highaltitude.log.20110531

[00:14] sbasuita (ssb@gateway/shell/anapnea.net/x-hzxqtnuqmgkcbozm) left irc: Ping timeout: 244 seconds
[00:14] sbasuita (ssb@gateway/shell/anapnea.net/x-zodnstgckobcwjbg) joined #highaltitude.
[00:19] Lunar_Lander (~lunar_lan@p54A07B6D.dip.t-dialin.net) joined #highaltitude.
[00:19] <Lunar_Lander> hello
[00:29] DarkCow (~DarkCow@5adefc6a.bb.sky.com) left irc: Ping timeout: 276 seconds
[00:33] <SpeedEvil> Lo
[00:48] Jasperw (~jasperw@93.89.81.29) left irc: Quit: Leaving.
[00:55] <polycarbonate1> eroomde: why thank you, I don't really have any explanation for it other than the 1 meaning I am the first, and there may be more
[00:55] <Lunar_Lander> polycarbonate1 seeing that you are new, welcome to UKHAS
[00:57] Action: polycarbonate1 nods to Lunar_Lander
[00:57] <polycarbonate1> I'm just lurking at the moment, but I'm interested in getting involved with Horus
[00:57] <Lunar_Lander> ah so you are in AUS
[00:58] <juxta> hi polycarbonate1
[00:58] <polycarbonate1> yeah, Adelaide, I'm a mate of Darkside's
[00:59] <Lunar_Lander> cool
[01:01] <juxta> that shady Darkside guy!
[01:01] <Darkside> i'm not that shady
[01:01] <Darkside> polycarbonate1: you're the shady one
[01:02] <Lunar_Lander> no
[01:02] <Lunar_Lander> he has a high transmission coefficient, try shining light through polycarbonate
[01:02] <Lunar_Lander> XD
[01:02] <Darkside> >_>
[01:18] spacekitteh (~TraumaPon@124-148-45-141.dyn.iinet.net.au) left irc: Disconnected by services
[01:19] spacekitteh (~TraumaPon@124-148-45-141.dyn.iinet.net.au) joined #highaltitude.
[01:42] Lunar_Lander2 (~lunar_lan@p54A07C6F.dip.t-dialin.net) joined #highaltitude.
[01:44] Lunar_Lander (~lunar_lan@p54A07B6D.dip.t-dialin.net) left irc: Ping timeout: 252 seconds
[02:51] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) left irc: Remote host closed the connection
[03:08] Lunar_Lander2 (~lunar_lan@p54A07C6F.dip.t-dialin.net) left irc: Quit: Lunar_Lander2
[03:32] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) joined #highaltitude.
[04:32] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Read error: Connection reset by peer
[04:36] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[04:37] kd0mto (~dago@74-94-125-113-illinois.hfc.comcastbusiness.net) left irc: Ping timeout: 258 seconds
[04:39] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[04:56] kd0mto (~dago@12.163.240.38) joined #highaltitude.
[05:04] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) left irc: Remote host closed the connection
[05:04] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) joined #highaltitude.
[05:09] kd0mto (~dago@12.163.240.38) left irc: Read error: Connection reset by peer
[05:11] <griffonbot> Received email: Terry Baume "Re: [UKHAS] interesting item on ebay?"
[05:38] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Read error: Connection reset by peer
[05:38] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[05:49] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Ping timeout: 252 seconds
[06:34] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[06:45] <Darkside> fuck shit fuck fuck
[06:46] <Darkside> the DDS i was using doesn't run at 3.3v
[06:46] <Darkside> now i have to redesign that part of my circuit
[06:46] <Darkside> lesson learnt: read the datasheet carefully
[06:51] <eroomde> whatcha using a DDS for?
[06:51] <eroomde> sounds interesting
[06:57] <SamSilver> direct digital synthesizer ???
[06:57] <SamSilver> dirty didgit stimulater ???
[06:58] <SamSilver> ;-)
[07:03] <Darkside> eroomde: hf transmitter
[07:03] <eroomde> ah cool
[07:03] <eroomde> what freq range?
[07:03] <Darkside> 0-25MHz or so
[07:04] <Darkside> i don't plan on going above 15MHz with it
[07:05] <eroomde> fun
[07:05] <eroomde> you could use it for 70cm too!
[07:05] <Darkside> no
[07:06] <Darkside> it will only generate signal between DC and 25MHz
[07:07] <eroomde> oh, that's annoying
[07:07] <eroomde> which device if you don't mind me asking?
[07:08] <eroomde> i'm sure you could extract a 70cm signal from it but it'd be like the 10th +ve image which would be pretty attenutaed
[07:08] <Darkside> nah
[07:09] <Darkside> AD9834
[07:09] <Darkside> and i've got a low pass filter after it
[07:09] <Darkside> 15MHZ cutoff
[07:10] <eroomde> yep you'd have to have some kind of siplexer to make this thing work
[07:10] <eroomde> diplaexer
[07:10] <eroomde> gah
[07:10] <Darkside> eh?
[07:10] <Darkside> no
[07:10] <eroomde> diplexer*
[07:10] <Darkside> this is a transmitter only
[07:10] <eroomde> oh no i mean the 70cm extraction
[07:11] <Darkside> i wouldn't do it anyway
[07:11] <Darkside> i'm planning on using this on HF, not UHF
[07:11] <Darkside> if i wanted to broadcast on 70cm, i'd make a 70cm transmitter
[07:11] <eroomde> sure, this is me just being armchair :)
[07:11] <eroomde> say, one dds transmitter on a trans-a balloon
[07:13] <Darkside> it could be done, but i wouldn't do it
[07:13] <Darkside> this board is designed to be a QRP transmitter, not necesarily for ballooning
[07:13] <Darkside> but it could fly
[07:18] CountX_ (~CountX@cpe-72-179-0-141.austin.res.rr.com) left irc: Quit: Leaving
[07:40] GW8RAK (~chatzilla@host217-43-78-244.range217-43.btcentralplus.com) joined #highaltitude.
[07:46] Elwell (~elwell@freenode/staff/elwell) left irc: Remote host closed the connection
[07:46] <SamSilver> bbl
[07:46] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Quit: Page closed
[07:49] Elwell (~elwell@freenode/staff/elwell) joined #highaltitude.
[08:37] TangoAlphaWork (~chatzilla@host217-34-50-32.in-addr.btopenworld.com) joined #highaltitude.
[09:05] <fsphil> weekend predictions still good
[09:05] Laurenceb_ (~Laurence@host86-172-188-211.range86-172.btcentralplus.com) joined #highaltitude.
[09:05] <fsphil> though variable
[09:09] <griffonbot> Received email: Steve Randall "Re: [UKHAS] interesting item on ebay?"
