highaltitude.log.20090316

[00:08] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) left irc: "Leaving"
[00:08] <Laurenceb> looks like a good launch
[00:11] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[01:33] sYx66 (n=wooh@bas1-kitchener06-1096765112.dsl.bell.ca) joined #highaltitude.
[03:34] sYx66 (n=wooh@bas1-kitchener06-1096765112.dsl.bell.ca) left irc: Read error: 110 (Connection timed out)
[04:01] sYx66 (n=wooh@bas1-kitchener06-1096647757.dsl.bell.ca) joined #highaltitude.
[06:46] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[07:21] Laurenceb (n=laurence@dyres221-35.surrey.ac.uk) left irc: Remote closed the connection
[07:21] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) joined #highaltitude.
[07:28] <jcoxon> morning all
[07:36] icez (n=icez@unaffiliated/icez) left irc: "Lost terminal"
[07:41] fergusnoble (n=fergusno@fn217.quns.cam.ac.uk) left #highaltitude.
[08:09] <natrium42> hey jcoxon
[08:15] <jcoxon> hey natrium42
[08:15] <jcoxon> hows it going?
[08:16] <natrium42> good, and you?
[08:16] <jcoxon> yeah
[08:16] <jcoxon> i've been thinking about these ballast tanks again
[08:17] <jcoxon> i reckon I could come up with a simple solution with a IR range sensor
[08:17] <jcoxon> it won't be that accurate but will show roughly how much
[08:18] <jcoxon> say in 1/10th of a tank
[08:18] <natrium42> that's probably good enough
[08:18] <jcoxon> thats what i'm thinking
[08:19] <jcoxon> but do we actually need a sensor? (we should have one)
[08:19] <jcoxon> as the proposed logic doesn't really car
[08:19] <jcoxon> e
[08:19] <natrium42> it's just for diagnostics
[08:19] <jcoxon> as it uses the response of ballast being dropped to decide what to do rather then the ballast level itself
[08:19] <natrium42> so that we know why mission failed, if it does
[08:19] <jcoxon> sure
[08:20] <jcoxon> the final thing - i beginning to see the appeal of not have it pressurized
[08:21] <jcoxon> 0.6ml/min isn't too bad and means that we don't have to worry about bursting
[08:21] <natrium42> so a hole at the top?
[08:22] <jcoxon> perhaps a tube out the top that runs away from the payload
[08:22] <jcoxon> don't want evaporated meths going near the batteries :-p
[08:24] <natrium42> what about a check valve?
[08:24] <jcoxon> yeah i had a look for one
[08:24] <jcoxon> didn't find anything that suitable
[08:25] <natrium42> perhaps just a piece of rubber over circular hole
[08:25] <natrium42> well, under in this case
[08:27] <jcoxon> yeah that'll do the job
[08:27] <jcoxon> i'll rig up my idea for a IR sensor this evening and take a few photos
[08:27] <natrium42> ok, cool
[08:27] <jcoxon> i'm worried about epoxy resin and denatured alcohol
[08:28] <natrium42> not compatible?
[08:29] <jcoxon> i'm not sure - guess i'll have to test
[08:30] <jcoxon> as vinegar can be used to clean it up
[08:31] Action: natrium42 spent a few hours determining how SPOT rounds up lat/lon
[08:31] <jcoxon> and...
[08:32] <natrium42> it's round, not ceil or floor :D
[08:32] <natrium42> anyway, we will have to lose 4 bits on lat and 4 bits on lon
[08:32] <natrium42> because of roundoff effects
[08:33] <natrium42> so it's a total of 5 bytes that can be transferred via one message
[08:33] <natrium42> i wrote a program to test it, and all values match
[08:34] <natrium42> altitude can fit in 2 bytes
[08:34] <natrium42> 2 meter precision is good enough, no?
[08:34] <natrium42> up to max of 32 km altitude
[08:34] <jcoxon> yeah thats fine
[08:34] <jcoxon> we just need to know roughly where it is
[08:35] <natrium42> btw, i tested 10 seconds frequency between messages
[08:35] <natrium42> worked
[08:35] <jcoxon> natrium42, might be worth sticking the beginnings of the ballast logic on the flight computer just to test - say loging to the sd card
[08:35] <natrium42> but of course a lot of messages get lost either way
[08:36] <natrium42> hmm
[08:36] <natrium42> could use a sim perhaps
[08:36] <natrium42> just something primitive
[08:37] <jcoxon> another question - how often do you think we should transmit with the radio?
