Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
longview
Dec 25, 2006

heh.
The cheapest way to get a scanner up and running is to get a USB TV tuner dongle on eBay and install some custom drivers with a program called Zadig. Basically the TV tuners for DVB have a programmable mixer and a reasonably fast ADC chip that can digitize and transfer about 2 MHz of spectrum anywhere between 25-1700 MHz or so, once it's in software doing things like decoding AM, FM, SSB and digital modes is actually pretty simple. More expensive cards connect via PCI-e and can do huge frequency spans at a time and even come with transmitters that are also software defined so any modulation form can be implemented.

Here's a screenshot I just took using a £10 dongle connected to a discone scanner antenna:


It shows the spectrum at the top and a waterfall (spectrum vs. time and amplitude as colour), the signals there are the local police/medical TETRA channels which are all encrypted, in the waterfall you can see the signal sometimes stops transmitting for a fraction of a second, that's because of the time division multiplexing, the client radios transmit in those time slots, so there can be multiple radios using the same channel without interference.

Here's a test I just did on 70cm transmitting some DTMF tones:


With a plugin it can be set up as a very good scanner, since it can immediately tune any channel within the 2 MHz, the entire european 2m band can be scanned at the same time and air-band with a just a few re-tunings.
SDR# only does one decoder at a time, but there's nothing stopping the software from decoding every valid signal in the 2 MHz band and playing it at the same time either, except that it would be impossible to listen to.

Another use I've found is adjusting deviation on microphones, and in fault finding since it's possible to pick up the local oscillator from a running radio directly and check it for stability, work out what frequency is tuned if the IF frequency is known and so on.

Fun fact: the local oscillator leakage can be severe enough in cheap radios that you can pick it up pretty far away, this is how the mythical TV-license van could detect a running TV by pointing a directional antenna at the aerial.

E: here's a SDR# detecting my old base station LO

In the waterfall you can see me changing frequencies by fiddling with the knobs, the 127 MHz signal is unrelated.

longview fucked around with this message at 17:16 on Feb 22, 2014

Adbot
ADBOT LOVES YOU

longview
Dec 25, 2006

heh.

spankmeister posted:

this is how radar detector detectors work also

yup, pretty much every radio will give off a detectable amount of LO leakage, I have seen pictures of old german receivers that used a ridiculous amount of tuned filters and avoided using a local oscillator altogether though, it was designed specifically to be undetectable to normal radio detectors

I can't find it on google unfortunately

longview
Dec 25, 2006

heh.
Jonny: I'm just gonna post about whatever, hope I don't mess up your plans

So what's the point of a local oscillator if the government can snoop on me with it?

Well, there's no need for a local oscillator if you want to listen to AM radio and only want the one channel, AM can be detected with a pencil and a razor-blade if you're a soldier in WW I, see the foxhole radio.

To detect AM you need a rectifier, it's based on detecting the change of amplitude in the AM radio signal, if there's more than one station on air you also need a set of filters to pass only one frequency to the rectifier diode/shaving kit, the output contains among other things the audio signal that can be connected to a set of headphones.

Unfortunately it's kind of a pain to re-tune this thing, since you'd have to wind a new coil and tune trimmer capacitors to make any major change in frequency. It's also going to pick up strong stations nearby no matter what since the selectivity (i.e. the ability to reject signals other than the one you want) is poor. Sensitivity can be improved by adding an amplifier to the RF side and to the audio-side.

It's possible to build increasingly complicated filter arrangements to select exact frequencies, and at HF frequencies it would even be possible to use crystal filters to get excellent selectivity at the exact frequency you designed for +- a tiny amount. You can also add an amplifier to boost the signal going in to the diode to make it more sensitive.

But we can do better, enter the superheterodyne receiver shamelessly borrowed from wikipedia:


Let's look at the normal kind of super heterodyne, this is the same for pretty much all receivers, for AM or FM the Demodulator contents change.
The antenna is on the left, this is followed by an amplifier and usually a coarse filter to prevent very strong out of band signals from messing things up for us. Next is a mixer, this is important: it is a multiplier of signals, so the output on the right is the Local Oscillator multiplied by the RF Amplifier output.
What this means in practice is we get the sum and difference frequencies out.
Using numbers: assume we want to receive Anime FM at 100.0 MHz, we'll decide we want to use 10.0 MHz as our reference frequency to keep numbers simple. 10 MHz is then the frequency output from the mixer, to make it output 10 MHz for a 100 MHz input we choose either 90 or 110 MHz for the Local Oscillator frequency.

If we want to listen to GBS.FM on our pirate FM transmitter at 96.0 MHz we can now choose a different LO frequency, instead of 90 we can use 86 MHz, and we still have the station we want at 10 MHz.
That's good news because we can design our IF Amplifier and Demodulator to only work at a single frequency, saving us a ton of headache! These things tend to be very sensitive to what frequency they're operated at, and indeed they generally perform better at a single fixed frequency rather than having to be tuned around the band.