[09:09] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[09:12] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 260 seconds
[09:18] Laurenceb_ (~Laurence@host86-172-188-211.range86-172.btcentralplus.com) left irc: Ping timeout: 255 seconds
[09:19] <griffonbot> Received email: Steve Randall "Re: [UKHAS] Re: CAA contact?"
[09:23] <griffonbot> Received email: Chris Foote (Spike) "Re: [UKHAS] interesting item on ebay?"
[09:30] rjharrison (~rjharriso@host81-129-33-61.range81-129.btcentralplus.com) joined #highaltitude.
[09:39] rjharrison (~rjharriso@host81-129-33-61.range81-129.btcentralplus.com) left irc: Quit: Colloquy for iPad - http://colloquy.mobi
[10:08] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 240 seconds
[10:14] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) left irc: Remote host closed the connection
[10:50] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[11:09] juxta_ (~juxta@ppp59-167-1-1.lns1.mel4.internode.on.net) joined #highaltitude.
[11:12] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 250 seconds
[11:14] <Laurenceb> SpeedEvil: ping
[11:17] <Laurenceb> do you still have that n900 gps recording from ages ago?
[11:18] <SpeedEvil> ?
[11:18] <SpeedEvil> ah
[11:18] <SpeedEvil> what for?
[11:19] <SpeedEvil> I have many GPS recordings
[11:19] <Laurenceb> i wanted to look at noise
[11:19] <Laurenceb> supposedly the fsa03 suffers from drafts effecting the tcxo temperature
[11:20] <Laurenceb> does n900 have EGNOS?
[11:20] <Laurenceb> i want to see if the fsa03 has more 1/f noise
[11:22] <SpeedEvil> From a stationary setting?
[11:22] <Laurenceb> yes
[11:23] <Laurenceb> i had to turn off SBAS/EGNOS to get higher update
[11:23] <Laurenceb> im not sure how much more noise that introduces
[11:23] <Laurenceb> and how significant tcxo temperature shifts are
[11:24] juxta__ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[11:24] juxta_ (~juxta@ppp59-167-1-1.lns1.mel4.internode.on.net) left irc: Ping timeout: 248 seconds
[11:26] <SpeedEvil> Unfortunately, I can't lay my hands on logs of stationary.
[11:26] <SpeedEvil> I've started it logging in the greenhouse now.
[11:27] <Laurenceb> cool thanks
[11:27] <Laurenceb> does it log velocity?
[11:27] <SpeedEvil> Not seperately.
[11:30] <Laurenceb> but its part of the log?
[11:31] <SpeedEvil> I mean - only as position delta
[11:31] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[11:32] <SpeedEvil> http://wiki.maemo.org/N900_Hardware_GPS I wrote this up - but the curves are probably not helpful to you, they are simply the percentage of points falling outside a radius.
[11:34] juxta__ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 240 seconds
[11:36] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[11:37] <Laurenceb> interesting
[11:39] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 276 seconds
[11:39] <SpeedEvil> aha - found old logs
[11:40] <SpeedEvil> the weeks logs
[11:44] <Laurenceb> does it have SBAS/EGNOS?
[11:45] <SpeedEvil> No.
[11:45] <SpeedEvil> http://www.mauve.plus.com/t/gps.tar.bz2 (in a minute)
[11:45] <SpeedEvil> This is the raw logs of a position with good view of the sky, over several hour logging periods most days for around a week.
[11:45] <SpeedEvil> Datum position was +-5cm or so
[11:46] <Laurenceb> ok, so camparible with the fsa03 with egnos off
[11:46] <SpeedEvil> Ok - there
[11:46] <Laurenceb> thanks
[11:46] <SpeedEvil> np!
[11:48] DarkCow (~DarkCow@5adefc6a.bb.sky.com) joined #highaltitude.
[11:49] <Laurenceb> what does gpxtocsv do?
[11:49] <SpeedEvil> It converts gpx files to a space seperated file
[11:49] <SpeedEvil> It's ... /me checks.
[11:50] <Laurenceb> ah
[11:50] <SpeedEvil> Ok - the file out is simply the timestamp, then the distance from my lat and lon datums.
[11:50] <Laurenceb> is there any velocity data?
[11:51] <SpeedEvil> Looking at the gpx - no
[11:52] <Laurenceb> thats a shame
[11:52] <SpeedEvil> Yeah - the velocity data is horribly quantised.
[11:54] <SpeedEvil> Oh - last two columns from 'out' are meters above my datum, and sats
[11:54] <Laurenceb> hmf octave doesnt like the time format
[11:55] <SpeedEvil> It's the standard time from GPX I think.
[11:55] <Laurenceb> its got : in it
[11:55] <Laurenceb> and multiple - ...
[11:55] <SpeedEvil> I wasn't interested in it in anything other than an index - so I diddn't bother to do more.
[11:56] Action: SpeedEvil checks when the nice man from tesco will be arriving.
[11:57] <SpeedEvil> Another hour. They've started texting you now giving you an hour slot close to the time - which is nice.
[12:00] <Laurenceb> awk is confusing
[12:00] <Laurenceb> how do i remove the time?
[12:00] <Laurenceb> which line of gpxtocsv?
[12:00] <SpeedEvil> print $1
[12:00] <SpeedEvil> that's print the first input field
[12:01] <SpeedEvil> so just nuke the $1
[12:01] <Laurenceb> ah
[12:01] <SpeedEvil> Sorry - I wan't writing it to be legible.
[12:01] <SpeedEvil> Just as a quick hack.
[12:02] <Laurenceb> ah works now
[12:10] <Laurenceb> interesting
[12:11] <Laurenceb> i took 'velocity' = delta position
[12:14] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) joined #highaltitude.
[12:16] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[12:18] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 250 seconds
[12:22] <Laurenceb> peak in spectrum around 0.05Hz
[12:22] <Laurenceb> and 0.005Hz
[12:23] <Laurenceb> then closing to white noise - looks like constellation changes and ionospheric noise
[12:29] <Laurenceb> http://i.imgur.com/7fYDz.png
[12:33] <cuddykid> built my soldering holder thing
[12:36] <Laurenceb> hmm i dont think thats all really velocity noise
[12:37] <Laurenceb> - i dont think itd all appear on the pll based gps velocity output
[12:38] <Laurenceb> but the ionospheric noise is probably real
[12:41] <Laurenceb> http://i.imgur.com/R9Cek.png
[12:41] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) left irc: Ping timeout: 240 seconds
[12:41] <Laurenceb> i think the noise above ~100 is dll tracking noise, then below 100 its ionopheric drift and constellation changes
[12:42] <Laurenceb> the peak around 50 im guessing is from the gps velocity estimate being incorporated into position
[12:42] <Laurenceb> that ionospheric doppler shifts
[12:42] <SpeedEvil> Interesting.
[12:43] cuddykid (~acudworth@78.145.195.228) left irc: Quit: cuddykid
[12:44] <Laurenceb> but slightly annoying
[12:44] <Laurenceb> kalman likes random guassian noise
[12:44] <SpeedEvil> Sounds plausible, yes.
