highaltitude.log.20110602

[00:20] juxta (~Terry@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 248 seconds
[00:30] MoALTz_ (~no@92.18.20.184) joined #highaltitude.
[00:31] MoALTz (~no@92.18.20.184) left irc: Ping timeout: 252 seconds
[00:39] jasonb (~jasonb@12.201.129.162) left irc: Ping timeout: 246 seconds
[00:41] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[00:49] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[00:50] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 246 seconds
[01:10] BrainDamage (~BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) left irc: Ping timeout: 246 seconds
[02:14] imrcly (~tim@74-128-123-149.dhcp.insightbb.com) left irc: Quit: leaving
[02:48] DarkCow (~DarkCow@5adefc6a.bb.sky.com) left irc: Ping timeout: 276 seconds
[03:22] kd0mto (~dago@12.163.240.17) left irc: Excess Flood
[03:33] juxta_ (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 244 seconds
[04:00] TangoAlpha (~TangoAlph@cpc5-rdng22-2-0-cust160.15-3.cable.virginmedia.com) left irc: Read error: Connection reset by peer
[04:16] TangoAlpha (~TangoAlph@cpc5-rdng22-2-0-cust160.15-3.cable.virginmedia.com) joined #highaltitude.
[04:48] shipit (~sumeet@204-15-2-155-static.ipnetworksinc.net) left irc: Remote host closed the connection
[04:48] shipit (~sumeet@204-15-2-155-static.ipnetworksinc.net) joined #highaltitude.
[06:01] m0lep (~irc-clien@41.215.92.62) joined #highaltitude.
[06:19] m0lep (~irc-clien@41.215.92.62) left irc: Quit: Leaving
[06:20] spacekitteh (~TraumaPon@203-214-99-19.dyn.iinet.net.au) joined #highaltitude.
[06:44] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[06:44] pericynthion (~henry@c-174-62-68-92.hsd1.ca.comcast.net) left irc: Ping timeout: 252 seconds
[06:58] Nick change: spacekitteh -> trollytroll
[06:58] Nick change: trollytroll -> spacekitteh
[07:58] mattltm-alt (~mattltm-a@mail.icm2.org.uk) joined #highaltitude.
[08:08] GW8RAK (~chatzilla@host86-134-243-150.range86-134.btcentralplus.com) joined #highaltitude.
[08:16] <SamSilver> bbl
[08:16] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Quit: Page closed
[08:22] shipit (~sumeet@204-15-2-155-static.ipnetworksinc.net) left irc: Remote host closed the connection
[08:29] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[09:01] <griffonbot> Received email: Steve Randall "Re: [UKHAS] Re: UK Balloon Flight Guidelines"
[09:58] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[10:00] BrainDamage (~BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) joined #highaltitude.
[10:45] GW8RAK (~chatzilla@host86-134-243-150.range86-134.btcentralplus.com) left irc: Read error: Connection reset by peer
[11:00] RocketBoy (~steverand@217.47.75.8) joined #highaltitude.
[11:25] RocketBoy (steverand@217.47.75.8) left #highaltitude.
[11:29] <SpeedEvil> http://fatpita.net/?i=4380 (on topic)
[11:30] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) joined #highaltitude.
[11:48] GW8RAK (~chatzilla@host86-134-243-150.range86-134.btcentralplus.com) joined #highaltitude.
[11:49] Laurenceb_ (~Laurence@host86-172-188-211.range86-172.btcentralplus.com) joined #highaltitude.
[11:56] Nick change: MoALTz_ -> MoALTz
[12:03] <Randomskk> DanielRichman: http://www.bbc.co.uk/news/education-13627415
[12:04] nelly11 (8afa0b0a@gateway/web/freenode/ip.138.250.11.10) joined #highaltitude.
[12:05] <nelly11> Hi All. What is the name of the "case thing" that normally surrounds a cutdown mechanism? Is is normally purchased or custom made
[12:10] <SpeedEvil> Custom.
[12:10] <SpeedEvil> There aren't any real standard cutdowns.
[12:11] <nelly11> Ok :)
[12:13] DarkCow (~DarkCow@5adefc6a.bb.sky.com) joined #highaltitude.
[12:15] nelly11 (8afa0b0a@gateway/web/freenode/ip.138.250.11.10) left irc: Quit: Page closed
[12:15] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) left irc: Ping timeout: 276 seconds
[12:16] <Laurenceb_> ive used hot resistor currers
[12:27] <Laurenceb_> hmf my cal code isnt working
[12:27] <Laurenceb_> seems it cant stably find the rotation matrix between the magno and accel sensors
[12:59] <Laurenceb_> rotation is confusing
[13:00] <griffonbot> Received email: Philip Heron "[UKHAS] hadie:3 launch this weekend (again)"
[13:12] <fsphil> I bet something stupid happens between now and saturday
[13:12] jgrahamc (58d33ade@gateway/web/freenode/ip.88.211.58.222) joined #highaltitude.
[13:13] RocketBoy (~steverand@217.47.75.8) joined #highaltitude.
[13:13] RocketBoy (steverand@217.47.75.8) left #highaltitude.
[13:13] <mattltm-alt> Have faith Phil :)
[13:13] <fsphil> haha
[13:14] <mattltm-alt> Images look good.
