high power! And leaves room for other models in case I make more of them.
Please support PocketQube's and my entry!
http://hackaday.io/project/1726-QubeCast-Max
Give me a skull if you like the project!
Monday, 23 June 2014
PocketQube Radio Ramblings (#2)
I have just come across a new product that might be interesting to all the PocketQube builders out there (or anyone who is interested in small RF links in general).
It is a new module from a company called "Nice RF" . This is different to HopeRF , so it seems that competition is alive and well in Shenzhen, China!
The module is the RF4463F30 .
If the part number looks familiar to those in the know, it is because it is based on the Silicon Labs Si4463 Wireless IC. The difference with this module is the F30 on the end of the part number....(Sorry, had to go there :-) ) Whereas the Si4463 is rated at +20dBm output, this module adds a power amplifier to boost the output to +30dBm! That's right, one whole watt! The module costs around $18USD, and is selling on AliExpress
Now HopeRF have a similar module , the RFM23BP, based on the Si4463's predecessor, the Si4432. This module can be purchased for around $9USD from places such as Anarduino
Remember, the Si4432 is the IC inside the venerable RFM22B, as used in $50Sat, which has now been continuously operating in space for more than 6 months! While the RFM22B has an output of +20dBm, the RFM23BP has an output of +30dBm. The RFM23BP also requires a 5V supply ( but supposedly works down to 3.3V)
To work out the differences between the RFM23BP and the RF4463F30, we'll need to dig into the datasheets. Now at the moment, the datasheet for the RF4463F30 is a little sparse, so I'll have to wait til my in-depth queries are answered by their tech support people. But on the surface, the differences are as follows:
Max Data Rate: 256kbps (RFM23BP), 1000kbps (RF4463F30)
Receive mode sensitivity: -120dBm (RFM23BP), -126dBm(RF4463F30)
Note the higher sensitivity of the RF4463F30 is based on a lower bit rate and smaller frequency deviation.
Receive mode current: 25mA (RFM23BP), 10-13.5mA (RF4463F30) - which seems to have 2 different sensitivity modes - High and Low
Transmit mode current (max power): 550mA (RFM23BP), 540mA (RF4463F30)
Size: 33mmx18mm(RFM23BP), 38mmx20mm(RF4463F30)
Now as the RF4463F30 is based on the Si4463, all the software commands that apply to the Si4463 can be used on the RF4463F30. Bearing in mind that 2 of the GPIO pins are used internally in the module for antenna switching functions.
So is this new module worth double the cost of the RFM23BP? Remembering we're talking about wireless modules that cost under $20 each, but it depends on your scenario. If you need to have (relatively) high power, both modules fit the bill, with similar power output and power usage. The newcomer seems to use less on the receive side, but if you want lower power usage for receiving, you may want to get the regular Si4463 module without the high power option, as the sensitivity is the same. The HopeRF module uses a register based firmware, whereas the Si4463 uses a new API approach.
I think some testing may be in order to really understand how the modules perform.
It is a new module from a company called "Nice RF" . This is different to HopeRF , so it seems that competition is alive and well in Shenzhen, China!
The module is the RF4463F30 .
Now HopeRF have a similar module , the RFM23BP, based on the Si4463's predecessor, the Si4432. This module can be purchased for around $9USD from places such as Anarduino
Remember, the Si4432 is the IC inside the venerable RFM22B, as used in $50Sat, which has now been continuously operating in space for more than 6 months! While the RFM22B has an output of +20dBm, the RFM23BP has an output of +30dBm. The RFM23BP also requires a 5V supply ( but supposedly works down to 3.3V)
To work out the differences between the RFM23BP and the RF4463F30, we'll need to dig into the datasheets. Now at the moment, the datasheet for the RF4463F30 is a little sparse, so I'll have to wait til my in-depth queries are answered by their tech support people. But on the surface, the differences are as follows:
Max Data Rate: 256kbps (RFM23BP), 1000kbps (RF4463F30)
Receive mode sensitivity: -120dBm (RFM23BP), -126dBm(RF4463F30)
Note the higher sensitivity of the RF4463F30 is based on a lower bit rate and smaller frequency deviation.
Receive mode current: 25mA (RFM23BP), 10-13.5mA (RF4463F30) - which seems to have 2 different sensitivity modes - High and Low
Transmit mode current (max power): 550mA (RFM23BP), 540mA (RF4463F30)
Size: 33mmx18mm(RFM23BP), 38mmx20mm(RF4463F30)
Now as the RF4463F30 is based on the Si4463, all the software commands that apply to the Si4463 can be used on the RF4463F30. Bearing in mind that 2 of the GPIO pins are used internally in the module for antenna switching functions.
