highaltitude.log.20080812

[00:07] <Laurenceb> but with an integration time of a minute or so and fluctuation of ~ +-10 degrees it looks like you can get the signal to noise ratio above 10 or so
[00:08] <SpeedEvil> I just question trying to pull so much data out of a derived measurement.
[00:08] <Laurenceb> you'd start off using the wind speed recorded before cutdown
[00:08] kamaji (n=kamaji@handtomouse.demon.co.uk) left irc: "leaving"
[00:09] <Laurenceb> then slowly give the derived wind vector a higher weighting as you descend
[00:09] <SpeedEvil> I'd favour something nice and simple to work out direction - take the northings and eastings, and feed into high Q filter
[00:09] <SpeedEvil> Look at the relative phases to work out the current pointing.
[00:09] <Laurenceb> thats almost what I'm saying, but its not quite as wimple as that
[00:09] <SpeedEvil> You'd want to do clever things with past measurements to make it so that the number isn't invalidated over turns, but not really that clever.
[00:10] <Laurenceb> the tricky thing is keeping SRAM use down
[00:10] <SpeedEvil> how much do you have?
[00:11] <Laurenceb> most of the code would have to run in the kalman filter ISR
[00:11] <Laurenceb> 1K
[00:11] <Laurenceb> approx 500 spare ATM but I plan to add a few bits and pieces like I2C eeprom support
[00:12] <SpeedEvil> 128K*8 I2C MRAM
[00:12] <Laurenceb> haha
[00:12] <Laurenceb> does it even exist?
[00:12] <SpeedEvil> yes
[00:12] <Laurenceb> amazing, where?
[00:12] <SpeedEvil> http://www.freescale.com/webapp/sps/site/homepage.jsp?nodeId=015424
[00:13] <SpeedEvil> hmm - I'm sure I've seen I2C somewhere
[00:14] <Laurenceb> madness
[00:14] <Laurenceb> but no I2C there
[00:14] <Laurenceb> I was just thinking of some microchip I2C eeprom, I've used it before
[00:15] <SpeedEvil> http://www.ramtron.com/products/nonvolatile-memory/
[00:15] <SpeedEvil> aha
[00:16] <Laurenceb> omg chestfield
[00:16] <Laurenceb> *chesterfield even
[00:17] <Laurenceb> theres a reseller there
[00:17] <Laurenceb> crazyness, thats just a few miles away from me
[00:18] <SpeedEvil> 32k*8 parts - $7 or so.
[00:18] <SpeedEvil> Not bad at all.
[00:18] <Laurenceb> thats pretty cool, I'll have to stick some MRAM on now, if just for the 1337 factor
[00:18] <SpeedEvil> Though I'm doing sillyness with SDRAM.
[00:18] <SpeedEvil> Bit-banging it :)
[00:19] <Laurenceb> but it still couldnt be used very efficiently
[00:19] <SpeedEvil> It's not actually that bad in my case, I can clock a block in quite fast - 10M/second or so.
[00:19] <Laurenceb> mucking about with the I2C registers whenever you want a byte will take all day
[00:20] <SpeedEvil> yeah
[00:20] <Laurenceb> I think an AVR can go almost that fast, but its still a bottleneck
[00:20] <Laurenceb> the SPI certainly can go to F_CPU/2
[00:20] <Laurenceb> =10MHz
[00:21] <SpeedEvil> Read more of the STM32 datasheet today.
[00:21] <SpeedEvil> Diddn't realise on first read how low it can clock.
[00:21] <Laurenceb> groo sounds like a tedious read
[00:21] <SpeedEvil> Down to 125KHz - at which it uses 0.5mA at 2V
[00:21] <Laurenceb> wow
[00:22] <Laurenceb> I have the atmega168 datasheet next to me - fills an entire folder
[00:22] <SpeedEvil> A 32 bit processor, 20K RAM, 32-256K flash, ...
[00:23] <SpeedEvil> Not Too spendy.
[00:24] <Laurenceb> I'll be getting familiar with it next year
[00:25] <Laurenceb> whats the best way to implement a digital filter?
[00:26] <SpeedEvil> Laurenceb: get someone in india to do it?
[00:26] <Laurenceb> lol
[00:26] <SpeedEvil> Sorry - haven't done much digital filtration.
[00:26] <SpeedEvil> memory is going to be annoying.
[00:29] <Laurenceb> have to filter things as you go along, not use buffers or arrays anywhere
[00:35] <akawaka> IIR or FIR
[00:35] <akawaka> low memory requirements
[00:35] <Laurenceb> I guess just multiply by sin and cos terms
[00:35] <Laurenceb> your only looking for one frequency
[00:36] <Laurenceb> in there a fast "pseudosinusiodal" function?
[00:36] <Laurenceb> rather than using a lookup table
[00:37] <akawaka> depends on how fast
[00:37] <akawaka> you can get away with a pretty small lookup table
[00:38] <Laurenceb> well avr-gcc benchmarks has floating point multiplication approx 150 clock cycles, but trig functions are over 1000
[00:38] <akawaka> what are you trying to do?
[00:38] <Laurenceb> so something approximate that has a clock cycle count somewhere inbetween would be nice
[00:38] <Laurenceb> wobble a servo approximately sinusoidally
[00:39] <akawaka> how much memory can you afford?
