DocScrutinizer05 changed the topic of #qi-hardware to: Copyleft hardware - http://qi-hardware.com | hardware hackers join here to discuss Ben NanoNote, atben / atusb 802.15.4 wireless, and other community driven hw projects | public logging at http://en.qi-hardware.com/irclogs and http://irclog.whitequark.org/qi-hardware
dos11 has quit [Ping timeout: 252 seconds]
Luke-Jr has quit [Remote host closed the connection]
Luke-Jr has joined #qi-hardware
Luke-Jr has quit [Remote host closed the connection]
Luke-Jr has joined #qi-hardware
Luke-Jr has quit [Client Quit]
Luke-Jr has joined #qi-hardware
hkm has quit [Quit: Leaving]
atommann has joined #qi-hardware
dandon_ has joined #qi-hardware
dandon has quit [Ping timeout: 276 seconds]
dandon_ is now known as dandon
Luke-Jr has quit [Remote host closed the connection]
Luke-Jr has joined #qi-hardware
gbraad_ has quit [Ping timeout: 265 seconds]
zcrc|drunk has joined #qi-hardware
zcrc|drunk is now known as pcercuei
wolfspraul has joined #qi-hardware
jekhor_ has joined #qi-hardware
sb0 has joined #qi-hardware
dos11 has joined #qi-hardware
jekhor_ has quit [Read error: Connection reset by peer]
nicksydney has quit [Remote host closed the connection]
nicksydney has joined #qi-hardware
viric has quit [Ping timeout: 240 seconds]
viric has joined #qi-hardware
Ornotermes has quit [Quit: leaving]
Ornotermes has joined #qi-hardware
jekhor has joined #qi-hardware
sb0 has quit [Read error: Operation timed out]
sb0 has joined #qi-hardware
jekhor has quit [Read error: Operation timed out]
viric has quit [Ping timeout: 240 seconds]
viric has joined #qi-hardware
wej has joined #qi-hardware
atommann has quit [Quit: Leaving]
FrankBlues has joined #qi-hardware
wej has quit [Ping timeout: 260 seconds]
wej has joined #qi-hardware
solrize has joined #qi-hardware
<solrize> are there still any ben nanonotes around?
<FrankBlues> I have one... kinda gathering dust atm
<solrize> neat
<FrankBlues> I've been thinking about running emulators on it - I used to have VICE working, and I thought someone had one of the older Mac emulators working
<solrize> nice! i didn't know about vice
<ysionneau> newbie question but, would it be very hard to make a simple 2.4 GHz GFSK demodulator?
<ysionneau> I guess that the fact that some part of the circuit would need to handle high frequency (GHz) is what would be the big issue for DIY
<ysionneau> but if you buy some IC to do the PLL part for instance, maybe it's feasible, right?
<FrankBlues> There's a Dingux port of VICE that used to run on Ben, last time I tried it had strange dependency issues that I never got around to fixing.
<solrize> ysionneau, i imagine something like an AM detector to get rid of the baseband, then a hardware or software PLL? but i have no idea how stuff works at those frequencies
<solrize> bjj
<ysionneau> I mean, as long as I have a PLL to lock on the 2.4 GHz carrier
<solrize> oops
<ysionneau> and then I can demodulate by mixing the input signal with the PLL locked output
<ysionneau> all of this is the "hard part" because it runs at 2.4 GHz
<ysionneau> after that, it's baseband indeed
<solrize> you have a 2.4 ghz pll? that's what i didn't realize was possible, at least in the dinky equipment (e.g. bluetooth) that i imagine you're dealing with
<ysionneau> how can you demodulate a bluetooth signal without a 2.4 GHz pll ?
<ysionneau> I thought this was mandatory
<larsc> sample _all_ the baseband ;)
<ysionneau> sample with 4.8 GHz ADC?
<ysionneau> and then do everything as SDR
<ysionneau> but you need very expensive electronics
<solrize> " i have no idea how stuff works at those frequencies" => pure speculation but i was imagining some crude LC filter made of pcb trace, then a diode detector
<larsc> ysionneau: there are lots of SDR platforms which allow you to sample at 2.4GHz with a limited bandwidth
<solrize> you should probably ask on #electronics where there might be some analog rf folks hanging out
<larsc> say a few MHz
<solrize> larsc, is that stuff big and complicated? i was presuming this is for cheap low performance gear like wifi or bluetooth
<ysionneau> larsc: I am searching for very low price solutions
<larsc> solrize: depends on your definition of big and complicated
wej has quit [Ping timeout: 252 seconds]
<larsc> ysionneau: but you want the digital bt signal?
