Some more work on the digimode interface for the FT-857 radio: setting it up on a real circuit board, to be built into a plastic casing. This time I heard a strange ticking sound when recording audio from PI3UTR but again it seems this clicking sound is a local problem normally filtered out from the audio path of the radio. When I picked up PI3UTR on the Wouxun KG-UVD1P radio I heard the same ticking noise. New respect for the audio filtering in the FT-857, but things like this make me question the digimode interface every time I'm testing it.
Friday evening I had time to work on the 'digimode' interface for the FT-857 radio. I set up a breadboard with the audio transformers and started making cables from the transformers to the USB audio interface. Then I added the cable to the radio on the other side. All of the earlier mentioned interfaces had resistors to regulate the signal level so I started with 12K resistors between the radio and the transformer in the radio to computer audio path. This gave me a weak signal when recording the result with audacity on the computer so I switched to a 6k8 resistor which improved audio but it still wasn't great. So the resistor was replaced by a simple wire which gave me good audio when recording from the PA00NEWS transmission on the PI3UTR repeater. I did notice some low hum while recording, but I realized that was just the CTCSS tone on the repeater output. Normally this tone is filtered out when hearing the audio through the speaker of a radio.
The cable and parts for a 'digimode' interface for the new FT-857 radio are on their way. Such an interface will allow me to get audio from the radio directly into a computer and audio from the computer directly into the radio. This will allow digital radio modes such as PSK31, RTTY or AFSK. These modes allow bits (text, or databits) to be exchanged over radio. The simplest interface like FT-817 AFSK cable couples these directly but based on the advice of fellow radio amateurs I will use audio transformers to decouple the computer and radio and keep stray radio signals out of my computer and any interference from the computer out of my radio and I decided to use a cheap USB audio interface. What I will build is based on Digital VOX sound card interface but without the 'VOX' part and iPhone / Baofeng interface (schema) and El Cheapo AFSK (e.g. RTTY) USB interface (for FT-8×7) - remco.org. Update: Parts have arrived, time to build something on a development board first.
Yesterday evening I tried to make some contacts during an SO-50 amateur satellite pass and twice the CT-62 USB interface to the radio crashed, with messages like:Aug 1 20:33:15 machiavelli kernel: [48075.216289] hub 6-0:1.0: port 2 disabled by hub (EMI?), re-enabling... Aug 1 20:34:45 machiavelli kernel: [48165.984146] hub 6-0:1.0: port 2 disabled by hub (EMI?), re-enabling...I had to unplug and replug the cable every time to get /dev/ttyUSB0 available to rigctld again. A few hours later the proverbial light above my head went on: EMI means electromagnetic interference, maybe transmitting quite close to the laptop is the problem. So this evening I created the same setup and tried transmitting so I could attack the problem and see if some ferrite cores would help. The problem decided to not return, even with the CAT cable and the antenna cable laying parallel. I'm still going to use at least one ferrite core to try to keep the USB interface from crashing.
I soldered the cable for a stereo headphone on the mono phone output on the FT-857 this morning and went to listen for the SO-50 pass with tuning done by gpredict. And I missed half of the satellite pass because gpredict has the 'preprogrammed' frequency and the satellite downlink frequency seems to be drifting away from this frequency, far enough to fall out of the FM receiver passband. I kept hearing nothing so I switched back to manual frequency control with the doppler-shifted frequencies preprogrammed in the radio and I found it again, somewhat shifted. I switched back to letting gpredict control the frequency but used the tuning dial on the radio to find the right spot, after which gpredict kept track of what I did. I still find this an awesome feature in gpredict, the two-way tracking of frequencies. It was as busy as could be expected on an FM satellite on a Sunday afternoon pass with nice weather, so I could not find a 'hole' in which to call CQ or answer some call I heard. The headphones do help with hearing the audio from the radio, so a good thing I made that cable.
I was looking for options for a pre-amplifier to amplify the incoming antenna signal for receiving amateur satellites in the 70cm band (430-440 MHz). Commercially available units seem to be quite expensive like this one: DBA 270 Duo-Band-Preamp. 2m + 70cm - SSB for 298 euro. A cheap option seemed to be Low Cost 440 MHz Receiver Preamplifier Kit - Ramsey Electronic Kits for US dollar 9.95. Oh, and US dollar 57 for shipping it to the Netherlands. I'll skip. Other options sofar seem to be quite expensive or require SMD soldering. I decided I have enough problem seeing normal soldering with a magnifying glass so I'll skip SMD soldering for now. A cheap but not ideal trick is (ab)using a cable TV amplifier: those include the 70cm band because they include everything from around 88 MHz to 1000 MHz. I found this suggested at KickSat Ground Station. So I walked into the local electronics store and found a cheap cable TV amplifier with a special sticker "does not support digital interactive TV" which means it doesn't support the returnchannel. Good, just a simple amplifier. The specific impendance in cable TV networks is 75 Ohm and I want 50 Ohm so that will probably have to be fixed too. It came with a small power supply which will be replaced with a battery like in the kicksat page.