[00:39] <Laurenceb> I guess a power series expansion would work
[00:39] <Laurenceb> maybe 50 bytes
[00:39] <akawaka> you could do a decent lookup table in there i suppose
[00:40] <akawaka> 2 degree accuracy
[00:40] <Laurenceb> yeah, probably simpler
[00:41] <akawaka> 45 entries
[00:43] <SpeedEvil> Wouldn't a simple triangle do?
[00:44] <Laurenceb> hmm maybe
[00:45] <Laurenceb> oh well I'd better get some sleep, cay all and thanx for the help
[00:45] <Laurenceb> I'll put a page on the wiki with some diagrams if I have time
[00:45] Laurenceb (n=Laurence@host86-133-69-206.range86-133.btcentralplus.com) left irc:
[00:45] <akawaka> SpeedEvil: so why the 32-bit processor?
[00:51] <SpeedEvil> akawaka: Why not?
[00:52] <SpeedEvil> akawaka: it's a flexible core of several things that I intend to sell on ebay.
[00:52] <akawaka> the M3 is awesome
[00:52] <SpeedEvil> akawaka: I figure that instead of trying to cram the big ones into a restricted micro, I can probably live with a 2-3 quid more expensive board, and easier design.
[00:53] <akawaka> what kind of things are you making?
[00:53] <SpeedEvil> akawaka: from ships and shoes and sealing wax, to cabbages and kings.
[00:53] <SpeedEvil> Though more seriously, I'm basically at the lots of ideas stage, I want to get a nice working board.
[00:54] <akawaka> gotcha
[00:54] <SpeedEvil> That is flexible enough to let me try stuff.
[00:54] <SpeedEvil> then I'll work out detailed designs for each.
[00:54] <akawaka> i feel kind of limited not being able to work with surface mount stuff
[00:55] <SpeedEvil> akawaka: why, have your hands been ripped off by a furious bear?
[00:56] <akawaka> it just seems really difficult
[00:56] <SpeedEvil> SMD really isn't that scary.
[00:56] <SpeedEvil> In some ways it's easier.
[00:56] <SpeedEvil> NO MORE DRILLING!
[00:56] <SpeedEvil> Yay!
[00:56] <akawaka> drilling?!
[00:57] <SpeedEvil> I was assuming you're doing PCBs
[00:59] <akawaka> no, just perfboard and stuff
[00:59] <akawaka> do you make your own pcbs?
[01:02] <SpeedEvil> It's not really hard.
[01:02] <SpeedEvil> http://max8888.orconhosting.net.nz/pcbs.htm
[01:04] <akawaka> fuck
[01:04] <akawaka> really?
[01:04] <SpeedEvil> really
[01:05] <akawaka> that doesn't seem so bad
[01:05] <SpeedEvil> http://www.qsl.net/ve2emm/pcb/pcb2e.html
[01:05] <akawaka> now i see what you mean about drilling:)
[01:06] <SpeedEvil> http://openhardware.net/Misc_Stuff/ToasterSMD/
[01:06] <SpeedEvil> too
[02:12] <natrium42> drilling is damn annoying
[02:12] <natrium42> i need a CNC drill
[02:44] jnd (n=jenda@unaffiliated/jnd) left irc: Read error: 110 (Connection timed out)
[03:00] natrium42 (n=alexei@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc:
[03:03] natrium42 (n=natrium4@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[03:40] natrium42 (n=natrium4@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left #highaltitude.
[03:44] natrium42 (n=alexei@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[04:20] phatmonkey (n=ben@scooby.firshman.co.uk) left irc: Remote closed the connection
[06:52] akawaka (n=akawaka@external.treyarch.com) left irc: Read error: 110 (Connection timed out)
[07:42] borism (n=boris@195-50-197-174-dsl.krw.estpak.ee) joined #highaltitude.
[07:51] Ebola (n=Ebola@unaffiliated/ebola) joined #highaltitude.
[07:55] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[07:59] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) joined #highaltitude.
[08:27] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) left irc:
[08:35] akawaka (n=akawaka@cpe-76-173-152-142.socal.res.rr.com) joined #highaltitude.
[08:45] Shanuson1 (n=Peter@p54A94CB0.dip.t-dialin.net) joined #highaltitude.
[08:49] edmoore (n=edmoore@smtp.vorticity-systems.com) joined #highaltitude.
[09:04] Shanuson (n=Peter@p54A97C14.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out)
[09:16] akawaka (n=akawaka@cpe-76-173-152-142.socal.res.rr.com) left irc: Read error: 110 (Connection timed out)
[10:20] icez (n=icez@ip68-98-34-247.ph.ph.cox.net) left irc: "Lost terminal"
[10:26] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) joined #highaltitude.
[10:41] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) left irc: Remote closed the connection
[10:44] Nick change: Shanuson1 -> Shanuson
[11:05] <edmoore> natrium42: ping for when you're up
[11:07] jcoxon (n=jcoxon@85-210-73-114.dsl.pipex.com) joined #highaltitude.
[11:10] <jcoxon> morning edmoore
[11:10] <edmoore> morning jcoxon
[11:10] <edmoore> results come?
