highaltitude.log.20090819

[00:07] Matt_APEX (n=chatzill@87.115.47.212.plusnet.pcl-ag01.dyn.plus.net) left irc: Read error: 110 (Connection timed out)
[00:11] Matt_APEX_ (n=chatzill@host86-154-241-231.range86-154.btcentralplus.com) left irc: Read error: 113 (No route to host)
[00:21] edmoore (n=ed@oort.chu.cam.ac.uk) left irc:
[01:02] Laurenceb (n=laurence@host86-159-200-131.range86-159.btcentralplus.com) left irc: Remote closed the connection
[01:49] DanielRichman (n=DanielRi@cpc3-cbly2-0-0-cust261.glfd.cable.ntl.com) left irc: Remote closed the connection
[02:27] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 104 (Connection reset by peer)
[02:40] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[02:45] Nick change: KingJ -> kingj
[06:11] ChezaWho (n=chezawho@adsl-63-197-2-193.dsl.snfc21.pacbell.net) left irc:
[06:20] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[06:20] simon__ (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[06:32] hallam (n=hallam@c-67-169-9-130.hsd1.ca.comcast.net) left irc: Read error: 113 (No route to host)
[06:51] simon__ (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[06:51] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[06:51] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[07:16] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 110 (Connection timed out)
[07:19] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[07:20] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Client Quit
[07:42] edmoore (n=ed@oort.chu.cam.ac.uk) joined #highaltitude.
[07:51] edmoore (n=ed@oort.chu.cam.ac.uk) left irc:
[08:21] edmoore (n=ed@oort.chu.cam.ac.uk) joined #highaltitude.
[08:54] edmoore (n=ed@oort.chu.cam.ac.uk) left irc:
[10:05] Laurenceb (n=laurence@host86-159-200-131.range86-159.btcentralplus.com) joined #highaltitude.
[10:13] GeekShadow (n=Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[10:16] icez (n=icez@unaffiliated/icez) left irc: Remote closed the connection
[10:22] <Laurenceb> http://www.microbrushless.com/products3.htm <- nice
[10:25] Laurenceb (n=laurence@host86-159-200-131.range86-159.btcentralplus.com) left irc: Remote closed the connection
[10:49] rjharrison (n=rharriso@gateway.hgf.com) joined #highaltitude.
[11:45] edmoore (n=ed@chu-gw.churchillcambridge.co.uk) joined #highaltitude.
[11:48] <edmoore> hi rjharrison
[11:52] DanielRichman (n=DanielRi@cpc3-cbly2-0-0-cust261.glfd.cable.ntl.com) joined #highaltitude.
[11:59] herabot (n=herabot@92.29.73.106) left irc: Remote closed the connection
[12:00] herabot (n=herabot@92.29.73.106) joined #highaltitude.
[12:04] <rjharrison> hi edmoore
[12:04] <edmoore> aha hi
[12:04] <edmoore> did you get my email?
[12:11] sbasuita (n=sbasuita@unaffiliated/drebellion) joined #highaltitude.
[12:14] <edmoore> CUSF launch probably tomorrow
[12:14] <DanielRichman> What are you launching?
[12:14] <edmoore> sacrificing to the north sea, but a good opportunity to listen
[12:14] <edmoore> badgercub kamikaze
[12:16] <DanielRichman> eh. why?
[12:16] <rjharrison> North sea landing
[12:17] <edmoore> too windy
[12:17] <edmoore> try our predictor
[12:17] <DanielRichman> why are you sacrificing your payload?
[12:18] <edmoore> that bit where i said it's too windy
[12:18] <edmoore> too = too
[12:18] <edmoore> windy = windy
[12:18] <edmoore> put them together
[12:18] <edmoore> about to loose batt
[12:19] <DanielRichman> exactly - you know it's going in the drink, and you're doing it deliberately
[12:19] <edmoore> basically it's too windy to get it back - mega jet stream
[12:19] <edmoore> and we want to range test
[12:19] <DanielRichman> ahh ok
[12:19] <edmoore> and it's cheap
[12:20] edmoore (n=ed@chu-gw.churchillcambridge.co.uk) left irc:
[12:24] Matt_APEX (n=chatzill@87.115.165.107) joined #highaltitude.
[12:36] <ms7821> oh disaster, poor rat http://www.hobbyspace.com/nucleus/?itemid=14670
[12:42] DanielRichman (n=DanielRi@cpc3-cbly2-0-0-cust261.glfd.cable.ntl.com) left irc: Remote closed the connection
[12:43] DanielRichman (n=DanielRi@cpc3-cbly2-0-0-cust261.glfd.cable.ntl.com) joined #highaltitude.
[12:49] Xenion (n=robert@ip-90-186-66-51.web.vodafone.de) joined #highaltitude.
[12:57] Xenion (n=robert@ip-90-186-66-51.web.vodafone.de) left #highaltitude ("Verlassend").
[13:07] nickname (n=robert@ip-90-186-193-90.web.vodafone.de) joined #highaltitude.
[13:32] Action: SpeedEvil wonders why the met office doesn't produce nice graphs.
[13:32] <SpeedEvil> http://www.yr.no/place/United_Kingdom/Scotland/Glenrothes/hour_by_hour.html
[13:45] ChezaWho (n=chezawho@adsl-63-197-2-193.dsl.snfc21.pacbell.net) joined #highaltitude.
[14:02] rjmunro (n=chatzill@87.127.91.51) joined #highaltitude.
[14:06] rjharrison (n=rharriso@gateway.hgf.com) left irc:
[14:09] Futurity (n=Futurity@cpc1-cmbg15-2-0-cust43.5-4.cable.virginmedia.com) joined #highaltitude.