Of course there's a problem here, like I said the mixer gives sum and difference, it's basically symmetrical, so we run into two problems:
If jocks beat us up and stole our FM transmitter, then they would know we'd tune to Anime.FM while posting the sickest burns we could think of on SA, if they set up their transmitter to run at 80 MHz, then we'd be in big trouble. Since it's symmetrical the 80 MHz signal would be subtracted from 90 MHz and would interfere with our 100 MHz signal! Fortunately that's what those filters in the RF Amplifier are supposed to prevent, but this is a real concern and frequently a problem with low end wide-band receivers. 80 MHz would be a mirror frequency for this radio. The IF frequency must be selected so that this mirror frequency is always outside the normal frequency range of the radio, otherwise it would never work right.

Another problem, is if they set up a strong signal at say 101 MHz, this signal being much stronger than the Anime.FM transmitter (MOM won't let the sysop put a GP antenna on the roof) it could overwhelm the IF amplifier and cause both interference (when an amplifier is overloaded it can also act as a mixer) or desensing, where the sensitivity to signals is reduced due to strong signals nearby. Receivers also typically use automatic gain control to prevent the overloading, this can in practice cause desensing, never trust a radio with no analog RF gain knob.

An FM signal covers less than 200 kHz of spectrum, so what we can do if we're really smart is to go buy a filter and put it between the mixer and the IF amplifier, normally we use crystals or other mechanical filters to provide this filtering because they can reach extremely high Q factors (Q for quality) meaning they would pass 9.1 MHz and 10.1 MHz but attenuate anything above and below that by a ridiculous amount. One measure of a good radio is having very good quality filters.

Often there will be a second stage of mixing and filtering to go down to an even lower frequency, all FM receivers I've seen have a second IF frequency of 455 kHz, but some AM and SSB receivers do the demodulation at the first IF frequency.

The demodulator section is its own effortpost, but for an AM receiver it can still be a diode, but it might also be a second mixer. Since an AM signal is generated with a mixer in the first place (it just is), we could generate a 10 MHz signal, mix it with our (now AM) signal and the output would actually be audio frequency. If we filter some more, we can decode SSB with this method.
The method for decoding AM like this is usually called a synchronous detector, and it comes with a circuit to align the 10 MHz frequency automatically (otherwise frequency drift would change the pitch of the audio like in SSB).

Now I said that the blocks are the same for FM and AM, there's one difference: in FM we don't need to be linear, so we can actually push the gain as far up as we want provided we've filtered out unwanted signals, the frequency changing is what encodes the audio signal. For AM, SSB and as mentioned, PSK modes, we need a linear amplifier chain and automatic gain control becomes more important to be able to handle both strong and weak signals.

To demodulate FM we need something more complicated, which always involves a phase locked loop frequency synthesizer, this is also normally how the Local Oscillator is generated based on a crystal time-base in all modern radios, they're great fun to work on, but I might get back to that.

Ps: those willing to think about this might also wonder: if we can decode AM synchronously by mixing our 10 MHz signal with the 10 MHz AM signal, why do we need the heterodyning stuff before it?
The reason is that the local oscillator will overpower our received signal through all sorts of leakage, making the sensitivity very poor, and we lose the benefit of the extra filtering, because of those problems the direct conversion superhet was very short lived.

Pps: I want to watch fox news without the obamacare drones picking it up, what can be done to shield LO leakage?

I'm glad you asked, hypothetical poster: I happen to be working on a commercial grade analog mobile telephone network receiver that tried very hard to prevent mirror frequencies from leaking and also prevents LO leakage. The solution is to shield the entire chassis of the radio and ground it, and also to install tuned filters on the antenna input connector, the NMT 450 (Ericsson designed) receiver I have uses what's called helical filters that are tuned to exactly the frequency it's supposed to receive, this means I can barely detect the LO with the covers closed, but when they're open it's fair game.

longview
Dec 25, 2006

heh.
as an addendum to the above: I chose a LO frequency of 10 MHz, 10.7 is the standard FM radio IF frequency, they standardize primarily to simplify design and make filters cheaply. 21.4 MHz is another old timey standard used in commercial gear. The base station I used earlier has 16.9 which was common for ham radio gear in the 70s/80s.
Conveniently there can be leakage at transmitting frequency + IF frequency from the first mixer, 16.9 MHz + 145 MHz works out to the area local PD used for their analog repeater inputs, I know someone who had a model similar to my base station who had some filter issues and possibly a crystal problem while mobile, and was then pulled over by the police demanding to know why he was talking on their repeater.