[13:14] <mattltm-alt> Do you think I gould decode them from here?
[13:15] <fsphil> that I'm not sure about. I got fragments of text from the last 300 baud launch in cambridge
[13:15] <fsphil> via the colinear
[13:15] <fsphil> so with a good yagi, it's possible
[13:18] <jgrahamc> I'm going to try to listen to the hadie:3 launch from London on Saturday.
[13:19] <fsphil> the antenna should remain firmly attached this time, it's just a piece of coax :)
[13:23] <fsphil> thanks jgrahamc - what sort of setup do you have?
[13:24] <jgrahamc> I will be using a 6 element Yagi plus my Yaesu FT-790R II
[13:25] Dan-K2VOL (~Dan-K2VOL@74-143-44-42.static.insightbb.com) joined #highaltitude.
[13:26] <Upu> what time you planning on launching fsphil ?
[13:27] <fsphil> 11:00 BST
[13:27] <fsphil> (ish)
[13:27] <Upu> k
[13:29] <cuddykid> good luck fsphil, great weather for launch
[13:29] <cuddykid> i'll try and listen in for a bit
[13:29] <fsphil> nice day here too- forecast is for a bit of cloud
[13:31] <Darkside> http://www.seeedstudio.com/depot/gprs-shield-p-779.html gprs shield back in stock!
[13:31] <Darkside> cheap too!
[13:32] <cuddykid> fed up with this notam... been waiting months! Launch window is in about 4 weeks, hopefully he'll send it soon
[13:40] delllat (7aad3a91@gateway/web/freenode/ip.122.173.58.145) joined #highaltitude.
[13:40] delllat (7aad3a91@gateway/web/freenode/ip.122.173.58.145) left irc: Client Quit
[13:41] jkominar (~justin@64.235.97.218) joined #highaltitude.
[13:42] GW8RAK (~chatzilla@host86-134-243-150.range86-134.btcentralplus.com) left irc: Quit: ChatZilla 0.9.87 [Firefox 4.0.1/20110413222027]
[13:45] Dan-K2VOL (~Dan-K2VOL@74-143-44-42.static.insightbb.com) left irc: Quit: Leaving.
[13:50] <W0OTM> anyone work with thelinkbox?
[13:53] <eroomde> cuddykid: I sympathise
[13:53] <eroomde> he's not rhe fastest
[13:53] imrcly (~tim@74-128-123-149.dhcp.insightbb.com) joined #highaltitude.
[13:55] jasonb (~jasonb@adsl-66-124-73-250.dsl.sntc01.pacbell.net) joined #highaltitude.
[13:55] Dan-K2VOL (~Adium@74-143-44-42.static.insightbb.com) joined #highaltitude.
[14:04] jgrahamc (58d33ade@gateway/web/freenode/ip.88.211.58.222) left irc: Quit: Page closed
[14:17] <Laurenceb_> eroomde: ever tried calibrating an imu?
[14:19] <eroomde> not to the homeric levels that it seems you have been!
[14:20] <Laurenceb_> heh
[14:20] <Laurenceb_> seems im wasting my time
[14:20] <Laurenceb_> unless i can be bothered with 20 points or os
[14:21] <Laurenceb_> just been doing some tests and it seems basic 6 dof bias/gain cal gives the best results
[14:22] <Laurenceb_> was actually better than even with just cross axis added - tho i only had 12 calibration points
[14:23] <Laurenceb_> the fitter becomes rather unstable and is prone to introduce excessive cross axis and similar
[14:23] <SpeedEvil> You you get mal-fits basically as you go closer to the noise?
[14:24] <SpeedEvil> What's the problem - the models of magnetic field not matching reality?
[14:24] <SpeedEvil> Moving during cal?
[14:24] <Laurenceb_> its the way im trying to do it with random points
[14:24] <Laurenceb_> and newton raphson - it becomes rather unstable as you do to more dof
[14:25] <SpeedEvil> So you tend to get 'holes' in the solution?
[14:25] <SpeedEvil> Which the solver fills with bogosity?
[14:25] <Laurenceb_> unless you increase the number of cal points well beyond necessary
[14:25] <Laurenceb_> im thinking of it more as the goodness of fit function in state space has multiple minimal and manifolds of good fit
[14:25] <Laurenceb_> the solver eaily ends up in the wrong place
[14:26] <Laurenceb_> with 6 dof the fit function in state space is less structured and the solver finds the correct solution
[14:27] <Laurenceb_> its just less accurate as theres no axis misalignment
[14:27] <Laurenceb_> im looking at ~0.25degree rms error
[14:27] <SpeedEvil> Hmm.
[14:27] <SpeedEvil> Yeah - if you have proper axis misalignment solving, it adss a _lot_ of DOG
[14:28] <SpeedEvil> DOF
[14:28] <Laurenceb_> the alternative is a proper alignment rig - thats how this is done commercially aiui
[14:28] <SpeedEvil> Are you cheating - and inputting real axis misalignments? - for example, assuming that stuff on the same chip is orthogonal?