[14:24] nickname (n=robert@ip-90-186-193-90.web.vodafone.de) left irc: Read error: 110 (Connection timed out)
[14:28] petrinm_ (n=petrinm@C231-152.surffi.net) joined #highaltitude.
[15:31] rjmunro (n=chatzill@87.127.91.51) left irc: Read error: 110 (Connection timed out)
[16:18] Nick change: kingj -> KingJ
[16:21] ChezaWho (n=chezawho@adsl-63-197-2-193.dsl.snfc21.pacbell.net) left irc:
[16:58] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[16:59] tty1 (n=freemo@unaffiliated/electric-penguin/x-9957366) joined #highaltitude.
[16:59] tty1 (n=freemo@unaffiliated/electric-penguin/x-9957366) left #highaltitude.
[17:11] hallam (n=hallam@c-67-169-9-130.hsd1.ca.comcast.net) joined #highaltitude.
[17:12] <russss> http://www.theregister.co.uk/2009/08/19/sanswire_sts_111_announcement/
[17:16] Action: SpeedEvil had this idea _ages_ ago
[17:17] <SpeedEvil> I was trying to workout a lightweight propane+air actuator
[17:17] <SpeedEvil> Oh - not quite
[17:19] <SpeedEvil> This was a 'swimming' airship
[17:23] <SpeedEvil> (lousy efficiency apart from at truly huge scales - more a toy)
[17:26] <SpeedEvil> http://www.theregister.co.uk/2009/03/20/top_model_stampede/ :)
[17:41] ms7821 (n=ms@87-194-6-74.bethere.co.uk) left irc: Nick collision from services.
[17:41] ms7821 (n=ms@87-194-6-74.bethere.co.uk) joined #highaltitude.
[17:57] ChezaWho (n=chezawho@adsl-99-38-181-246.dsl.pltn13.sbcglobal.net) joined #highaltitude.
[18:18] petrinm_ (n=petrinm@C231-152.surffi.net) left #highaltitude.
[18:32] Xenion (n=robert@p579FCCA3.dip.t-dialin.net) joined #highaltitude.
[18:36] ChezaWho_ (n=chezawho@adsl-99-38-181-238.dsl.pltn13.sbcglobal.net) joined #highaltitude.
[18:44] ChezaWho (n=chezawho@adsl-99-38-181-246.dsl.pltn13.sbcglobal.net) left irc: Read error: 60 (Operation timed out)
[18:46] ChezaWho_ (n=chezawho@adsl-99-38-181-238.dsl.pltn13.sbcglobal.net) left irc: Read error: 145 (Connection timed out)
[18:54] petrinm_ (n=petrinm@C231-152.surffi.net) joined #highaltitude.
[18:54] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 110 (Connection timed out)
[18:59] ChezaWho (n=chezawho@adsl-99-38-181-246.dsl.pltn13.sbcglobal.net) joined #highaltitude.
[19:17] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[19:27] Futurity (n=Futurity@cpc1-cmbg15-2-0-cust43.5-4.cable.virginmedia.com) left irc:
[19:29] icez (n=icez@unaffiliated/icez) joined #highaltitude.
[20:15] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 110 (Connection timed out)
[20:16] RocketBoy (n=Steve@217.47.75.8) joined #highaltitude.
[20:22] petrinm_ (n=petrinm@C231-152.surffi.net) left #highaltitude.
[20:46] rjharrison_home (n=rharriso@80.176.172.227) joined #highaltitude.
[20:50] sbasuita (n=sbasuita@unaffiliated/drebellion) left irc: Read error: 104 (Connection reset by peer)
[20:51] sbasuita (n=sbasuita@unaffiliated/drebellion) joined #highaltitude.
[20:55] <rjharrison_home> asda have 2xAA li's free at 6.99 a pack of 6
[21:04] nickname_ (n=robert@p579FCBBF.dip.t-dialin.net) joined #highaltitude.
[21:06] <RocketBoy> so thats 6 (including the free ones) for £6.99 - if so very good - and these are enigizer ultimate lithiums?
[21:20] Xenion (n=robert@p579FCCA3.dip.t-dialin.net) left irc: Read error: 101 (Network is unreachable)
[21:30] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[21:44] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[21:46] <nickname_> Gute Nacht // Good Night :-)
[21:46] nickname_ (n=robert@p579FCBBF.dip.t-dialin.net) left irc: "Verlassend"
[21:49] <Matt_APEX> hey, out of interest, which RF connectors do you tend to use on the balloon end?
[21:50] <hallam> SMA is nice and lightweight, BNC is easy and cheap but heavier
[21:50] <hallam> making up SMA cables can be annoying, especially at first
[21:50] <Matt_APEX> is there any chance a sma could unscrew itself?
[21:51] <hallam> you could add a drop of loctite if you want to be sure, but I've never seen it
[21:51] <Matt_APEX> ok thanks
[21:51] <hallam> I guess you should try to strain relieve the cable such that it's never going to be in that sort of situation
[21:51] <Matt_APEX> a SMA connector would be better on the pcb as you can get them in smd
[21:51] <Matt_APEX> havnt found a smd bnc connector yet
[21:54] <hallam> http://www.sparkfun.com/commerce/product_info.php?products_id=593 I really like these connectors, they sandwich the PCB between the pins
[21:55] <hallam> very easy and very secure
[21:55] <hallam> I'd be wary of a non-edge-mount surface mount connector, whether BNC or SMA, because it could rip the tracks off the PCB easily
[21:55] <Matt_APEX> ya thats the sorta thing i wanted
[21:56] <Matt_APEX> also is the transmitter module usually soldered to the pcb rather than using sockets?