i also chose to put the LO below the receiving signal, this is common at higher frequencies for practical reasons, but for HF bands it might make more sense to put the LO frequency far above the HF band because it simplifies filtering, according to wikipedia this is often done with three conversions (presumably 1-30 MHz -> something above 30 MHz -> 16.9 or 10.7 -> 455 kHz). analog mobile telephones also did this, using a 70ish MHz IF to get better mirror rejection.
models with wide-band RX also have to do this, my Icom HT has a first IF of about 64 MHz.

anyway, here's my VHF rig:

this is the stately 10W VHF AM FM SSB and CW FDK-Multi 2700, a 10W PA stage gives it real puch

Here's the schematic: linky

of note is an excellent microphone with adjustable compressor, 230V/12V dual power input and with a mod, automatic failover to battery, the very special OSCAR module which activates a transverter in the receiver that allows you to receive 10 meter signals and transmit on 2 meters for operating a now defunct series of OSCAR satellites (AFAIK there are no 2m/10m OSCARs still operating), being built like a truck and the fact that it has both a PLL and a VCO for tuning, presumably the VCO was intended for use with the FM center meter when operating OSCAR

reliability and quality was key for FDK rigs like this, it was a state of the art transceiver in 1978, all the primary signal paths are either standard bipolar transistors or a very standard dual gate MOSFET for mixers ensuring this thing will never stop working as long as I'm willing to repair it, the PA stage is built for 100% duty cycle as a base station, the RF transistor in the output stage is rated at 18W with basically a shorted antenna, there's an optional fan slot, and it has the best squelch circuit I've ever heard, it pretty much never has a squelch tail

and because I'm a total sperg I use a H-250 military surplus handset connected to the accessory port on the back

longview
Dec 25, 2006

heh.

Sharktopus posted:

I have a tv tuner *TRIGGER WARNING*dongle*TRIGGER WARNING* and want to start recording some of hte local airwaves. My tuner has a big band gap though and also only goes up to 1ghz or something. Is there any cheap hardware that will get into the 2.4ghz range?

The master plan here is to try and accumulate a data set of radio broadcasts and see if I can do some speech to text

what do you want to pick up at 2.4 GHz? there are no analog broadcasts to pick up above the 23cm band (and that's not exactly crowded) that I know of, if you want people talking I'd start at 144 MHz and 430 MHz and tune up the repeaters there

longview
Dec 25, 2006

heh.
I suppose there might be cordless phones around that use 2.4, over here all that stuff is at 800 MHz and will soon have to fight with 4G LTE transmitters.

SDR# is supposed to be fairly easy to write plugins for, if you want to receive multiple frequencies you might want to set it up as a scanner or just get some more dongles for $10 a piece and a T piece to connect the same antenna like I did (it's not perfect but it works just fine)

the connector on those TV dongles is usually a MCX connector, an adapter to SMA would let you connect a more standard antenna

you probably also want to adjust the gain in the settings to the point where the noise floor starts rising, as well as making sure the ppm adjustment (frequency correction) is somewhat reasonable, in the picture I posted of the UHF transmission it was pretty far off center

one of the dongles I have can actually be so far off it can't receive a narrow signal without adjustment

longview
Dec 25, 2006

heh.
LOVE that vintage of Icom radios, especially the ones with my favourite display type: the VFD, when I have some spare cash one of my mates wants to sell me an old Icom that looks like that but 2m/70cm, forgot the number though

the only other radios I have are HTs (Icom IC-T70 and ID-51, both great at what they do, and a few chinese brands for fun) and my D-Star hotspot running on an old Icom IC-25, I have an old FT-747Gx in storage but the PA's busted and I don't have room for any antenna here anyway

strictly speaking I'm not allowed to put antennas up at all but nobody's complained about a mobile whip and a discretely installed Royal Discone on a tripod, and I hid a slim jim for the D-Star in a corner

longview
Dec 25, 2006

heh.
that's a cool trick, I guess that's why some scanners have IF outputs, would save a little time

my proudest radio hacker moment was sitting in a dark hotel room on a business trip at 11 pm and working out that my Tecsun PL-660 receiver was misaligned by 1 kHz from the factory and using a leatherman charge and SignalScope on my iphone to align the BFO with some shady russian pictures as a reference, then aligning it by tuning +-1 kHz from RWM at 4995 kHz and checking the AF frequency on the audio analyzer app

checked it after, perfectly aligned :c00lbert:

longview
Dec 25, 2006

heh.