[11:10] <jcoxon> yeah
[11:10] <jcoxon> all good
[11:11] <jcoxon> now i can relax
[11:11] <jcoxon> someone at hte door bbl
[11:11] <jcoxon> oh maybe not
[11:11] <jcoxon> edmoore, i've been building a laptop tablet
[11:12] <edmoore> oooh
[11:13] <jcoxon> but i'm having difficulty on deciding how to do the inputs
[11:13] <jcoxon> touchscreen kits are expensive
[11:13] <edmoore> yeah
[11:13] <jcoxon> at the moment i'm using a wireless mouse
[11:14] <edmoore> well I'm playing with FPGAs at the moment. I am totally totally taken with software defined radio
[11:14] <edmoore> it's ludicrously powerful
[11:14] <jcoxon> :-)
[11:14] <edmoore> james, have a look at this, you will love it
[11:15] <edmoore> drag the tuning and bandwidth and USB/ssb to your heart's content - you will see just how much more powerful it is as a technology that what we've been using
[11:15] <edmoore> http://websdr.ewi.utwente.nl:8901/
[11:15] <edmoore> and that's just a basic implimentation
[11:15] <gordonjcp> jcoxon: can I bug you about AFSK telemetry at some stage?
[11:16] <edmoore> so jcoxon - you have that, decoding 2 balloons at once from one telem stream is easy
[11:16] <edmoore> or 200 balloons all at once, even
[11:17] <edmoore> with just the one radio
[11:17] <edmoore> send up 10 radiometrix modules, all 10mW, break no laws, but have massive badnwidth
[11:17] <edmoore> although very poor show on the etiquette from
[11:17] <edmoore> you'd upset other hams
[11:17] <edmoore> whilst being perfectly legal
[11:17] <gordonjcp> m
[11:18] <jcoxon> gordonjcp, ooh yes, i'll have a go later
[11:18] <gordonjcp> what about trying to work amsats from the balloon?
[11:18] <gordonjcp> ;-)
[11:18] <jcoxon> edmoore, so where does the fpga go in all this?
[11:18] <edmoore> jcoxon: what do you think? I'm salivating. FPGAs are the future
[11:18] <edmoore> ok
[11:18] <jcoxon> i'm confused
[11:18] <jcoxon> hehe
[11:18] <edmoore> so in software defined radio you have a really really high speed ADC
[11:18] <edmoore> to directly sample what comes off the antenna
[11:19] <edmoore> you then use the FPGA to down-convert the signal to the point that the data rate is low enough to be sent over USB, where the computer (capable of 2 billion floating point ops a second) can do the fft and further processing
[11:19] <edmoore> but the fpga does the front end, basically
[11:19] <jcoxon> right
[11:19] <gordonjcp> well hang on
[11:20] <jcoxon> so the fpga converts whats coming from the radio into usuable stuff
[11:20] <gordonjcp> couldn't you do much the same thing with the DAC in a video capture card, and an IF of some sort?>
[11:20] <edmoore> gordonjcp is probably about to give a better explanation
[11:21] <edmoore> well the FPGA does DDC in this case, and over a massive range
[11:21] <edmoore> but yeah, you can make more speciic implimentations
[11:21] <gordonjcp> like, make a rather broadband IF to shift (say) 430-435MHz to 1-6MHz, sample with your video capture card, and decode?
[11:21] <edmoore> gordonjcp: yes indeed
[11:23] <edmoore> the advantage of doing it this way is the flexibility - you want to try and decode freeview one day, radio astronomy the next, play with rfid tags the next - all with the same pcb
[11:23] <gordonjcp> jcoxon: http://www.gjcp.net/~gordonjcp/co55.mp3 <- warning - loud and noisy
[11:24] <gordonjcp> jcoxon: that was from Oscar-55 quite low in the sky so fairly weak
[11:24] <gordonjcp> does that sound about right?
[11:24] <jcoxon> ummmm, i don't know!
[11:24] <jcoxon> what sort of transmission do you want?
[11:25] <jcoxon> my packet radio position beacon?
[11:25] <gordonjcp> well it's apparently 1200 baud AFSK, which is what I thought you were using
[11:25] <jcoxon> hmmmm mine sounds different
[11:25] <jcoxon> its a lot shorter as well
[11:25] <gordonjcp> oh, ok
[11:25] <jcoxon> i'll record some now
[11:25] <gordonjcp> cool
[11:29] <edmoore> bbiab, computer has been on for 4 months and is beginning to slow slightly
[11:31] edmoore (n=edmoore@smtp.vorticity-systems.com) left irc:
[11:34] edmoore (n=edmoore@smtp.vorticity-systems.com) joined #highaltitude.
[11:36] <edmoore> back
[11:37] <edmoore> I had a 10GB swap file
[11:44] <jcoxon> not bad
[11:57] <edmoore> jcoxon: http://www.wired.com/science/discoveries/news/2006/06/70933
[11:57] <edmoore> a decent intro
[11:59] <jcoxon> thansk
[11:59] <jcoxon> expensive bit of ket
[11:59] <jcoxon> well not proportionally
[12:01] Robyn (n=Ebola@149.254.192.192) joined #highaltitude.
[12:02] <edmoore> That's why you built it yourself
[12:02] <edmoore> far more fun
[12:02] <edmoore> hi Robyn
[12:02] Ebola (n=Ebola@unaffiliated/ebola) left irc: Nick collision from services.
[12:02] Nick change: Robyn -> Ebola
[12:04] jcoxon_ (n=jcoxon@85-210-73-114.dsl.pipex.com) joined #highaltitude.
[12:04] jcoxon (n=jcoxon@85-210-73-114.dsl.pipex.com) got netsplit.
