#qi-hardware IRC log for Sunday, 2015-06-14

DocScrutinizer05interesting module/EVB: http://www.globalstar.com/shop/index.php?main_page=product_info&cPath=23&products_id=8114:11
whitequarkooh neat14:12
eintopfthe question is, can you put it into the neo900?14:13
eintopfand the other question is, will it blend?14:13
DocScrutinizer05simplex upstream14:22
wpwrakstandby 0.5 W typical. oh dear !14:22
DocScrutinizer05yep, of course14:27
DocScrutinizer05you hardly can find much better for sat communications14:28
DocScrutinizer05well, their phones are allegedly 2 days standby14:28
DocScrutinizer05but then I wonder what "standby" means for a device that requires unfolding and extending the antenna before it can enter active mode14:29
DocScrutinizer05generally sat phones are not really meant to _receive_ calls14:30
wpwrakbtw, philae is back ! (article in german) http://www.heise.de/newsticker/meldung/Hallo-vom-Kometen-Lander-Philae-meldet-sich-zurueck-2690194.html14:30
whitequarkI wonder if the .5W are for something like TCXO14:32
DocScrutinizer05hmm, I don't think they are that pciky about their clock frequency14:36
DocScrutinizer05it's rather simply continuous mode RX, not optimized by any sleep modes or similar14:37
Action: DocScrutinizer05 idly wonders what been GTA02/calypso power rating when applying sleep=0 #1024 bug work around14:38
DocScrutinizer05so you could guess like: battery capacity ~4Wh, 6-8h of standby without proper deep-sleep => 0.5W 14:45
DocScrutinizer05no need for a crystal heating to burn power ;-)14:46
eintopfI remember my old smartphone, when I did to turn on the GPS it was extremely hot14:46
eintopfand my current smartphone, when it's hot _or_ cold then touchscreen doesn't work14:47
DocScrutinizer05wl1251 wlan chipset in AP scan mode (which forbids any powersaving (aka deepsleep) mode) also is known to be a terrible power hog14:47
eintopfI think, I need a neo90014:47
DocScrutinizer05hurry, we might have no more available for preorder at end of month ;-)14:48
DocScrutinizer05kidding aside, we might want to start N900 sourcing next month, and it's highly questionable if we could source further N900 for late customers14:49
DocScrutinizer05when you only want a NeoN board to upgrade your own N900, don't worry14:50
eintopfI don't have a N90014:51
DocScrutinizer05then your deadline is more relaxed14:51
eintopfI have a motorola milestone14:51
eintopfand now one without a keyboard and I want my keyboard back14:51
eintopf480 eur is a little bit too expensive for me :/14:51
DocScrutinizer05that's even only a down payment14:52
eintopfthen wpwrak will kill me, I should buy a frequency spectrum analyzer for that money14:52
DocScrutinizer05o.O spectrum analyzer? what for?14:52
eintopf802.15.4 stuff14:53
DocScrutinizer05I doubt *I* could properly use a spectrum analyzer for such stuff14:53
DocScrutinizer05also isn't that 2.4GHz?14:54
wpwrakyup, 2.4 GHz14:54
DocScrutinizer05I wonder where you get which spectrum analyzer for 480 bucks that can do 2.4GHz14:54
DocScrutinizer05and provide useful readouts14:55
DocScrutinizer05get a RTL2032 ;-)14:55
whitequarkcan't you lower it from 2.4GHz to some IF and then use something with a few tens of MHz of bandwidth?14:55
eintopfthat's what wpwrak told me14:55
wpwraka wee bit more expensive: http://www.aaronia.com/products/spectrum-analyzers/HF-2025E-V3-Low-Cost-Spectrum-Analyzer/14:55
DocScrutinizer05prolly just as good for the purpose you faintly mentioned14:55
wpwrakheh :)14:56
eintopfI google RTL2032 and got results for a stick which helps me to put my clothes on14:57
DocScrutinizer05oops 2832 sorry14:58
DocScrutinizer05mmmph, 1850MHz14:59
eintopfmhh, if this device can do 2.4 Ghz spectrum of 802.15.4 then I could put a firmware on which implements 802.15.4 standard and using it as a transceiver?14:59
eintopfor is the modulation differs, or that doesn't matter?15:00
DocScrutinizer05it's RX only15:00
DocScrutinizer05but it has no demodulator, it's a SDR15:00
DocScrutinizer05it provides I/Q digital data via USB15:00
eintopfI thought I could make atusb215:01
eintopfwith open firmware15:01
eintopfand atusb protocol ;-)15:01
DocScrutinizer05I *guess* you could use a cheap analog sat receiver for forntend as well15:01
eintopfyou want to tell me now, that when I have some LNB and an analog receiver and I set the frequency to the right. Then I can see some patterns in the "analog snow screen" ?15:02
eintopffrom my 802.15.4 transceiver15:02
eintopfI do some transmitting then15:03
Action: eintopf can't believe that15:03
eintopfso analog tv is about  9502150 MHz15:04
eintopf950 - 215015:05
eintopfand my transceivers are 2405 - 248015:05
DocScrutinizer05yes, I already noticed it's a 300MHz off15:09
DocScrutinizer05awesome: https://www.youtube.com/watch?v=-k2ZuhAz0ac15:19
DocScrutinizer05eintopf: yep, analog sat receivers have a IF of 9502150 MHz15:52
DocScrutinizer05anyway I wonder why you need a spectrum analyzer for those 802.15.415:53
eintopfdetermine xtal trims, I am not an expert. But the xtal pins of the transceiver has the feature to set the pin capacity15:54
eintopfthey doing PLL and then I can set these capacity pins to the right value, when I see the result15:55
eintopfthat's I think how it works, this is what wpwrak teach me :/15:56
DocScrutinizer05you need a frequency counter for this, or a very precise receiver, not a spectrum analyser15:59
eintopfbut when I see the spectrum I can see the piek16:00
eintopfand the piek should be something like the frequency and around is some gauss curve16:00
