PICAXE is in Space
- Thread starter srnet
- Start date
@Stuart
I've tried listening, but I still can't "hear" it.
Do you have any more reception advice please,
which might also help others?
I've got a good easterly "visual view" from ~ 0º to ~ 145º
Orbitron software with your up-to date data.
I've tried with an old Yupiteru MVT-7100 receiver (ssb / usb / fm) with a "rubber-duck" antenna.
Also, I've tried with a Yaesu FT-7800E (fm, 5k steps) with a roof mounted "white stick" antenna.
I've tried audio and also using an old version of Spectran.
Still no joy. :- (
e
I've tried listening, but I still can't "hear" it.
Do you have any more reception advice please,
which might also help others?
I've got a good easterly "visual view" from ~ 0º to ~ 145º
Orbitron software with your up-to date data.
I've tried with an old Yupiteru MVT-7100 receiver (ssb / usb / fm) with a "rubber-duck" antenna.
Also, I've tried with a Yaesu FT-7800E (fm, 5k steps) with a roof mounted "white stick" antenna.
I've tried audio and also using an old version of Spectran.
Still no joy. :- (
e
This is what I did this morning, pass due of Unisat-5 at 11:34 UTC, closest 648km, 69degrees.
Turn on Yaesu FT60 handheld, its got a Nagoya NA-666 VHF\UHF antenna on it, so unlikley to have any gain.
Tune handheld to 437.505Mhz, its on 5khz tuning steps and only receives FM.
Listen.
At approx 11:38 Morse beacon heard. By this time $50SAT ought to be well past the closest approach.
Audio recording on the Dropbox, in the pass information folder.
Turn on Yaesu FT60 handheld, its got a Nagoya NA-666 VHF\UHF antenna on it, so unlikley to have any gain.
Tune handheld to 437.505Mhz, its on 5khz tuning steps and only receives FM.
Listen.
At approx 11:38 Morse beacon heard. By this time $50SAT ought to be well past the closest approach.
Audio recording on the Dropbox, in the pass information folder.
We believe that $50SAT\Eagle2 may be object 2013-066W, which can now be tracked online at;
http://www.satview.org/?sat_id=39436U
Bearings made yesterday with a yagi and checks today on the doppler shift changes, do indicate that object W is within 20 seconds or so of $50SAT
The other morning I was using my UHF handheld and 7 element yagi and heard enough of the beacon audio (just) to get a fix at a range of 2,200km.
http://www.satview.org/?sat_id=39436U
Bearings made yesterday with a yagi and checks today on the doppler shift changes, do indicate that object W is within 20 seconds or so of $50SAT
The other morning I was using my UHF handheld and 7 element yagi and heard enough of the beacon audio (just) to get a fix at a range of 2,200km.
That's good news Stewart. Looking on space-track.org the gap in the Norad catalogue numbers is down from 17 to 7 now.
As you originally said, size makes for difficult tracking or at least the initial unique identification. That would mean that pending any further revelations that it looks as though EAGLE 2, $50SAT, is Norad catalogue 39436, and the currently assigned SATNAME "DNEPER OBJECT W", INITLDES "2013-066W".
Once firmly established, I don't know how long the process of the satellite name being reassigned to something more like the satellite operator's project title, but it does indicate that it's all far from being just a point and shoot game, even for the large professional set-ups that watch our heavens.
As you originally said, size makes for difficult tracking or at least the initial unique identification. That would mean that pending any further revelations that it looks as though EAGLE 2, $50SAT, is Norad catalogue 39436, and the currently assigned SATNAME "DNEPER OBJECT W", INITLDES "2013-066W".
Once firmly established, I don't know how long the process of the satellite name being reassigned to something more like the satellite operator's project title, but it does indicate that it's all far from being just a point and shoot game, even for the large professional set-ups that watch our heavens.
We should be able to tell fairly quickly if the TLE for object W is correct.
The doppler shift changes very rapidly as the satellite is at its closest to the listener, thus if you let the computer adjust the tune frequency of the radio (using the TLE) you can easily see on the received signal spectrum display if the tuned frequency stays in the centre of the received frequencies.