<wpwrak_> ysionneau: what do you want to accomplish in the end ? build an SDR system ? receive BT ? something else
<ysionneau> receive BTLE
<solrize> larsc, big/complicated = you can't build it into a 5 dollar consumer gizmo
<larsc> the cheapest solution is probably a off the shelf bt receiver
<ysionneau> well, in fact I want to learn RF
<ysionneau> I think I will start with a simple FM demodulator
<larsc> then it's not going to be cheap ;)
<wpwrak_> ysionneau: have you been following what i'm doing on anelok ?
<ysionneau> wpwrak_: yes :)
<wpwrak_> and do you just want to receive or also send ?
<ysionneau> pretty cool stuff
<wpwrak_> then you already know some chips you can use :)
<ysionneau> yes
<ysionneau> but I would like to replace the chip with home made circuitery
<ysionneau> when it's possible
<ysionneau> to learn the RF part
<wpwrak_> oh, i see. ah yes, there are components for that, too. and fpgas may come in handy, too
<solrize> larsc this is the copyleft hardware channel, using off the shelf bt doesn't count ;-). i've been wondering about this myself because of the suspicious code blobs in all the bt processors
<larsc> solrize: where do you draw the line?
<ysionneau> larsc: I want to demodulate, then I get the baseband analog signal, then I can sample with cheap ADC and get the digital stream
<ysionneau> it's just for fun, no big precise goal
<larsc> ysionneau: pll + demodulator
<ysionneau> larsc: does it sound like a very complex project?
<solrize> ysionneau, this rf is kind of outside the computer zone... but there are some arrl publications? arrl.org
<ysionneau> for sure I guess I need to buy the PLL since it will operate at high frequency
<ysionneau> but the rest of the board... I guess I can drop discrete components
<larsc> there are also demoulators with a built-in pll
<solrize> ysionneau, larsc does bt gear actually use plls?
<wpwrak_> does anything not use a PLL ? :)
<wpwrak_> PLLs are about as common in modern electronics as electrons :)
<ysionneau> ah nice , thanks larsc :)
<solrize> wpwrak_, i dunno, we're talking about something with a design range of a few meters and in fact typical handsfree earpieces work a lot better if you hold the phone closer than keeping it in your pocket
<solrize> so they are willing to throw away a lot of the signal to keep the hw small and cheap
<solrize> that's why i was thinking of amplitude detection
<wpwrak_> solrize: well, you can build an analog radio system, sure. old school :)
<wpwrak_> not sure how quickly you'll bump into regulatory limit, though. but then if your sender is weak enough, maybe nobody will notice.
<ysionneau> for GFSK I guess I don't care about the I/Q output, right?
<ysionneau> I just take one of the two
<ysionneau> I/Q is useful for QAM, no?
<solrize> wpwrak_, yeah i was thinking if you make it crappy enough, it doesn't need much hardware
<larsc> having both makes the demodulation dead easy
<ysionneau> I/Q is the output of the chip, so it's already baseband, right?
<ysionneau> so already demodulated
* ysionneau newbie, sorry
wej has joined #qi-hardware
<larsc> yes
rz2k has joined #qi-hardware
<ysionneau> by trying to understand some analog circuits, I feel I'm putting a toe in an ocean of unknown stuff
<ysionneau> "what's this ... ?" + wikipedia/google , then end up with explanation which raises another "what's this ?" etc
* ysionneau affraid
* whitequark is completely unable to grok analog circuits
<DocScrutinizer05> >>SDR platforms which allow you to sample at 2.4GHz<< wut?