[14:28] <Laurenceb_> no im not
[14:28] <griffonbot> Received email: Mission Excelsior "Re: [UKHAS] hadie:3 launch this weekend (again)"
[14:28] <Laurenceb_> every sensor axis has its own 3 component sensitivity vector and scalar offset
[14:29] <SpeedEvil> Wow. That's a _lot_ of degrees of freedom. What - 36
[14:29] <Laurenceb_> 21 for the accel and magno
[14:30] <Laurenceb_> the gyro is solved later
[14:30] <Laurenceb_> and theres 3 that cant be solved for - mounting of the accel to the pcb
[14:31] <Laurenceb_> it simplifies down to bias, gain cross axis and misalignment between magno and accel
[14:32] <Laurenceb_> i had to take it down to 12, or 6 for each sensor - just bias and gain to make it work best
[14:34] <Laurenceb_> the misalignment rotation between accel and magno was particularly unstable - cross axis might work ok with about 20 cal points
[14:36] <Laurenceb_> geckosenator could probably offer some advice - hes made this work
[14:36] <SpeedEvil> Do you actually see meaningfull cross errors intra-device?
[14:37] <SpeedEvil> I don't think he was going for this sort of accuracy
[14:37] <SpeedEvil> I mean misalignment
[14:37] <SpeedEvil> Oh - I suppose you can't really disentangle
[14:38] <Laurenceb_> im not sure - ive read specs of +-0.1degree for die placement in devices
[14:38] <Laurenceb_> so if thats the case its of the same ballpark i was aiming for, so not reall an issue
[14:39] <SpeedEvil> I suppose a couple of bearings, a couple of servos, and a relay to turn the servos on and off, and you pretty much have it
[14:39] <SpeedEvil> (for a random align jig)
[14:40] <Laurenceb_> yeah - turning it by hand is a pita
[14:40] <Laurenceb_> always seems to be slowly slipping when i set it down - that makes the averaging useless
[14:41] <SpeedEvil> I would suggest use a nice swivel vice - but...
[14:44] <Laurenceb_> the misalignment between gyro and magno+accel could be significant
[14:44] <Laurenceb_> https://github.com/Laurenceb/Dactyl/raw/master/Hardware/Assembly/photo2.JPG
[14:44] <Laurenceb_> looks pretty small to me
[14:48] <SpeedEvil> cant?
[14:48] <SpeedEvil> I suppose it's not going to be big.
[14:48] <Laurenceb_> looks like couple of degrees max to me
[15:22] m1x10 (m1x10@ppp046177092129.dsl.hol.gr) joined #highaltitude.
[15:36] Ikarus (~ikarus@46-129-66-189.dynamic.upc.nl) left irc: Ping timeout: 252 seconds
[15:41] <m1x10> Hi punks
[15:43] jasonb (~jasonb@adsl-66-124-73-250.dsl.sntc01.pacbell.net) left irc: Ping timeout: 240 seconds
[15:56] m1x10 (m1x10@ppp046177092129.dsl.hol.gr) left irc: Ping timeout: 240 seconds
[15:59] Ikarus (~ikarus@46-129-66-189.dynamic.upc.nl) joined #highaltitude.
[16:02] m1x10 (m1x10@ppp046177022190.dsl.hol.gr) joined #highaltitude.
[16:04] <Laurenceb_> Randomskk: ping
[16:04] <Randomskk> hi
[16:04] <Laurenceb_> EXTI lines are grouped as they go into the NVIC?
[16:05] <Randomskk> think so, can't remember for sure
[16:05] <Laurenceb_> the reference manual isnt very clear
[16:05] <Laurenceb_> seem to be looking at the code
[16:05] <Laurenceb_> thanks
[16:08] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[16:15] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Quit: Page closed
[16:16] fsphil (~phil@2001:8b0:34:1:21f:c6ff:fe44:b25b) left irc: Ping timeout: 260 seconds
[16:19] fsphil (~phil@2001:8b0:34:1:21f:c6ff:fe44:b25b) joined #highaltitude.
[16:30] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) joined #highaltitude.
[16:51] shipit (~sumeet@c-67-180-23-104.hsd1.ca.comcast.net) left irc: Remote host closed the connection
[16:56] m1x10 (m1x10@ppp046177022190.dsl.hol.gr) left irc: Quit: If you run you only gonna die tired
[17:02] <eroomde> jonsowman: ping
[17:13] MrCraig (~craig@host109-156-178-239.range109-156.btcentralplus.com) joined #highaltitude.
[17:15] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 244 seconds
[17:17] pericynthion (~henry@c-174-62-68-92.hsd1.ca.comcast.net) joined #highaltitude.
[17:27] DarkCow (~DarkCow@5adefc6a.bb.sky.com) left irc: Remote host closed the connection
[17:36] jasonb (~jasonb@12.201.129.162) joined #highaltitude.
[17:41] <cuddykid> is anyone competing in the N project here?
[17:42] <SpeedEvil> Laurenceb had a look at it.
[17:42] <SpeedEvil> He's got some simulations.
[17:42] <SpeedEvil> As far as I'm aware he's not actually planning to do it though.
[17:43] <cuddykid> looks an incredible project, however, I can imagine it would take some very good engineering skills and a reasonable budget
[17:43] <cuddykid> cool
[17:43] <SpeedEvil> Well - no.
[17:43] <jonsowman> eroomde: pong
[17:43] <SpeedEvil> It can't take a reasonable budget - owing to the limits.
[17:46] <cuddykid> yeah, but a larger budget than the average hab flight!