[08:38] <natrium42> not sure, what did snox use?
[08:38] <jcoxon> they had a very weird setup
[08:38] <jcoxon> which recognised propegation
[08:39] <jcoxon> okay found it
[08:39] <jcoxon> during the day every 5 mins
[08:39] <jcoxon> at night every 30mins
[08:40] <natrium42> interesting
[08:40] <jcoxon> and a CW beacon every minute not transmittign something else
[08:41] <jcoxon> http://dtrc.utarc.org/status.php
[08:47] <jcoxon> perhaps every 5min during the day every 10mins at night
[08:47] <natrium42> sounds good to me
[08:48] <jcoxon> 30mins at night is a long time
[08:48] <jcoxon> considering its the most dangerous part
[08:48] <jcoxon> and SPOT how often?
[08:48] <natrium42> why does it need to be longer at night anyway?
[08:48] <jcoxon> i think the assumption is that less people are listening
[08:48] <jcoxon> in some ways we'll get better propegation at night i think on 40m
[08:49] <natrium42> SPOT: probably each 5 mins too -- bursts of a couple of packets
[08:49] <jcoxon> okay cool
[08:50] <jcoxon> i don't mind for the radio every 5mins all the tim
[08:50] <jcoxon> e
[08:50] <jcoxon> if we have the power
[08:58] <jcoxon> this would be a cool flight:
[08:58] <jcoxon> http://spacenear.us/gfs/2009-03-16-3.png
[08:59] <jcoxon> avoids the US nicely
[08:59] <natrium42> hehe, neat
[09:00] <natrium42> might even end up in spain
[09:00] <jcoxon> get Meteotek guys to get it
[09:00] <natrium42> yeah
[09:00] <natrium42> :)
[09:01] <jcoxon> for the 434 module do you want to drive that direct from the flight computer?
[09:02] <jcoxon> say with 2 pins and resistors to do rtty?
[09:03] <natrium42> hrm, can has codez?
[09:03] <jcoxon> i've got arduino code
[09:04] <natrium42> i mean, it's fine if you give me the code :)
[09:04] <jcoxon> rjharrison_ has code
[09:04] <jcoxon> for avr
[09:04] <natrium42> will it work at 4MHz clock?
[09:04] <jcoxon> http://www.pegasushabproject.org.uk/wiki/doku.php/ideas:notes
[09:04] <natrium42> it should right
[09:04] <jcoxon> if you play with the BAUD_RATE
[09:04] <natrium42> looks easy
[09:05] <natrium42> yeah, no need to put extra micro just for that
[09:05] <jcoxon> even given you the resistors :-p
[09:05] <natrium42> lol
[09:05] <jcoxon> its what i wrote down for josh
[09:05] <jcoxon> i should credit rob on that oops
[09:07] <jcoxon> yeah then when we reach the UK just put EN to high and off you go...
[09:24] mc- (n=mfcastle@cpc3-glfd1-0-0-cust439.glfd.cable.ntl.com) joined #highaltitude.
[09:25] <mc-> natrium42, you might get in trouble with SPOT if you send messages at 10s intervals
[09:25] <natrium42> i won't
[09:25] <natrium42> it was just a test
[09:26] <jcoxon> mc-, did you get steve's email about the rockmite?
[09:26] <mc-> I did some testing at 7mins, and they didn't like it.
[09:27] <mc-> jcoxon, yes, will try and send it tomorrow
[09:27] <jcoxon> great
[09:27] <natrium42> mc-, using spot?
[09:27] <natrium42> or stx2?
[09:28] <mc-> yes using STX2
[09:28] <natrium42> ah, did they contact you?
[09:28] <mc-> my mite has crystals at the US QRP freq. I think it's 7.040MHz
[09:29] <mc-> I'm trying to get a product using STX2 approved by them
[09:29] <natrium42> it's weird i am getting quite a few messages lost
[09:29] <natrium42> it's normal, right?
[09:29] <mc-> yes, sats not always in view
[09:30] <natrium42> besides, i was sitting in car with one side blocked by house
[09:30] <mc-> that's why they want 3 transmissions with 5 min intervals
[09:30] <mc-> was it out of the window, could the SPOT see upwards?