unclouded has quit [Ping timeout: 245 seconds]
<solrize> those exist but not for bluetooth ;)
<solrize> they are for fancy pants mobile base stations
<whitequark> there's some dongle which can sniff bluetooth, but it's absurdly expensive
<DocScrutinizer05> I doubt *any* SDR is sampling with 4.8Gsamples/s
<DocScrutinizer05> you always use a Local Oscillator to downbeat-mix the input to some Intermediate Frequency
<DocScrutinizer05> for <100MHz you might do primary sampling
<larsc> I meant the baseband pll frequency, not the sampling frequency
<larsc> but there are also systems that have such a sampling frequency
<DocScrutinizer05> good luck with processing that data stream in realtime X-P
<wpwrak_> larsc: every hobbyist should have one of those ;-)
<larsc> DocScrutinizer05: you'd typically not process it on a CPU
<DocScrutinizer05> aha
<larsc> in realtime
<DocScrutinizer05> so where's the S in SDR then?
<larsc> in the FPGA
<DocScrutinizer05> sure, a FPGA easily is processing a datastream of ~10GByte/s, probably via FFT
<DocScrutinizer05> maybe you get away even with 5GB/s
<sb0> yeah, and then some more. ask the radiotelescope folks :-)
<DocScrutinizer05> when using 8bit samples
jekhor has joined #qi-hardware
<DocScrutinizer05> ysionneau: I/Q is basically the 90° phase skewed digitized samples of the IF. All (well, most) SDR is based on that
<DocScrutinizer05> consider sampling a frequency of 2Pi with a A/D that samples (average in) windows of Pi. You'd either get a maximum output signal of +max -max +max... for frequency in phase with sampling, or a 0, 0, 0, when frequency is 90° off in phase
<DocScrutinizer05> with I and Q sampling with a skew of 90° phase between the two A/D, you have supfficient info about your signal to do SDR on it, no matter what#s the modulation
<DocScrutinizer05> basically for AM you would do I+Q, for FM you do I-Q (simplified picture)
<DocScrutinizer05> since sampling at air frequency is simply impossible for all but the highest budget, at e.g. 2.4GHz, you first mix the input frequency with a Local Oscillator in a ring modulator that creates f(in)+f(LO) and f(in)-f(LO) as output
<DocScrutinizer05> then you usually filter out the f(in)-f(LO) band with a bandpass of a few MHz or kHz width, and then sample it or do analog decoding
<DocScrutinizer05> you also want to have a bandpass at "antenna" to avoid another input frequency sneak in, f1(in)-f(LO) == f2(in)+f(LO)
<DocScrutinizer05> that's called the mirror frequency
<larsc> the trick with I/Q modulators acutally is, that there is no image
<DocScrutinizer05> often SDR works with mixing down to zero, means LO frequency is (almost) equal to input frequency
<whitequark> the cake is a lie
FrankBlues has quit [Remote host closed the connection]
<DocScrutinizer05> for the LO usually PLL is used (actually an analog oscillator, a reference XTAL oscillator, and two programmable dividers feeding their output to a PLL that adjusts the analog oscillator), which you program to create a LO frequency of f(in)-$some-offset. This $some-offset is the Intermediate frequency you want to use. In shortwave receivers this IF is 455kHz usually, on FM receivers (80-200MHz) it's usually 10.7MHz
<DocScrutinizer05> so for FM the mirror freq is 2 * 10,7MHz
<DocScrutinizer05> ...off from your intended receiving frequency
<DocScrutinizer05> you want a bandpass at antenna that gives a sufficient dB attenuation for frequencies that are this far from your center freq, so even strong local transmitters at mirror freq don't interfere with your signal
<DocScrutinizer05> and you definitely don't want to build all that from discrete components for BTLE ;-)
wej has quit [Ping timeout: 260 seconds]
<DocScrutinizer05> when you want to learn about that stuff, maybe RTLSDR is a good starting point
<DocScrutinizer05> the dongles are still available for a 10 to 20 bucks
<DocScrutinizer05> and some of them should even be able to reach 2.4GHz
<DocScrutinizer05> http://rtlsdr.org/
wej has joined #qi-hardware
<ysionneau> thanks for the big sum up :)
<DocScrutinizer05> yw
<DocScrutinizer05> NB that RTL dongles lack most of the filters/bandpasses I mentioned. The cheat on it
<DocScrutinizer05> they*
<ysionneau> so I/Q output of the demodulator are still signals at the carrier frequency?