I decided to try aprsdroid on my android device. First and easiest should be to select the option AFSK via Speaker/Mic and I hoped it would work when the speaker and microphone of the portable radio are near the microphone and speaker of the android device. No luck: no transmitted APRS packet was found via aprs.fi call PD4KH and no incoming packet was decoded. I guess there was too much distortion and interference. I looked at the aprsdroid settings a lot since I noticed outgoing audio seems to be fixed at the highest volume, which can also be an issue. In the video Get Started with APRS for only $30! - youtube.com video a simple cable from the android device to the radio is used, so it can be done. Going for a solution like an Mobilinkd is more expensive, but maybe interesting in the long run when I want to do more with APRS. For now, an interim solution would be nice: building the right cable to get audio from the radio to the android device and back. I found the specs for such a cable at iPhone / Baofeng interface bouw en schema - PA4TW which can be adjusted for the android device and Wouxun. And PA4TW has used it for APRS in Aprs via porto en iPhone - PA4TW. Old analog telephone equipment can be a source of 600 ohm 1:1 transformers, so I guess some old equipment will have to donate those when I go for such a cable.
De frequentie-upconverter die ik als kit had aangeschaft en in elkaar gezet is af genoeg voor de eerste testen. Dus de metalen behuizing dichtgemaakt met een elastiekje en eens testen. Om het wat makkelijker te maken heb ik eens windows geboot en de rtl-sdr quick start guide gevolgd om snel signalen te kunnen decoderen en te zoeken naar de juiste frequenties. SDRsharp heeft een nette user-interface en je bent erg snel aan de gang. Er is vergelijkbare software voor Linux waar ik zeker naar wil kijken: Gqrx sdr. De tip van van Dijken voor het testen van de HF upconvertor was om sterke AM omroepzenders te proberen. Een uitgebogen paperclip was niet zo'n goede AM antenne, maar ongeveer een meter draad uit de junkbox werkte beter. De eerste zender die lukte was Radio Maria op 675 kHz. De offset bleek ongeveer 5 kHz te zijn, wat niet slecht is voor een rtl-sdr en/of een HF converter. Dus: beide kits werken! Even een goed moment voor mijn zelfvertrouwen in electronica in elkaar zetten: ik kan het nog en het werkt. Daarna wat andere frequenties gezocht, onder andere kwam ik de audioloop tegen op 828 kHz AM dat Radio 10 (Gold) er mee stopt op AM en overgaat op alleen FM frequenties en DAB+. Het mooiste is als straks de actieve HF antenne (mini-whip) ergens buiten gemonteerd is zodat deze beter ontvangt en de converter en rtl-sdr dan binnen (in de schuur) hangen. Als ik de beschrijvingen van rtl_tcp goed lees kan met een recente versie van deze software de hele aansturing (frequentiekeuze) ook remote over het netwerk.
Some soldering work on the HF convertor kit for the rtl-sdr stick today. I noticed I haven't mentioned the progress of this electronics building work yet.
HF active antenna kit PCB printed side.The main board of the HF active antenna kit (a mini-whip, based on the excellent work by Roelof PA0RDT has done) is finished. Soldering the antenna-plate to the amplifier has failed sofar: soldering two bigger areas of copper together with a soldering iron designed for electronics work fails. And a small inductor between antenna plate and amplifier might be a good idea too reading the design, it would improve reception above 20 MHz where I want to have a listen to the 10 meter amateur band too (28.000 MHz - 29.700 MHz). There is also an indicator led included which I will solder in place once the print is tested and ready to be mounted in a housing (PVC pipe seems to be quite popular for this). I plan to mount the led in such a way the housing doesn't have to be opened to check the status.
HF active antenna kit PCB component side.
HF upconvertor kit PCB printed side.Next soldering project is the HF upconvertor kit. I'm working on it at the moment, finding time for soldering and working very carefully. Not careful enough: the kit contains one length of wire for winding 4 coils and I found out at the fourth coil that I took a bit too much wire for the first three ones. Time to visit the electronics shop again, to get wire for winding coils. I checked the collection of parts at the radio club but there was no winding coil wire of 0.8mm. I ordered the kits from Van Dijken Electronica, a Dutch shop in electronics, quite popular with amateur radio hobbyists.
HF upconvertor kit component side.
And yes, it's more difficult than first thought to take usable pictures of printed circuit boards.
