highaltitude.log.20100504

[00:00] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[00:08] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 252 seconds
[00:10] SpeedEvil (~user@tor/regular/SpeedEvil) joined #highaltitude.
[00:10] Snomi (snomi@unaffiliated/snomi) left irc: Ping timeout: 276 seconds
[00:10] sbasuita (ssb@unaffiliated/sbasuita) left irc: Ping timeout: 276 seconds
[00:28] DanielRichman (~DanielRic@unaffiliated/danielrichman) left irc: Remote host closed the connection
[00:40] Laurenceb (~laurence@host86-136-234-181.range86-136.btcentralplus.com) left irc: Ping timeout: 240 seconds
[01:00] MoALTz_ (~no@92.19.16.70) left irc: Quit: Leaving
[01:15] Jasperw (~jasperw@92.40.119.76.sub.mbb.three.co.uk) left irc: Quit: Leaving.
[01:33] juxta (fourtytwo@203.122.193.94) joined #highaltitude.
[02:09] SpeedEvil (~user@tor/regular/SpeedEvil) left irc: Quit: Leaving.
[02:13] SpeedEvil (1000@mauve.plus.com) joined #highaltitude.
[02:13] SpeedEvil (1000@mauve.plus.com) left irc: Changing host
[02:13] SpeedEvil (1000@tor/regular/SpeedEvil) joined #highaltitude.
[02:46] jasonb (~jasonb@dsl027-180-244.sfo1.dsl.speakeasy.net) left irc: Ping timeout: 260 seconds
[03:03] ki6ymz (~dave@c-67-190-100-35.hsd1.co.comcast.net) joined #highaltitude.
[03:16] jasonb (~jasonb@m450536d0.tmodns.net) joined #highaltitude.
[03:44] jasonb (~jasonb@m450536d0.tmodns.net) left irc: Ping timeout: 264 seconds
[04:00] ki6ymz_ (~dave@c-67-190-100-35.hsd1.co.comcast.net) joined #highaltitude.
[04:00] ki6ymz (~dave@c-67-190-100-35.hsd1.co.comcast.net) left irc: Read error: Connection reset by peer
[04:00] Nick change: ki6ymz_ -> ki6ymz
[05:25] stilldavid (~dave@68-64-214-18.static.forethought.net) got netsplit.
[05:25] Cleo (~Cleo@unaffiliated/cleo) got netsplit.
[05:27] stilldavid (~dave@68-64-214-18.static.forethought.net) returned to #highaltitude.
[05:27] Cleo (~Cleo@unaffiliated/cleo) returned to #highaltitude.
[05:38] ki6ymz (~dave@c-67-190-100-35.hsd1.co.comcast.net) left irc: Quit: ki6ymz
[06:05] natrium42 (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Quit: My other car is a cdr.
[06:48] Upu (~Upu@ubn.upuaut.net) joined #highaltitude.
[06:53] G8DSU (~chatzilla@cpc3-mort4-0-0-cust192.croy.cable.virginmedia.com) left irc: Quit: ChatZilla 0.9.86 [Firefox 3.5.9/20100315083431]
[06:53] Simon-MPFH (~simon@phantom.mpfh.co.uk) joined #highaltitude.
[07:25] junderwood (~John@adsl.jcu.me.uk) joined #highaltitude.
[07:43] <earthshine> morning
[07:47] kc0wys (~adwiens@75-130-209-194.dhcp.stls.mo.charter.com) joined #highaltitude.
[07:49] adwiens (~adwiens@resnet-pat25.nts.wustl.edu) left irc: Ping timeout: 248 seconds
[07:58] kc0wys (~adwiens@75-130-209-194.dhcp.stls.mo.charter.com) left irc:
[08:02] mct (~mct@entropia.netisland.net) left irc: Ping timeout: 246 seconds
[08:02] mct (~mct@entropia.netisland.net) joined #highaltitude.
[08:02] mct (~mct@entropia.netisland.net) left irc: Changing host
[08:02] mct (~mct@unaffiliated/mct) joined #highaltitude.