[21:56] <hallam> the NTX2? I've soldered it all but one time
[21:57] <Matt_APEX> no BiM1
[21:57] <Matt_APEX> BiM2 even
[21:58] <Matt_APEX> no NiM2 :p
[21:58] <hallam> looks like that would be quite easily socketed
[21:58] <hallam> oh
[21:59] <hallam> make up your mind :P
[21:59] <Matt_APEX> yea got it right eventially
[21:59] <Matt_APEX> this http://www.radiometrix.co.uk/products/nim2.htm
[22:00] <hallam> the NTX2 was fragile on a socket, this would be less so because it has a couple rows of pins
[22:01] <hallam> fwiw, I still think you're taking an expensive gamble trying to do a datalink this way, but good luck
[22:01] <Matt_APEX> the problem is the pins are thicker than usual ICs so it wont fit in to all sockets
[22:01] <Matt_APEX> we are going to transmit rtty
[22:01] <Matt_APEX> just thats not the only thing we're gonna transmit
[22:01] <hallam> ok
[22:01] <hallam> if it's a hassle to socket it, then just solder it :)
[22:02] <Matt_APEX> yea
[22:02] <Matt_APEX> also the socket creates extra loss on the rf pin
[22:03] <hallam> a little, but that's much more of an issue for receive than transmit
[22:03] <RocketBoy> some of the radiometrix modules dont have the anagloge input charatistics of the NTX2
[22:04] <rjharrison_home> RocketBoy yep energizer lithiums
[22:04] <rjharrison_home> 699 for 6
[22:04] <rjharrison_home> dp missing
[22:04] <rjharrison_home> nights all
[22:04] rjharrison_home (n=rharriso@80.176.172.227) left irc:
[22:05] <RocketBoy> check the NIM2 datasheet says somthing like "The NTX2’s TXD input is normally driven directly by logic levels but will also accept analogue drive
[22:05] <RocketBoy> (e.g. 2-tone signalling)."
[22:05] <Matt_APEX> i think the NTX2 and NiM2 have the same input stage, its just the one we have is a transciever
[22:05] <Matt_APEX> er ok
[22:06] <Matt_APEX> i think it is analogue input
[22:06] <Matt_APEX> thats the impression i get from it
[22:07] <Matt_APEX> as it talks about putting a series resistor into the input so not to overmodulate
[22:07] <Matt_APEX> so suggesting analogue in
[22:07] <Matt_APEX> and it says that sentence too :)
[22:07] JimAM (n=jim@client-81-107-217-193.glfd.adsl.virgin.net) joined #highaltitude.
[22:08] <JimAM> did anyone see bang goes the theory?
[22:08] <RocketBoy> yes
[22:08] <JimAM> pretty impressive shame the batteries ran out
[22:09] <JimAM> don't know why they didn't stabilise the descent
[22:10] <RocketBoy> with what
[22:12] <JimAM> a chute?
[22:13] <RocketBoy> it had a chute deployed - its just that the doll on the end of the line was in a almost flat spin
[22:14] <RocketBoy> the camera was mounted in a box at the dolls feet - pointing down
[22:14] <RocketBoy> when the doll was released the parachute was deployed more or less immediatly
[22:15] <JimAM> how do you know that?
[22:15] <RocketBoy> but the drag on the doll + box was such that it eneded up at an angle - almost horizontal
[22:15] <RocketBoy> cos I did it
[22:15] <JimAM> you were involved?
[22:15] Action: hallam chuckles
[22:15] <RocketBoy> i was the guy in the show
[22:16] <RocketBoy> in the read coat
[22:16] <JimAM> how did you calculate the volume of the hydrogen
[22:16] <hallam> but Steve, you didn't really do anything apart from say "ok" ;) ;) ;)
[22:16] <RocketBoy> helium was used - 1Kg of lift per cum approx
[22:17] Action: SpeedEvil tries to resist bad puns.
[22:17] <RocketBoy> this is true henry
[22:17] <JimAM> the laptop was getting live data from high altitude, what transmission method did they use
[22:17] <RocketBoy> actully a saaid a load on camera in a studio - but that was all cut from the show
[22:18] <RocketBoy> licence exempt low power RTTY
[22:18] <RocketBoy> narrowband
[22:19] edmoore (i=836f0142@gateway/web/freenode/x-iujxniiumlthliui) joined #highaltitude.
[22:19] <DanielRichman> you can see fldigi on the laptop if you look closely
[22:19] <hallam> hi edmoore
[22:19] <edmoore> so the payload has just successfully received a string uploaded from the ground station
[22:20] <hallam> awesome!
[22:20] <edmoore> 'by the power of grayskull, I have the strap-on'
[22:20] <hallam> errr
[22:20] <hallam> what a string
[22:20] <edmoore> we'll be taking requests for jokes for tomorrow's launch
[22:20] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 60 (Operation timed out)
[22:20] <hallam> flying tomorrow?
[22:20] <hallam> badgercub test?
[22:20] <edmoore> currently yes. could all go wrong yet though
[22:20] <edmoore> yep
[22:21] <hallam> any backup transmitter onboard?
[22:21] <edmoore> nope. it's going into the sea
[22:21] <hallam> I guess you could wait for badger 1 to be posted back
[22:21] <hallam> fair dos
[22:21] <SpeedEvil> any potential of reaching .nl or somewhere?
[22:21] <hallam> why the splash? also testing something else?
[22:22] <JimAM> so are these personal projects?
[22:23] <SpeedEvil> JimAM: basically
[22:24] <SpeedEvil> JimAM: JimAM there is little sponsorship for the bulk of the payload - though some have gotten bits donated.