[12:04] <jcoxon_> wow something went wrong there
[12:04] Nick change: jcoxon_ -> jcoxon
[12:04] Possible future nick collision: jcoxon
[12:04] <jcoxon> gordonjcp, http://www.pegasushabproject.org.uk/wiki/doku.php/missions:haps:haps-1#radio
[12:04] <jcoxon> there is a example mp3
[12:07] <gordonjcp> jcoxon: yeah, multimon decodes that
[12:07] <jcoxon> excellent
[12:07] <jcoxon> the data is all good?
[12:07] <gordonjcp> jcoxon: I'm reasonably happy that I could probably get *something* off the balloon ;-)
[12:08] <jcoxon> excellent
[12:08] <gordonjcp> AFSK1200: fm HAPS-0 to BASE-0 UI^ pid=F0
[12:08] <gordonjcp> GOLIATH,4,1,105538.00,0.000000,0.000000,,,
[12:08] <gordonjcp> AFSK1200: fm HAPS-0 to BASE-0 UI^ pid=F0
[12:08] <gordonjcp> GOLIATH,6,1,105603.00,0.000000,0.000000,,,
[12:08] <gordonjcp> (sorry for the floodage)
[12:08] <jcoxon> oops i thought i had a lock
[12:09] <jcoxon> gordonjcp, what OS do you use?
[12:09] <gordonjcp> jcoxon: Linux
[12:09] <jcoxon> on i386?
[12:09] <gordonjcp> yup
[12:09] <jcoxon> hmmmm
[12:09] Action: jcoxon has a crazy plan
[12:09] <gordonjcp> yay, I like crazy plans!
[12:10] <jcoxon> wine + truetty + dxlog + some scripts to upload data...
[12:10] <jcoxon> the reason i say truetty is that it works well on wine
[12:10] <jcoxon> actually
[12:10] <jcoxon> there would be a suitable linux version
[12:11] <gordonjcp> well
[12:11] <jcoxon> using multimon, logging + upload
[12:11] <gordonjcp> I'm planning on hacking on multimon to let me do that
[12:11] <jcoxon> bingo
[12:11] <gordonjcp> also maybe libax25
[12:12] <jcoxon> thats what i'm doing on the gumstix
[12:12] <jcoxon> using ax25 and beacon to just downlink the data
[12:14] <gordonjcp> oh right
[12:15] <gordonjcp> how does the gumstix transmit afsk, just soundmodem?
[12:16] <jcoxon> yeah its soundmodem
[12:16] <gordonjcp> cool
[12:16] <gordonjcp> I've not had a lot of luck getting that going yet
[12:16] <jcoxon> nightmare to cross compile but works well
[12:17] <gordonjcp> all the howtos seem to be out of date
[12:17] <jcoxon> yeah
[12:17] <jcoxon> one sec i'll grab my conf and scripts for you
[12:17] <gordonjcp> istr Fred van Kempen was heavily involved in it, but he's dropped off the face of the world
[12:17] <gordonjcp> rumours abound, but you know how it is
[12:31] <jcoxon> hmmm perhaps we could add another variable to the tracker
[12:31] <jcoxon> which is which base station it was received at
[12:32] <jcoxon> so that we can all upload to the tracker
[12:41] <gordonjcp> yup
[13:05] Action: gordonjcp idly wonders how far away the balloon launch sites are
[13:05] <gordonjcp> about 500km seem about right?
[13:05] <gordonjcp> I finally found a practical use for SOHCAHTOA
[13:07] <gordonjcp> quick back-of-an-envelope calculation, at 500km horizontal distance and 32km alt
[13:08] <gordonjcp> my slant range is a bawhair over 501km, and elevation is around 31 degrees
[13:09] <gordonjcp> sound about right?
[13:10] <SpeedEvil> No.
[13:11] <SpeedEvil> Assuming a flat earth, 31km/500km is about .06 radians, or 3-4 degrees
[13:12] <gordonjcp> hm
[13:13] <gordonjcp> oh well, that's dyscalculia for you
[13:13] <gordonjcp> if it's that close to the horizon I'll need to go somewhere quiet with clear views to the south
[13:27] dusty (n=dusty@pdpc/supporter/sustaining/dusty) joined #highaltitude.
[13:35] borism (n=boris@195-50-197-174-dsl.krw.estpak.ee) left irc: Read error: 145 (Connection timed out)
[13:38] <edmoore> a mountain
[13:40] <gordonjcp> basically
[13:40] <gordonjcp> fortunately, I'm in Scotland, so mountains are easy
[13:45] <jcoxon> gordonjcp, i'll put a cycle of morse code as well - that should be easier to at least pick up
[13:45] <gordonjcp> cool
[13:51] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) joined #highaltitude.
[13:52] <SpeedEvil> For tomorrow!
[13:52] <SpeedEvil> http://www.sciencedaily.com/releases/2008/08/080811094611.htm
[13:53] <SpeedEvil> (atom thick balloon)
[13:55] <gordonjcp> incidentally who's the admin for the ukhas site?
[13:55] <gordonjcp> you've got *shitloads* of spam...
[13:55] <gordonjcp> http://www.ukhas.org.uk/news/28/pong-1-found
[14:11] <jcoxon> gordonjcp, oh yeah thats me
[14:12] <jcoxon> really need to sort that out
[14:12] <jcoxon> reckon i'll just turn off comments
[14:13] <SpeedEvil> No easy way to require an ack from the emai
[14:13] <SpeedEvil> l?