[17:46] <SpeedEvil> cuddykid: Not much.
[17:46] <SpeedEvil> cuddykid: Say couple of hundred for HAB
[17:46] <SpeedEvil> Maybe 10 launches worth
[17:47] <cuddykid> yeah
[17:48] <Laurenceb_> erm they are kind of running out of time :P
[17:48] <Laurenceb_> N prize is a bit of a joke
[17:49] <SpeedEvil> In principle, it's possible.
[17:49] <SpeedEvil> But the cost of time dwarfs the cost of materials.
[17:49] <SpeedEvil> It's like in principle you can make a carbon-carbon composite structure for a tenner.
[17:49] <SpeedEvil> Because pitch is cheap, and carbon fibre tow is cheap.
[17:50] <Laurenceb_> http://www.youtube.com/watch?v=JYNRhzX6gw8
[17:54] <SpeedEvil> http://www.youtube.com/watch?v=-RP19fnff_c - while we're on random stuff
[17:55] <Laurenceb_> lulwuyt
[17:56] <SpeedEvil> osteogenesis silver fracture
[17:56] <SpeedEvil> oops
[17:58] <pericynthion> you guys are silly
[18:01] <Laurenceb_> pericynthion: long time no see
[18:01] <Laurenceb_> what are you up to these days?
[18:05] <SpeedEvil> http://en.wikipedia.org/wiki/Photoacoustic_imaging_in_biomedicine - interesting.
[18:05] <pericynthion> the usual - airborne wind energy, low cost high performance GPS
[18:05] <pericynthion> you?
[18:06] <cuddykid> is sellotape alright to affix the harmless notice (paper) to outside of payload or will it fall off due to cold temps?
[18:06] <SpeedEvil> pericynthion: What do you mean by high performance?
[18:06] <pericynthion> Laurenceb_ used bluetack to attach his parachute, so sellotape for the notice is probably fine
[18:06] <SpeedEvil> More accurate positions, or high dynamics?
[18:07] <Laurenceb_> lol when did i use bluetack?
[18:07] <cuddykid> pericynthion thanks!
[18:07] <pericynthion> RTK accuracy (~2cm within a few km of the base station), and high dynamics, and multiple antennas, and high update rate
[18:07] <SpeedEvil> pericynthion: Shiny!
[18:07] <cuddykid> lol Laurenceb_
[18:07] <Laurenceb_> are you still working on high altitude wind energy?
[18:07] <SpeedEvil> pericynthion: Got it down to 20 quid yet? :)
[18:07] Hiena (~boreger@81.93.195.181.datatrans.hu) joined #highaltitude.
[18:07] <pericynthion> Laurenceb_: yes, SpeedEvil: not quite
[18:08] <Laurenceb_> nice, is there a company site?
[18:08] <pericynthion> www.makanipower.com
[18:08] Action: Laurenceb_ is presently working on https://github.com/Laurenceb/Dactyl/raw/master/Hardware/Assembly/photo2.JPG
[18:08] <SpeedEvil> I was sort-of-wondering about that - a DIY test isn't _completely_ impossible.
[18:08] <SpeedEvil> As 10km of the best fishing line is - for example - a suitable anchor - if you do it right.
[18:08] <pericynthion> SpeedEvil: BOM cost for the single antenna version is around £45
[18:08] shipit (~sumeet@204-15-2-155-static.ipnetworksinc.net) joined #highaltitude.
[18:08] <SpeedEvil> And it's barely liftable by a HAB
[18:09] <SpeedEvil> pericynthion: You're using some sort of GPS->bits chip and then jamming it into USB or somethign?
[18:09] <SpeedEvil> Or is it a 'proper' GPS
[18:09] <Hiena> Anybody has experience about the DDR memories refresh cycle?
[18:09] <pericynthion> have spent >$100k on development though, so we're going to sell them with a considerable margin, but still a lot cheaper than anything on the market
[18:09] <SpeedEvil> pericynthion: :)
[18:09] <pericynthion> it's independent of PC
[18:09] <SpeedEvil> pericynthion: Nice!
[18:09] <Laurenceb_> custom GPS decoder?
[18:09] mixio (mixio@ppp046177022190.dsl.hol.gr) joined #highaltitude.
[18:10] <SpeedEvil> pericynthion: I got partway through a GPS design - using stm32 and software decoding, but then stalled.
[18:10] <pericynthion> I have a version that uses a Blackfin DSP to do software reception
[18:10] <Laurenceb_> ah your blackfin thingy
[18:10] <mixio> hi
[18:10] <Laurenceb_> so you got it working in the end?
[18:10] <SpeedEvil> Hiena: What do you mean?
[18:10] <pericynthion> currently working on FPGA version which supports a multi-bit frontend for more EMI immunity
[18:10] <Laurenceb_> nice
[18:10] <SpeedEvil> Hiena: you more or less need to refresh every line of every bank every 4ms or so
[18:10] <pericynthion> it worked to control one of our turbines doing 60m/s circles at 12g accel
[18:10] <SpeedEvil> Hiena: self-refresh works
[18:11] <Laurenceb_> pericynthion: autogyro type device?
[18:11] BrainDamage1 (~BrainDama@dynamic-adsl-94-36-235-37.clienti.tiscali.it) joined #highaltitude.