The doppler shift changes very rapidly as the satellite is at its closest to the listener, thus if you let the computer adjust the tune frequency of the radio (using the TLE) you can easily see on the received signal spectrum display if the tuned frequency stays in the centre of the received frequencies.
Just received a series of valid 1kbps data packets from $50SAT as it passed over the UK.
Using a handheld 7 element yagi and a low noise amplifier in front of the RFM22B.
No adjustments to receiver frequency for doppler, it was just left on the centre frequency.
Estimated distance to $50SAT, 740km.
Using a handheld 7 element yagi and a low noise amplifier in front of the RFM22B.
No adjustments to receiver frequency for doppler, it was just left on the centre frequency.
Estimated distance to $50SAT, 740km.
That's good news, you must be pleased as punch . . . deservedly so.
To put things into context, an FM receiver isn't likely to be as sensitive as an SSB receiver, but it is less sensitive to mistuning, which here is used to good advantage. The data packets were doubtless taken whilst the satellite was close to the nearest overhead point, at which time the transmission will appear at or about the nominal frequency, with least amount of Doppler shift. Even though at that time the Doppler will be changing at it's maximum rate during the pass.
The insensitivity of FM to tuning errors, the timing of the packet transmissions near to the overhead position have all contributed to produce a good result, which might have been less likely had an SSB receiver been used on a fixed frequency.
Arguably with a good signal level produced through the use of a gainy antenna, a wider reception window might be engineered by tuning anything up to a couple of kHz above or below the nominal frequency, because the Doppler is moving more slowly the further away from the nominal frequency it is.
If the pass is at a lower elevation, and so slightly further away, the trade off is slower effective Doppler shift versus a potentially weaker signal.
These are all points that Stewart and others are using to their advantage, but to those with no previous experience they are effects that may not be instantly obvious.
without a degree of frequency tracking. A high data rate requires only a narrow reception window at a fixed frequency but is more affected by Doppler
To put things into context, an FM receiver isn't likely to be as sensitive as an SSB receiver, but it is less sensitive to mistuning, which here is used to good advantage. The data packets were doubtless taken whilst the satellite was close to the nearest overhead point, at which time the transmission will appear at or about the nominal frequency, with least amount of Doppler shift. Even though at that time the Doppler will be changing at it's maximum rate during the pass.
The insensitivity of FM to tuning errors, the timing of the packet transmissions near to the overhead position have all contributed to produce a good result, which might have been less likely had an SSB receiver been used on a fixed frequency.
Arguably with a good signal level produced through the use of a gainy antenna, a wider reception window might be engineered by tuning anything up to a couple of kHz above or below the nominal frequency, because the Doppler is moving more slowly the further away from the nominal frequency it is.
If the pass is at a lower elevation, and so slightly further away, the trade off is slower effective Doppler shift versus a potentially weaker signal.
These are all points that Stewart and others are using to their advantage, but to those with no previous experience they are effects that may not be instantly obvious.
without a degree of frequency tracking. A high data rate requires only a narrow reception window at a fixed frequency but is more affected by Doppler
You might think that, but it is not so.
I went back through the reception log, the 'receiver' is just one of the PICAXE boards with an RFM22 on it, and it logs to terminal what its up to.
Since I know the time between groups of packets, and even if I assume that the two groups of packets were spaced (in time) equal distance around the point of nearest approach, one of the packets came in from at least 821km, and the received signal was then doppler shifted by 7khz away from centre, 437.505mhz. The automatic frequency control (AFC) of the RFM22B copes very well with small frequency offsets. I did a lot of testing for this, and with the settings I was using the RFM22B performance only started to degrade when the frequency offset between TX and RX was more than 10khz.
The PICAXE receive program does allow the RFM22B frequency to be shifted manually by 1khz at a time, I will be trying that later to see if it makes a difference. It might, but my guess is not as much as you might think.
If you think of the normal application for the device, very cheap short range data stuff, then a good AFC performance becomes essential as it avoids the need to calibrate each device to ensure RX and TX frequencies match. You also dont need to use an expensive close tolerance crystal.