[12:44] <Laurenceb> you need to update gnuplot
[12:44] <Laurenceb> proper fonts :P
[12:44] <SpeedEvil> ?
[12:45] <SpeedEvil> Ah - I have a fairly old version. :)
[12:45] <SpeedEvil> I last updated it for colour plots.
[12:45] <Laurenceb> heh
[12:45] <SpeedEvil> Err - coloured points
[12:46] <Laurenceb> i guess the delta position based velocity estamte isnt really telling us much
[12:46] <Laurenceb> as most of the position 'information' comes from the dll delay tracker
[12:46] <Laurenceb> not the velocity estimate from doppler tracking
[12:47] BrainDamage (~BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) joined #highaltitude.
[12:47] <Laurenceb> i can output the estimated velocity error from the ublox, but as the noise is concentrated around 0.05hz, it wont work very well in the kalman
[12:48] <Laurenceb> guess all i can do is increase the noise to account for that
[12:48] <SpeedEvil> I don't suppose treating the GPS as two recievers helps?
[12:48] <SpeedEvil> Say one high frequency, and one low?
[12:49] <Laurenceb> by a factor of sqrt(1/0.05*2.5)
[12:49] <SpeedEvil> Oh.
[12:49] <Laurenceb> - for 5hz gps smaple rate
[12:49] <SpeedEvil> That works for stationary use-cases. :)
[12:49] <Laurenceb> well two receiver would work
[12:49] <Laurenceb> one EGNOS receiver -> egnos info to correct the ublox
[12:50] <Laurenceb> that takes out most of the ionospheric ~10+ second cycle stuff
[12:50] <SpeedEvil> Can you get raw psuedoranges off the ublox?
[12:50] <Laurenceb> unfortunately egnos hammers the processor on the ublox and it cant go past 4hz
[12:50] <Laurenceb> not the fsa03
[12:50] <Laurenceb> some of the modules can
[12:51] <Laurenceb> mediatec has egnos and 10hz
[12:51] <Laurenceb> but you cant configure the onboard kalman filter so its rather laggy
[12:51] <Laurenceb> you cant win :/
[12:52] <SpeedEvil> I suppose adding a GPS solver into your code just makes it more messy.
[12:52] <SpeedEvil> Is it really that much of an issue, in general?
[12:52] <Laurenceb> DGPS off the ground would work
[12:52] <SpeedEvil> If you're not trying to fly to a given spot.
[12:53] <SpeedEvil> I mean - yes - you rely on the GPS to desaturate the gyro+mag+accel drift
[12:54] <SpeedEvil> But doesn't much of that cover the worst of the GPS errors yuou'd hit?
[12:54] <SpeedEvil> Other than absolute position.
[12:54] <Laurenceb> yeah... i guess for fixed wing having something more responsive helps
[12:54] <Laurenceb> but i wonder if a quadcopter would get better results using mediatec
[12:55] <Laurenceb> thats what openpilot decided
[12:55] <Laurenceb> due to the egnos/waas support
[12:55] <Laurenceb> they like to hover in really stable spots
[12:56] <SpeedEvil> I suppose the question is on the velocity spectrum - WRT quadcopter. At what point does the summed velocity of the GPS errors get curtailed by the accel.
[12:57] <Laurenceb> yeah - with quads they arent doing anything particularly dynamic most of the time
[12:57] <SpeedEvil> 10s should - for example - be a few dozen cm drift - from the accel.
[12:57] <SpeedEvil> If I've got the numbers right.
[12:58] <Laurenceb> so the mediatec filter glitchyness shouldnt effect them too much
[12:58] <Laurenceb> yes - its not the end of the world having to average over 10s
[12:58] <Laurenceb> the openpilot forum is interesting on thsi topic
[12:59] <Laurenceb> they started with ublox5 and then went to mediatec, but the waas/egnos
[12:59] <Laurenceb> s/but/for
[13:01] Jasperw (~jasperw@93.89.81.29) joined #highaltitude.
[13:01] <Laurenceb> ok ill take that back http://www.vimeo.com/groups/openpilot/videos/24235678
[13:02] juxta__ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[13:03] <Randomskk> wow, those are nimble
[13:04] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 250 seconds
[13:04] <Laurenceb> would be interesting to see what the mediatec gps was saying during that
[13:04] <SpeedEvil> On reflection, I suppose quadcopters have bugger all control authority for lateral moves.
[13:04] <SpeedEvil> As they've got to tilt first.
[13:05] <Randomskk> that is really very nimble. I want to get my quadcopter working again now :|
[13:05] jkominar (~justin@64.235.97.218) joined #highaltitude.
[13:07] <SpeedEvil> I want to get my convertable UAV done too.
[13:07] <SpeedEvil> Convertable as it's either a VTOL UAV, or you can clip a couple of wings on, and it's a thrust vectored ducted fan.
[13:07] <Laurenceb> i like that idea
[13:08] <Laurenceb> harder to break and cheaper
[13:08] <griffonbot> Received email: Steve Randall "Re: [UKHAS] interesting item on ebay?"
[13:10] <Laurenceb> http://www.foxtechfpv.com/fy3zt-uav-controller-p-275.html
[13:11] Dan-K2VOL (~Adium@74-143-44-42.static.insightbb.com) joined #highaltitude.
[13:13] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[13:13] juxta__ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 240 seconds
[13:13] <Laurenceb> http://www.foxtechfpv.com/fh10z-pal10x-optical-zoom-rc-controlled-camworld-smallest-7-p-166.html
[13:13] <Laurenceb> niiiiceee
[13:15] cuddykid (~acudworth@78.145.195.228) joined #highaltitude.
[13:16] imrcly (~tim@74-128-123-149.dhcp.insightbb.com) joined #highaltitude.
[13:16] <Laurenceb> its almost as good as a predator
[13:16] <Laurenceb> servos are a bit jittery for the camera
[13:17] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) joined #highaltitude.
[13:19] <cuddykid> anyone used servoshop.co.uk before?
[13:20] <Laurenceb> http://www.foxtechfpv.com/fhx46-quadcopter-hexacopter-controller-p-286.html
[13:20] <Laurenceb> intriguing - seems to be stm32, freescale accel and invensense analogue gyros
[13:20] <cuddykid> that's awesome
[13:21] <Randomskk> why is the switch labelled "ON" and "1"
[13:21] <Randomskk> :|
[13:21] <cuddykid> would be cool as a recovery vehicle .. attached to bottom of payload that kicks in when a couple of kms up!
[13:22] <Laurenceb> stuffs pretty expensive there
[13:22] <Laurenceb> seems to be based on multicopter.de
[13:22] <BrainDamage> probability of wire tangling on the fans is =1
[13:23] <Laurenceb> i like the camera tho, thats insane
[13:23] <Laurenceb> they really need better servos for it tho
[13:23] juxta__ (~juxta@ppp59-167-3-105.lns1.mel4.internode.on.net) joined #highaltitude.