[18:11] BrainDamage (~BrainDama@dynamic-adsl-78-12-224-175.clienti.tiscali.it) left irc: Disconnected by services
[18:11] <SpeedEvil> I was looking at something doing 1km circles at 22g the other day. Fun.
[18:11] Nick change: BrainDamage1 -> BrainDamage
[18:11] <SpeedEvil> (dynamic soaring with r/c gliders)
[18:11] <pericynthion> Laurenceb_: not exactly autogyro
[18:11] <pericynthion> it's a lifting wing
[18:11] <SpeedEvil> pericynthion: You've had stuff flying?
[18:12] <pericynthion> yes
[18:12] <SpeedEvil> And I read that as mankinipower.
[18:12] <SpeedEvil> :)
[18:12] <Laurenceb_> http://www.makanipower.com/2010/06/airborne-wind-turbine/
[18:12] <Laurenceb_> ah i see
[18:12] <Laurenceb_> genius
[18:12] <pericynthion> speedevil: in two years I'd never thought of that
[18:12] <pericynthion> you have a special talent
[18:12] <Hiena> SpeedEvil: I'm in the middle setting up a new config, and thinking about buying some high clock rate module for a cheap motherboard.
[18:12] <Laurenceb_> so you have a turbine on the back to generate the power
[18:13] <SpeedEvil> Hiena: The chipset deals with refresh then.
[18:13] <Laurenceb_> im currently using a lamish ublox5 https://github.com/Laurenceb/Dactyl/raw/master/Hardware/Assembly/photo1.JPG
[18:13] <pericynthion> several rotors that do {generation, VTOL propulsion, active stabilisation}
[18:13] <Hiena> The board rated to 800MHz and i thought to install 1800MHz module.
[18:14] <Laurenceb_> how big is the gps?
[18:14] <SpeedEvil> pericynthion: Interesting! I see problems with the 'but birds won't see it' crowd.
[18:14] <pericynthion> single antenna version is 50x50mm without the radio for differential corrections
[18:14] <Laurenceb_> not bad
[18:14] <pericynthion> SpeedEvil: also icing, lightning, airspace, insurance
[18:14] <pericynthion> fatigue, maintenance, redundancy
[18:15] <pericynthion> it's never going to be ublox sized
[18:15] <Laurenceb_> space solar will be economical soon
[18:15] <pericynthion> so what's your Dactyl got apart from a ublox?
[18:16] <Laurenceb_> 9dof imu, 434mhz radio, stm32
[18:16] <Laurenceb_> smps, sd, pitot, baro, servo pwm in/out and rs485
[18:16] <Laurenceb_> sort of a does everything uav board
[18:17] <pericynthion> neat
[18:17] <Laurenceb_> presently developing firmware - mems sensor calibration is a pita
[18:17] <pericynthion> have you seen LISA/M?
[18:17] <Laurenceb_> no?
[18:18] <Laurenceb_> http://paparazziuav.org/wp-content/uploads/2010/12/Lisa_m_bot_small.jpg
[18:18] <Laurenceb_> aha - yeah ive gone with the same gyro, but combined magno accel from st
[18:18] <Laurenceb_> same micro
[18:18] <pericynthion> https://www.jobyrobotics.com/product.php?id_product=54
[18:18] <pericynthion> right
[18:18] <Laurenceb_> ST will have a (DOF IMU ic out soon
[18:19] <Laurenceb_> - 2 months top
[18:19] <Laurenceb_> *9
[18:19] <pericynthion> really?
[18:19] <Laurenceb_> but their gyro design has low resonant frequency and suffers from vibration
[18:19] <pericynthion> our ST rep didn't mention that
[18:19] <Laurenceb_> so an ST rep told me
[18:19] <pericynthion> yeah I know that's been an issue with the ST gyros
[18:19] <pericynthion> still, neat to have it all in one chip
[18:20] <Laurenceb_> it may have been cancled :-/ they keep changing their mems roadmap
[18:20] <pericynthion> Aspirin is a pretty nice IMU IMO
[18:20] <pericynthion> of course I am biased since we make it
[18:20] <Laurenceb_> good news is you can now get an lsm303dlh in a lga14
[18:21] <Laurenceb_> we?
[18:21] <pericynthion> I work for Joby Robotics and Makani
[18:21] <Laurenceb_> i though you worked for makani
[18:22] <Laurenceb_> ah
[18:22] <Laurenceb_> nice
[18:22] <pericynthion> they are related
[18:22] <Laurenceb_> can i pick your brains on IMU calibration?
[18:23] <Laurenceb_> im trying to write cal code that uses only a set of random orientations
[18:23] <Laurenceb_> unfortunately its rather badly behaved
[18:24] <Laurenceb_> do you think its possible to find a misalignment rotation matrix between a magno and accel knowing only what the dot product of B and G should be
[18:24] kd0mto (~dago@sdrc.com-162.fuse.net) joined #highaltitude.