At around 11:38 UTC on the 8th December I sent a series of test packets to $50SAT.
The packets were from a RFM22B and PICAXE ground station amplified up to 20W, antenna was a 7 element yagi. Distance to $50SAT 1000km.
$50SAT received the packet and replied by sending acknowledge packets and the RSSI level in slow Morse.
There will be further tests in the coming days to see how much uplink power I need to use, no need to use more than required.
The packets were from a RFM22B and PICAXE ground station amplified up to 20W, antenna was a 7 element yagi. Distance to $50SAT 1000km.
$50SAT received the packet and replied by sending acknowledge packets and the RSSI level in slow Morse.
There will be further tests in the coming days to see how much uplink power I need to use, no need to use more than required.
That's really impressive! Good stuff.
Any chance of a pdf document for others to read about the project?
A design overview.
Testing done.
Total cost from scratch (exc time).
Total time - including assistance from others.
Things you would change - knowing what you know now.
Any radio licensing/qualifications required.
Any chance of a pdf document for others to read about the project?
A design overview.
Testing done.
Total cost from scratch (exc time).
Total time - including assistance from others.
Things you would change - knowing what you know now.
Any radio licensing/qualifications required.
When time allows, as you can imagine putting documents like that together can be a major effort on its own.
The completed satellite cost around £125 in parts, most of that went on the solar panels, which were given to us anyway.
Total cost, no idea, several prototypes for sure, from the processor\radio board perspective, maybe 4 of them at around £20 each in parts.
Not even going to try and count the amount of time spent.
Some one holding an Amateur radio license needs to be responsible for the Satellite in order to get a frequency allocated.
A lot of testing I did was at full power, even though it was only 100mW, an Amateur radio license is required for that.
Possibly.
Its a relativly simple build and non of the parts difficult to find, a combination of Farnell, eBay and Rev Ed.
The both the ground station board and $50SAT board can also act as a balloon tracker transmitter, so you can either into near space at least.
It should easily have enough range for balloon tracking, even when you restrict the power output to 10mW to satisfy UK requirements.
Oh crumbs I can imagine it could be a whopper.
I wasn't thinking of a War&Peace
In terms of time I wasn't expecting to the nearest hour, but people may like to get an idea whether it took weeks/months depending on their spare time available.
Cost; well, you mention RF amplifier and yagi. Do all those bits'n'bobs get included in your costs?
It could be a show-stopper for some.
Your 20W ground station; would I need a license to broadcast that power (+yagi)?
I was just thinking of all the add-ons that little Johnny would have to get to carry out something similar.
I wasn't thinking of a War&Peace
In terms of time I wasn't expecting to the nearest hour, but people may like to get an idea whether it took weeks/months depending on their spare time available.
Cost; well, you mention RF amplifier and yagi. Do all those bits'n'bobs get included in your costs?
It could be a show-stopper for some.
Your 20W ground station; would I need a license to broadcast that power (+yagi)?
I was just thinking of all the add-ons that little Johnny would have to get to carry out something similar.
Not really, that is stuff many Amateur radio practitioners would have anyway.
For a school for instance, they will get good signals with a Funcube Dongle (£125) and a home made QF Helix (£10 in bits).
Reception will improve quite a bit if you use a masthead low noise amplifier, £125+. I have tried a few much lower cost priced kits without a lot of success, so I use a commercial one.
The power amplifier was a kit, about £75, but its license holders only of course. Amplifiers like this need a fair bit of care in construction and you need basic equipment to measure harmonics etc, these things can put out a lot of interference if not done well. You can buy commercial ones of course,
I will be testing in the next few days to see how much power is needed to uplink, without the yagi. Its a lot more convenient to use the omni directional QF helix. I can upgrade the power amplifier to 60W if I need to.