[09:31] <natrium42> it was under the windshield
[09:31] <mc-> not optimal
[09:31] <mc-> good work on the Sarantel ant with the GPS
[09:31] <natrium42> i am thinking 2 mins delay, what do you think?
[09:31] <natrium42> thx
[09:32] <mc-> you could use a patch if you want to get the cost down?
[09:32] <mc-> 2 mins delay will lose messages from time to time.
[09:32] <natrium42> why?
[09:33] <mc-> sat take 15 mins from horizon to horizon
[09:33] <natrium42> how does that prevent you sending messages each 2 min?
[09:33] <mc-> so it might only get data through when sat is +30 deg elevation
[09:34] <mc-> nothing prevents you from sending messages every 2 mins
[09:34] <mc-> but SPOT might cut you off
[09:35] <natrium42> i could be sending them each 2 minutes manually
[09:35] <mc-> its worth a try...
[09:35] <mc-> how did you get altitude in the message?
[09:35] <natrium42> is there a problem with satellites sending too many messages?
[09:36] <natrium42> perhaps they will only keep one of the messages?
[09:36] <mc-> they don't expect SPOT users to be sending a message every 2 mins, is there any small print?
[09:36] <natrium42> nope
[09:36] <natrium42> i could press the button each 2 mins
[09:36] <natrium42> with fingers
[09:37] <natrium42> btw, i noticed that the messages come in bursts for some reason
[09:38] <mc-> I think that's just the server latency
[09:39] <natrium42> http://share.findmespot.com/shared/faces/viewspots.jsp?glId=0iYBQsZKupoOBTJq8x6cWA4PRJUYRi1ir
[09:40] <natrium42> :)
[09:41] <natrium42> if you go to 5th message from top, that's when i was sending messages each minute
[09:41] <natrium42> but quite a few are missing
[09:42] <mc-> btw are you using the expensive batts in the SPOT?
[09:42] <natrium42> i connected it to flight computer now
[09:43] <natrium42> it's powered by my DC/DC converter
[09:46] <natrium42> time to get some sleep
[09:46] <natrium42> g'nite
[09:47] <mc-> you're in Africa...
[09:50] <jcoxon> night natrium42
[09:53] mc- (n=mfcastle@cpc3-glfd1-0-0-cust439.glfd.cable.ntl.com) left irc:
[10:00] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[10:06] <jcoxon> morning edmoore
[10:08] <edmoore> morning
[10:29] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) left irc: "Leaving"
[10:35] <SpeedEvil> jcoxon: running the radio synced to the GPS - if you have GPS fix - say the 37th and 23rd minute of the hour may be an idea
[10:57] <SpeedEvil> For those interested in imaging, http://www.framos.eu/ is a slightly interesting site.
[10:57] <SpeedEvil> They've got some CCDs/cmos stuff that's hard to obtain in other places
[10:59] <SpeedEvil> MT9T031 2048 x 1536 1/2" 12 fps 23 euro
[10:59] <SpeedEvil> b ut a sucky login-only webstore
[11:03] <SpeedEvil> Oooh. http://www.framos.eu/csc12m25bmp19.html?&no_cache=1 Me wantee.
[11:03] <SpeedEvil> Though me no affordee.
[11:32] <SpeedEvil> Wacky.
[11:32] <SpeedEvil> I bought some DS18B20s on ebay. I'm now being offered extended warranty.
[11:32] <SpeedEvil> (through ebay)
[11:58] shea_ (n=shea@pool-71-164-88-29.albyny.fios.verizon.net) joined #highaltitude.
[11:59] MOTEK (n=ed@chu-gw.churchillcambridge.co.uk) joined #highaltitude.
[12:00] shea|up (n=shea@pool-71-164-88-29.albyny.fios.verizon.net) left irc: Read error: 110 (Connection timed out)
[12:03] MOTEK (n=ed@chu-gw.churchillcambridge.co.uk) left irc: Client Quit
[13:24] sYx66 (n=wooh@bas1-kitchener06-1096647757.dsl.bell.ca) left irc: "bye"
[13:29] gordonjc1 (n=gordonjc@symmetria.fi) joined #highaltitude.
[13:40] gordonjcp (n=gordonjc@symmetria.fi) left irc: Read error: 110 (Connection timed out)
[13:42] jcoxon (i=zeusbot@lister.antycip.co.uk) joined #highaltitude.
[13:44] hallam (i=836ff82e@gateway/web/ajax/mibbit.com/x-eaadd3ec8da7500e) joined #highaltitude.