[22:24] <JimAM> disappears into the sea with a whole host of electronics on board?
[22:24] <JimAM> expensive waste
[22:24] <hallam> that's not usually the plan, you normally try to get it back
[22:24] <hallam> but sometimes it's expedient to splash something
[22:24] <hallam> more often it happens by mistake
[22:24] <JimAM> so what's the main purpose behind the projects?
[22:24] <JimAM> most of them are balloons I take it?
[22:24] <sbasuita> JimAM, pretty much
[22:25] <sbasuita> JimAM, just because we can
[22:25] <SpeedEvil> It's either splash - or wait weeks or more to get a launch that can recover.
[22:25] <hallam> jim: see www.ukhas.org.uk and, e.g. www.cuspaceflight.co.uk for the group that edmoore and I are part of
[22:26] <JimAM> most of your gear analogue or do you use microcontrollers?
[22:26] <hallam> micros
[22:27] <edmoore> back
[22:27] <JimAM> and you just detach your equipment from the balloon normally?
[22:27] <hallam> either that or the balloon pops
[22:27] <JimAM> what sort of altitudes do you get to?
[22:27] <edmoore> yes, we're going into the sea as this flight computer board is a bit damaged already and not really flight worth for its intended use, but it can still perform all the basic subsystem tasks
[22:27] <hallam> ~100,000 ft
[22:27] <SpeedEvil> 32km
[22:27] <SpeedEvil> ish
[22:27] <edmoore> + it's very very windy tomorrow, but we really want to launch
[22:27] <edmoore> so in combination, we think going into the sea is ok
[22:28] <JimAM> how do you get any data back from 100,000ft though?
[22:28] <hallam> radio
[22:28] <edmoore> it's purpose it to do a range test anyway, and that always involves going out to sea
[22:28] <SpeedEvil> http://wiki.ukhas.org.uk/guides:tracking_guide
[22:28] <hallam> edmoore: are you hoping for uplink at full altitude?
[22:28] <SpeedEvil> see that
[22:28] <SpeedEvil> JimAM:
[22:28] <edmoore> of course!
[22:28] <Matt_APEX> what power do you uplink at?
[22:29] <edmoore> we'll experiment, but my ic-7000 will go up to 30W
[22:29] <hallam> sweet
[22:29] <hallam> so the test just now was with the ic-7k
[22:29] <hallam> what modulation etc?
[22:29] <edmoore> yes
[22:29] <edmoore> OOK
[22:29] <hallam> heh
[22:29] <edmoore> power turned right down, no antenna on badgercub
[22:29] <hallam> well, it works
[22:29] Action: hallam is impressed/excited
[22:30] <JimAM> so this is part of undergraduate or postgraduate research?
[22:30] <edmoore> the cc1111 is a little bitch if you don't want to use it the way it wants to be used, but OOK seems to work with a soft serial port
[22:30] <edmoore> JimAM: undergraduate
[22:30] <edmoore> (currently)
[22:30] <JimAM> in what field?
[22:30] <hallam> engineering
[22:30] <sbasuita> JimAM, not all the projects involve universities
[22:30] <edmoore> Well, there's three of us in the room now, I'm systems and control (+ electronics), Iain is an aero guy, and fergus is a physicist
[22:31] <hallam> edmoore, what sort of launch time are you planning, and will it be on the web tracker?
[22:31] <edmoore> but this isn't 'for' undergraduate studies, it's entirely extra-curricular
[22:31] <edmoore> noon-ish, and yes, i hope so
[22:32] <hallam> great
[22:32] <hallam> I'll get up early :)
[22:32] <edmoore> :)
[22:32] <JimAM> you don't use matlab to do your software do you?
[22:32] <hallam> we've used matlab for some research/modelling, but not for flight software
[22:33] <JimAM> I was based at the SEIC at loughborough uni for a short while, and all they ever went on about there was matlab
[22:33] <JimAM> never used it
[22:33] <hallam> there is a matlab/simulink compiler for embedded targets so you could go that route, but it sounds kind of painful to me
[22:33] <edmoore> it's great for analysis, but it's not a general purpose programming language
[22:34] <hallam> edmoore: I'm wrestling the last parts of the gps from matlab's clutches as we sepak
[22:34] <hallam> speak*
[22:34] <hallam> only the navigation filter is still there
[22:34] <edmoore> there's Octave which is basically free matlab (minus some fringe features) which is useful
[22:34] <hallam> uggggh
[22:35] <JimAM> can you use gps at those altitudes or do you rely on barometers?
[22:35] <edmoore> or as perhaps hallam will tell you, matlab and octave are divided by a common language
[22:35] <JimAM> for altitude
[22:35] <edmoore> JimAM: you can use GPS units, as long as they work at the altitudes you want to go to
[22:35] <edmoore> the wiki contains this information
[22:37] <RocketBoy> jimAM: GPS
[22:38] <SpeedEvil> hallam: :)
[22:38] <SpeedEvil> hallam: so you've got the code working on the blackfin?
[22:38] <hallam> I had the tracking loops going on the blackfin a while back
[22:39] <hallam> I'm going for gumstix rather than blackfin now
[22:39] <hallam> might change my mind if it ends up being a bitch
[22:39] <SpeedEvil> So you're basically outputting psuedoranges at the moment?
[22:39] <SpeedEvil> ah
[22:39] <SpeedEvil> For any reason?
[22:39] <SpeedEvil> just tat you can buy it easily?