[13:24] Action: Laurenceb wonders how long before someone flies it over a nudist camp
[13:26] <Laurenceb> that camera seems designed purely for perverted uses
[13:26] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 260 seconds
[13:26] <Randomskk> I can think of non-perverted uses
[13:26] <Randomskk> on the other hand, it could also hover outside windows :P
[13:27] <Laurenceb> as i was saying
[13:27] <Laurenceb> also has nightvision :S
[13:27] <BrainDamage> need shower curtain vision
[13:28] <Randomskk> it's kinda tempting to just buy one
[13:28] <Randomskk> but no
[13:28] <Randomskk> I will press on :|
[13:28] <BrainDamage> mmm, maybe a terahertz body scanner sensor
[13:29] <Laurenceb> i have dactyl boards spare
[13:29] juxta__ (~juxta@ppp59-167-3-105.lns1.mel4.internode.on.net) left irc: Ping timeout: 240 seconds
[13:30] x-f (~x-f@taburete.zetcom.lv) left irc: Ping timeout: 260 seconds
[13:35] x-f (~x-f@taburete.zetcom.lv) joined #highaltitude.
[13:42] <cuddykid> I'm obviously being thick, but... what's the difference between a servo and a motor?
[13:42] <Randomskk> Laurenceb: I still need to make my motor controller :P
[13:43] <Laurenceb> does it take pwm?
[13:43] <Randomskk> also I still want to try this single-pcb concept: one large pcb making up the entire airframe, with perhaps some metal cross sections bolted on the bottom for rigidity
[13:43] <Randomskk> I doubt it will. probably just going to do spi or i2c or something
[13:43] <Laurenceb> ok
[13:43] <Randomskk> plenty of COTS bldc escs that do though
[13:43] <Laurenceb> i have all three broken out
[13:44] <BrainDamage> cuddykid: a servo has a motor, and a position sensor
[13:44] <BrainDamage> usually also a torque converter
[13:44] <cuddykid> ahh right, cheers!
[13:45] <Laurenceb> lol
[13:45] <Laurenceb> i think nomal people call it a gearbox
[13:46] <cuddykid> found a great curved piece of plastic ideal for a window for camera to view all around.. will have a look for some servos to turn camera
[13:46] <Dan-K2VOL> Cuddykid a window in front of a cam on a high altitude balloon will most often frost over
[13:47] <cuddykid> ooh yeah, hmm.. I was working towards a picture of a project I had seen with a curved window.. I'll try and find
[13:47] <Dan-K2VOL> Cuddykid also most common servos are limited to a single rotation or half rotation of travel
[13:48] <Dan-K2VOL> But can be modified to do full rotation
[13:48] <Randomskk> the modification turns their position input into an angular velocity input, though
[13:48] <cuddykid> would it be ok, just to move the camera around say 45 degrees?
[13:48] <Randomskk> yes
[13:48] <cuddykid> cool
[13:48] <cuddykid> http://ukhas.org.uk/projects:apex-i
[13:49] <cuddykid> 1st pic is the one I was aiming for
[13:49] <Dan-K2VOL> it's a lot easier to just include two cams
[13:49] <Dan-K2VOL> These days they're so cheap and easy
[13:49] <cuddykid> yeah, good point
[13:49] kd0mto (~dago@sdrc.com-162.fuse.net) joined #highaltitude.
[13:50] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[13:57] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[13:59] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 255 seconds
[14:00] <eroomde> certainly agreed
[14:00] <eroomde> servos in space is a bit of a faff
[14:02] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[14:03] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 252 seconds
[14:06] <cuddykid> just adjusting my bas script for a570.. from past experience is 5 mins enough to lid payload box, glue it down and prepare for launch etc? Thinking it might be better if it was ~10mins or so
[14:18] <eroomde> 30
[14:18] <eroomde> cuddykid: ^
[14:19] <cuddykid> oh wow!
[14:19] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) left irc: Ping timeout: 276 seconds
[14:19] <cuddykid> was going to try and prep balloon etc before turning on payload stuff
[14:19] <eroomde> never do that
[14:19] <eroomde> do the balloon last
[14:20] <eroomde> once everything else is working
[14:20] <eroomde> and sealed
[14:20] <cuddykid> oh right, glad I've got you guys with experience to advise me!
[14:20] <eroomde> once the balloon is inflated it wants to go, you don't want to be hanging around
[14:20] <eroomde> especially when it doesn't get a gps lock 'even though it worked on the bench last night'
[14:21] <cuddykid> yeah, right, will put in 30mins
[14:21] <cuddykid> lol.. that's what I dread haha
[14:21] <eroomde> though i wouldn't program a gap between turning on and taking pictures
[14:21] <eroomde> that'll just cause you problem
[14:21] <eroomde> i'd just start taking media from the off and get a big enough sd card
[14:22] <cuddykid> the problem was the battery life
[14:22] <eroomde> otherwise what will happen is you'll be ready to let go, the gps will loose lock, you'll decide to do a restart, the gps will start working again, but you'll have to wait 30 mins for the cameras to start. which would be a disaster
[14:22] <eroomde> on an a570? really?
[14:22] <cuddykid> for some reason (I know the lithiums will last much longer) but the alkaline was running out after an hr or so
[14:22] <eroomde> we've had them run for hours just fine
[14:22] <eroomde> you're using energizer lithium ultimates?
[14:22] <eroomde> ah right
[14:23] <cuddykid> haven't trialed with the lithiums yet
[14:23] <eroomde> well, sod alkalines, they're not a realistic test environment and you don't want to use them in flight anyway
[14:23] <cuddykid> got them waiting for the big day
[14:23] <eroomde> try it with lithiums :)
[14:23] <cuddykid> yeah, will do :)
[14:23] <cuddykid> 4gb sd card should do nicely
[14:23] <eroomde> it's worth a full soak test with flight hardware, in flight configuration
[14:23] <eroomde> it'll throw up so many issues you wouldn't otherwise catch
[14:23] <cuddykid> I think it was 2100 photos
[14:23] <cuddykid> yeah
[14:28] Darkside (~Darkside@hpavc/darkside) left irc: Ping timeout: 260 seconds
[14:29] Darkside (~Darkside@hpavc/darkside) joined #highaltitude.
[14:30] Laurenceb (~laurence@vlsi1.eee.nottingham.ac.uk) left irc: Remote host closed the connection
[14:44] Darkside (~Darkside@hpavc/darkside) left irc: Ping timeout: 244 seconds
[14:46] Darkside (~Darkside@hpavc/darkside) joined #highaltitude.
[14:47] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[14:48] <SamSilver> Dan-K2VOL: long time no see, how you doing?