[13:50] <jcoxon> afternoon hallam
[13:52] Action: SpeedEvil reposts.
[13:52] <SpeedEvil> http://www.framos.eu/csc12m25bmp19.html?&no_cache=1
[13:52] <SpeedEvil> site selling random CCD and CMOS imagers
[13:52] <SpeedEvil> some fun stuff, though lousy login only shop
[13:57] <jcoxon> bbiab
[13:57] jcoxon (i=zeusbot@lister.antycip.co.uk) left irc: "leaving"
[14:06] Laurenceb (i=83e34f19@gateway/web/ajax/mibbit.com/x-b58fc20cb561f93a) joined #highaltitude.
[14:06] <Laurenceb> hello
[14:07] gordonjc1 (n=gordonjc@symmetria.fi) left irc: Read error: 60 (Operation timed out)
[14:10] gordonjcp (n=gordonjc@symmetria.fi) joined #highaltitude.
[14:14] gregHome (n=gleblanc@75.108.33.75) joined #highaltitude.
[14:14] gregHome (n=gleblanc@75.108.33.75) left #highaltitude.
[14:14] jcoxon (i=zeusbot@lister.antycip.co.uk) joined #highaltitude.
[14:23] hallam (i=836ff82e@gateway/web/ajax/mibbit.com/x-eaadd3ec8da7500e) left irc: "http://www.mibbit.com ajax IRC Client"
[14:48] jcoxon (i=zeusbot@lister.antycip.co.uk) left irc: "leaving"
[14:49] <edmoore> gah
[14:50] <Laurenceb> hes clever
[14:50] <Laurenceb> can match G8KHW and RocketBoy
[14:52] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[15:36] fergusnoble (n=fergusno@fn217.quns.cam.ac.uk) joined #highaltitude.
[15:36] fergusnoble (n=fergusno@fn217.quns.cam.ac.uk) left irc: Client Quit
[17:03] hallam (i=836fc8c8@gateway/web/ajax/mibbit.com/x-fe7015f3fe677520) joined #highaltitude.
[17:08] <hallam> hey all
[17:08] <hallam> SpeedEvil: thanks for the imager link
[17:08] <hallam> Nova 10 is on for Saturday atm
[17:11] <Laurenceb> nice
[17:11] <Laurenceb> whats launching?
[17:11] <Laurenceb> star tracker?
[17:11] <hallam> yes
[17:12] <hallam> still want your Rogallo on it?
[17:12] <Laurenceb> cool, so we can rig up the rogallo?
[17:12] <Laurenceb> yeah
[17:12] <Laurenceb> I'll have to work out a means to transport it
[17:12] <hallam> can you provide the release please?
[17:12] <Laurenceb> need a large cardboard box, it'll have to go in the tube
[17:12] <Laurenceb> the cutdown?
[17:12] <hallam> yeah
[17:12] <Laurenceb> yeah theres one on it
[17:13] <hallam> cool
[17:13] <hallam> will it self-cut-down at a suitable altitude?
[17:13] <Laurenceb> its about 500grams mass
[17:13] <hallam> should be fine
[17:13] <Laurenceb> yeah prob 10Km
[17:13] <hallam> ok
[17:13] <hallam> then I'll have a redundant cutdown controlled from Nova blow at 15km
[17:13] <Laurenceb> cool
[17:14] <hallam> will you come to the launch?
[17:14] <Laurenceb> I'm going top try and find a large cardboard box
[17:14] <Laurenceb> I'll take the train
[17:14] <hallam> great
[17:14] <hallam> which freq is it again?
[17:14] <Laurenceb> will we have internet at the lanuchsite?
[17:14] <Laurenceb> 434.075
[17:15] <hallam> there'll be internet at the prep site (churchill bar) which is a 5 minute walk from the launch site
[17:15] <hallam> there might be wifi on the field, I don't know
[17:15] <Laurenceb> hmm ok
[17:15] <Laurenceb> my php tracking script tries to relay stuff on to natriums tracker
[17:15] <Laurenceb> be a shame if we cant do that
[17:16] <Laurenceb> but if we're launching from churchill... I might want to set it to land somewhere else
[17:16] <Laurenceb> at the moment its targeted at ears
[17:17] <Laurenceb> grr so much to do... I need to compile the code with the ubuntu gcc and see if tis the same size
[17:17] <hallam> you can always track from the bar
[17:17] <hallam> or close enough to it to get wifi
[17:17] <hallam> brb
[17:17] <Laurenceb> yeah
[17:17] <Laurenceb> me too
[17:19] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[17:20] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc: Client Quit
[17:21] <Laurenceb> back
[17:22] <Laurenceb> ok... is it sensible to try and land at churchill?