[18:24] <pericynthion> hm I am not the best person to talk to about that, but I know one of our guys did something similar and it worked
[18:24] <Laurenceb_> and having a set of random orientations
[18:24] <Laurenceb_> interesting
[18:24] <pericynthion> I think it is possible
[18:24] <Laurenceb_> it seems to be unstable
[18:24] <Laurenceb_> but a- ive found some typos in my jacobians, and b-i only have 11 cal points
[18:25] <Laurenceb_> hmm i think it should be possible too - in theoretical land, interesting if its been done
[18:27] <Laurenceb_> for the time being im just doing bias and gain cal - adding cross axis and misalignment will only give me 0.5 degrees more accuracy or so
[18:27] <Laurenceb_> have you seen openpilot?
[18:28] Action: SpeedEvil wonders if his idea is applicable to wind turbines.
[18:28] <Laurenceb_> im using their EKF code :P
[18:29] <SpeedEvil> I had the thought of a weight-shift plane. You have a wing, with a mass swinging around it in the yaw axis. The centre of gravity of the wing/mass system moves up and down as the mass spins.
[18:29] <pericynthion> I don't think our nicest attitude estimators are open but we did backport some stuff to Paparazzi
[18:30] <SpeedEvil> The wing is wiggled according to the position in the cycle of the mass, so you get a net lift.
[18:30] <pericynthion> SpeedEvil: the mass is orbiting? or just free to swing?
[18:30] <Laurenceb_> openpilot is nice in terms of speed
[18:30] <SpeedEvil> I suppose you could run it backwards, and get power out.
[18:30] <SpeedEvil> pericynthion: orbiting.
[18:30] <Laurenceb_> they inlined tons of stuff - the code is horrendous for very quick
[18:30] <Laurenceb_> s/for/but
[18:30] <pericynthion> I don't quite get the picture
[18:31] <SpeedEvil> pericynthion: Imagine a horizontal wing - flying horizontally - with the centre of the wing in the fore-aft direction cut out, and a weight rotating around it.
[18:32] <SpeedEvil> pericynthion: So you get the weight moving from forward, to down, to back, to top of the wing.
[18:32] <SpeedEvil> This moves the wing up and down, as the centre of mass varies.
[18:32] <pericynthion> ok
[18:33] <SpeedEvil> You then have a cam arrangement or something that tilts the wing relative to the motion at different phases.
[18:33] <pericynthion> sounds a bit like an ornithopter
[18:33] <SpeedEvil> So the only energy you're putting in is the energy you need to make up the aerodymanic losses, and the amount of air moved is much higher.
[18:33] <SpeedEvil> Exactly.
[18:33] <SpeedEvil> But with a rigid wing.
[18:36] <pericynthion> cute
[18:36] <pericynthion> and yes, you could use it as an airborne turbine if you had a tether
[18:37] <SpeedEvil> I suppose in many ways it doesn't really matter what you use - the hardest part will be the politics.
[18:38] <SpeedEvil> I also wonder how the mandatory fenced-off-areas would work with right-to-roam legislation and such.
[18:39] <pericynthion> we're mostly looking at offshore
[18:39] <SpeedEvil> Makes sense.
[18:39] <pericynthion> pain in the ass though
[18:40] <SpeedEvil> I imagine landings are fun too :)
[18:40] <pericynthion> VTOL to a perch
[18:40] <SpeedEvil> That'd work well in low wind shutdowns - high wind would be amusing though.
[18:41] <SpeedEvil> Actually - I suppose normally you 'fly' at several times the windspeed.
[18:41] <SpeedEvil> So there is a fair bit of reserve power there.
[18:41] <SpeedEvil> I hope it works.
[18:42] <SpeedEvil> Cheaper wind turbines would be a good thing.
[18:42] <SpeedEvil> (though I'm still in favour of a massively upscaled nuclear system)
[18:42] <Laurenceb_> space solar ftw
[18:43] <SpeedEvil> That has fun issus of its own of course.
[18:43] <SpeedEvil> Solar concentrated, with thermal storage does look interesting.
[18:43] <Laurenceb_> how to slew the thing round is a pita
[18:43] <SpeedEvil> But the UK is not best placed for concentrated solar.
[18:43] <Laurenceb_> 24/7 space solar is harder than first appears
[18:44] <SpeedEvil> Are you assuming GEO, or MEO? or LEO?
[18:44] <Laurenceb_> geo
[18:45] <SpeedEvil> I vaguely recall a BOTE of mine was a sun-staring array, with a structure like a corner-cube on the back for steering the emissions.
[18:45] <SpeedEvil> Actively of course.
[18:48] <Laurenceb_> yeah thats the problem
[18:48] <Laurenceb_> you need a seperate strucutre for transmitting
[18:49] <SpeedEvil> But you can electrically steer.
[18:53] BrainDamage (~BrainDama@dynamic-adsl-94-36-235-37.clienti.tiscali.it) left irc: Read error: Operation timed out
[18:54] <mixio> just soldered the new board and it works
[18:54] <SpeedEvil> :) http://www.makanipower.com/technology/structures/ - the top drawing almost exactly matches one I have here for the silly wing thing.
[18:54] <SpeedEvil> (though it's very different if you look at all the views)
[18:58] <Laurenceb_> hmm openpilot hardcoded G in
[18:58] <SpeedEvil> pericynthion: Won't it be rather louder than a conventional turbine of the same megawattage. (though admittedly - you don't care about that so much if it's a t sea)
[18:58] <Laurenceb_> bad openpilot, bad
[18:58] <SpeedEvil> Laurenceb: reasonable if you've not got very good sensors.