[14:14] <jcoxon> might as well turn it off, no one ever leaves a commetn
[14:14] <SpeedEvil> :(
[14:18] <jcoxon> wow there are 10700 spam posts
[14:18] <gordonjcp> jcoxon: akismet
[14:27] Robyn (n=Ebola@149.254.192.192) joined #highaltitude.
[14:27] Ebola (n=Ebola@unaffiliated/ebola) left irc: Nick collision from services.
[14:27] Nick change: Robyn -> Ebola
[15:02] phatmonkey (n=ben@scooby.firshman.co.uk) joined #highaltitude.
[15:04] Ebola (n=Ebola@unaffiliated/ebola) left irc: "Leaving"
[15:06] Ebola (n=Ebola@unaffiliated/ebola) joined #highaltitude.
[15:48] <edmoore> it's getting kinda dramatic out me window
[15:51] simon__ (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[15:51] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: Network is unreachable
[16:34] robert1971 (n=user@212.183.134.66) joined #highaltitude.
[16:34] <robert1971> Hi all
[16:35] <robert1971> Made it to Amsterdam :0 .oO
[16:38] <edmoore> congrats
[16:38] <edmoore> tired?
[16:49] natrium42 (n=alexei@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 110 (Connection timed out)
[16:49] natrium (n=alexei@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[16:53] Ei5GTB_ (n=Paul@78.16.10.117) joined #highaltitude.
[16:54] <robert1971> Yep and a sore arse from sitting on the saddle for a hundered miles. It's been a year since I last rode the bugger
[16:55] <robert1971> Out of interest i was in a place called Noordwijk today which is home of the European Space Agency
[16:56] <robert1971> I popped in to lool at their space exhibition and noted that a couple of old communications satelites they had kicking around had standard RS232 port on them
[16:56] <robert1971> lool = look (and I haven't even smoked any pot)
[16:58] <robert1971> They had gold contacts but they even had the screw in connectors on each side of the port. I was tempted to power it up and plug my computer in to see what was being sent out. Funny if it was a linux login prompt
[17:01] edmoore (n=edmoore@smtp.vorticity-systems.com) left irc:
[17:07] <soneil_> amsterdam made me hate bicycles heh
[17:08] <soneil_> when it feels like you're the only person in the city that's not on one, you're constantly watching three directions for the next one to mow you off the path
[17:10] EI5GTB (n=Paul@213-202-133-145.bas502.dsl.esat.net) left irc: Read error: 110 (Connection timed out)
[17:11] <jcoxon> woohoo, just found canon A60 on ebay with a cracked screen - perfect
[17:12] <jcoxon> not very good resolution but better then a webcam
[17:13] <natrium> jcoxon, nice
[17:13] <jcoxon> only 3.99
[17:13] <soneil_> score
[17:13] <jcoxon> quite heavy though
[17:13] <jcoxon> 215g without batteries of cf card
[17:13] <jcoxon> hmmm
[17:14] <soneil_> I wonder how much you can strip it without breaking lightseal
[17:14] <soneil_> it's not like you need it to be pretty
[17:15] <jcoxon> indeed
[17:17] <gordonjcp> surely being a digital camera the optical block will be fairly well enclosed?
[17:17] <jcoxon> they usually are
[17:18] Action: jcoxon has disassembled a lot of cameras
[17:27] <SpeedEvil> Dust is a nightmare to them.
[17:27] <SpeedEvil> So they tend to be well sealed.
[17:31] <natrium> i have tried repairing smashed lens assembly on my A70
[17:31] <natrium> didn't work out too well :P
[17:31] <SpeedEvil> I have a Dimage X1 - well two of - one with a workin cam, one with a working display but buggered dock connector.
[17:32] <SpeedEvil> My experiences with opening lenses haven't generally worked out well
[17:35] <robert1971> natrium I have been meaning to ask was you camera damaged beyond repair when your payload hit the deck? I'm talking about the video on your website in case you have done some other launches
[17:35] <SpeedEvil> That's the other issue.
[17:35] <natrium> not the flight2 one
[17:35] <SpeedEvil> Pull bits off the camera, and it'll get less robust.
[17:36] <natrium> on flight1 i smashed my A70... it still worked, but lens needed help to extend :)
[17:36] <robert1971> looked like it had a hard imact on flight2
[17:36] <natrium> yah, but camera was attached in a more flexible way
[17:36] <robert1971> #impact
[17:36] <natrium> only via the servo
[17:36] <natrium> so the servo axle bent
[17:36] <natrium> but camera was fine
[17:37] <natrium> i guess rigidity isn't always good :D
[17:37] <robert1971> I have gone for an aluminium tube on the otherside for some extra support
[17:37] <natrium> yeah, just have pieces of styro shield the camera
[17:37] <natrium> by extending over it
[17:38] <SpeedEvil> Or bubble-wrap
[17:38] <natrium> mine were too short
[17:38] <natrium> lol SpeedEvil :P
[17:38] <robert1971> Good point and that insulation should help too
[17:39] <robert1971> Not bubble wrap though as the bubbles will shrink
[17:39] <SpeedEvil> expand
[17:39] <robert1971> or pop even
[17:39] <SpeedEvil> Actually, I doubt it.
[17:39] <robert1971> The cycling has got to me
[17:39] Action: SpeedEvil does some calibrated experiments
[17:40] <robert1971> In a vacum chamber? Candle in a pot with the stopper on?