atomicthumbs posted:


here's some ~art photos~ I took in the receive site building, which is open on saturdays (I think) and maybe other days too


loving it, when i'm old I'm gonna move to the country and just be a cranky old hermit with tons of antennas and stuff, surrounded by teletypes

spankmeister posted:

this overly enthusiastic australian dude visited an analog TV transmitter station recently, p interesting stuff:

https://www.youtube.com/watch?v=mR_wJkxKSXU

EEVBlog is awesome, I prefer Mike's Electric Stuff videos when he posts them because they're always straight to the point, like today he posted a video about filling a tub with water and enameled wire to make a 800W dummy load
https://www.youtube.com/watch?v=WECW88rJYrE

anyway today I cursed a lot when I did some work on my D-Star hotspot setup, replacing the wire-in-plug cable between the controller and the radio with an actual cable
it was awful because the DV-RPTR guys hosed up the pinout on the PS/2 style connector (my least favourite connector) so I just had to try, measure, try again etc for about 30 minutes and the only solder I had around was the devils own lead free non eutectic so all the connections look like poo poo
this is the second time I have had to deal with this exact same problem, the first time it took us about 2 hours to work out that the pinout was useless, then another hour to get it to work

also improved the range by about 2x by installing a 455 kHz notch filter on the IF output tap I use to get the straight FM discriminator output from this old rear end radio

over all the DV-RPTR is a pretty neat bit of kit, it's basically just a GMSK packet radio modem, so it doesn't need to decode the actual DV voice signal, instead it just looks at the stream-headers to read out commands and to/from fields, and sends the AMBE signal over the internet (ethernet port, uses pro as gently caress telnet to connect to the gateway controller) to other repeaters
the expensive version includes the AMBE codec chip and a user interface so it can be used with a 9600 compatible radio to make a D-Star radio instead of just acting as a repeater controller and node adapter

longview
Dec 25, 2006

heh.

spankmeister posted:

yeah eevblog owns my only problem is that the guy Dave likes to ramble a bit, stretching a 10 minute video into 40 minutes.

mikeselectricstuff owns too my only gripe with that one is that he needs to post more

:justpost:

i saw his vid on the dummy load earlier, the idea is brilliant

i sort of gave up on dave's lab videos when he posted a rant about his ~ultra precision~ current source didn't work like in the datasheet when he built it on a loving solderless breadboard

double sided copper clad board and a dremel or bust, :cmon:

longview
Dec 25, 2006

heh.
Nobody will talk to someone without a call sign anyway, it's the first thing you get asked for so you wouldn't get very far

longview
Dec 25, 2006

heh.

rotor posted:

can i just make one up?

Probably, there's always going to be somebody checking the lists to see if it's legit, this can be fun if your call sign isn't on the lists yet.

Calling B0OB

longview
Dec 25, 2006

heh.

Werthog 95 posted:

anybody know any good books on the history of this poo poo? after reading that big ol' PSK post now i wanna see how this poo poo evolved into cell phones

cell phones (GSM) use a modulation form that's pretty similar to PSK, it's called GMSK and it's not a linear modulation so it's closer to FM

it's basically frequency shift keying with a specially chosen ratio of 1 and 0 frequencies that changes frequency slowly instead of rapidly (fast frequency changes = wide bandwidth), the demodulator is more complicated than straight FSK but I don't think it's much more complicated than the PSK demodulator except it's harder to build because the carrier frequency is higher

in addition to that, there's a channel equalizer system to prevent reflections from interfering too much (each packet contains a certain set of bits that can be analyzed in real time to correct for reflections), a very efficient vocoder, encryption, time division and frequency division multiplexing (mobiles communicate in time slots that are assigned by the base station and also change frequency several times a second)
there's also the roaming system where cells listen to nearby cells channels to handle the handover

D-Star radios use the same kind of GMSK modulation at similar bitrates, but in D-Star they simplified the modem design by making an audio-frequency GMSK signal and then wrapping that in a normal FM carrier, I think the main attraction is to make it cheaper and easier to retrofit older designs with a DV mode
D-Star also doesn't have any of the channel equalization (it cant because it can't see the actual FM signal on the air), nor does it have time division to handle multiple users per channel

E: to put it in ham radio terms, GSM is basically transmitted as an SSB signal, whereas in D-Star they use the same modulation but put it inside an FM carrier and rely on the FM capture effect to reject interference and multipath

longview fucked around with this message at 19:21 on Feb 24, 2014

longview
Dec 25, 2006

heh.
I'll try to write up something about D-Star later tonight, including the ways of making hotspots and repeaters, it's not necessarily that expensive to set up a hotspot with decent power

DMR radio is also starting to become more popular but it kind of sucks because it's only the repeater admin that can decide where to link repeaters, whereas in dstar all valid users can link repeaters to reflectors and other repeaters, the actual air-interface and (I think) voice codec are better though

longview
Dec 25, 2006

heh.
I know you're not allowed to use encrypted modes, like DMR or TETRA with encryption, some people think any kind of encoding is also illegal but that's not the case.
If you're on an ARES assignment and need to transfer patient data or something like that then I suspect encrypting that data and transferring it using D-RATS would probably not cause any problems.