[18:59] <SpeedEvil> More modern ones can pick up global variations.
[18:59] <Laurenceb_> yeah exactly
[18:59] <SpeedEvil> Also - won't work on venus.
[18:59] <Laurenceb_> heh
[18:59] <Laurenceb_> i may more it out and have a G free frame
[18:59] <Laurenceb_> then fudge the vertical velocity externally
[19:02] <pericynthion> SpeedEvil: Noise is definitely a concern for onshore ops although I don't know if it is definitely louder than a conventional turbine
[19:03] <SpeedEvil> I suppose the shed noise from the turbines will be in a more audible region - but the turbulence noise may be less - watt for watt - as it's higher.
[19:03] <SpeedEvil> Looks like a deeply fun project from the technical side anyway.
[19:04] <pericynthion> yeah it's pretty interesting
[19:09] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) joined #highaltitude.
[19:14] Jasperw (~jasperw@93.97.176.250) joined #highaltitude.
[19:15] mixio (mixio@ppp046177022190.dsl.hol.gr) left irc: Ping timeout: 244 seconds
[19:19] mixio (mixio@ppp046177037084.dsl.hol.gr) joined #highaltitude.
[19:21] <Laurenceb_> why the hell is there no decent bmp085 datasheet :(
[19:23] mixio (mixio@ppp046177037084.dsl.hol.gr) left irc: Ping timeout: 240 seconds
[19:29] mixio (mixio@ppp079166249161.dsl.hol.gr) joined #highaltitude.
[19:35] <pericynthion> Laurenceb_: look up the SMD500 instead?
[19:37] <Laurenceb_> hmm thats a thought
[19:37] <Laurenceb_> i think ive got some code - just have to trawl the ardutard forums
[19:52] mixio (mixio@ppp079166249161.dsl.hol.gr) left irc: Read error: Connection reset by peer
[19:59] mixio (mixio@ppp046177075138.dsl.hol.gr) joined #highaltitude.
[20:04] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) joined #highaltitude.
[20:10] SamSilver (2985f564@gateway/web/freenode/ip.41.133.245.100) left irc: Quit: Page closed
[20:12] <griffonbot> @fsphil: hadie:3 high altitude balloon launch planned for Sat @ 10:00 UTC (11:00 BST) from An-Creagan in N.Ireland http://bit.ly/mdrMzD #ukhas #arhab [http://twitter.com/fsphil/status/76380717348687872]
[20:12] mattltm (~mattltm@81.134.144.179) joined #highaltitude.
[20:13] <griffonbot> @cuspaceflight: RT @fsphil: hadie:3 high altitude balloon launch planned for Sat @ 10:00 UTC (11:00 BST) from An-Creagan in N.Ireland http://bit.ly/mdrM ... [http://twitter.com/cuspaceflight/status/76380947255279616]
[20:19] RocketBoy (~steverand@217.47.75.8) joined #highaltitude.
[20:20] <mattltm> Ha! that bit.ly link links to a nasty video!
[20:20] <jonsowman> oh no! how unfortunate
[20:21] <fsphil> bit.ly failed a bit there
[20:22] <RocketBoy> cuddykid: PM
[20:25] RocketBoy (steverand@217.47.75.8) left #highaltitude.
[20:34] mattltm (~mattltm@81.134.144.179) left irc:
[20:37] kd0mto (~dago@sdrc.com-162.fuse.net) left irc: Ping timeout: 240 seconds
[20:42] kd0mto (~dago@sdrc.com-162.fuse.net) joined #highaltitude.
[20:57] kd0mto (~dago@sdrc.com-162.fuse.net) left irc: Quit: AFK, I'm IRL now.
[20:59] <fsphil> hadie:3 payload running on rechargeable batteries now for 14 hours and still working
[21:00] Hiena (~boreger@81.93.195.181.datatrans.hu) left irc: Remote host closed the connection
[21:04] Dan-K2VOL (~Adium@74-143-44-42.static.insightbb.com) left irc: Ping timeout: 250 seconds
[21:08] mixio (mixio@ppp046177075138.dsl.hol.gr) left irc:
[21:09] pericynthion (~henry@c-174-62-68-92.hsd1.ca.comcast.net) left irc: Quit: pericynthion
[21:14] spacekitteh (~TraumaPon@203-214-99-19.dyn.iinet.net.au) left irc: Ping timeout: 260 seconds
[21:28] <fsphil> anyone know how to stop CHDK taking raw images?
[21:29] Jasperw (~jasperw@93.97.176.250) left irc: Quit: Leaving.
[21:34] <MrCraig> um - in my case that would have been don't turn on raw in the first place.. I seem to recall it being in the config menus
[21:34] <MrCraig> couldn't tell you where though, sorry.
[21:37] <fsphil> ah, yea think I found it
[21:42] <MrCraig> Aaaah if you're implementing a stack and you do [push value] [inc pointer] the correct way to pop would be [dec pointer] [pop value] not [pop value] [dec pointer] *doh*
[21:49] <fsphil> hmpf, it's still taking raw images
[21:53] <fsphil> muhaha, victory is mine!
[21:54] <MrCraig> I never doubted it
[21:54] <fsphil> under the raw menu, there's a toggle. obvious enough - but it would never stay off.