[17:40] <SpeedEvil> It looks like burst pressure may be 2atm or so
[17:40] <SpeedEvil> No.
[17:40] <SpeedEvil> I took a nearby bit of bubble wrap, and estimated the force to pop a bubble.
[17:41] <robert1971> How much presssure is 1atm in netwons/cm2
[17:41] <SpeedEvil> 1atm = 10^5Pa - 10^5N/m^2
[17:42] <SpeedEvil> = 1Kg/cm^2
[17:42] <robert1971> I hope you googled that
[17:42] <SpeedEvil> no
[17:42] <robert1971> I thought not
[17:43] <robert1971> Attaching a servo to a camera is a tricky task
[17:44] <robert1971> I ended up making an aluminium frame and screing it on with tiny bolts
[17:44] <robert1971> //s/screwing
[17:44] Ebola (n=Ebola@unaffiliated/ebola) left irc: "home time"
[17:44] <robert1971> /s/screing/screwing
[17:45] <SpeedEvil> hotmelt++
[17:45] <jcoxon> SpeedEvil, oh bubble wrap pops
[17:45] <jcoxon> i've done it
[17:45] <robert1971> Experiance... Counts for a lo
[17:45] <robert1971> lot
[17:45] <SpeedEvil> jcoxon: large ones?
[17:45] <SpeedEvil> jcoxon: or small ones.
[17:45] <jcoxon> small ones
[17:46] <jcoxon> we left a bit of bubble wrap in one of the payload cases when i was flying ZPs in Africa
[17:46] <SpeedEvil> Some definately (based on ground-based tests) will not pop.
[17:46] <SpeedEvil> As it's really robust and thick
[17:46] <jcoxon> perhaps rubbish bubble wrap
[17:46] <jcoxon> its a bit like closed cell foam
[17:46] <jcoxon> that goes a bit funny at very low pressures
[17:48] <SpeedEvil> I suppose it won't be less insulative when burst
[17:48] <jcoxon> bubble wrap
[17:48] <jcoxon> ?
[17:48] Action: SpeedEvil wonders if EPP would expand at altitude.
[17:48] <SpeedEvil> yes
[17:48] <SpeedEvil> As long as the layers are not compressed
[17:49] <jcoxon> yeah, probably
[17:50] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) joined #highaltitude.
[17:50] <robert1971> I think i'm going to settle for open cell foam and some foil for the camera
[17:51] <robert1971> Nice gold foil. Anyone looked into solar pannels or is there nothing like enough power in the surface area of a payload?
[17:52] <SpeedEvil> there is plenty of power
[17:52] <SpeedEvil> But.
[17:52] <robert1971> I have no experience of photovoltic cells or what ever you call them
[17:52] <SpeedEvil> Li-po batteries are lightweight.
[17:52] <SpeedEvil> And if your flight is 3h or so, using a watt, then teh battery weighs 20g
[17:52] <robert1971> And given we only need a few hours of power not worth considering anything else
[17:52] <robert1971> beat me to it
[17:52] <SpeedEvil> And solar cells smash
[17:53] <SpeedEvil> Unless they are really rugged.
[17:53] <robert1971> = expensive
[17:53] <gordonjcp> ... and if they're really rugged they're expensive, and heavy
[17:53] <SpeedEvil> Not really - but expensive in terms of weight
[17:53] <gordonjcp> being made of big heavy bits of glass
[17:53] <gordonjcp> ... and they'll probably *still* smash
[17:53] <SpeedEvil> (if you make your own)
[17:53] Action: SpeedEvil puts on a faker scottish accent.
[17:53] <jcoxon> they'll come in useful once someone cracks the holy grail of ballooning
[17:53] <robert1971> I have a life to live too
[17:53] <SpeedEvil> 'Transparent aluminium!'
[17:54] <jcoxon> which is to keep the payload at altitude in the same place
[17:54] <jcoxon> i might add a list of challenges on the wiki
[17:54] <robert1971> Geostationary balloon.
[17:54] <jcoxon> yup
[17:54] <robert1971> There is money for that I'm sure
[17:54] <jcoxon> really useful
[17:54] <jcoxon> oh definitely
[17:55] <jcoxon> emergency services, launch a balloon with emergency comms
[17:55] <jcoxon> spying
[17:55] <jcoxon> military
[17:55] <jcoxon> other challenges:
[17:55] <jcoxon> rocket
[17:55] <robert1971> Coms in the main. Image how cheap it would be to roll out a 100Mb/s network
[17:55] <jcoxon> return to base payload
[17:55] <jcoxon> in the uk uplink and downlink
[17:55] <jcoxon> pictures during the flight ;-)
[17:56] <SpeedEvil> Has anyone looked into JDAM guidance packages for recovery? :)
[17:56] <jcoxon> hehe
[17:57] <SpeedEvil> I was looking at http://cgi.ebay.co.uk/6mm-Clear-Acrylic-Perspex-Plastic-Sheet-250x500mm_W0QQitemZ300247259368QQcmdZViewItem?hash=item300247259368&_trkparms=72%3A984|39%3A1|66%3A2|65%3A12&_trksid=p3286.c0.m14.l1318 - and similar for a ruggedised solar panel a while back
[17:57] <SpeedEvil> take individual small cells, and glue to panel with low-modulus silicone.
[17:57] <SpeedEvil> attach rigid structure in between cells.