[08:24] b3cft (~abrock@nat/yahoo/x-iyfrnecriedcadrh) joined #highaltitude.
[08:47] GW8RAK (~chatzilla@host86-134-241-131.range86-134.btcentralplus.com) joined #highaltitude.
[09:17] MoALTz (~no@92.19.16.70) joined #highaltitude.
[09:31] Daviey (~Daviey@ubuntu/member/daviey) left irc: Quit: leaving
[10:08] GW8RAK (~chatzilla@host86-134-241-131.range86-134.btcentralplus.com) left irc: Ping timeout: 265 seconds
[10:12] GW8RAK (~chatzilla@host86-134-241-131.range86-134.btcentralplus.com) joined #highaltitude.
[10:21] Laurenceb (~laurence@host86-136-234-181.range86-136.btcentralplus.com) joined #highaltitude.
[11:13] b3cft (~abrock@nat/yahoo/x-iyfrnecriedcadrh) left irc: Read error: Connection reset by peer
[11:13] b3cft (~abrock@nat/yahoo/x-rpdcinpgdvhtflwi) joined #highaltitude.
[11:15] <GW8RAK> Morning all. Does the spacenear.us tracking page update the path predictions in real time as a result of data coming in from the balloon?
[11:17] <junderwood> GW8RAK, I think so.
[11:18] <russss> yes, it does
[11:18] <GW8RAK> Thanks, I could see it adjusting itself as Alien came into land and presumed it does. Just need to add that into the planning stage for our launch.
[11:18] <junderwood> I think the model needs work, though. It was consistently predicting well to the west of the actual landing spot.
[11:18] <SpeedEvil> junderwood: I think that's more the weather data it has
[11:18] Cleo (~Cleo@unaffiliated/cleo) left irc: Ping timeout: 264 seconds
[11:19] <GW8RAK> Do you know how far out it was?
[11:19] <junderwood> Would be useful to have the ballistic coefficient of the payload as an input for the prdictor
[11:19] Cleo (~Cleo@unaffiliated/cleo) joined #highaltitude.
[11:19] <junderwood> Initial predicted landing spot was about 10 miles West of actual.
[11:19] <GW8RAK> It's a shame we can't track them right onto the ground, but without someone standing underneath it, it would be difficult.
[11:20] <SpeedEvil> Well - 300m!!!
[11:20] <SpeedEvil> that's damn near to the ground
[11:20] <junderwood> Chase car should be able to get within a few miles. I've managed to get within a mile twice.
[11:21] <SpeedEvil> Though this one was perhaps a bit lucky
[11:26] rharrison (~rharrison@gateway.hgf.com) joined #highaltitude.
[11:33] <Laurenceb> http://wyvernupenn.blogspot.com/
[11:33] <Laurenceb> ^ quadcopter with the sfe razor imu :P
[11:39] <Randomskk> someone has started a CUED society to make a quadcopter
[11:39] <Laurenceb> 120Hz operation at 8Mhz is pretty impressive
[11:39] <Laurenceb> Randomskk: I'm working on a comms usb stick now
[11:40] <Laurenceb> http://i.imgur.com/cElL4.png
[11:40] <Laurenceb> thats the last part of the main board done
[11:41] <Laurenceb> added a cutdown/brushed esc
[11:41] <Randomskk> sweet
[11:41] <Laurenceb> I'm thinking si4432 + ft2232D for the comms stick
[11:41] <Laurenceb> and a jst with serial out so you can plug it into the main board
[11:41] <Laurenceb> ft2232D has serial and spi
[11:42] <Laurenceb> libftdi support is very poorly documented for the spi, but theres very good C# example code for windoze
[11:43] <Randomskk> the ft chips are really clever
[11:44] <Laurenceb> yeah - it has 3v3 reg and level conversion built in
[11:44] <Laurenceb> meaning you can plug it into the main board and apply power to either side in any order and it wont get fried
[11:45] <SpeedEvil> Laurenceb: Can you do a comms microSD? thanx.
[11:45] Action: SpeedEvil sighs.