DocScrutinizer05huh? no, it should not have any gauss curve16:01
eintopfa gauss curve and a piek in the frequeceny which is set16:01
eintopfthen only a piek in the middle where the frequency is set the rest is random16:01
DocScrutinizer05a gauss curve or the like would be seen by modulated signals16:01
eintopfit's modulated16:02
DocScrutinizer05which create sideband frequencies16:02
eintopfQPSK or something16:02
DocScrutinizer05the xtal frequency is NOT modulated16:02
eintopfyea, but I cannot touch the pins inside the transceiver16:02
eintopfthen I need to listen to the frequency and the real result16:03
DocScrutinizer05even then you're simply not interested in that gauss curve, any arbitrary frequency counter will already pick up the "peek" for you16:05
DocScrutinizer05you say you need a spectrum analyzer to show the sidebands so you can ignore them, that's a tad awkward approach16:06
eintopfThen I could use maybe a atmega32 and use it as frequency counter16:06
DocScrutinizer05for 2.4GHz hardly16:06
eintopfit's PLL16:07
eintopfthere is some 16 Mhz crystal16:07
eintopfcan I not check this frequency?16:07
DocScrutinizer05but you should be able to touch one of the two XTAL pins of the chip with a high impedance probe16:07
DocScrutinizer05or even couple inductively to the RF emission of the xtal16:08
DocScrutinizer05the XTAL1/2 pins are output and input basically16:08
DocScrutinizer05you probably can touch the output without detuning the xtal oscillator too much, when you use e.g. a 10MOhm scope probe16:09
DocScrutinizer05but honestly why do you worry that much about exact tuning? just tune your one receiver so it works fine with the peer transmitter, so both use same (possibly not totally correct) frequency16:10
DocScrutinizer05you only worry about absolute correct frequency when building a 1000 or a million devices and all should be inter-operable with each other16:11
DocScrutinizer05and possibly even to 2nd source devices not built by you16:11
DocScrutinizer05or you worry when you need to pass approvals my FCC and the likes16:12
DocScrutinizer05sorry when I'm assuming (e.g. existence of external xtal and XTAL1(2 pins on chip) and your situation looks different. I have no info what you actually do and what's your circuit16:15
DocScrutinizer05but generally for adjusting frequency you don't use a spectrum analyzer16:16
DocScrutinizer05you may, when you already have one16:16
DocScrutinizer05and when the one you have is calibrated and pretty exact16:17
DocScrutinizer05for your average dirt cheap spectrum analyzer (500 bucks) odds are your own DIY circuit is more precisely tuned to the target frequency than the SA's internal reference clock16:18
DocScrutinizer05take a shortwafe radio, tune it to 16MHz and see if you can pick up the XTAL signal 16:21
DocScrutinizer05if you're lucky you may create interference with a nearby (in freq) broadcast station, and form the pitch/freq of resulting beat product tone you can know the difference in freq between your xtal signal and the broadcast station. Then you just need to find out which broadcast station this is you hear, and you easily can find their exact frequency they transmit on, which is pretty precise16:24
DocScrutinizer05whatever the chip you use, when it has a xtal oscillator then it also provides a means to tune that, by either providing the 16MHz to a test pin so you could run whatever freq counter you like on it (use a 1/16E6 divider and compare the number of seconds counted in one hour against an arbitrary watch), or the chip has built in freq offset evaluation and all it needs is a correct freq external 2.4MHz signal and the chip tells you in 16:29
DocScrutinizer05some diagnostic register how much the internal xtal is offset from the applied external reference signal16:29
DocScrutinizer05and for a last resort you can attach to the crystal mechanically/acoustically, using e.g. a piezo. This shouldn't detune your xtal oscillator frequency and will even work for (never seen yet) chips with integrated crystal16:32
DocScrutinizer05while on this topic: a generally unknown/neglected issue is mechanical-acoustical coupling and interference between different crystals SMT-mounted to same PCB16:33
DocScrutinizer05crystals would make for nice hidden/stealth microphones in an exploited circuit :-)16:35
Action: DocScrutinizer05 should find out if the crystal's frequency (or at least phase) also gets detuned/modulated from the acceleration by audible frequency mechano-acoustic waves introduced to the PCB16:38
whitequarkoh, cool idea16:38
whitequarkyou know tantalum caps are piezoelectric, right?16:38
whitequarkthat also works16:38
DocScrutinizer05err, not known of tantalum, no16:38
DocScrutinizer05X5R yes16:38
whitequarker, yes, ceramic of course16:39
whitequarkbrain fart16:39
DocScrutinizer05wouldn't feel surprised to see similar effects in generic electrolytic caps though16:39
DocScrutinizer05particularly when they are already a tad aged and dried, so some gas is inside between the electrodes16:40
DocScrutinizer05for gas-free electrolyte I dunno if we could expect much sensitivity to pressure / mechanical interference16:40
DocScrutinizer05(phase modulation of reference clock)  wouldn't be any threat in a PLL design, since the PLL has LPFs that filter this out completely16:44
DocScrutinizer05(never seen yet) sorry, incorrect. The legendary DALLAS chip must have had an integrated crystal iirc16:55
--- Mon Jun 15 201500:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!