<DocScrutinizer05> they don't use proper antenna bandpass so mirror frequencies may be a problem, and they don't use IF filter bandpass so the A/D for I/Q is prone to aliasing
<ysionneau> they are not located after the multiplication of the local oscillator ?
<DocScrutinizer05> no, they use a downbeat mixer tuner as described above
<DocScrutinizer05> the tuner output is some Intermediate Frequency of a few MHz which then gets digitized by the RTL2832 chip
wej has quit [Ping timeout: 252 seconds]
<DocScrutinizer05> the tuner has the PLL aka LO and the mixer, and some amps
<DocScrutinizer05> the RTL2832 sends I/Q data to PC via USB
<ysionneau> ah ok so it lowers the frequency a bit, but not until baseband
<ysionneau> ok :o
<DocScrutinizer05> depends on how you program it
<DocScrutinizer05> "baseband" is not demodulated signal though. What you call "baseband" would be IF of zero
<DocScrutinizer05> and it's actually usually doing that
<ysionneau> IF = input frequency?
<DocScrutinizer05> well, some setups use a very low IF of some kHz or so, to get rid of the DC that sneaks in to I/Q when you mix down to exactl zero
<DocScrutinizer05> Intermediate Frequency
<ysionneau> what I call baseband is when you mix the input signal with exactly the carrier frequency (by using a PLL to obtain it, in phase)
<DocScrutinizer05> the input to the RTL2832, coming from tuner output
<ysionneau> then you should get your demodulated tone, but centered at 0 Hz
<ysionneau> instead of centered on the carrier freq
<DocScrutinizer05> well, modulation not working like this
<DocScrutinizer05> for AM you might get away with it, yes
wej has joined #qi-hardware
<ysionneau> yes I think most of my "modulation" thinking is centered on QAM stuf
<DocScrutinizer05> but then it's pretty hard to filter away the LO signal from your input signal when you mix down to exact 0 to demodulate
<ysionneau> would you have some litterature to educate me on those topics ?
<ysionneau> modulation/demodulation
<DocScrutinizer05> hmm, maybe something comes to mind when I tthink a bit about it
<ysionneau> on several examples: FSK, QAM all sort of different modulation scheme
<DocScrutinizer05> right off top of my head I'm short of suggestions
<DocScrutinizer05> well, those are all already digital modulations
<ysionneau> book or website (pdf..) whatever
<DocScrutinizer05> FSK is digital FM
<ysionneau> I think I can start by digging up my old school book
<DocScrutinizer05> QAM is digital AM (more or less)
<ysionneau> yep I got that
<DocScrutinizer05> sorry, afk
<ysionneau> I'm mostly interested in digital communications
<ysionneau> np, thanks :)
wej has quit [Ping timeout: 245 seconds]
mth has joined #qi-hardware
wpwrak_ is now known as wpwrak
* wpwrak hates those *_
wej has joined #qi-hardware
wej has quit [Ping timeout: 260 seconds]
dandon has quit [Ping timeout: 252 seconds]
dandon has joined #qi-hardware
wej has joined #qi-hardware
<pcercuei> larsc, hey, do you know an internet provider in Germany that offers contracts without a minimal period?
<larsc> mnet
<larsc> that's what I use
<pcercuei> I didn't know you were into K-pop
<pcercuei> unless mnet.com is not the right website
<pcercuei> ah, .de
<pcercuei> of course
<larsc> I think I can refer you
wej has quit [Ping timeout: 252 seconds]
wej has joined #qi-hardware
wej has quit [Ping timeout: 252 seconds]
wej has joined #qi-hardware
wolfspraul has quit [Quit: leaving]
solrize has quit [Ping timeout: 276 seconds]
Jay7 has quit [Read error: Connection reset by peer]
Jay7 has joined #qi-hardware
Ornotermes has quit [Ping timeout: 240 seconds]
Ornotermes has joined #qi-hardware
Ornotermes has quit [Ping timeout: 240 seconds]
Ornotermes has joined #qi-hardware
nicksydney has quit [Remote host closed the connection]
nicksydney has joined #qi-hardware
jekhor has quit [Ping timeout: 255 seconds]
sb0 has quit [Quit: Leaving]
rz2k has quit []