- Bouwpakketje actieve antenne 10 KHz - 20 MHz(max 30 MHz) Kit mini-whip antenna
- Bouwpakket HF converter voor de rtl sdr sticks DC - 65 MHz Kit HF convertor
I took some pictures of the PA0FBK coax antenna in full action.
Ik heb er nog geen metingen aan kunnen doen maar de zelfgebouwde PA0FBK 2m/70cm portable coax antenne doet het in ieder geval beter op 70cm dan de standaard spriet van mijn Wouxun KG-UVD1P portofoon. De antenne is duidelijk gevoelig voor van alles in de omgeving: ophangen in de buurt van actieve LCD schermen leverde ook al heel slechte resultaten op, en een eerdere test hangend aan een verfroller-stok die van metaal bleek was ook al geen succes. Het toekomstig gebruik is dus duidelijk aan een lijntje de boom in of op een andere manier vrij ophangen ver van allerlei storingsbronnen. En nog iets van tijdens het bouwen: voldoende licht op de werkplek is nuttig. Ik had een keer 's avonds aan het op lengte maken en strippen van coax gewerkt en toen ik het resultaat goed bij daglicht zag wist ik dat ik al wat dingen over ging doen.
Misschien een interresant project om een 2m/70cm antenne te bouwen die makkelijk uit het raam kan hangen: PA0FBK's 2/70 portabele coax antenne. Ik kwam onder andere een foto met positieve beschrijving tegen van PD1JMB. 2m and 70cm Dualband Antenna according to PA0FBK - DL8KDL (engelstalig) geeft nog meer informatie en heeft zelfs serieus er aan gerekend. Ik ben meestal niet zo goed in mechanische dingen, maar dit ziet er uitvoerbaar uit, zeker als ik wat versteviging aan de connector maak. Ik ga ook zeker een ophangtouw er aan maken.
Unlicensed mobile radio (walkie-talkie like use) in Europe is possible with PMR446 devices which operate on frequencies between 446.0 MHz and 446.1 MHz. PMR446 was introduced in 1998. The original PMR446 standard gives 8 channels where FM analog audio is used. Using CTCSS or DCS codes extra options to distinguish users of PMR446 are available. But these are more 'calling certain users' options than real privacy, a radio scanner tuned to the PMR446 frequencies can receive all users. Analog PMR446 radios are quite cheap nowadays: when I browse conrad I can find a simple set of 2 PMR 446 radios for Eur 19.99 Audioline portofoon PMR-15 (Conrad Electronic). A begin of privacy comes at a price: the lowest price for a (single) PMR446 radio with 'scrambler' and a description which suggests extra software is needed to program that scrambler is Eur 135 Midland PMR portofoon set G14 (Conrad Electronic) or Stabo PMR portofoon Freetalk com (Conrad Electronic) or Eur 205 for the PMR-zendontvanger Kenwood TK-3301E (Conrad Electronic) which is advertised as 'professional' which is visible in the price. Diving into a manual shows that the scrambler has an 'on' or 'off' setting, which makes the given privacy limited to "can't be heard on a standard scanner" but a determined listener can determine the scrambling system and listen in. A newer development is Digital private mobile radio 446 (DPMR 446) which uses frequencies between 446.1 MHz and 446.2 MHz, not overlapping with analog PMR446. There are 16 channels with FSK (frequency shift keying) in a 6.25 kHz wide channel (half that of analog PMR). I can find exactly one device supporting digital PMR446 on sale: the ICOM IC-F4029SDR professional digital license free transciever and the price I see is 180 UK pound which is also a 'professional' price. The 'digital' part of DPMR could allow for real encryption, but for as far as I can find it is not implemented, and searching for details about this I find: Draft amended PMR446 ERC_DEC(98)25 which states:a)that it is not recommended that applications requiring encrypted speech should be used with PMR 446 radio equipment;It would seem (to me) that adding strong encryption on digital PMR446 wouldn't be too hard, but the standard doesn't have space for it (at the moment). There are options for group codes (just like ctcss or dcs codes in analog PMR446) but it is possible for a (specialized) scanner to ignore all those. ETSI TS 102 490 technical specification: Peer-to-Peer Digital Private Mobile Radio using FDMA with a channel spacing of 6,25 kHz with e.r.p. of up to 500 mW.
In the weekend I had some more time to play with rtl-sdr, the cheap software defined radio using rtl2832u based receivers. I tested with coupling rtl-sdr and gnuradio companion using a fifo, so I can do experiments 'realtime' and not with (huge) capturefiles. This works, which enabled me to listen to FM radio. I'm still looking what the minimum and maximum frequencies of the Fitipower FC0013 tuner are, but my guess is that the range might also be near 64 to 1700 MHz. The good news is also that there are the first signs of a gnuradio block for this receiver on the site above, which means I can use it directly in gnuradio and retune it on the fly.