[17:25] <Laurenceb> cnaces are something will go wrong... I'm hopeing for some controlled flight at any rate. But if we dont reach the target and land short we might end up in cambridge
[17:25] <Laurenceb> shall I keep the target at ears?
[17:29] <Laurenceb> there will be nonoe to see it land if it does work
[17:35] <Laurenceb> winds from the north east by the look of it
[17:37] <Laurenceb> hmm not much in that direction apart from the british antarctic survey
[17:40] <Laurenceb> should have added a motor :-/
[17:47] Action: SpeedEvil bought 5 ICX429ALL - exview HAD 1/2" CCDs from ebay to play with the star-tracker thing.
[17:47] <SpeedEvil> With free security cameras soldered to them.
[17:48] <SpeedEvil> (Though I'll probably be doing the security camera thing with 3 of them)
[17:53] <Laurenceb> 5
[17:53] <Laurenceb> crazy
[17:58] <SpeedEvil> they were 12 quid each
[17:58] <Laurenceb> not bad
[17:58] <Laurenceb> my rams hould be here tomorrow
[17:59] <SpeedEvil> and I need cameras - well want - anyway
[17:59] <SpeedEvil> And these are highly sensitive ones, that'll work well with small IR floods for extra coverage
[18:01] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[18:01] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc: Client Quit
[18:01] <Laurenceb> neat
[18:01] <Laurenceb> I need to get a few of these http://www.adaptivemodules.co.uk/index.cfm/fa/shopdetails/Product_ID/318/Category_ID/12/Sub_Category_ID/73
[18:02] <Laurenceb> (111 dBm at 1.2 kBaud, 868 MHz, 1% packet error rate)
[18:07] <Laurenceb> giving a range of around 400Km :D
[18:09] <Laurenceb> http://focus.ti.com/lit/ds/symlink/cc1100.pdf
[18:10] <Laurenceb> figure 2 is just so nice :P everything youd even need for a balloon link on a single IC
[18:13] <SpeedEvil> neat - how much?
[18:14] <Laurenceb> not sure, need to ask for a quote
[18:14] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) joined #highaltitude.
[18:14] <SpeedEvil> it's not even in too horrible a package
[18:15] <Laurenceb> http://uk.farnell.com/texas-instruments/cc1100rtk/single-chip-rf-transceiver/dp/1310184?_requestid=159441
[18:15] <Laurenceb> £3.23 at farnell
[18:15] <Laurenceb> thing is that etiny module uses it
[18:20] <natrium42> hi
[18:22] <Laurenceb> right I'm off
[18:22] <Laurenceb> might be online later
[18:22] <natrium42> later
[18:28] Laurenceb (i=83e34f19@gateway/web/ajax/mibbit.com/x-b58fc20cb561f93a) left irc: "http://www.mibbit.com ajax IRC Client"
[18:30] <jcoxon> hallam, launch this weekend?
[18:37] <natrium42> jcoxon, halo3 launch moved to wednesday
[18:38] <jcoxon> natrium42, okay
[18:38] <jcoxon> just walked back with a 5L bottle of water
[18:38] <jcoxon> and laurence is sending me the sensor
[18:39] <natrium42> ok, cool
[18:40] <jcoxon> im thinking everything should go through the cap
[18:40] <jcoxon> the outlet tube and also teh connections for the sensors
[18:42] <SpeedEvil> you can 'solder' the cap.
[18:42] <SpeedEvil> using a cool soldering ion
[18:42] <SpeedEvil> iron
[18:42] <SpeedEvil> assuming it's ldpe
[18:43] <jcoxon> SpeedEvil, as in melt it around
[18:43] <SpeedEvil> yeah.
[18:43] <SpeedEvil> Did that to one of my car headlights, and bumper, to repair it
[18:44] <jcoxon> interesting idea
[18:44] <jcoxon> will try it
[18:45] <SpeedEvil> makes robust through-holes a lot easier.