[21:55] <fsphil> discovered one of the buttons on the back turns it back on
[21:55] <fsphil> I'd been pressing it every time
[21:55] <MrCraig> ahh
[21:55] <MrCraig> I'd suspected maybe your script was nudging it on
[21:55] <fsphil> reminds me, I must make a backup of that this time
[21:57] <MrCraig> *nods, it's handy to have. I used a third party unmodified but know I have a copy safely stashed
[21:58] <fsphil> that's the payload sorted, just got to organise everything else
[21:58] <MrCraig> awesome, it's saturday this week right?
[21:59] <fsphil> yea
[21:59] <MrCraig> Once again I sincerely doubt reception from here, and I don't have any transport to get out on the hill... but, I'll try.
[22:01] <fsphil> always worth a try - this is probably pushing it a bit though :)
[22:01] <fsphil> you should definitely hear it, but decoding is another matter
[22:03] <fsphil> assuming it doesn't burst early that is
[22:04] <MrCraig> Well I'm not so convinced about hearing it, I'm in a valley, surrounded by flats. :-/ but may get something when it's high up.
[22:05] <MrCraig> you have a prediction?
[22:06] <fsphil> it'll go up, then down
[22:06] <MrCraig> nice - that's pretty sweet.
[22:06] <fsphil> more specifically: http://habhub.org/predict/#!/uuid=3fb98df577f88740bd7a4d1cb152dff6c86d6ad8
[22:07] <fsphil> so it'll basically land on a mountain :)
[22:07] <MrCraig> been rock climbing recently?
[22:08] <fsphil> not for .. *counts* ... ever
[22:08] <fsphil> I'll probably underfill it a bit, get a slower ascent so it should clear the worst bits
[22:08] <fsphil> http://habhub.org/predict/#!/uuid=f8bf778e218b19543a2eef489c412507a78bc9cd
[22:09] <SpeedEvil> First rule of mountain climbing - bring a helicopter.
[22:09] <MrCraig> Understanding a bit about your location and the predictions you had for hadie:2 and it's outcome, and now this prediction - I understand why you went to so much effort to beam back live image streams.
[22:09] <fsphil> haha
[22:09] <DanielRichman> Randomskk: rock bottom grade boundaries woooo! [delayed response is delayed]
[22:09] <SpeedEvil> fsphil: Damn - that's a good prediction (for you)
[22:09] <Randomskk> DanielRichman: :D
[22:09] <fsphil> don't I know it SpeedEvil :) why I'm taking it even with the risk of a mountain landing
[22:10] <DanielRichman> 42/70 mebe; even 40 is possible
[22:11] <SpeedEvil> It would even (just about) work for me.
[22:11] <SpeedEvil> http://habhub.org/predict/#!/uuid=693673976e55a6152f31b15b108e2cba5e5af037
[22:12] <fsphil> living life on the edge
[22:22] jkominar (~justin@64.235.97.218) left irc: Quit: It was a perfectly cromulent thing to do.
[22:37] Laurenceb_ (~Laurence@host86-172-188-211.range86-172.btcentralplus.com) left irc: Ping timeout: 250 seconds
[22:44] Action: MrCraig <--- good night from him.
[22:45] MrCraig (craig@host109-156-178-239.range109-156.btcentralplus.com) left #highaltitude.
[22:56] Lunar_Lander (~lunar_lan@p54A064FB.dip.t-dialin.net) joined #highaltitude.
[22:56] <Lunar_Lander> hello
[22:57] <fsphil> is it just me, or is rjharrison's server going realllllly slowly
[22:58] <fsphil> hullo Lunar_Lander
[22:58] <Lunar_Lander> hi fsphil
[23:00] <fsphil> nope, looks like it's just me
[23:01] <fsphil> just seems to be this machine
[23:03] <Lunar_Lander> yeah
[23:03] <Lunar_Lander> today I found something interesting in connection to a balloon experiment
[23:05] <Lunar_Lander> they had a cryosampler for sampling bacteria, in India, back in 2001
[23:05] <Lunar_Lander> around 2003, there was a split between the british and the indian part of the team
[23:05] <Lunar_Lander> because the british part thought, they had proven Panspermia (theory that life came from space) and that SARS could also come from space
[23:05] <Lunar_Lander> the indians wrote a reply to that and the british guy wrote a reply to them
[23:05] <Lunar_Lander> and he kinda bashed them and wrote "Time will tell who is right."
[23:05] <Lunar_Lander> and when I researched him, I found that he appearently supports Intelligent Design and stuff like that
[23:05] <fsphil> never a good sign
[23:06] <fsphil> one of the more ironic names in the world
[23:06] <Lunar_Lander> yeah
[23:06] <Lunar_Lander> and that Panspermia thing was invented by Fred Hoyle
[23:06] <Lunar_Lander> the man who denied the big bang
[23:08] <Lunar_Lander> and this man: http://en.wikipedia.org/wiki/Chandra_Wickramasinghe
[23:28] juxta (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[23:51] polycarbonate1 (~Stack@unaffiliated/polycarbonate1) left irc: Ping timeout: 250 seconds
[23:53] polycarbonate1 (~Stack@unaffiliated/polycarbonate1) joined #highaltitude.
[00:00] --- Fri Jun 3 2011