[14:53] <cuddykid> time to do some soldering
[15:02] Dan-K2VOL (~Adium@74-143-44-42.static.insightbb.com) left irc: Ping timeout: 244 seconds
[15:06] mattltm-alt (~mattltm-a@mail.icm2.org.uk) left irc:
[15:07] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[15:16] Laurenceb (~laurence@vlsi1.eee.nottingham.ac.uk) joined #highaltitude.
[15:17] Action: Laurenceb RAGGGEEEE
[15:17] <Laurenceb> stupid usb and windoze
[15:19] <cuddykid> ice cube test..
[15:20] Dan-K2VOL (~Adium@74-143-44-42.static.insightbb.com) joined #highaltitude.
[15:26] <cuddykid> will it kill my temp sensor putting it in ice water?
[15:27] TangoAlphaWork (~chatzilla@host217-34-50-32.in-addr.btopenworld.com) left irc: Quit: bye
[15:32] <cuddykid> oh wow...
[15:32] <SamSilver> What???
[15:32] <cuddykid> temp sensor is 4C out
[15:32] <cuddykid> that's huge
[15:32] <cuddykid> given +/- 0.5C presicion
[15:33] <Upu> same as mine then
[15:33] kevin_ (c67b3304@gateway/web/freenode/ip.198.123.51.4) joined #highaltitude.
[15:33] <cuddykid> yeah, thought it was a bit high!
[15:33] <SamSilver> is the water / ice mix "soft" or "hard" water?
[15:33] <SamSilver> low or high in salt?
[15:34] kevin_ (c67b3304@gateway/web/freenode/ip.198.123.51.4) left #highaltitude.
[15:34] <cuddykid> urm, not sure :S
[15:35] <cuddykid> think its hard but not entirly sure
[15:35] <cuddykid> right.. seems to be slowly decreasing in temp.. not reading 2.3C
[15:35] <cuddykid> *now
[15:36] jasonb (~jasonb@adsl-66-124-73-250.dsl.sntc01.pacbell.net) left irc: Ping timeout: 260 seconds
[15:39] <cuddykid> ahh... it keeps decreasing
[15:39] <Upu> they are very slow to respond
[15:39] <Upu> you got it working in the end then ?
[15:39] <cuddykid> yeah, 2.1 C now
[15:39] <Upu> what did you do ?
[15:40] <cuddykid> put it in a plastic bag and in a cup of ice water (with ice cubes)
[15:40] <cuddykid> been in there about 10mins
[15:40] <cuddykid> now reading 2C
[15:42] <W0OTM> Hello World
[15:42] <cuddykid> hello W0OTM!
[15:43] <Laurenceb> im not the world
[15:43] <cuddykid> im above the world.. in space haha
[16:06] juxta (juxta@ppp203-122-193-94.static.internode.on.net) left #highaltitude ("Leaving").
[16:18] GW8RAK (~chatzilla@host217-43-78-244.range217-43.btcentralplus.com) left irc: Remote host closed the connection
[16:29] <W0OTM> GREAT COMMENT! "Individuals are free to conduct their launches consistent with their
[16:29] <W0OTM> own interpretations of the regulations and knowledge of the local
[16:29] <W0OTM> terrain and airspace."
[16:30] <eroomde> how bizarre!
[16:36] kd0mto (~dago@sdrc.com-162.fuse.net) left irc: Ping timeout: 260 seconds
[16:37] jasonb (~jasonb@12.201.129.162) joined #highaltitude.
[16:38] jan_bangna (~jandetlef@ppp-115-87-79-211.revip4.asianet.co.th) joined #highaltitude.
[16:41] kd0mto (~dago@sdrc.com-162.fuse.net) joined #highaltitude.
[16:44] <TangoAlpha> what angle does anyone mount their camera at? Lens horizontal? Or angled down?
[16:44] <eroomde> slightly down tends to get more interesting results
[16:44] <eroomde> black sky is only so interesting
[16:44] <eroomde> maybe 20 degrees down?
[16:44] <TangoAlpha> that much?
[16:44] <eroomde> also if you want to do reconstruction from several frames, go vertical
[16:45] <eroomde> as you will get 360 degree spinning, you may aswell get more in the vertical field of view
[16:45] <eroomde> yes that much
[16:45] <TangoAlpha> i have two cameras...can't decide whether to mount one looking directly down or not
[16:45] <eroomde> surprising i know!
[16:46] <eroomde> http://www.nivnac.co.uk/files/HAPSD_NOVA8/HAPS-D_pan1_half.jpg
[16:46] <SpeedEvil> Directly down has benefits.
[16:46] <eroomde> nova 8 (so back in... 2008?) with a vertical camera angled i think 20 degrees down
[16:46] <eroomde> 360 degree panorama, taken just after dawn
[16:46] <SpeedEvil> But - consider also pointing so that the edge of the image is a little bit beyond vertical, with the picture vertical
[16:47] <TangoAlpha> do you know how often it was taking photos?
[16:47] <eroomde> i do!
[16:47] <eroomde> 20 secs
[16:47] <eroomde> oh maybe 15
[16:47] <eroomde> well, one of those two
[16:47] <TangoAlpha> well, much more often than i had initially planned
[16:47] <eroomde> TangoAlpha: so 2008 is way beyond the half life of my hard drives but I *might* have the chdk script somewhere - let me see if i can find out definitively for you
[16:48] <TangoAlpha> cheers
[16:48] <TangoAlpha> would love to get a result like that panorama
[16:49] <eroomde> that confirmed to me the benefits of just throwing all your data onto the internet
[16:49] <eroomde> that was made by a guy who saw the project and had a go
[16:49] <eroomde> just took our sd card dump and got to work. amazing!
[16:49] <TangoAlpha> and that's 20 degrees down?
[16:49] <eroomde> yes, as i recall
[16:50] <eroomde> again - give me 10 mins as i def wrote this all down at some point
[16:50] <eroomde> just a question of finding where
[16:50] <TangoAlpha> i guess i could actually work it out
[16:50] <eroomde> yep indeed
[16:50] fsphil (~phil@2001:8b0:34:1:21f:c6ff:fe44:b25b) left irc: Ping timeout: 260 seconds
[16:51] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) joined #highaltitude.
[16:52] jan_bangna (~jandetlef@ppp-115-87-79-211.revip4.asianet.co.th) left irc: Quit: jan_bangna
[16:55] fsphil (~phil@2001:8b0:34:1:21f:c6ff:fe44:b25b) joined #highaltitude.