Another grey area is using DD mode to transfer aprs.fi for example, or running an amateur radio wifi link.

longview
Dec 25, 2006

heh.

PuTTY riot posted:

thats kinda where i was going w/ this....


also if u use a repeater to make a voip call and the person u call starts speaking obscenities

almost a 100% chance that running a 1500W 2.4GHz link and taking out wifi for the surrounding county will get you into some kind of trouble, european limit is 300W on VHF and up but anything above ~1W will take out wifi over a pretty large area

another fun use of ham radio is keying up a 70cm HT and taking out OTA TV for everyone nearby using indoor antennas

doing this may be legal but it's a good way to make enemies

there was a case here in norway last year where a guy got the authorities to confiscate his neighbors plasma TV because it interfered with his HF rig, this wasn't exactly good for PR but a lot of hams I talked to said it was a good thing
going around acting like we own the radio waves and causing problems for normal people is a good way to lose frequencies, a large continuous span like the 70cm band is worth a lot of money on the free market

longview
Dec 25, 2006

heh.

Verdafolio posted:

i'm too stupid to understand how this works. is his signal so powerful that whatever wires used in his earbuds act as antennas?

in a word, yes

SSB and AM transmissions transmit information by varying signal strength in tune with the signal, semiconductors and other metal-metal junctions can act as the rectifier diode in the AM receiver I talked about a while ago, the effect would be that you get a receiver out of audio circuits, digital circuits can be affected but usually not enough to cause noticeable interference since they're usually shielded against external RF and to prevent the signal wires from generating interference as well

this was a huge problem with all 90s consumer audio equipment, and many cheap amplifiers today, when GSM phones are placed nearby the time division nature of the signal means it turns on and off the transmitter at a frequency you can hear when it's rectified in an inadvertent receiver

manufacturers eventually sorted out the GSM signals by shielding and using opamps with built-in RFI shielding (opamps with EMI shielding are like magic compared to ones without), the reason they did it was of course because people were noticing the problem

the only people who notice CB and Ham activity are people right near powerful transmitters, AM radio transmitters have the same problem but they can usually put the transmitter in a mountain somewhere and do a remote feed

FM signals have a continuous signal strength so when this rectification happens it'll just cause some DC to appear, often a bit of 100/120Hz hum from the power supplies on top of it but it's not as big a problem

TV transmitters might also cause problems, analog TV used a sort of SSB called VSB that varies with picture brightness and DVB-T transmitters use OFDM, a linear digital modulation form that can also cause this problem

to prevent this problem, you can direct the transmitter to direct power elsewhere or otherwise lower power, or try to work out what wires are causing problems and put clip on RF chokes on the wires, coiling up the wire for a few turns can help against higher frequency noise too
but usually if you're getting interference in there's either a very strong signal nearby exceeding the civilian EMC test spec(10V/m I think is the toughest civilian spec, mil-spec is about 200V/m of RF at different frequencies + often an EMP test) or the equipment designer or installer made a mistake

longview fucked around with this message at 21:06 on Feb 24, 2014

longview
Dec 25, 2006

heh.
~~D-Star~~

What is D-Star?
D-Star is a amateur radio standard for digital radio, it supports two modes, Digital Voice and Digital Data, DV is the only mode in any real use so I'll focus on that

D-Star was developed by the Japanese Amateur Radio League with government funding from the late 90s and standardized in the early 2000s, it is to my knowledge the only wide spread digital voice mode designed specifically for amateur radio requirements. it is an open sores protocol that is fairly well documented, it does use a proprietary voice codec that is the root of a lot of criticism
Additionally the only manufacturer to deliver working out of the box radios is Icom, they were part of the design process and worked out of lot of the bugs in the original system.

DD mode doesn't have a voice component (I think) and allows a 128 kbit data link, it's only supported in the 23cm band and only one radio, the Icom ID-1 supports it. this radio has an ethernet port and it's possible to use it to run normal TCP/IP traffic over this channel

At a technical level DV mode uses a 6.25 kHz channel as opposed to a 25 kHz FM Voice channel, and the more modern and lower quality 12.5 kHz narrow FM voice channel that is starting to be mandated
at the core it has a 4800 baud digital channel, of this a proprietary AMBE2020 codec is used to make a 2.4 kbit/s voice bitstream, 1.2 kbit/s is used for forward error correction and the other 1200 bauds are mostly left up to the implementer
in practice the 1200 baud channel supports digital data transfer at a glacial pace (D-RATS) and is also used to transfer a text message included with the callsign, and Icom supports D-PRS for position reporting over this channel too

the air interface is a GMSK signal encapsulated in an FM carrier, there is no time division or other multiplexing technique so only one person can talk per channel, this is an advantage because it means a lot of radios can technically support it as long as they have a 9600 baud data port or can be modified to have one, most radios with pure FM transmitters (not phase modulated) can be modified for D-Star support