So is this new module worth double the cost of the RFM23BP? Remembering we're talking about wireless modules that cost under $20 each, but it depends on your scenario. If you need to have (relatively) high power, both modules fit the bill, with similar power output and power usage. The newcomer seems to use less on the receive side, but if you want lower power usage for receiving, you may want to get the regular Si4463 module without the high power option, as the sensitivity is the same. The HopeRF module uses a register based firmware, whereas the Si4463 uses a new API approach.
I think some testing may be in order to really understand how the modules perform.
Other Modules
Looking at the NiceRF website, they have a few other modules that use either the Si4432 or the Si4463 Wireless IC's. Another of interest is the RF4432F27 . Looks like a +27dBm version of the RFM22B. Max TX current is 350mA @ 5V ( in case 500mA is too high)
They also have some test or demo boards with built-in Microcontroller and LCD.
All in all, it looks like there are now a few more options for ready made radio modules for PocketQube's!
Labels:
DIY Space,
HopeRF,
NiceRF,
PocketQube,
Radio
Tuesday, 17 June 2014
EPS Info for nanosats
Some more useful info for PocketQube builders relating to the ST SPV1040 IC. Turn out that it is being used in the US AMSAT Fox-1 Cubesat, CUBESTAR cubesat, NTNU Test Satellite ( NUTS), ESTCube-1 , and probably several others. If you read all the papers and articles about these, you should get a good idea about system designs.
This means the list of useful candidates is:
Linear Technology LTC3105
ST Microelectronics SPV1040
Spansion MB39C831QN
Texas Instruments BQ25504
They all have different features and limitations, so you'll need to look at the datasheets to make sure it's right for your project.
This means the list of useful candidates is:
Linear Technology LTC3105
ST Microelectronics SPV1040
Spansion MB39C831QN
Texas Instruments BQ25504
They all have different features and limitations, so you'll need to look at the datasheets to make sure it's right for your project.
Labels:
BQ25504,
EPS,
LTC3105,
MB39C831QN,
PocketQube,
Power,
SPV1040
Tuesday, 15 April 2014
PocketQube Power!
I thought the it was about time that I put something out there about the power subsystem for PocketQubes. As I've said previously, there aren't many COTS products available specifically for the PocketQube form factor. So to date, this means that everything from solar cells to EPS ( Electrical Power Systems) needs to be custom made. Cubesats are a different story. There are a myriad of vendors offering EPS boards with many different options, depending on the intended purpose and cubesat size. Clyde Space have a large range, starting from $3500.
Then there's solar cells. Once again, plenty of vendors. A common DIY cell is the Spectrolab TASC cell. This has been used in many Cubesats, including PhoneSat. The problem is that these aren't readily available outside USA
So back to a PocketQube sized budget. What are the options?
I'm sure someone will say "It depends on a range of factors........" but I like to simplify. The surface area of PocketQubes is fairly small, which seriously limits the amount of solar power that can be harnessed. So we're looking at fairly low power scenarios.
Take $50Sat for instance. The EPS board was built from regular COTS components. It is based on four LTC3105 max power point controllers, one for each solar panel, and some current measuring components etc. The solar panels themselves are TASC cells, with 6 on each of the 4 sides of the satellite. One of the other smaller sides is used by the antenna, and the other left vacant. Now I'm assuming that they are using pairs of the cells in series to generate approx 4.4V ( I could be wrong here, so don't take this as gospel), and 3 pairs in parallel. Looking at the datasheet, and based on the Max Power Point values, we're looking at 84mA, which totals approx 370mW per side. ( and yes, that's not based on the irradiance levels expected in space. Most datasheets use the terrestrial measure of 1000W/m2)
The place to look for power components here is the "Energy Harvesting" categories of the various chip makers. This is where you have to trawl through datasheets. Here's some that I've come across.
The LTC3105 is from Linear Technology. They have a few IC's in this area, but the LTC3105 seems to be a good match for PQ's in terms of input voltage range, and power efficiency. It has MPPT charging for maximum charging efficiency.
Then there's the SPV1050 from STMicroelectronics. It is also an MPPT design. It does seem limited in battery charging current though, so it may not be suitable.
Texas Instruments have a few in this area. The BQ25505 looks promising, and has a cool feature in that it can switch to a primary ( non rechargeable battery) if the rechargeable one is fully drained.
Spansion have MB39C831 . It is a little larger package than the others listed.