[22:39] <edmoore> it's on badger2 :)
[22:39] <SpeedEvil> ah
[22:39] <hallam> and usb host and some other nice stuff
[22:39] <hallam> and more ram
[22:39] <hallam> than the blackfin dev board I was using
[22:40] <SpeedEvil> Can it run matlab though? :)
[22:40] <hallam> heh
[22:40] <hallam> matlab could probably compile for it
[22:40] <edmoore> i'm in ur sudoranjez, stealin ur cycles
[22:40] <hallam> but screw that
[22:40] <hallam> ed are you a little bit tired?
[22:40] <edmoore> tired + a press of the most amazing coffee
[22:40] <edmoore> so carl and I went halves on an i-Roast 2
[22:41] <edmoore> and, if i say so myself, we have just about got the roasting profile for these particular beans spot on
[22:41] <edmoore> and it is *delicious*
[22:41] <hallam> not bad
[22:41] <edmoore> but the problem is that I've had about 4 mugs of espresso-strength coffee now
[22:41] <hallam> I see
[22:41] <hallam> for calibration purposes
[22:42] <edmoore> and late-night layout purposes
[22:42] <edmoore> oh I managed to re-lyric 'hey big spender' for pcb layouts
[22:42] <edmoore> 'i could see you were a topological impossibilityyyyyyyyy
[22:42] <edmoore> hey auto-router
[22:43] <edmoore> layoooouut this PCB for meeeeee
[22:43] <edmoore> etc
[22:43] <hallam> thanks for that
[22:43] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[22:43] <hallam> 6 layers = win
[22:44] <SpeedEvil> more layers is easy.
[22:44] <edmoore> and exponentially more expensive
[22:45] <SpeedEvil> yeah - that too.
[22:45] <edmoore> 4 is fine, just needs some consideration
[22:45] <SpeedEvil> I have a project that looks like it's going to involve BGAs on both sides.
[22:45] <SpeedEvil> For V3 at least.
[22:46] <hallam> if your time has value then more layers are cheap
[22:46] <edmoore> I'm a student
[22:47] <hallam> I thought you were a contractor
[22:48] <hallam> it's ok, I know about the devil on your shoulder saying "go on, route me. You don't need four layers, you can do it in two. I bet if you just threaded this signal over here and moved that IC a little bit you could use one fewer via"
[22:48] <hallam> make the board smaller!
[22:49] <JimAM> so when your equipment crash lands, do you discover it based on signal strength or lat and long coords from some final transmission?
[22:49] <hallam> if it really crash lands it's difficult to find
[22:50] <edmoore> JimAM: usually you you can get close enough based on final transmission heard to then hear it on the ground and get the landing co-ordinates
[22:50] <hallam> if it lands as it's supposed to, you go to the predicted landing site (prediction based on coordinates transmitted during descent) and that's usually less than a mile from the actual site
[22:50] <hallam> right, that
[22:50] <edmoore> should that for some reason not work, you can just traingulate based on signal strength, yes
[22:51] <hallam> some people also fly mobile phones that SMS the landing lat/lon
[22:53] <JimAM> that sounds tricky
[22:53] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 60 (Operation timed out)
[22:53] <JimAM> finding a phone open enough to allow you to use its messaging api, and then developing software for it was well as the excess payload
[22:54] <sbasuita> JimAM, no it is easy to use phones
[22:54] <sbasuita> JimAM, a lot just act as modems
[22:54] <sbasuita> JimAM, the trickiest part is finding one with a real hardware serial interface
[22:55] <DanielRichman> JimAM, T68i makes it easy
[22:55] <DanielRichman> well.. to a certain extent. the soldering is a bit of a pain but could be worse
[22:55] <JimAM> the serial interface will allow you to send an SMS?
[22:55] <hallam> the downside is really weight rather than difficulty
[22:55] <hallam> yes
[22:55] <edmoore> using AT commands
[22:55] <edmoore> peasy sqeezy
[22:56] <edmoore> but rare on modern phones
[22:56] <DanielRichman> idd. the code is quite simple
[22:56] <DanielRichman> modern phones only have usb on the bottom (absolute pain).
[22:56] <sbasuita> JimAM, you can pretty much access all of the phone's functionality over the serial
[22:56] <JimAM> I was just about to ask, what might be the point in that in a phone
[22:56] <edmoore> scour ebay for old 6310 or similar - they're ideal
[22:57] Matt_APEX (n=chatzill@87.115.165.107) left irc: "ChatZilla 0.9.85 [Firefox 3.5/20090624025744]"
[23:02] <JimAM> so is it purely for the photography or is there any kind of data analysis you can do up there?
[23:03] <SpeedEvil> pics and comms pretty much.
[23:03] <edmoore> there's unsuprisingly plenty you can do that's interesting in an environment like that
[23:03] <SpeedEvil> yeah
[23:03] <SpeedEvil> but the sky is the limit :)
[23:04] <hallam> some people are working on autonomous gliders, suborbital or even orbital rockets launched from the balloon
[23:05] ShellEvil (n=user@tor/regular/SpeedEvil) joined #highaltitude.
[23:05] <JimAM> i wouldn't mind having a go at building a quad copter
[23:05] <ShellEvil> For various values of 'working'.
[23:06] <hallam> quads are fun
[23:06] <JimAM> make it autonomously stable, the one I saw in use looked precarious to fly
[23:07] <hallam> without some kind of controller onboard they are ~impossible to fly
[23:07] <edmoore> that's a fun control challenge
[23:07] <hallam> it just depends what you mean by "stable"
[23:07] <JimAM> just that it will attempt to hold its position
[23:08] <hallam> that's about the third level of complexity
[23:08] <hallam> 1 = hold a steady rate of rotation
[23:08] <JimAM> the first and second being?