[18:46] <jcoxon> i've got a drill bit thats just the right size for the tubing
[18:48] <SpeedEvil> you want a very tight fit
[18:48] <SpeedEvil> though at low temps, I'd really want a proper hose connector
[19:14] borism_ (n=boris@195-50-205-173-dsl.krw.estpak.ee) joined #highaltitude.
[19:19] borism (n=boris@195-50-205-184-dsl.krw.estpak.ee) left irc: Read error: 145 (Connection timed out)
[19:23] Laurenceb (n=laurence@dyres221-35.surrey.ac.uk) joined #highaltitude.
[19:27] <jcoxon> bbl
[19:27] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) left irc: "This computer has gone to sleep"
[19:28] <Laurenceb> hi
[19:28] <Laurenceb> hallam: about?
[19:30] <shellevil> last talked with you earlier
[19:34] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: Read error: 104 (Connection reset by peer)
[19:35] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[19:43] icez (n=icez@unaffiliated/icez) joined #highaltitude.
[20:46] RocketBoy (n=Steve@217.47.75.8) joined #highaltitude.
[20:51] <RocketBoy> rjharrison: any plans for a launch this w/e
[20:51] <RocketBoy> ?
[20:57] Xenion (n=robert@p579FC999.dip.t-dialin.net) joined #highaltitude.
[21:02] hallam (i=836fc8c8@gateway/web/ajax/mibbit.com/x-fe7015f3fe677520) left irc: "http://www.mibbit.com ajax IRC Client"
[21:04] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) joined #highaltitude.
[21:05] <jcoxon> RocketBoy, theres a nova flighjt
[21:05] <jcoxon> flight*
[21:05] Xenion (n=robert@p579FC999.dip.t-dialin.net) left irc: "Verlassend"
[21:10] <Laurenceb> hi jcoxon
[21:10] <Laurenceb> fraid I missed the post office
[21:10] <Laurenceb> I'll get some stamps from the corner shop then drop it off in a postbox tomorrow morning
[21:10] <Laurenceb> should get to you just as fast anyway
[21:11] <Laurenceb> - by wednesday morning
[21:11] <jcoxon> no worries Laurenceb
[21:11] <Laurenceb> right I'd better head off to catch the shop
[21:11] <Laurenceb> cya
[21:12] <jcoxon> cya
[21:32] <Laurenceb> back
[21:45] <jcoxon> Laurenceb, i read that the rogallo might actually get off the ground! ;-)
[21:46] <SpeedEvil> Nevah!
[21:47] <Laurenceb> yes
[21:47] <Laurenceb> not sure if we'll have internet
[21:47] <Laurenceb> also not sure if I'll set it to land at churchill
[21:47] <jcoxon> churchill launch?
[21:47] <jcoxon> i don't think churchill is a good place to land
[21:47] <Laurenceb> looks like it would approach over the british antarctic survey
[21:48] <jcoxon> as it would need to be tres accurate
[21:48] <Laurenceb> so its about the best direction
[21:48] <Laurenceb> well if it does work I expect it to be
[21:48] <jcoxon> best to be safe...
[21:49] <Laurenceb> I think it'll either fail for some hardware reason, or there will be some bug causing it to land some distance away
[21:49] <jcoxon> what sort of range do you get?
[21:49] <Laurenceb> the chances of it correctly making it to churchill then landing more than 10m off target are slim
[21:49] <Laurenceb> however it may run out of altitude just before landing
[21:50] <Laurenceb> in which case it would probably land around the antarctic survey buildings
[21:50] <Laurenceb> ~40Km
[21:50] <jcoxon> Laurenceb, there are some busy roads around
[21:50] <jcoxon> that is what concerns me
[21:50] <Laurenceb> but not sure how far it will drift
[21:50] <SpeedEvil> Laurenceb: can you do multiple targets?
[21:50] <Laurenceb> hmm yeah thats a point
[21:50] <Laurenceb> SpeedEvil: yeah I could
[21:50] <SpeedEvil> Laurenceb: points
[21:50] <Laurenceb> its not in the current firmware
[21:50] <jcoxon> Laurenceb, how about aiming for suffolk or somewhere?
[21:50] <Laurenceb> but wouldnt be too hard to add
[21:50] <jcoxon> or do you want to watch it in?
[21:51] <Laurenceb> course :P
[21:51] <SpeedEvil> have you ghot the ground control bit working?