Then there's solar cells. Once again, plenty of vendors. A common DIY cell is the Spectrolab TASC cell. This has been used in many Cubesats, including PhoneSat. The problem is that these aren't readily available outside USA
So back to a PocketQube sized budget. What are the options?
I'm sure someone will say "It depends on a range of factors........" but I like to simplify. The surface area of PocketQubes is fairly small, which seriously limits the amount of solar power that can be harnessed. So we're looking at fairly low power scenarios.
Take $50Sat for instance. The EPS board was built from regular COTS components. It is based on four LTC3105 max power point controllers, one for each solar panel, and some current measuring components etc. The solar panels themselves are TASC cells, with 6 on each of the 4 sides of the satellite. One of the other smaller sides is used by the antenna, and the other left vacant. Now I'm assuming that they are using pairs of the cells in series to generate approx 4.4V ( I could be wrong here, so don't take this as gospel), and 3 pairs in parallel. Looking at the datasheet, and based on the Max Power Point values, we're looking at 84mA, which totals approx 370mW per side. ( and yes, that's not based on the irradiance levels expected in space. Most datasheets use the terrestrial measure of 1000W/m2)
The place to look for power components here is the "Energy Harvesting" categories of the various chip makers. This is where you have to trawl through datasheets. Here's some that I've come across.
The LTC3105 is from Linear Technology. They have a few IC's in this area, but the LTC3105 seems to be a good match for PQ's in terms of input voltage range, and power efficiency. It has MPPT charging for maximum charging efficiency.
Then there's the SPV1050 from STMicroelectronics. It is also an MPPT design. It does seem limited in battery charging current though, so it may not be suitable.
Texas Instruments have a few in this area. The BQ25505 looks promising, and has a cool feature in that it can switch to a primary ( non rechargeable battery) if the rechargeable one is fully drained.
Spansion have MB39C831 . It is a little larger package than the others listed.
Ideally, you want to design the system to maximise efficiency. The diminutive size of PocketQubes
Thursday, 6 March 2014
Receiving Telemetry from Satellites - DIY Style
Why would you want to?
I'm sure everyone could appreciate the reasons for satellite operators to receive the signals or transmissions from their own satellites, but why do people want to receive signals from other peoples's satellites? I'm not talking about the signals of the Satellite TV stations. I'm talking about the morse code, and other encoded data transmitted from many nanosatellites currently in orbit.
I call it Citizen Science. And because it doesn't take a whole lot of gear to get it happening.
Alot of Cubesats and PocketQubes are from organisations that don't have the global communications networks like NASA or ESA. They are often based in a single location in a single country. Consider how often a satellite will pass within range of that location. At the altitude most nanosatellites orbit the earth, they only have line-of-sight contact with the groundstation for about 9 minutes or so. Nanosatellites also have a limited speed that they can transfer data at. If that organisation could only transfer data from the satellite when it is in range of that single ground station, the total amount of data transferred would be very low. The satellite would have to store information from its many orbits, then try and dump all the data when it is in range.
Now imagine if that satellite operator had ground stations spread all the way round the world. The measurements taken by the instruments onboard could be downloaded much more often. It the information was then collated, it would be much more useful.
An example at the moment is the PocketQube $50SAT. This little satellite transmits in the UHF band, at 437.505Mhz. The transmitter is only 100mw, which is pretty low compared to most small satellites. They have published a document in their dropbox account that describes how to receive their telemetry. Each transmission contains the actual realtime readings from various spacecraft sensors. If no-one receives the transmission, that info is lost. Anyone with the right gear can receive these signals and help the $50SAT team with their measurements.
Aren't HAM radio's expensive?
Good ones are, but we're going to use some modern tech that is cheap, and then you can either buy or build an antenna to your liking. It all depends on how you want to use it.
What do I need to do?
We'll use $50Sat as an example. The published communications guide contains alot of info, but before you get to that, you'll need about 5 things. A USB DVB-T Dongle, an adapter, an antenna cable, an antenna, and a computer to install a few pieces of software on.
USB DVB-T Dongle.
Often found on eBay. Search for RTL2832U or E4000. Or go to http://sdr.osmocom.org/trac/wiki/rtl-sdr to get a list of compatible dongles. I used one like this
Often found on eBay. Search for RTL2832U or E4000. Or go to http://sdr.osmocom.org/trac/wiki/rtl-sdr to get a list of compatible dongles. I used one like this
Antenna adapter.
Depending on the type of dongle you purchased, you'll need to adapt the antenna connector to one more readily used for the frequencies we're interested in. Also on eBay. The connector I had was an MCX connector, so I got an MCX male to SMA Female adapter.