Last edited:
kranenborg
Senior Member
Hi Srnet,
Is there a charging problem that may affect $50SAT on the long run? I do not fully understand the issue:
http://amsat-uk.org/2013/12/09/50dollarsat-seeks-help-from-radio-hams/ (includes very nice zoomable pictures)
http://st2nh-blogger.blogspot.se/2013/12/50sat-received.html
/Jurjen
http://www.kranenborg.org/electronics
Is there a charging problem that may affect $50SAT on the long run? I do not fully understand the issue:
http://amsat-uk.org/2013/12/09/50dollarsat-seeks-help-from-radio-hams/ (includes very nice zoomable pictures)
http://st2nh-blogger.blogspot.se/2013/12/50sat-received.html
/Jurjen
http://www.kranenborg.org/electronics
Last edited:
We know its getting plenty of charge, it would only have lasted a day or so without it, but we would like some reception reports from the Southern Hemisphere which will tell us how quickly the satellite is warming up, and the actual amount of charge its getting. This will all help to benchmark the performance of such small satellites.
"Your every wish is my command" -down under reporting in! I'm now back in high summer Wellington, but have yet to hear (in spite of assorted Yagi's & Ham UHF gear) anything of the little darling I'm afraid. No doubt this relates to my monitoring too early/late?
Stan. (ZL2APS)
Stan. (ZL2APS)
No idea.
The TLEs on the Dropbox are about right, within a few seconds.
And the object we think it is can be tracked on the online trackers.
We know its getting down South because we have had a reception report from Buenos Aires
Its only visible above the horizon for about 13 minutes in total. So you wont hear it for more than maybe 5 minutes either side of the point of closest approach.
Best to set up for UTC and check online what the current UTC is, any other time reference is way to confusing for everybody.
Fortunately (for me) at this time of the year UTC = UK local Time.
Best to set up for UTC and check online what the current UTC is, any other time reference is way to confusing for everybody.
Fortunately (for me) at this time of the year UTC = UK local Time.
Its due your way at 22:37:43 UTC, range 722km.
You should hear it on a HT and simple whip.
See attached graphic.
Attachments
manuka
Stuart is correct; the pass starting at 22:30 UTC (10:30 AM local time) will come within 704 km (based on gpredict with the latest TLEs) of your location. I am not sure what you are using for a radio, but the calculations we did using an Yaesu FT-60 and a Diamond RH77-CA (about 2 to 3 dBi) indicate the link margin goes to zero at about 1100 km slant range. This means you should hear it when the elevation is about 30 degrees or higher.
You may have to tune upto +/- 10 KHz on your radio to compensate for doppler as well as output frequency drift due to temperature. Between -10 and 0 degrees C the output frequency will be high by about 7 KHz.
Once you are able to hear it, we hope you might be able to provide some telemetry reports. Because of the orbit $50SAT is in and the fact it is approaching winter here in the northern hemisphere, it is still too cold to enable solar power when it makes a pass over our locations. Stations in the southern hemisphere have the best chance to receive telemetry with $50SAT at or near it maximum solar power generating capacity, which is one of the last items left on our post-launch checklist.
For all you folks in the northern hemisphere, do not let that stop you from listening to $50SAT and reporting telemetry. One of the other items on our post-launch checklist is an estimate of minimum and maximum temperature. Thus far, we have seen temperatures as high as 26 degrees C and low as -21 degrees C. Stations in the northern hemisphere can probably get us a low temperature reading, especially on an evening pass.
Michael Kirkhart
KD8QBA
$50SAT/Eagle-2 team member (one of Stuart's "collegues")
Stuart is correct; the pass starting at 22:30 UTC (10:30 AM local time) will come within 704 km (based on gpredict with the latest TLEs) of your location. I am not sure what you are using for a radio, but the calculations we did using an Yaesu FT-60 and a Diamond RH77-CA (about 2 to 3 dBi) indicate the link margin goes to zero at about 1100 km slant range. This means you should hear it when the elevation is about 30 degrees or higher.
You may have to tune upto +/- 10 KHz on your radio to compensate for doppler as well as output frequency drift due to temperature. Between -10 and 0 degrees C the output frequency will be high by about 7 KHz.