[21:51] <Laurenceb> I was going to go for ears
[21:51] <Laurenceb> yes
[21:51] <Laurenceb> so you can guide it in under rc control
[21:51] <Laurenceb> from about 4Km away
[21:51] <jcoxon> EARS is good apart from the A14
[21:51] <Laurenceb> - I've got some very decent hardware
[21:51] <Laurenceb> yeah
[21:52] <jcoxon> can it go against the wind?
[21:52] <Laurenceb> if it gets near to churchill and I'm not happy I can always try the RC
[21:52] <Laurenceb> should be able to
[21:52] <SpeedEvil> a light wind
[21:52] <Laurenceb> I'm not sure about the trim
[21:52] <Laurenceb> its a bit hard to test
[21:52] <SpeedEvil> anything more than ~10MPH or so, and it'll be going backwards
[21:52] <Laurenceb> appears to want to glide something around 15mph
[21:53] <Laurenceb> yeah I'd say 10mph max for the wind
[21:54] <SpeedEvil> Laurenceb: can the model youv'e got inside predict how likely it is to hit a given landing spot?
[21:54] <Laurenceb> nope
[21:54] <Laurenceb> yeah I want to add that sort of thing
[21:54] <Laurenceb> but I dont want to get too complex all at once
[21:54] <SpeedEvil> what does it do at the moment?
[21:54] <Laurenceb> it can pilot it using gps and gyro
[21:54] <SpeedEvil> just steer towards the target, taking into account relative wind?
[21:55] <Laurenceb> and fly to one spot
[21:55] <Laurenceb> using an optimal route
[21:55] <Laurenceb> thats recalculated at 1hz
[21:55] <Laurenceb> it also logs data at 1Hz
[21:55] <SpeedEvil> direct to the target?
[21:55] <Laurenceb> runs filter at 50Hz, sends data to the radio and checks for pwm from the rc receiver
[21:56] <Laurenceb> not direct
[21:56] <Laurenceb> an optimal route would only be direct if theres no wind
[21:56] <SpeedEvil> if it's calculating the optimal route, don't you know if you've hit the ground in the algorithm ?
[21:56] <Laurenceb> no it runs in reverse
[21:57] <Laurenceb> starting from the ground then running up
[21:57] <Laurenceb> it doesnt know how well we can glide
[21:57] <SpeedEvil> ah
[21:58] <Laurenceb> the radio is only operable below a few hundered meters altitude
[21:59] <Laurenceb> I could alter that but really want to keep it as a constraint in software
[21:59] <Laurenceb> but I think if it approached cambridge from the southwest, there would be more than enough time to take control with the RC
[21:59] <Laurenceb> the range is like 3miles
[22:00] <Laurenceb> probably more with the altitude it'll be at
[22:00] <Laurenceb> I've got a dsp based receiver
[22:01] <jcoxon> Laurenceb, its something i'd love to see
[22:01] <jcoxon> it flying in - would be amazing
[22:01] <Laurenceb> ideally I'd use the signal integrity output from the receiver to control the flight computer, but I built it to use any receiver so it sanity checks the pwm atm
[22:01] <Laurenceb> but I'll probably do that for next time
[22:02] <Laurenceb> along with back up landing spots and landing spot probability of success calculation, and use of weather models instead of wind found during ascent
[22:05] <Laurenceb> I'm not sure how to do uptimised landing route estimation with a 3D array of wind vectors
[22:06] <SpeedEvil> and can you store it?
[22:07] <Laurenceb> the wind data?
[22:07] <Laurenceb> sure
[22:11] <jcoxon> night all
[22:11] jcoxon (n=jcoxon@host86-145-224-2.range86-145.btcentralplus.com) left irc: "Leaving"
[22:25] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[22:57] <Laurenceb> ok... think I've worked something out...
[22:57] <Laurenceb> not sure if I'm mad enough to try and code it for the launch
[22:58] <Laurenceb> you use a layered atmospheric model to seed a monte carlo based model
[22:59] <Laurenceb> or something like that...