Depending on the type of dongle you purchased, you'll need to adapt the antenna connector to one more readily used for the frequencies we're interested in. Also on eBay. The connector I had was an MCX connector, so I got an MCX male to SMA Female adapter.
Cable.
Now this depends on what type of antenna you're going to use as well. I'll assume for now, you're going to wave something round the lounge room, until you work out exactly how to consistently receive the transmissions. So you'll need a cable anywhere from 1-3M. The type if cable can vary. Use RG59 for short lengths, or something like LMR200 for lower cable loss. SMA Male connector on one end, and the other end will depend on your antenna. Usually it'll be an N Male. Mine was purchased with the antenna. I went with 3M of LMR200 cable.
Now this depends on what type of antenna you're going to use as well. I'll assume for now, you're going to wave something round the lounge room, until you work out exactly how to consistently receive the transmissions. So you'll need a cable anywhere from 1-3M. The type if cable can vary. Use RG59 for short lengths, or something like LMR200 for lower cable loss. SMA Male connector on one end, and the other end will depend on your antenna. Usually it'll be an N Male. Mine was purchased with the antenna. I went with 3M of LMR200 cable.
Antenna.
There are many books on the subject. You could either make or buy one, depending on how keen you are. I went with a custom tuned 4 element Yagi from an Australian supplier called ZCG Scalar. The model is the Y404 . I requested the tuning for a center frequency of 437Mhz. RRP for this antenna was $199.
Here's a British Interplanetary Society article showing a few other options they are using for receiving signals from KickSats.
Computer
First, you'll need some software to use with the SDR dongle. 2 that I use are SDR# and HDSDR . SDR# has a guide on how to install the required drivers on Windows. The HDSDR site has instructions for specifics to their software.
Basically, don't let windows detect and install a driver. Click cancel to check online etc. Then run the Zadig software to associate the proper driver with the Bulk-In Interface 0. Don't worry about Interface 1.
Then put the correct dll files into the program file folder.
Like I said, just follow the instructions for the software you're going to use. You may want to read the user guides for each bit of software to become familiar with how they operate and how you set the tuning frequencies etc.
Next, you'll need to find out where the satellites are! You can do this with a program called Orbitron. Once again, read the manual for how to use it. You'll need to get the latest TLE's for the satellites from Celestrak. Orbitron has this function built in. TLE's are also known as Two Line Elements. They describe the orbital characteristics of a satellite. Load the Cubesat.txt TLE file, then select Eagle 2 from the list. This is the alternative name for $50SAT. Once selected, it will show you the current location of the satellite!
Now set your home location, then run the prediction setup and enter the required parameters. Go to the Prediction tab then press Predict. It will show you when the satellite will be in range next.
Now you can either keep recording while the satellite is in range, or move to the next step.
There are many books on the subject. You could either make or buy one, depending on how keen you are. I went with a custom tuned 4 element Yagi from an Australian supplier called ZCG Scalar. The model is the Y404 . I requested the tuning for a center frequency of 437Mhz. RRP for this antenna was $199.
Here's a British Interplanetary Society article showing a few other options they are using for receiving signals from KickSats.
Computer
First, you'll need some software to use with the SDR dongle. 2 that I use are SDR# and HDSDR . SDR# has a guide on how to install the required drivers on Windows. The HDSDR site has instructions for specifics to their software.
Basically, don't let windows detect and install a driver. Click cancel to check online etc. Then run the Zadig software to associate the proper driver with the Bulk-In Interface 0. Don't worry about Interface 1.
Then put the correct dll files into the program file folder.
Like I said, just follow the instructions for the software you're going to use. You may want to read the user guides for each bit of software to become familiar with how they operate and how you set the tuning frequencies etc.
Next, you'll need to find out where the satellites are! You can do this with a program called Orbitron. Once again, read the manual for how to use it. You'll need to get the latest TLE's for the satellites from Celestrak. Orbitron has this function built in. TLE's are also known as Two Line Elements. They describe the orbital characteristics of a satellite. Load the Cubesat.txt TLE file, then select Eagle 2 from the list. This is the alternative name for $50SAT. Once selected, it will show you the current location of the satellite!
Now set your home location, then run the prediction setup and enter the required parameters. Go to the Prediction tab then press Predict. It will show you when the satellite will be in range next.
Receiving signals from space!!!!
Once you've got the hang of Orbitron and the SDR software, you're ready to track down the satellite. When Orbitron says the satellite will be in range, start the SDR software and configure it to record the raw input signal RF. This file gets big very quick!