[16:56] <eroomde> TangoAlpha: every 10s
[16:57] <TangoAlpha> thanks
[16:57] <TangoAlpha> guess i need to test my camera battery life
[16:57] <eroomde> yep, definitely worth doing a soak test
[16:58] <eroomde> and ideally a *cold* soak test if you can manage such a thing
[16:58] <TangoAlpha> i'd done one but at 1 minute
[16:58] <eroomde> i.e. chuck it in the freezer
[16:58] <TangoAlpha> yeah, i'd planned to do that at some point
[16:58] <eroomde> well, more photos is a good thing from the pov of getting the 9Whrs of an AA cell into the payload box more quickly
[16:58] <eroomde> but sure yes, batt life takes a hit
[16:59] <TangoAlpha> aa cell?
[16:59] <eroomde> AA energizer lithiums
[17:00] <TangoAlpha> i have two a3000is canons with their own batteries
[17:00] <eroomde> ah fine
[17:00] <eroomde> the a560's we used to use were AAs, that's all
[17:00] <eroomde> not to worry
[17:00] <TangoAlpha> well, not the original batteries, some ebay 1200mAh ones instead of the original 740mAh
[17:00] <eroomde> hmm, that's definitely less capacity
[17:00] <TangoAlpha> i'm hoping that's enough
[17:00] <eroomde> def do a check in the freezer
[17:01] <TangoAlpha> it's done 3 hours @ 1 minute and 3hrs @ 2 minutes
[17:01] <TangoAlpha> so it would seem it's not the actually taking of the photos that takes the power
[17:02] <eroomde> i guess the actual photo taking is infrequent enough that's it's dominated by just being 'on'
[17:02] <TangoAlpha> i need to wire a switch into the lcd circuit
[17:02] <SpeedEvil> I assume you've turned the screen off?
[17:02] <TangoAlpha> hehe
[17:02] <SpeedEvil> Ah - that'll do it.
[17:02] <TangoAlpha> haven't found how to disassemble to rear
[17:03] <TangoAlpha> front parts come off easily, the back not so easy
[17:04] Hiena (~boreger@81.93.195.181.datatrans.hu) joined #highaltitude.
[17:09] <eroomde> TangoAlpha: can't find the downward angle yet i'm afriad
[17:09] <eroomde> but i guess try 15 - you can't really go wrong
[17:16] <Dan-K2VOL> TangoAlpha normal rechargeable are all likely to stop working entirely between 0C and 20C. May want to keep the internal temp up if you can't use the AA lithiums
[17:17] <eroomde> taking photos more often is a good way to do that
[17:17] <eroomde> when we arrived at nova 8's landing site the cameras were about 45 celcius each
[17:18] <SpeedEvil> Indeed.
[17:18] <SpeedEvil> You only need to be 30C over ambient to get to ~-10C
[17:19] <SpeedEvil> At which point batteries are - mostly - fine
[17:19] <eroomde> well, some LiPoly's will actually be very unforgiving at -10
[17:19] <eroomde> we noticed very large decreases in capacity when we flew the blimp on cold mountain mornings
[17:20] <eroomde> we had to gun it for a while (big voltage drops, only 20A instead of expected 80) until they warmed up a bit
[17:21] <TangoAlpha> i have some handwarmers for internal heating if needed
[17:21] <SamSilver> bbl
[17:21] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Quit: Page closed
[17:25] <TangoAlpha> the batteries i have are marked Li-ion, is that the same as the AA lithiums?
[17:27] <eroomde> no
[17:27] <eroomde> the aa lithiums are single use - a different chemistry
[17:27] Jasperw (~jasperw@93.89.81.29) left irc: Quit: Leaving.
[17:27] spacekitteh (~TraumaPon@124-148-45-141.dyn.iinet.net.au) left irc: Ping timeout: 276 seconds
[17:27] <eroomde> they have quite a lot higher capacity and better thermal performance
[17:29] <SpeedEvil> hmm
[17:29] <TangoAlpha> right, freezer test this evening then
[17:29] <SpeedEvil> Higher mAh
[17:29] <SpeedEvil> But lower voltage
[17:29] <SpeedEvil> (for the 1.5V AA lithium)
[17:29] <SpeedEvil> So the total stored energy is more similar.
[17:29] <TangoAlpha> battery = 3.6v 1200mAh 4.3Wh Li-ion
[17:30] <TangoAlpha> according to the printing on it
[17:39] <Dan-K2VOL> TangoAlpha here's some info with links to data sheets on the properties of the energizers: TangoAlpha
[17:39] <Dan-K2VOL> http://wiki.whitestarballoon.com/doku.php?id=compendium&s[]=energizer#batteries
[17:42] <eroomde> Dan-K2VOL: how's tricks with the payload?
[17:43] <eroomde> I've been away for a bit, lost track
[17:57] mattltm (~mattltm@81.134.144.179) joined #highaltitude.
[18:05] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Read error: Connection reset by peer
[18:09] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[18:12] jcoxon (~jcoxon@cpc1-lanc4-0-0-cust720.3-3.cable.virginmedia.com) joined #highaltitude.
[18:18] <Dan-K2VOL> hey eroomde, nothin new to report, we've all been taking a recuperation break, with the jet stream down for the season. Hope to pick things back up this week again with the sat modem antenna investigation
[18:19] <Dan-K2VOL> however i did recruit an actual ex- Air Traffic Controller guy to be on our team
[18:19] <Dan-K2VOL> which is fantastic
[18:20] <jcoxon> evening all
[18:20] <Dan-K2VOL> Hi James
[18:20] Action: jcoxon is smiling at the GPSL debate going on about a 'code of conduct'
[18:20] <Dan-K2VOL> ahh yes!
[18:20] <Dan-K2VOL> I did the same
[18:20] <Dan-K2VOL> you should pipe up, I haven't had a chance to
[18:20] <Dan-K2VOL> if you want to
[18:21] <Dan-K2VOL> I'm glad that people are finally getting more toward that position
[18:22] <jcoxon> Dan-K2VOL, our rules are a bit uk centric
[18:26] <fsphil> "maybe" a launch this weekend - predictions behaving themselves so far
[18:27] <jcoxon> fsphil, hehe is always maybe
[18:27] Jasperw (~jasperw@93.97.176.250) joined #highaltitude.
[18:27] <jcoxon> its*
[18:28] <Dan-K2VOL> jcoxon, good point, but I think it's similar enough to consider when thinking about making others in other countries
[18:28] <fsphil> ain't it lol - every time I say it's going ahead, the prediction changes or someone can't make it
[18:29] <fsphil> there will *definitely* be a launch the weekend after next, regardless of prediction
[18:38] Laurenceb_ (~Laurence@host86-172-188-211.range86-172.btcentralplus.com) joined #highaltitude.
[18:44] <cuddykid> haha, can't wait to see if I can pick up from worcs... will probably be fine
[18:48] <griffonbot> @adamcudworth: Calibrated temp sensors for flight.. were a whole 2C out! Need to code in correction now.. #HABE1 #UKHAS [http://twitter.com/adamcudworth/status/75634641951932416]
[18:53] <fsphil> I think the record for 300 baud signals is about 350km
[18:55] <fsphil> the error correction might help too
[18:58] mattltm (~mattltm@81.134.144.179) left irc:
[19:07] <Laurenceb_> better tracking would help way more
[19:07] <Laurenceb_> fldigi doesnt have very good dll/fll code
[19:11] <fsphil> dll/fll? like a pll?