[22:59] <Laurenceb> there two major unknowns - flight speed and glide ratio
[23:00] <Laurenceb> its probably more worthwhile finding them with more accuracy than going for monte carlo on a 3D array of wind vectors
[23:00] <Laurenceb> my exisitng code uses a layered model and runs kalman filters to find the two coefficients
[23:01] <Laurenceb> 3D array of wind vectors would probably only be worth it for correct gps velocities
[23:01] Action: shellevil wonders how much a slow 360 would lose alt
[23:01] <shellevil> to get absolute flight speed
[23:01] <Laurenceb> not worth it
[23:01] <shellevil> yeah
[23:02] <Laurenceb> you can use cicrle fitting to your data with as much sucess
[23:02] <Laurenceb> but atmospheric models and wind recorded during ascent have higher accuracy than that
[23:03] <shellevil> I see
[23:03] <Laurenceb> in very turbulent conditions monte carlo may help
[23:03] <Laurenceb> but you dont want to fly on those days
[23:03] <Laurenceb> on the other hand... if this is a long range glider ... :P
[23:03] <shellevil> 3d wind vectors would surely be out of date rapidly?
[23:03] <Laurenceb> from weather models
[23:04] <shellevil> well - won't l/d be something like 1:3?
[23:04] <shellevil> actually
[23:04] <shellevil> from 10km, that is 30km
[23:04] <Laurenceb> yeah, but compared to wind which may be 10% out
[23:04] <Laurenceb> 3.3:1 is quite possible
[23:05] <Laurenceb> its easier to be out with cruise velocity and glide ratio
[23:05] <Laurenceb> I actually hope for about 4:1
[23:05] Action: shellevil wonders what the best achieved l/d is for a rollago
[23:05] <Laurenceb> just a sec I have a video somewhere
[23:05] <Laurenceb> as you move c of g forward it gets better
[23:06] <Laurenceb> until it becomes unstable and dives
[23:06] <Laurenceb> I tried with a scale model and found the optimal point
[23:07] <shellevil> nvm - having connection problems
[23:08] <shellevil> video not work well
[23:09] <Laurenceb> http://video.google.co.uk/videoplay?docid=4858664474879256299
[23:09] <Laurenceb> there you go anyway
[23:10] <Laurenceb> got to be about 4:1
[23:11] <Laurenceb> once its stable
[23:11] <shellevil> :)
[23:11] <shellevil> have you dropped this thing from somewhere high, and got it to terminal velocity, and made sure it flies OK?
[23:14] <Laurenceb> not the one I'm flying :-/
[23:15] <Laurenceb> scared of breaking it
[23:15] <Laurenceb> maybe I could use a bike
[23:16] <Laurenceb> I've tried running down hills and releasing then catching it
[23:16] <Laurenceb> but cant let go for more than a second as it flies faster than I can run
[23:18] <Laurenceb> need an area with very long grass
[23:20] <shellevil> or just a close replica - identical wing and a box
[23:20] <Laurenceb> yeah
[23:20] <shellevil> probably a good idea to have replacement bits anyway
[23:20] <Laurenceb> but it would help to be running the heading hold loop at the same time
[23:21] <shellevil> hmm
[23:21] <Laurenceb> my motorised one managed to oscillate itself out of control in the air
[23:21] <shellevil> stilts!
[23:21] <shellevil> run faster
[23:21] <Laurenceb> did a few 360 rolls then went chaotic
[23:22] <Laurenceb> looked very cool but it broke the wing spar then smashed several more bits when it hit the ground
[23:22] <shellevil> :/
[23:22] RocketBoy (n=Steve@217.47.75.8) left irc: "Leaving"
[23:23] <Laurenceb> if you tim them right motorised rogallos fly really fast
[23:23] <Laurenceb> its pretty awsome to fly
[23:24] <Laurenceb> but not very efficient - you need at least 1.5:1 thrust to weight to get high performance
[23:24] <shellevil> might as well lose the wings :)
[23:24] <Laurenceb> erm 1thrust to 1.5 weight
[23:24] <Laurenceb> hehe
[23:24] <Laurenceb> not as easy to fly
[23:25] <shellevil> yeah
[23:26] <Laurenceb> also yo dont need full throttle all the time
[23:27] <shellevil> seems a relatively modest motor would be enough to stretch your glide
[23:28] <Laurenceb> I used a $3 one off a chinese site
[23:28] <Laurenceb> together with a tower pro esc costing about $5 :P
[23:29] <Laurenceb> which uses an AVR and has open firmware avaliable as well :D
[23:30] <Laurenceb> you can use I2C on the tower pro controllers
[23:38] <shellevil> you mean to set speed?
[23:41] <Laurenceb> yes
[23:42] <shellevil> neat
[23:52] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[00:00] --- Tue Mar 17 2009