Once you are able to hear it, we hope you might be able to provide some telemetry reports. Because of the orbit $50SAT is in and the fact it is approaching winter here in the northern hemisphere, it is still too cold to enable solar power when it makes a pass over our locations. Stations in the southern hemisphere have the best chance to receive telemetry with $50SAT at or near it maximum solar power generating capacity, which is one of the last items left on our post-launch checklist.
For all you folks in the northern hemisphere, do not let that stop you from listening to $50SAT and reporting telemetry. One of the other items on our post-launch checklist is an estimate of minimum and maximum temperature. Thus far, we have seen temperatures as high as 26 degrees C and low as -21 degrees C. Stations in the northern hemisphere can probably get us a low temperature reading, especially on an evening pass.
Michael Kirkhart
KD8QBA
$50SAT/Eagle-2 team member (one of Stuart's "collegues")
manuka
Thanks for the reminder about daylight savings - I completely forgot about it! Sometimes I wonder if we would all be better off if the entire planet used UTC.
Looking back at the link budget numbers, it seems I was a bit too generous. With just a whip antenna like the RH77CA, the link margin is much tighter; at 700 km slant range, it is only 1.0 dB. While you will probably be able to still hear it some time before/after time of closest approach, you will not likely be able to hear it down to 30 degrees elevation on just a whip antenna. If you have a handheld yagi (such as an Arrow), this should allow you to hear it upto 1900 km slant range, which corresponds to an elevation of about 10 degrees or so.
This, of course, is all theory. There is a saying that goes something like: "In theory, there is no difference between theory and practice. But in practice, there is."
Michael Kirkhart
KD8QBA
$50SAT/Eagle-2 team member (one of Stuart's "collegues")
Thanks for the reminder about daylight savings - I completely forgot about it! Sometimes I wonder if we would all be better off if the entire planet used UTC.
Looking back at the link budget numbers, it seems I was a bit too generous. With just a whip antenna like the RH77CA, the link margin is much tighter; at 700 km slant range, it is only 1.0 dB. While you will probably be able to still hear it some time before/after time of closest approach, you will not likely be able to hear it down to 30 degrees elevation on just a whip antenna. If you have a handheld yagi (such as an Arrow), this should allow you to hear it upto 1900 km slant range, which corresponds to an elevation of about 10 degrees or so.
This, of course, is all theory. There is a saying that goes something like: "In theory, there is no difference between theory and practice. But in practice, there is."
Michael Kirkhart
KD8QBA
$50SAT/Eagle-2 team member (one of Stuart's "collegues")
I heard it at 2200km (but only just) with my FT60 and Arrow yagi.
This is indeed a useful bit of info Michael and suggests that sitting 5kHz above the nominal 437.505MHz rather than 7kHz, 8kHz or slightly further out (to maximise signal strength at first contact rather than a slightly earlier intercept) using an SSB(USB) receiver may not be the best strategy after all? Often with these things the devil is in the detail. I had no idea that the temperature fluctuations produced that much of a difference. Having now been told, of course it really doesn't surprise me. Lack of local imagination at my QTH I guess. Too used to believing that these things are absolutes and side-tracked by too much thinking about Doppler rather than Doppler and temperature effects.
Just for reference, the following graph is from a test we did on the RFM22 output frequency vs temperature back in May of 2012:
It seems I misspoke yet again. The total variation in output frequency over the -25 degree C to 70 degree C temperature range is about 7 KHz. The peak frequency error is about +6 KHz at -10 degrees C. It goes to 0 at about 45 degrees C, which indicates there is some non-zero frequency offset error in the oscillator. When this test was run, the oscillator frequency error was not trimmed out like it was on $50SAT; Stuart took the time to calibrate each individual CPU board, including the flight model board, for frequency error, at least at room temperature (25 degrees C). We should have run a similar test on the flight model board, but we were running tight on time.