D-Star originally only operated in the 23cm, 70cm and 2m bands, but the IC-7100 supports running D-Star on all HF band plus 6m, I think most use is restricted to 10m and higher since the bandwidth needed is big compared to SSB and it might not be the best protocol for HF propagation IMO

What can I do with it?
D-Star DV mode supports both slow data transfer (position reporting for example) and voice, the neat thing about D-Star is it's digital so it can be stuffed into the internet pretty easily
this allows multiple repeaters to be linked up to the same chat room, typically called reflectors or rooms, when anyone at any linked repeater starts talking, all the other repeaters transmit this signal
the most popular set of rooms right now is the DCS system, which has rooms for every country with more than a few users, each country usually has a set of rooms for regions, testing, emcomm and so on
DCS013 (Norway) and probably others now have Echolink (a similar system for linking FM repeaters) gateway rooms too, I tested this yesterday and it worked pretty well
another neat feature is basically SELCAL, this system allows you to automatically connect your repeater to another repeater using the callsign of the person you want to reach (technically the last repeater this person was registered at, this happens whenever they key the transmitter)
unfortunately it doesn't work that well in the original implementation because call-sign routing (that's what it's called) automatically disconnects anyone else using the target repeater and it requires that the answering party hit a button to reverse-route otherwise it won't be a two way link

fortunately the germans who built the DCS system also built what's called the CCS system, this implements call sign routing by handling the linking in a different way where the target repeater is still connected to the room it was in, AND the calling repeater, allowing the caller to listen to active QSOs before making a call

so basically you can chat all over the world or locally, using the internet


Some myths
D-Star is old - it is, but it's also the only standard with more than a handful of users, but then again most standards are old by the time they're in widespread use

D-Star is Icom only - it isn't, the AMBE2020 codec is proprietary but is also widely available, Icom is the only manufacturer to provide a set of radios and complete repeater stacks to run D-Star but several projects offer Icom-free D-Star repeaters and radios if you're willing to DIY a bit
Yaesu and Kenwood are playing a political game, all these manufacturers were able to make interoperable NEXEDGE/IDAS radios using very similar tech for example

I can't decode D-Star on my Baofeng/IC-02AT, it's clearly encrypted and illegal
Nope, the codec is available at a low cost and volume to anyone willing to buy the AMBE2020 chip, the rest of the standard is open source

Other modes are more efficient - not really, Yaesu being Yaesu has a competing standard but I've literally never heard of anyone using it for anything except testing it once, this mode supports 9600 kbit but it uses twice the bandwidth so not really a big win, it does support dedicated data links at 9600 kbit so it's much faster than DV at data transfer
DMR and MOTOTRBO is starting to gain traction, oddballs in the UK are starting to use TETRA radios
these standards have some very nice features, like DMR using a 12.5 kHz channel with TDMA to run two voice channels at the same time, typically channel 1 is international and channel 2 is local chat for example. the disadvantage is the radios for this mode are not designed for amateur radio use, they often aren't keypad programmable and the programming software is usually expensive as hell if you can get it (vertex standard/yaesu being the worst at this)
additionally repeaters are linked to whatever the sysop set it to whereas D-Star lets all valid users link the repeater to where they want it

as an aside, if DMR was designed for amateur radio it would use the TDMA system on the repeater side to eliminate the need for duplex filters and splits, but they didn't because pro users don't care about buying a duplex filters, TETRA apparently supports using mobile radios as repeaters for HTs though

D-Star has bad voice quality - well, ok maybe, it's good for being a 2400 baud link ok? people sometimes sound like they have a cold, and when signals are choppy the vocoder output turns into beep and bloops, but over all it's not bad

Internet linking isn't real DXing, when the internet goes down the whole system goes down!
I won't comment on DXing or not, depends on the definition you use. When the internet goes down so does systems like Echolink, but then it's not like the FM repeater will reach any farther than the D-Star one anyway
D-Star makes it easy to link up repeaters and hotspots across bands and distances when the internet does work, which is a big advantage for many emcomm operations.
using hotspot nodes it is also possible to run cross-band repeaters on HF bands (think 40 meters NVIS) without generation loss like you'd get on a long chain of analog cross bands

So what's a hotspot, how do I make a repeater?

To make a repeater you buy the ID-RP2C repeater controller, then you choose what bands you want, most users want a 70cm module, the ID-RP4000V, and many want the ID-RP2000V 2m module and at only the price of a cheap car it's a real bargain, don't forget duplex filters!