[17:58] <SpeedEvil> http://cgi.ebay.co.uk/25mm-Clear-Acrylic-Perspex-Plastic-Sheet-500x1000mm_W0QQitemZ290211960038QQihZ019QQcategoryZ26258QQtcZphotoQQcmdZViewItemQQ_trksidZp1742.m153.l1262 Hmm. They have thick stuff quite cheap.
[18:08] <robert1971> I see a few people a looking at getting paragliders to remain stableish. I guess it's going to be tricky nut to crack
[18:08] <jcoxon> indeed
[18:09] <jcoxon> but i think we'll see some progress soon
[18:10] <edmoore> how would you keep the thing still at high alt?
[18:10] <edmoore> ... back to stationary balloons, this is
[18:10] <jcoxon> oh right
[18:11] <jcoxon> high efficiency propellors?
[18:11] <jcoxon> JP Aerospace have developed something along those lines
[18:11] <edmoore> hrm
[18:11] <edmoore> well, they flew a propellor
[18:11] <jcoxon> hehe
[18:11] <jcoxon> true
[18:12] <jcoxon> gas jets? rocket engines?
[18:12] <jcoxon> who knows - hence the challenge
[18:13] <robert1971> Bit of string from the moon :)
[18:14] <robert1971> How stationary does a baloon need to be? Within 1K circle. I guess it depends on the height
[18:15] <robert1971> I the prinary uplink /dlink could track it, but the users would need it fairly fixed
[18:21] robert1971 (n=user@212.183.134.66) left irc: "Leaving."
[18:32] jnd (n=jenda@unaffiliated/jnd) joined #highaltitude.
[18:43] natrium (n=alexei@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Remote closed the connection
[18:49] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) left irc: Read error: 113 (No route to host)
[18:59] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) joined #highaltitude.
[19:11] akawaka (n=akawaka@external.treyarch.com) joined #highaltitude.
[19:13] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) left irc: "Connection reset by a tapir"
[19:20] jcoxon (n=jcoxon@85-210-73-114.dsl.pipex.com) left irc: Read error: 110 (Connection timed out)
[19:21] jnd (n=jenda@unaffiliated/jnd) left irc: Read error: 110 (Connection timed out)
[19:22] jnd (n=jenda@unaffiliated/jnd) joined #highaltitude.
[19:33] Nick change: Ei5GTB_ -> EI5GTB
[19:56] edmoore_ (n=edmoore@37.124-84-212.ippool.ndo.com) joined #highaltitude.
[20:14] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) left irc: Read error: 110 (Connection timed out)
[20:38] edmoore_ (n=edmoore@37.124-84-212.ippool.ndo.com) left irc:
[20:51] G8KHW (n=Steve@217.47.75.27) joined #highaltitude.
[20:55] Hiena (n=Hiena@81.93.195.181.datatrans.hu) joined #highaltitude.
[21:07] Shanuson (n=Peter@p54A94CB0.dip.t-dialin.net) left irc: Read error: 104 (Connection reset by peer)
[21:09] icez (n=icez@ip68-98-34-247.ph.ph.cox.net) joined #highaltitude.
[21:27] Hiena (n=Hiena@81.93.195.181.datatrans.hu) left irc: "-=Halt! Hammerzeit!=-"
[21:30] simon__ (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[22:47] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) joined #highaltitude.
[22:58] walters_ (n=walters@flasheart.xs4all.nl) joined #highaltitude.
[23:00] phatmonkey (n=ben@scooby.firshman.co.uk) left irc:
[23:02] <walters_> Goodevening, I'm trying to use james' soundmodem port on gumstix. Sending beacons no problem, excellent...
[23:02] <walters_> receiving, no luck...
[23:03] <walters_> Did anyone get the reception side working on 1200bps. I'm using verdex XL6P with gpsstix
[23:03] <walters_> any comment greatly appreciated
[23:07] <gordonjcp> evening
[23:07] <gordonjcp> walters_: I'm quite interested in getting AX25 going too
[23:08] <walters_> excellent, I have a working setup on a laptop, receiving aprs at the moment. Thought I'd just copy the config. Was amazed by initial success with sending...
[23:08] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) left irc: Remote closed the connection
[23:09] <walters_> unfortunately the reception side doesn't seem to work on the gumstix. Already looked into many things.
[23:09] <walters_> suspect the problem is in oss emulation in alsa. However, i'm guessing here. So any verification it could work would be nice.
[23:10] <walters_> Do you know if the reception side should work ?
[23:10] MetaMorfoziS (n=eee@53d83582.adsl.enternet.hu) joined #highaltitude.
[23:10] <walters_> Also,... I'm very keen to know how (and if) the push-to-talk line is available on the gumstix
[23:11] <walters_> I'm already a happy man though, now that I've seen it beaconing ;-)
[23:12] <akawaka> walters
[23:12] <akawaka> oss emulation could be the issue
[23:13] <akawaka> also just the levels
[23:13] <akawaka> i imagine you don't have the ability to run soundmodemconfig on the gumstix?
[23:15] <walters_> no, I'd be surprised if john has been able to ressurect the old glade soundmodemconfig app. No I'm afraid I can't use it.
[23:16] <walters_> however, I already verified that my hardware is working using arecord and aplay. Also played around with alsamixer and made sure to select the 'line' input and not the 'mic' input (default)
[23:16] jcoxon (n=jcoxon@85-210-73-114.dsl.pipex.com) joined #highaltitude.