Michael Kirkhart
KD8QBA
It seems I misspoke yet again. The total variation in output frequency over the -25 degree C to 70 degree C temperature range is about 7 KHz. The peak frequency error is about +6 KHz at -10 degrees C. It goes to 0 at about 45 degrees C, which indicates there is some non-zero frequency offset error in the oscillator. When this test was run, the oscillator frequency error was not trimmed out like it was on $50SAT; Stuart took the time to calibrate each individual CPU board, including the flight model board, for frequency error, at least at room temperature (25 degrees C). We should have run a similar test on the flight model board, but we were running tight on time.
Michael Kirkhart
KD8QBA
Stuart
Was this measured with your Funcube Dongle Pro+, or your FT-60, or both?
I had no idea you started a new forum thread - I just stumbled on "PICAXE is in Space" from a chance Google search. Lots and lots of good discussions here. We need to continue to recruit $50SAT observers/telemetry gatherers.
Michael Kirkhart
KD8QBA
Was this measured with your Funcube Dongle Pro+, or your FT-60, or both?
I had no idea you started a new forum thread - I just stumbled on "PICAXE is in Space" from a chance Google search. Lots and lots of good discussions here. We need to continue to recruit $50SAT observers/telemetry gatherers.
Michael Kirkhart
KD8QBA
My FT60, FT817 and portable frequency counter are all within 200hz of each other.
I let the Funcube warm up for 30 mins or so and stabilize, it does drift a bit, then use the FT60 to calibrate and adjust the PPM offset of the Funcube.
Nothing (?) heard during the apparent pass over Wellington (NZ) late morning today. I was away from base so simply used a Baofeng UV5R & it's supplied antenna, a combination that is normally VERY effective indeed at UHF. (The UV5R is rated to an astoundingly sensitive ~ -130dBm at 430 MHz). However digital TV etc may have desensitised the receiver - I'm only ~10km LOS from a very powerful UHF transmitter...
This silence contrasts with the ready reception of numerous OSCAR satellite beacons I've long received with similar "Ya-Ken-Com" ham gear. The challenge lead me to rustle up a DIY 8 el. Yagi for the next decent pass.
I may also be interested in penning an informative article on the whole project. Stan. (ZL2APS)
This silence contrasts with the ready reception of numerous OSCAR satellite beacons I've long received with similar "Ya-Ken-Com" ham gear. The challenge lead me to rustle up a DIY 8 el. Yagi for the next decent pass.
This posting was noted at the Yahoo Forum, & I'm interested. Hacked TV dongles of course tempt, but my experiences show they'e pretty insensitive. What sort of approaches & skinflint budget had you in mind?
I may also be interested in penning an informative article on the whole project. Stan. (ZL2APS)
Well skinflint in that a Baofeng and a yagi made out of sticks and wire is plenty for hearing the beacon. You can then get data such as battery voltage from the Morse beacon, and solar charge data from the fast Morse.
But for skinflint data, use one of those deaf as a post DVB TV dongles ($10) and a PC to pick up the FSK RTTY. The key to this low cost receiver approach being viable is a low noise amp, add one of these and the (low signal) performance comes up to that of a proper comms reciever, such as my FT817 or Funcube dongle.
Then of course you can use the LNA in front of the Baofeng too.
Or use the LNA and stick and wire yagi with the $50SAT ground station receiver (£20 in bits) to receive the data packets from $50SAT
That last options sounds very attractive to an old skinflint like me. I recently purchased a Baofeng and have just started trying to receive signals from the $50SAT, but would like to construct a dedicated ground station for decoding data on a budget. It would therefore be really good if srnet, and/or manuka, could find the time to write an article covering this approach (details of the LNA and antenna ext). After all, there is a Christmas holiday coming up and I'm sure they will have lots of spare time.
Google "w3kh quadrifilar antenna" for details of the antenna.
You need to think a bit about how its wired, but its cheap to build and works very well.
As I said earlier a key to the skinflint approach is finding a decent (low cost) LNA that is not an oscillator, there are plenty of good commercial ones out there, but expect to pay at least £150.
You need to think a bit about how its wired, but its cheap to build and works very well.
As I said earlier a key to the skinflint approach is finding a decent (low cost) LNA that is not an oscillator, there are plenty of good commercial ones out there, but expect to pay at least £150.