[23:08] <hallam> 2 = hold a steady attitude
[23:08] <JimAM> really I've never heard that before
[23:08] <JimAM> who defined those
[23:08] <hallam> me, just now
[23:08] <JimAM> :)
[23:09] <hallam> the first of those makes it flyable by a human
[23:09] <JimAM> I was hoping you were going to refer me to a paper
[23:09] <hallam> the second makes it flyable by a monkey
[23:09] <JimAM> and your definition of the third then?
[23:10] <ShellEvil> hold a steady position
[23:10] <hallam> hold position, like you said
[23:10] <hallam> there's a 1/s term between each of those
[23:10] <JimAM> I see
[23:10] <hallam> (to first order - it's the higher order terms that make it interesting)
[23:10] herabot (n=herabot@92.29.73.106) left irc: Read error: 145 (Connection timed out)
[23:11] <hallam> remember, always mount a scratch monkey
[23:11] Nick change: hallam -> hallam|autonomou
[23:11] Nick change: hallam|autonomou -> hallam|autokite
[23:11] <JimAM> I have a pilot lined up don't worry
[23:11] <JimAM> I've owned him since I was about 14, he has goggles and a scarf and everything
[23:16] <edmoore> JimAM: if you want papers, check out the stanford quadcopter research
[23:16] <JimAM> ok will do thank you
[23:17] <edmoore> i think as long as you know what hallam means by '1/s', it's all pretty straightforward
[23:17] <JimAM> is there a particular reason you use gumstix or blackfin over more readily available micros like PIC or ARM
[23:18] <JimAM> he meant reciprocal yes?
[23:18] <edmoore> try making a software gps on a pic
[23:18] <edmoore> we do use arm for the basic stuff
[23:18] <edmoore> it does all the house keeping
[23:18] <edmoore> the gumstix runs stuff which needs much more horse power
[23:18] <edmoore> he didn't mean reciprocal, he meant an integrator
[23:18] <edmoore> pick up any introductory text on control theory and you'll be sorted
[23:19] <edmoore> 1/s being the representation of an integrator in the laplace space
[23:21] <JimAM> what are some of the more complex software routines you perform which require the horsepower?
[23:21] <hallam|autokite> edmoore: coincidentally the stanford quadrotor guy was visiting here this morning
[23:21] <hallam|autokite> he's a pretty bright chap
[23:21] <hallam|autokite> they use two dspics as their flight computer
[23:21] <edmoore> I can imagine
[23:22] <edmoore> JimAM: a software defined GPS, for example
[23:22] Action: edmoore passes conch to hallam|autokite
[23:22] <JimAM> I thought most GPSs offered a serial string?
[23:23] <ShellEvil> they do.
[23:23] <hallam|autokite> this isn't most GPSs
[23:23] <ShellEvil> They're not much good if you want to go >1000knots
[23:23] <ShellEvil> or for rocketry
[23:23] <edmoore> sometime an off the shelf gps isn't good enough for what you need to do
[23:23] <ShellEvil> for a number of reasons
[23:23] <hallam|autokite> which we needn't go into right now
[23:23] <JimAM> I see so you need more receiving hz?
[23:23] <ShellEvil> JimAM: no
[23:23] <hallam|autokite> the GPS receiver is implemented in software rather than hardware
[23:24] <ShellEvil> JimAM: there are hard limits built into 'all' GPSs
[23:24] <JimAM> oh ok
[23:24] <hallam|autokite> it's basically antenna, amplifier, ADC, computer
[23:24] <ShellEvil> JimAM: >60000ft and >1000knots - and you don't get a signal
[23:24] SpeedEvil (n=user@tor/regular/SpeedEvil) left irc: Connection timed out
[23:24] <JimAM> so you feed three analogue receivers into an ADC and do the triangulation and other computation yourself?
[23:25] <hallam|autokite> no, one receiver
[23:25] <hallam|autokite> it's a lot more complicated than triangulation
[23:25] <hallam|autokite> GPS is the cleverest thing that I've ever seen and understood
[23:26] <JimAM> more so than fourier transforms? -- I'm currently lost in translation reading about those
[23:26] <hallam|autokite> yes more so than them
[23:26] <JimAM> ouch
[23:26] <hallam|autokite> my receiver uses fourier transforms during acquisition, gives it an edge over the hardware receivers
[23:27] <ShellEvil> GPS uses time broadcast from each satellite.
[23:27] <hallam|autokite> the trimble website has a nice explanation
[23:27] <ShellEvil> The satellites also broadcast their orbits.
[23:27] <hallam|autokite> that makes it seem a lot easier than it is
[23:28] <ShellEvil> You take the time and the orbit - which gives you teh sats current position , and then work out where you must be - based on the speed of light
[23:28] <ShellEvil> and lots of maths
[23:29] <JimAM> where does fourier play a part in that? I imagined it was to receive data from multiple satellites simultaneously?
[23:29] <ShellEvil> It's to detect 'where' in the frequency range the satellites are.
[23:29] <ShellEvil> broadly.
[23:29] <hallam|autokite> it lets you do a circular cross-correlation in a parallel way, so you can test all 4092 possible codephases simultaneously
[23:30] <hallam|autokite> ShellEvil, nah, it's much more efficient to use FFTs to parallelise the codephase search than the freq search
[23:30] <ShellEvil> I was using frequency in its broadest and slightly inaccurate sense :)
[23:30] <hallam|autokite> heh
[23:30] <hallam|autokite> don't expect that to make sense until you've read about GPS for a few hours
[23:30] <ShellEvil> yeah.
[23:30] <ShellEvil> And even then.
[23:31] <hallam|autokite> it won't really make sense until you've coded it
[23:31] <ShellEvil> Much of hte stuff doesn't 'gel' untill you try to implement it
[23:31] <ShellEvil> at least on paper.