Select LSB , and set the tuning frequency just below the frequency of $50SAT ( 437.495 ish) The exact figure doesn't matter as the dongle isn't 100% accurate, and you can adjust later.
Watch the waterfall around where you've set the tuning frequency, and with any luck, you'll see the telltale blips leading up to the RTTY transmission. You can set the tuning frequency just above the signal, so that the highlighted area covers the radio signal. The RTTY looks like 2 parallel lines running at an angle up the screen. Here's an HDSDR screenshot.
Decoding It!
Now here's the very low tech bit. You'll need a 3.5mm Stereo to 3.5mm Stereo cable, long enough to connect your microphone jack to the headphone jack.
Open the sound mixer on your computer and go to the input settings. Turn off any microphone boost checkboxes, and set the volume down just under 1 segment. Plug in the cable.
In HDSDR, open the WAV file you created earlier ( usually in MY Docs / HDSDR). Loop through the file and set the tuning frequency so that the whole transmission is contained within the highlighted area, then go back to just before the transmission starts, then pause the playback. Open a sound recording program, such as the built in Windows Sound Recorder. Hit record, then unpause the HDSDR playback. Look at the sound recorder to see if the audio level as high or low enough. You want it high enough so that the volume is good, but no clipping or distortion. ( Remember to unplug the headphone plug to hear the playback!!!)
If the levels are good, save the file as 48Khz , 16 bit mono audio. ( Must be mono)
Then, as described in the $50SAT comms document, you'll need another bit of software called "fldigi".
Setup the software as described in the document. One thing that isn't mentioned in the document is some settings that are visible in the latest version of the software. At the bottom left of the screen are 2 boxes with arrows either side - "Upper Signal" and "Signal Range". I set these to -28 and 17 respectively, but this can vary depending on the volume level of the output file.
Select File > Audio > Playback, then select the file you created with Sound Recorder. It should start a waterfall display down the bottom of the screen. If you've setup the software correctly, you should be able to click in the centre of the rtty transmission, and the red bars go over the red part of the waterfall plot.
If the signal is right, you should start seeing the telemetry being decoded in the top window!
You may need to manually click the centre frequency "up the slope" to get it to decode different sections of the file, in case the Auto Frequency Correction doesn't work.
Voila! You're now receiving and decoding data from a satellite.
If you validate the data correctly, submit it to the $50SAT team to add to the other reports that people are making. Hopefully you'll get a warm fuzzy feeling that you've helped someone, and a sense of achievement that you've built a functional ground station!
Open the sound mixer on your computer and go to the input settings. Turn off any microphone boost checkboxes, and set the volume down just under 1 segment. Plug in the cable.
In HDSDR, open the WAV file you created earlier ( usually in MY Docs / HDSDR). Loop through the file and set the tuning frequency so that the whole transmission is contained within the highlighted area, then go back to just before the transmission starts, then pause the playback. Open a sound recording program, such as the built in Windows Sound Recorder. Hit record, then unpause the HDSDR playback. Look at the sound recorder to see if the audio level as high or low enough. You want it high enough so that the volume is good, but no clipping or distortion. ( Remember to unplug the headphone plug to hear the playback!!!)
If the levels are good, save the file as 48Khz , 16 bit mono audio. ( Must be mono)
Then, as described in the $50SAT comms document, you'll need another bit of software called "fldigi".
Setup the software as described in the document. One thing that isn't mentioned in the document is some settings that are visible in the latest version of the software. At the bottom left of the screen are 2 boxes with arrows either side - "Upper Signal" and "Signal Range". I set these to -28 and 17 respectively, but this can vary depending on the volume level of the output file.
Select File > Audio > Playback, then select the file you created with Sound Recorder. It should start a waterfall display down the bottom of the screen. If you've setup the software correctly, you should be able to click in the centre of the rtty transmission, and the red bars go over the red part of the waterfall plot.
If the signal is right, you should start seeing the telemetry being decoded in the top window!
You may need to manually click the centre frequency "up the slope" to get it to decode different sections of the file, in case the Auto Frequency Correction doesn't work.
Voila! You're now receiving and decoding data from a satellite.
If you validate the data correctly, submit it to the $50SAT team to add to the other reports that people are making. Hopefully you'll get a warm fuzzy feeling that you've helped someone, and a sense of achievement that you've built a functional ground station!
Labels:
AMSAT,
HopeRF,
PocketQube,
SDR
Subscribe to:
Posts (Atom)