The Noxon DAB USB stick I got as a cheap DAB/DAB+ experimenting device has hidden powers. A big part of the 'cheap' is because this device leaves all of the work to software. If I understand the discussions about the Noxon correctly, there are drivers to receive dvb-t signals with the same stick. Most development for the linux driver seems to be on the dvb-t side too. I came across a webpage which puts this all in a new light: The whole device turns out to be a wideband tuner and an analog to digital convertor. Software has to do the rest. And now someone has written that software. Some clever thinking and discussion started by Antti Palosaari turned into rtl-sdr, a software defined radio using the RTL2832U chip. Together with my other thoughts about software defined radio where I thought about receiving and decoding D-STAR ham radio signals with for example a funcube dongle this means I already have a software defined radio. Time to capture some I/Q data and learn about gnuradio. The rtl-sdr software has already been used to receive Tetra signals (frequencies UHF 380 MHz .. 476 MHz) and GMR signals (satellite phone, L-band 1600 MHz). So I guess 70cm D-STAR should not be a problem. Update 2012-03-21:Played with this, and I managed to receive an FM radio station and decode the stereo audio from it using the grc (gnuradio companion) rtl2832 to fm scripts mentioned here. Hearing a bit of radio received and decoded by hardware which absolutely wasn't sold to me with that capability is fun. Everything gnuradio is all new to me so most of the time I have no idea what I'm looking at and the noise in the Noxon stick is causing issues as usual. Update 2012-03-22: Played some more, and added the plot. And I heard audio from some of the stronger stations. But I still have serious noise issues, even when moving the usb stick away from the computer with a long usb extension cable. As you can see in the plot there is not much difference between the noise level and a strong local station. The wideband FM receiver module available in gnuradio companion couldn't even decode the signal at 93.8 MHz (Slam!FM Noordbrabant) in the captured data.
Simple FFT (spectrum analysis) of data coming from the NOXON dab usb stick. Works in the VHF FM broadcast band, I can see various signals and verify their frequencies. With stronger stations I can use the wideband FM decoder in gnuradio and get audio. Click for full screen capture from gnuradio FFT.
This evening I hooked up the radio scanner to the UHF TV antenna to see if using that might get me one of those '35 cm pirates' which should be active in the 856 - 862 MHz and 865 - 868 MHz ranges which is UHF TV channel 69/70. So the preamp should work.It took a bit of searching through the junkbox to find parts for a cable to get from the output of the UHF TV antenna (through the power inserter) which is a Belling-Lee connector (as shown in the picture) to the input of the scanner which is a BNC connector. My guess is these transmitters would be active on a Sunday evening. But, the only thing I heard in wide FM mode was a wireless headset.
IEC-169-2 connector, picture by Colin from Wikimedia commons (CC-BY-SA license)
Fun with rechargeable batteries: two GP NiMh 2300 mAh batteries. One spent months in my MP3 player, sometimes playing stuff. The other was sitting in a drawer. When the battery in my MP3 player stopped I wanted to change to the other one, but that one was empty enough too that the MP3 player would not boot. So self-discharge takes more out of the batteries than my MP3 listening I guess.
Ik heb van het weekend eens voor de aardigheid een proefexamen radio amateur novice gedaan. Zonder enige studie vooraf een score van 27 van de 40, om het te halen is een score van 29 nodig. Blijkbaar is er genoeg van mijn MTS electronica blijven hangen. Op specifieke radio techniek en de regels moet ik nog wel het nodige studeren als ik een licentie zou willen halen.
Seen today: #twatch Open Hardware Networked LCD Screen - Slashdot Hardware. Quite tempting! Affordable (USD 45) and it can do cool stuff. By default it displays real-time topic trends from Twitter on an LCD. But with a bit of playing with LCDproc it can also show system statistics, RSS feeds, mail notifications, and more such as ... METARs.
Hope someone of the LCDproc development team picks this up soon: Although the page assumes there is a way to redirect a serial port to a network stream in linux this is a bit harder.
Finally I counted a few light pulses with the one-wire counter. No big circuit with the Velleman MK120R kit but just a photodiode and a resistor hooked up to the counter module. I got the idea from looking at the schematics for the Hobby Boards 1-Wire Lightning Detector. The counting circuit is a phototransistor (in an optocoupler) and a resistor. With a bit of tweaking on the resistor I eventually got the counting circuit to count 2 light flashes from a flashlight. With some more tweaking of the resistor value I think I could count red flashes from the electricity meter.
I also installed the lightning detector under the roof. But it is too close to the wiring of the house I think: sofar all counts are related to me being in the attic and turning on the light. In a 'real' setup I think it needs to be away from the house. Something to keep in mind for the project sundial weather station.