If Icom won't give you the repeater stack for free then you're not going to buy it, fortunately the germans really love D-Star and have made a repeater controller, the DV-RPTR. This is a GMSK modem, a powerful DSP and some interface hardware for keying radios, USB and Ethernet.
The DV-RPTRv2 is usually used as a node adapter, or a modem. To make a working hotspot (simplex internet gateway), you need a radio with a 9600 baud packet radio port and you have to make your own interface cable, after that's done it's plug and play.
If you're like most people you'll then get a raspberry pi running some obscure linux distro and some software called irddbgateway and dvrpt-controller, if you're a shaggarian übermensch like me you get a windows server 2012r2 vm running in hyper-v and run it there instead.

To make a repeater you need two radios or a suitable repeater and duplex filters, but the same DV-RPTR can be used provided it has internet.
When we set up our local 70cm repeater we used two FT-7900s and then a second DV-RPTR configured as a hotspot to remotely feed the repeater the same way echolink is usually connected.

The DV-RPTR can also be bought with an AMBE codec and a display, the same setup then works as a normal D-Star radio only more clunky and about as expensive, but it does support directly connecting to the internet at the same time.

So to end this, here is what D-Star isn't:
Professional grade modern digital radio - it isn't, pro users have other requirements and would find D-Star massively confusing to operate, TETRA, DMR and P25 are far better options for pro users but are too limited for amateur radio
TDMA FDMA and CDMA packet oriented (except DD) high bandwidth true diversity roaming - it's none of those things except one radio supports diversity receive, it's an improved analog FM, not 4G LTE
Encrypted - there is no support for encryption except for what you put in the data-channel

longview
Dec 25, 2006

heh.
It's $330 for a norwegian license, there's no tiers though so everyone can do all bands and the same power limits

longview
Dec 25, 2006

heh.

Jonny 290 posted:

poo poo i wish

that's actually really clean

it's clean but you can tell it's basically a DC circuit by all the right angle bends all over the place

air-wired over a copper ground plane except for one part I saw in the power supply, don't know what you'd call that

longview
Dec 25, 2006

heh.
http://hamgear.wordpress.com/2014/01/04/chinas-best-and-worst-of-2013/

longview
Dec 25, 2006

heh.
IMO the HT antenna to get is the Comet SMA-24, it's basically the same as the nagoya but much thinner and lighter

also the SMA-209 for when you're in good coverage areas, Icom FA-S270C is pro as gently caress though

and the Diamond RH-770 if you want to show off, it's fragile as gently caress but it performs really well at both 2m and 70cm

longview
Dec 25, 2006

heh.
I really really hate tuning SSB with SDR software, maybe I should buy one of those glowy hockey puck things

http://utaharc.org/rptr/synchronous_62.html this is one of the cooler repeater projects I've seen, two repeaters at two sites on the same frequency, with a very slight frequency offset and a duplex link + voting circuit they implemented a repeater system with diversity receive and transmit to normal radios

longview
Dec 25, 2006

heh.
when I read up on them I noticed that most western nations have stopped using them in the last decade, but the russians and other east block countries seem to still operate them

presumably the more technologically sophisticated spy agencies have switched to horse ebook twitters and other more advanced communication methods, which makes sense since these days having short wave radios is pretty rare compared to 30 years ago

longview
Dec 25, 2006

heh.

EMILY BLUNTS posted:

you triangulate it and it's probably just some antennas out in a field and a little shack you can't get to

They have done, usually they're in some RAF facility in Cyprus or the Ukraine

longview
Dec 25, 2006

heh.
lol if you're not rolling all N and SMA with semi rigid coax loving everywhere and RG-316 (proest coax) for flexible jumpers

just lol

longview
Dec 25, 2006

heh.
never had a problem with eBay coax and SMA/N, RS and Farnell charge ridiculous amounts for coax

RG-316 is the best ham radio coax because most old timers have no idea what it actually is (hint: it's a teflon version of RG-174)

longview
Dec 25, 2006

heh.

Jonny 290 posted:

dude is a whiz, and it really shows how simple these are. ADC into an FPGA. thats pretty much it. I don't know a god drat thing about FPGA's though so I don't know how he's utilizing it

i'm really impressed with his active antenna's performance and it is SO COOL to be able to listen to longwave broadcasts, we don't get poo poo in NA there, it's just noise.

http://dl1dbc.net/SAQ/miniwhip.html

yeah that antenna seems really interesting, turns out I have most of the parts except that JFET either here or at work

gonna try the cantenna solution, I have an aluminium project box of just the right size with plastic endplates

longview
Dec 25, 2006

heh.
just built my mini-whip antenna today (took like four hours)

first impressions: I need to re-bias the thing since I used different parts, but over all it does work, it seems to pick up less noise than the radios built in whip and also less than the loop antenna