[23:16] <walters_> So,.... using alsa tools I get sound.
[23:17] <jcoxon> evening all
[23:17] <akawaka> walters_: okay
[23:17] <walters_> evening
[23:17] <akawaka> are you able to recompile soundmodem?
[23:18] <jcoxon> oooooo soundmodem
[23:18] <walters_> no, although I recompiled it (probably fixing the exact code james fixed aroung xml2 and autotools) earlier on laptop using FC9
[23:18] <jcoxon> hey walters_ you using soundmodem then?
[23:18] <walters_> yes
[23:19] <jcoxon> its not a happy puppy when you compile it
[23:19] <akawaka> the problem with soundmodem is the crc check
[23:19] <akawaka> there is no option to disable it
[23:19] <walters_> jcoxon, I just used you binaries and installed on gumstix XL6P
[23:20] <jcoxon> oh excellent
[23:20] <jcoxon> glad it worked
[23:20] <akawaka> so you can't see if you are getting 99% correct data with some slight hiccups caused by some unknown issue
[23:20] <walters_> Sends fine, reception is a different story.
[23:20] <jcoxon> oh i've never tried reception
[23:20] <jcoxon> doesnt' really fit in with my immediate plans
[23:20] <akawaka> walters_: i would record a wav on the gumstix
[23:20] <akawaka> and run it through soundmodemconfig on another machine
[23:20] <jcoxon> however i'd be very interested in it working
[23:21] <akawaka> and see if soundmodemconfig can decode it
[23:21] <walters_> akawaka, good plan, use clean sample.
[23:21] <walters_> jcoxon, so you've not seen it work on gumstix then ?
[23:21] <akawaka> i had a lot of problems with reception
[23:21] <jcoxon> reception?
[23:21] <jcoxon> i haven't tried
[23:22] <walters_> akawaka, did you ever receive on gumstix ?
[23:22] <akawaka> mainly because it was so difficult to configure without a graphical interface
[23:22] <jcoxon> but for the balloon i'm working on i'm only doing a downlink
[23:22] <akawaka> walters_: never tried on a gumstix, i was using a fitpc, a small linux box
[23:22] <akawaka> had to use an external usb soundcard because the built in sound hardware was so noisey
[23:23] <walters_> yeah, I noticed a lot of noise in the gumstix recordings as well.
[23:24] <jcoxon> yeah its noisey - even more noise using a goliath
[23:24] <jcoxon> where the gsm lines are tied to the audio (i think they are)
[23:24] <jcoxon> walters_, out of interest whats the gumstix being used for?
[23:24] <walters_> okay, thanks for the info, it's very good to know I'm not doing something obviously stupid. Will investigate a bit further to see what I can find out.
[23:25] <akawaka> my fitpc was receiving almost perfectly
[23:25] <akawaka> but there was always 2-3 corrupt characters in a packet
[23:26] <walters_> jcoxon, well, it's just playing around at the moment, but it's candidate for inclusion in my motorcycle ;-)
[23:26] <jcoxon> oh cool
[23:26] <jcoxon> reception is something i have planned in the future
[23:26] <jcoxon> getting soundmodem to work was a bit of a marathon hack so i haven't taken it any further
[23:27] <jcoxon> was pleased that it was working
[23:27] <jcoxon> sstv has been my latest challenge to get working
[23:28] <walters_> jcoxon, I'm also looking into a tool called multimon (by same author as soundmodem, but a bit more recent). Unfortunately lacks ax25 device binding though
[23:28] <jcoxon> yeah i've seen multimon
[23:28] <jcoxon> gordonjcp was using it earlier today on his linux i386 box
[23:29] <walters_> other tool I'm interested in testing on gumstix is gnuradio
[23:29] <jcoxon> yes indeed
[23:29] <jcoxon> edmoore (who is usually around) is interested in that
[23:30] <walters_> excellent,.... I will return. Now my bed is calling out to me ;-) Thanks for the info, and a very good evening.
[23:30] <jcoxon> no problem
[23:30] <jcoxon> cya
[23:30] <akawaka> not sure how useful ax25 is at the end of the day
[23:31] <akawaka> given my problems with soundmodem
[23:31] <akawaka> a more robust, less integrated solution would suit me fine
[23:31] <jcoxon> it certainly isn't easy
[23:31] <jcoxon> and there are pretty good tnc's on a single chip which would do the job just fine
[23:32] walters_ (n=walters@flasheart.xs4all.nl) left irc: "Leaving"
[23:32] <akawaka> i don't even want a tnc
[23:33] <akawaka> just the software to do the mod/demod
[23:33] <akawaka> and give me a character device to read and write from
[23:33] <jcoxon> hfkernel?
[23:33] <akawaka> and avoid all the other kernel level stuff
[23:33] <akawaka> hmm
[23:38] <jcoxon> right, i'm off
[23:38] <jcoxon> night all
[23:38] jcoxon (n=jcoxon@85-210-73-114.dsl.pipex.com) left irc: "Leaving"
[23:41] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) joined #highaltitude.
[23:41] edmoore (n=edmoore@37.124-84-212.ippool.ndo.com) left irc: Client Quit
[23:57] jnd (n=jenda@unaffiliated/jnd) left irc: Read error: 110 (Connection timed out)
[23:58] G8KHW (n=Steve@217.47.75.27) left irc: "Leaving"
[00:00] --- Wed Aug 13 2008