[19:11] <jonsowman> http://en.wikipedia.org/wiki/Frequency-locked_loop
[19:12] <jonsowman> http://en.wikipedia.org/wiki/Delay-locked_loop
[19:12] <Laurenceb_> yes
[19:12] <jonsowman> :)
[19:12] <Laurenceb_> fll tracks frequency
[19:12] <Laurenceb_> ive coded this in matlab using a correlator to find packet headers
[19:12] <Laurenceb_> managed about a 10dB improvement of fldigi
[19:12] <Laurenceb_> *on
[19:13] <fsphil> nice
[19:13] <fsphil> does it handle the frequency drift of the ntx2?
[19:14] <Laurenceb_> yes
[19:15] <Laurenceb_> massive hack, none of my code is very usable :P
[19:15] <Laurenceb_> it needs writing in c, never got round to that
[19:15] <fsphil> got the t-shirt :)
[19:15] <Laurenceb_> also the correlator was brute force
[19:16] <Laurenceb_> you can do it much faster with fft
[19:16] <Laurenceb_> libfft or whatever its called
[19:17] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) left irc: Ping timeout: 240 seconds
[19:17] <fsphil> I could transfer the data faster without the start or stop bits, but fldigi needs them for the frequency tracking to work
[19:18] <Laurenceb_> if you code a dll and use a prn type header you can drop them
[19:19] <Laurenceb_> then you can stick in a ton of fec :D
[19:19] <Laurenceb_> i wouldnt be suprised in +20dB on fldigi is possible
[19:21] <jcoxon> Laurenceb_, do it!
[19:21] <Laurenceb_> heh
[19:21] <Laurenceb_> ive got a bazillion things to do
[19:21] <Laurenceb_> its a week or twos work, but not very hard
[19:22] <jcoxon> do it do it do it
[19:22] <fsphil> lol
[19:22] <jonsowman> hi jcoxon
[19:22] <Laurenceb_> fldigi suffers massively as it tries to sync off start stop bit edges
[19:23] <fsphil> that it does
[19:23] <Laurenceb_> you can easily lose several characters
[19:23] <Laurenceb_> from one tiny bit of noise
[19:23] <fsphil> I got about 5 extra image packets from a recording of the last flight
[19:23] <fsphil> by using stricter timing
[19:23] <Laurenceb_> dll avoids that
[19:23] <Laurenceb_> and you dont need start stop bits
[19:23] <Laurenceb_> interesting
[19:24] <Laurenceb_> the fll on fldigi is _slgihtly_ better
[19:24] <Laurenceb_> but it still seems rather buggy in my experience
[19:24] <Laurenceb_> also you need to set it up by hand
[19:24] <Laurenceb_> you should be able to enter the packet header and the decoder will find it by correlating
[19:25] <Laurenceb_> then you could have multiple payloads at once as well
[19:30] <Laurenceb_> fsphil: so my code worked by taking audio stream and having a rolling buffer one bit period long
[19:31] <Laurenceb_> then i took dft of that
[19:31] <Laurenceb_> each time a sample came in
[19:32] <Laurenceb_> then go through frequency taking amplitude(f+deviation)-amplitude(f)
[19:33] <Laurenceb_> http://i.imgur.com/smd1K.png <-looks like that (data down the middle)
[19:34] <fsphil> that the dft of a bit-wide buffer?
[19:34] <Laurenceb_> not exactly
[19:35] <Laurenceb_> amplitude(dft(f))-amplitude(dft(f+deviation))
[19:35] <fsphil> aah I see the bits
[19:35] <Laurenceb_> as a waterfall
[19:35] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[19:35] <fsphil> gotcha
[19:36] <Laurenceb_> on each frequency you runa correlator to find headers -> thats a header found at 900 http://i.imgur.com/GNAMG.png
[19:36] <Laurenceb_> can you see why this eats the cpu yet :P
[19:36] <fsphil> slightly lol
[19:37] <fsphil> my laptop's going to hate you :p
[19:37] <Laurenceb_> theres lots of ways to speed it up with fft, then use libfft and write it in c
[19:37] <Laurenceb_> my matlab is about 10% real time
[19:38] BrainDamage (BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) left #highaltitude ("Leaving.").
[19:38] <Laurenceb_> then you fire up a dll+fll on the correct freuqency and start time
[19:38] <Laurenceb_> you start it at the start of the header so it can tune up
[19:38] <Laurenceb_> http://i.imgur.com/psuhl.png
[19:39] <Laurenceb_> squares is dll offset - its oscillating a bit :P
[19:39] BrainDamage (~BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) joined #highaltitude.
[19:40] <Laurenceb_> and you get some data -> http://i.imgur.com/NfQHN.png
[19:41] <Laurenceb_> that had about 50% character error rate, fldigi couldnt get anything and there was ~nothing on the fldigi waterfall
[19:41] <fsphil> one of my samples? :)
[19:41] <Laurenceb_> no, from one of jcoxons floaters
[19:42] <Laurenceb_> i recorded it from 450km with my sdr and a crappy ant
[19:42] <Laurenceb_> its important to use dft as you need closely spaced frequency bins to get a good signal
[19:43] <Laurenceb_> but you could use multiple offset ffts
[19:43] <Laurenceb_> offset in frequency
[19:45] <fsphil> I'll try to remember to record the signal of the next flight, see what I can get from it
[19:45] <fsphil> libfft or fftw?
[19:48] <Laurenceb_> oh yeah fftw
[19:49] <SpeedEvil> Just rewrite it to use the GPU
[19:50] <Laurenceb_> lol
[19:50] <fsphil> just :)
[19:50] <Laurenceb_> fpga ftw
[19:51] <fsphil> mmm dedicated telemetry receiver
[19:52] shipit (~sumeet@204-15-2-155-static.ipnetworksinc.net) joined #highaltitude.
[20:00] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Quit: Page closed
[20:16] mattltm (~mattltm@81.134.144.179) joined #highaltitude.
[20:28] DarkCow (~DarkCow@5adefc6a.bb.sky.com) left irc: Remote host closed the connection
[20:29] Hiena (~boreger@81.93.195.181.datatrans.hu) left irc: Remote host closed the connection
[20:29] fsphil-laptop (~phil@2001:8b0:34:1:219:d2ff:fe09:a6b9) joined #highaltitude.
[20:38] <Laurenceb_> interesting - if i dont include Coriolis the accel drift is significantly worse
[20:38] <SpeedEvil> ?
[20:39] <SpeedEvil> What do you mean?