[23:31] <JimAM> well my first guess was codephase meant some sort of identifier, my second guess was phase of data transmission, ie. lat long, time,
[23:31] <ShellEvil> nope
[23:31] <ShellEvil> It's the 'offset' of the bitstream - 50bps - containing hte satellite info.
[23:31] <hallam|autokite> it's for the CDMA modulation
[23:32] <ShellEvil> the precise - to the nanosecond - offset - that's needed for accurate positioning.
[23:32] <hallam|autokite> ShellEvil, it's the offset of the PRN chipstream not the nav bits
[23:32] <JimAM> ah
[23:32] <hallam|autokite> nav bits are irrelevant to acquisition
[23:32] <hallam|autokite> except that they get in the way sometimes
[23:32] <ShellEvil> You however do need the offset to the navbits
[23:32] <ShellEvil> and I was again trying to simplify :)
[23:32] <hallam|autokite> sorry :P
[23:33] <ShellEvil> np
[23:33] <ShellEvil> I need to get my coding started in anger.
[23:33] <hallam|autokite> acquisition = "where" (in frequency space and codephase space, not physical position) are the satellites
[23:33] <hallam|autokite> tracking = what are they saying, and when did they say it
[23:33] <hallam|autokite> navigation = turning that information into a position solution
[23:34] <ShellEvil> I have some ideas to parallelise stuff on normal processors using SIMD/bitslicing techniques.
[23:34] <hallam|autokite> normal = x86?
[23:34] <ShellEvil> no - normal in this case = stm32
[23:34] <hallam|autokite> oh
[23:34] <hallam|autokite> yeah that would be sweet
[23:34] <ShellEvil> I think it can do it.
[23:34] <hallam|autokite> I'm just getting started using the STM
[23:35] <ShellEvil> It's not very trivial.
[23:35] <hallam|autokite> acq will be slow
[23:35] <ShellEvil> I'm unsure.
[23:35] <ShellEvil> I've a few other ideas for that that seem to work on paper.
[23:35] <ShellEvil> I need to reduce them to actual code, and backup my paper scribblings though.
[23:35] <hallam|autokite> why are you trying to do it - just the challenge?
[23:36] <ShellEvil> hallam|autokite: I want to sell it! :)
[23:36] <RocketBoy> nights
[23:36] <hallam|autokite> but to whom
[23:36] RocketBoy (n=Steve@217.47.75.8) left irc: "Leaving"
[23:36] <hallam|autokite> night Steve
[23:37] <ShellEvil> hallam|autokite: I have an idea for a STM32 based platform - that can do many things that I can sell on ebay. The GPS is just an add-on that I'm not really banking on working.
[23:37] <hallam|autokite> what application would this be preferred to a baseband ASIC
[23:37] <hallam|autokite> oh ok
[23:37] ChezaWho (n=chezawho@adsl-99-38-181-246.dsl.pltn13.sbcglobal.net) left irc: Read error: 60 (Operation timed out)
[23:37] <ShellEvil> If I can get GPS working on it - it'd mean I could easily add GPS to several other ideas - even low performance GPS is useful sometimes.
[23:38] <ShellEvil> And worstcase, if I can't - I can simply do the cheat - and sample a seconds GPS and jam it on flash for later reduction.
[23:38] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[23:38] <hallam|autokite> yeah, the geotate thing
[23:38] <hallam|autokite> I really like that
[23:39] <edmoore> JimAM: If you're working on the Fourier Transform, you'll find the laplace transform (the 1/s business) a fairly simple conceptual leap. Infact for fun perform the final value theorem on a laplace transform and see what you get...
[23:39] <ShellEvil> And also some things that I haven't seen being done - for example waking up for a few tens of milliseconds every 30s to maintain lock.
[23:39] <ShellEvil> For fun?
[23:39] Action: ShellEvil prefers to sandpaper his genitalia.
[23:39] <hallam|autokite> I'll be impressed if you can propagate for 30s
[23:40] <ShellEvil> hallam|autokite: It depends on the clock of course.
[23:40] <JimAM> when I get to that stage I'll certainly look into it, I have a lot of reading to do on the subject
[23:40] <ShellEvil> I've forgotten most of the maths I once knew.
[23:41] <ShellEvil> I need a refresher.
[23:41] <hallam|autokite> I would have thought you'd be better off just slapping a ublox AMY on there, it will be smaller than your front end
[23:42] <hallam|autokite> for me, software gps is all about unusual applications like high dynamics, or differential or carrier phase
[23:42] <edmoore> as soon as I graduate i can start forgetting it all too
[23:42] <ShellEvil> hallam|autokite: Perhaps. It's basically an interesting possibility that I'm not relying on. I'm assuming for a first cut I'll be doing geotate-like stuff.
[23:42] <hallam|autokite> that's fun
[23:42] <hallam|autokite> if you want a nice easy product
[23:43] <hallam|autokite> build a bare-minimum geotate-style logger
[23:43] <hallam|autokite> very tiny
[23:43] <JimAM> are you studying electronics also edmoore?
[23:43] <ShellEvil> hallam|autokite: I'm not quite going for bare minimum - but I'm aiming at the order of 2*1cm or so - so not large
[23:43] <hallam|autokite> ok, that size is fine
[23:43] <edmoore> well, control theory and signal processing by title, but really than manifests itself in practice in embedded stuff
[23:43] <hallam|autokite> but all it does is log the IF data, and upload it to your PC when you plug it into USB
[23:44] <ShellEvil> yeah.