I built mine differently than the reference, putting actual RF transformers (wideband 50/50 ohm) at both the antenna and inside the power supply, then I run the power on a stereo audio cable using the L/R as power and the shield grounded to power return and coax shield at the power supply
also added several common-mode inductors to suppress power supply noise

longview
Dec 25, 2006

heh.
that being said, looks like I still need to add a series resistor to the antenna element, those JFET gates are ridiculously easy to break

longview
Dec 25, 2006

heh.
I think any kind of diode on the gate will destroy the signal due to the capacitive effect of the diode, the element is about 1-10 pF and adding a several pF diode to ground or VCC is probably a bad idea

I'll probably just add a 100k-1M resistor, the input is high impedance so shouldn't matter except for the increase in thermal noise

nah, I have a reel full of mini-circuits RF transformers (15kHz-~1000MHz) in my office, for the chokes I had wurth 744220s lying around

I'll post pictures tomorrow when I've fixed it

longview
Dec 25, 2006

heh.
actually, the source of the gate is at the same potential as the gate (source follower configuration), adding a low capacitance diode from the gate to the source should result in a net zero added capacitance

longview
Dec 25, 2006

heh.
here you go, my idiot tiny antenna:

i made this PCB wayyy to big for the parts, whip (chassis) connects at the left, output to the balun on the right
of note: pro as gently caress 1210 package capacitors soldered to the ground plane acting as solder turrets for the power distribution
ground for the antenna is connected to the mounting bracket


i cut a piece of ~RG-316~ to the same length as the stereo audio cable I use for power, the PSU box also has a 3,5mm jack RF output to interface with scrub tier receivers like the Tecsun PL-660 (actually pretty good)



the power box is just some line filters and another balun, nothing special

longview
Dec 25, 2006

heh.
just did a comparison with the sony an-lp1 loop, it gives out a stronger signal but has worse SNR than this mounted in a non-ideal spot but with good grounding

if you want to go all out, power the preamp through the coax and use a relay to disconnect the amp, that way you could transmit on the same antenna

i might do that for my discone actually... but I'd need a much more broad band amplifier (25-1700 MHz)
in fact, i'll check what parts might be suitable tomorrow, a nice broad-band IC with a few dB gain would be very nice to have

longview
Dec 25, 2006

heh.
http://www.sprut.de/electronic/pic/projekte/adsb/adsb_en.html

try this one, it's really easy to build if you have some stiff wire around and I've had good experience with collinear antenna designs assuming the lengths are correct

i forgot to mention i also put a poly-fuse in the power box, in case of shorts (had several due to the horrible design of the 3,5mm jack chassis connectors)

longview
Dec 25, 2006

heh.
just get two dongles, less work than fiddling with pin diodes or coax relays

longview
Dec 25, 2006

heh.
also you want the pi in a box on the ground, unless you feel like shielding the power supplies (12V->5V module for example), pi board and the sdr dongles

it's very sensitive to electric fields so probably want to shield things regardless of distance if you're using it outside where there's nothing attenuating the noise sources

longview
Dec 25, 2006

heh.
here's a 5 minute project for the lazy ham: tripod HT antenna mount





only issue i had is first go i got solder all over the threads

longview
Dec 25, 2006

heh.
http://www.ti.com/product/ths9001

im gonna try using this thing for my active discone, should be able to fit inside a tiny box possibly inside the discone mounting tube if i build it right

challenging part will be building a 1 GHz bias-tee from discretes, but oh well, needs must

Adbot
ADBOT LOVES YOU

longview
Dec 25, 2006

heh.
here's some pdfs describing the apollo missions S-band radio system
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19660018739.pdf

http://www.hq.nasa.gov/alsj/19650010806_1965010806.pdf

from wikipedia:

quote:

Allocating uplink/downlink frequency pairs in a fixed ratio of 221/240 permitted the use of coherent transponders on the spacecraft. The spacecraft tracked the uplink carrier with a phase locked loop and, with a series of frequency dividers and multipliers, multiplied the uplink carrier frequency by the ratio 240/221 to produce its own downlink carrier signal.
When no uplink was detected, the transponder downlink carrier was generated from a local oscillator at the nominal frequency.
This "two-way" technique allowed extremely precise relative velocity measurements (in centimeters/sec) by observing the Doppler shift of the downlink carrier without a high accuracy oscillator on the spacecraft, although one was still needed on the ground.

[...]
A typical Apollo S-band spacecraft transmitter produced 20 watts; a typical uplink transmitter produced 10 kW

and a retelling of how the parkes receiver station got the actual OTA signals of the first steps
http://www.parkes.atnf.csiro.au/news_events/apollo11/pasa/on_eagles_wings.pdf

  • Locked thread