[20:39] <Laurenceb_> G isnt vertical
[20:39] <SpeedEvil> Doesn't that just constantly offset?
[20:39] <SpeedEvil> Or you mean in a mag+accel solution.
[20:39] <Laurenceb_> yes.. its intriguing
[20:40] <Laurenceb_> openpilot just adds a offset to velocity after each time step
[20:40] <Laurenceb_> so you treat it as a frame thing
[20:40] <Laurenceb_> guess its a non inertial frame
[20:45] <Laurenceb_> hadnt thought of it like that before - makes it simpler
[20:53] mattltm (~mattltm@81.134.144.179) left irc:
[20:59] jkominar (~justin@64.235.97.218) left irc: Quit: It was a perfectly cromulent thing to do.
[21:00] <SpeedEvil> Hmm/.
[21:00] Dan-K2VOL (~Adium@74-143-44-42.static.insightbb.com) left irc: Quit: Leaving.
[21:00] <SpeedEvil> This only happens if you've got more than just accels?
[21:05] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Ping timeout: 258 seconds
[21:06] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[21:15] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Quit: The cake is a lie !
[21:17] fsphil-laptop (~phil@2001:8b0:34:1:219:d2ff:fe09:a6b9) left irc: Ping timeout: 260 seconds
[21:34] <Laurenceb_> yes
[21:34] <SpeedEvil> Can the direction of the required corrections act as a gyrocompass/
[21:36] <Laurenceb_> from the gps?
[21:36] <Laurenceb_> possibly
[21:36] <Laurenceb_> youd have to be moving about a bit
[21:36] <Laurenceb_> as in moving and changing direction
[21:36] <SpeedEvil> yeah - I suppose so.
[21:37] <Laurenceb_> but gps velocity is pretty good if you have the righ receiver
[21:37] <Laurenceb_> e.g. can pick up me shaking it on the desk
[21:38] <SpeedEvil> Oh - sure - I wasn't imagining it'd be a good source of data - just wondering if it was usable at all.
[21:39] jcoxon (~jcoxon@cpc1-lanc4-0-0-cust720.3-3.cable.virginmedia.com) left irc: Quit: Leaving
[21:39] <Laurenceb_> itd work with good enough sensors
[21:40] <Laurenceb_> i think it might be borderline possible to detect earths rotation with the itg3200
[21:41] <Laurenceb_> its about 0.1lsb
[21:41] <SpeedEvil> :)
[21:41] <SpeedEvil> You can almost get 1/r^2 coming into play for g too.
[21:41] <SpeedEvil> But the resolution sucks.
[21:42] <Laurenceb_> noise is 0.6lsb
[21:42] <Laurenceb_> at 125sps
[21:44] <Laurenceb_> so 1 second averaging or so
[21:46] <Laurenceb_> the itg3200 is insane - almost an order of magnitude lower noise than in the datasheet
[21:52] <eroomde> Laurenceb_: you like it
[21:53] <eroomde> am putting it in to a new design soon
[21:59] <Laurenceb_> i suspect theres some parts binning at play
[22:00] <Laurenceb_> my best guess - insensce found many of their itg3200 sensors had much lower noise
[22:00] <Laurenceb_> so they created imu6000
[22:00] <Laurenceb_> which has noise matching my sensor
[22:01] kd0mto (~dago@sdrc.com-162.fuse.net) left irc: Quit: AFK, I'm IRL now.
[22:02] <SpeedEvil> Laurenceb: But it's really the same as the itg
[22:02] <SpeedEvil> The itg is the gyro part?
[22:04] stentor (~stentor@tarsonis.acad.rose-hulman.edu) left irc: Ping timeout: 276 seconds
[22:04] <Laurenceb_> yes
[22:04] <Laurenceb_> i wonder if you buy an itg from after the imu came out itll have higher noise
[22:08] <SpeedEvil> Assuming they sell enough of them.
[22:12] <Laurenceb_> think ive got the calibration worked out
[22:12] <Laurenceb_> just needs 7 randomly arranged points
[22:12] <SpeedEvil> Awesome!
[22:12] <SpeedEvil> You leave it stationary in those points for a bit?
[22:13] <Laurenceb_> then it calibrates the accel and magno to give |G|, |B|, and G.B
[22:13] <Laurenceb_> yes
[22:13] <Laurenceb_> its autodetects those points
[22:13] stentor (~stentor@tarsonis.acad.rose-hulman.edu) joined #highaltitude.
[22:13] <Laurenceb_> or at least ive got all this working in seperate files... need to put it all together
[22:14] <Laurenceb_> needs the USGS model sticking in there - ideally you have a config file with your location/enter it at the command line
[22:14] Lunar_Lander (~lunar_lan@p54883905.dip.t-dialin.net) joined #highaltitude.
[22:14] <Laurenceb_> and itll run and output the header file to incorporate into the build
[22:14] <Lunar_Lander> hello everyone
[22:14] <Laurenceb_> some work to go yet :P
[22:16] <Laurenceb_> probably needs to be done outside as well - tried recording data at work in a reinforced concrete building and |B| is 15% lower
[22:18] stentor (~stentor@tarsonis.acad.rose-hulman.edu) left irc: Ping timeout: 240 seconds
[22:21] Action: Laurenceb_ -> zzzzzz
[22:23] <SpeedEvil> http://cgi.ebay.co.uk/370420607007 - for a case? :)
[22:23] <SpeedEvil> Night!
[22:23] <SpeedEvil> Sounds like interesting stuff!
[22:28] Laurenceb_ (~Laurence@host86-172-188-211.range86-172.btcentralplus.com) left irc: Ping timeout: 255 seconds
[22:35] kd0mto (~dago@74-94-125-113-illinois.hfc.comcastbusiness.net) joined #highaltitude.
[22:48] stentor (~stentor@tarsonis.acad.rose-hulman.edu) joined #highaltitude.
[23:03] TylerD (~TylerD@unaffiliated/tylerd) left irc: Ping timeout: 248 seconds
[23:03] TylerD (~TylerD@unaffiliated/tylerd) joined #highaltitude.
[23:07] <Lunar_Lander> SpeedEvil yea interesting dice
[23:18] [STAR]Atanyi|Wor (~Upu@ubn.upuaut.net) joined #highaltitude.
[23:19] BrainDamage (~BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) left irc: Read error: Operation timed out
[23:20] Upu (~Upu@ubn.upuaut.net) left irc: Ping timeout: 240 seconds
[23:51] Lunar_Lander (~lunar_lan@p54883905.dip.t-dialin.net) left irc: Ping timeout: 240 seconds
[23:54] imrcly (~tim@74-128-123-149.dhcp.insightbb.com) left irc: Ping timeout: 240 seconds
[23:55] imrcly (~tim@74-128-123-149.dhcp.insightbb.com) joined #highaltitude.
[00:00] --- Wed Jun 1 2011