[23:44] <hallam|autokite> so many applications for that
[23:44] <hallam|autokite> you could sell it for $300 and they'd go like hotcakes
[23:44] <hallam|autokite> with postprocessing you can deal with ~any dynamics
[23:44] <edmoore> JimAM: may I ask what's your background?
[23:45] <JimAM> I have a bachelors in industrial computing -- which included embedded
[23:45] <ShellEvil> hallam|autokite: yeah - I'm aiming at various niches that are poorly served ATM.
[23:45] <hallam|autokite> it's perfect for amateur rockets, gliders (my friend is doing dynamic soaring, getting his R/C models up over 500mph)
[23:45] <hallam|autokite> GPS research in education
[23:45] <JimAM> we programmed in one particular environment though on a motorola 68000
[23:45] <hallam|autokite> they're not making the GN3S any more
[23:46] <ShellEvil> hallam|autokite: yeah - stupid stuff like just logging the bitstream to a SD.16G of microSD will go a long way
[23:46] <hallam|autokite> right
[23:46] <JimAM> I know more about scheduling theory and algorithms and formal verification
[23:46] <hallam|autokite> all the amateur rocket guys would love the idea of something that tiny that can give them not just perfect apogee information but the whole flightpath, acceleration, the works
[23:46] <hallam|autokite> I wish I had time to productize it myself
[23:46] <JimAM> I dont have a background in calculus or electronics sadly
[23:47] <JimAM> and apart from that I am interested in aviation
[23:47] <ShellEvil> hallam|autokite: Annoyingly I've been doing rather more in the garden than electronically recently.
[23:47] <ShellEvil> hallam|autokite: _lots_ of membrane put down.
[23:47] <hallam|autokite> I saw the photo
[23:47] <ShellEvil> And rotavating stuff, and ...
[23:48] Laurenceb (n=laurence@host86-159-200-131.range86-159.btcentralplus.com) joined #highaltitude.
[23:48] <hallam|autokite> wouldn't it be great if there were 36 hours in the day
[23:48] <ShellEvil> Slowly getting more sane.
[23:48] <hallam|autokite> hi sarah
[23:48] <Laurenceb> hi
[23:48] Action: Laurenceb is knackered
[23:48] <Laurenceb> also metposts are a rip off
[23:48] <ShellEvil> Laurenceb: Indeed.
[23:49] <edmoore> JimAM: sorry got called away. so sounds like you'll pick this stuff up pretty easily
[23:49] <Laurenceb> £5 each for some pressed and spot welded steel
[23:49] <edmoore> can't recommend enough tring to code a discrete fourier transform, say in octave
[23:49] <edmoore> it makes it all click
[23:49] Action: Laurenceb coded it on a ti86
[23:49] <Laurenceb> along with fft
[23:50] <edmoore> you intuitively understand that it's a sort of correlation with a whole set of wee rotating phasors
[23:50] <Laurenceb> well thats one way of looking at it
[23:50] <Laurenceb> hallam|autokite: you out testing?
[23:51] <hallam|autokite> actually no wind today
[23:51] <Laurenceb> ah that kind of screws that plan
[23:51] <hallam|autokite> well, there is wind
[23:51] <hallam|autokite> but not enoughj
[23:52] <hallam|autokite> the boundary layer is a bitch like that
[23:52] <Laurenceb> how do you land it when the wind drops?
[23:52] <hallam|autokite> carefully
[23:52] <ShellEvil> Was really quite windy up here.
[23:52] <hallam|autokite> you basically just dive it and then flare
[23:52] <Laurenceb> hmm
[23:52] <Laurenceb> that would be the expense
[23:52] <Laurenceb> and air exclusion zone
[23:52] <Laurenceb> personally I prefer CSP
[23:52] <ShellEvil> R/C kite?
[23:52] <ShellEvil> I mean wire
[23:52] <hallam|autokite> well in the real app, we can send power up the wire
[23:52] <ShellEvil> computer
[23:52] <edmoore> i remember learning how to autorotate my RC heli
[23:53] <hallam|autokite> and use turbines as props
[23:53] <edmoore> that was an expensive day
[23:53] ChezaWho (n=chezawho@adsl-99-38-181-246.dsl.pltn13.sbcglobal.net) joined #highaltitude.
[23:53] <Laurenceb> yeah
[23:53] <JimAM> eventually I wanted to do it for a continuous time signal to create a spectograph, but I'd be pleased enough with the octave idea!
[23:53] <Laurenceb> octave is good
[23:53] <Laurenceb> I wrote a rocket launch simulator in it
[23:53] <Laurenceb> - for n prize
[23:54] <edmoore> discrete fourier first (you get the pricipal) then try fast fourier. trying to figure out what is going on straight from the fast fourier transform is tricker. it's a gorgeous algorithm though
[23:54] <edmoore> i can't really visualise in my head the radix-4 algorithm - I just write the code and smile when it seems to work.
[23:56] <Laurenceb> http://www.bsdmicrorc.com/index.php?productID=703 <- looks good for a cmg
[23:57] <hallam|autokite> I'm working with OpenBLDC
[23:57] <hallam|autokite> it's not ready yet
[23:57] <hallam|autokite> but when it is, you'll want to use it
[23:58] <hallam|autokite> http://www.goldmine-elec-products.com/prodinfo.asp?number=G17270 <-- porn for Laurenceb and ShellEvil
[23:58] <ShellEvil> I've got lots of those.
[23:58] <ShellEvil> Disk motors out of 130MB drives
[23:58] <ShellEvil> PCMCIA
[23:59] <hallam|autokite> oh man someone just crashed a quadrotor into the cat
[23:59] <hallam|autokite> I think the cat won
[00:00] --- Thu Aug 20 2009