[11:45] <SpeedEvil> I wish microSD were that leetle bit larger
[11:45] <SpeedEvil> It would make making your own barely possible.
[11:46] <Laurenceb> transceiver in a microsd ?
[11:46] <SpeedEvil> yes
[11:46] <SpeedEvil> I want to have a nice gyro and magnetometer in a microSD
[11:46] <SpeedEvil> It's 'easily' possible in terms of footprint
[11:46] <Laurenceb> hmm think the qfn packages themselves are too thick - 0.9 to 1.1mm
[11:46] <SpeedEvil> And I can get it to about 2mm thickness without _too_ much pain
[11:46] <SpeedEvil> but then there is the thickness
[11:47] <SpeedEvil> - n900 has a microsd slot which can be trivially used as GPIO
[11:48] <Laurenceb> pcb stuck in?
[11:48] <SpeedEvil> yeah
[11:48] <Laurenceb> with cable to stuff
[11:48] <SpeedEvil> it's easy to make a connector for an external thing
[11:48] <SpeedEvil> but - it's a phone
[11:48] <Laurenceb> yeah
[11:49] <SpeedEvil> Same way I wish there was a nice cheap bluetooth HID chip I could find docs on.
[11:50] <SpeedEvil> I want to try making a waterproof case with an external window, which is a touchscreen
[12:00] <Laurenceb> bbl
[12:03] b3cft (~abrock@nat/yahoo/x-rpdcinpgdvhtflwi) left irc: Ping timeout: 240 seconds
[12:04] Laurenceb (~laurence@host86-136-234-181.range86-136.btcentralplus.com) left irc: Ping timeout: 240 seconds
[12:37] b3cft (~abrock@nat/yahoo/x-nricxaynmlarxrkf) joined #highaltitude.
[12:49] MoALTz (~no@92.19.16.70) left irc: Quit: Leaving
[13:22] Jasperw (~jasperw@skeleton2.london.iofc.org) joined #highaltitude.
[14:11] juxta (fourtytwo@203.122.193.94) left irc: Ping timeout: 276 seconds
[14:28] juxta (fourtytwo@203.122.193.94) joined #highaltitude.
[14:38] juxta (fourtytwo@203.122.193.94) left irc: Ping timeout: 264 seconds
[14:42] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[14:44] <N900evil> slightly offftopic. I'm looking for CO2 cheap. As in I'd want tens of l eventually - where might I get this?
[14:45] <N900evil> and does anyone recall ballpark prices?
[14:49] <russss> CO2 should be really cheap
[14:49] <russss> BOC will probably be the cheapest
[14:50] <russss> helium is only crazy expensive because it's rare
[14:56] <N900evil> yeah
[14:56] <Randomskk> yea we should just use hydrogen :P
[14:57] <N900evil> again, yeah :)
[14:57] <N900evil> if launchig from own property
[14:57] <N900evil> others may have largely unfounded concerns as to safety
[15:04] Futurity (~bill@212.44.16.158) joined #highaltitude.
[15:06] haptlK (~alsodongs@157.140.112.178) left irc: Ping timeout: 258 seconds
[15:08] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 252 seconds
[15:10] haptlK (~alsodongs@157.140.112.178) joined #highaltitude.
[15:25] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[15:32] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 240 seconds
[15:35] jasonb (~jasonb@m350536d0.tmodns.net) joined #highaltitude.
[15:45] jasonb (~jasonb@m350536d0.tmodns.net) left irc: Ping timeout: 264 seconds
[15:56] AlexBreton (~Alexander@client-80-5-40-26.cht-bng-014.adsl.virginmedia.net) joined #highaltitude.
[15:56] <AlexBreton> afternoon
[15:56] <AlexBreton> writing to the local press
[16:02] jasonb (~jasonb@adsl-66-124-73-250.dsl.sntc01.pacbell.net) joined #highaltitude.
[16:02] b3cft (~abrock@nat/yahoo/x-nricxaynmlarxrkf) left irc: Quit: b3cft
[16:12] <earthshine> why?
[16:16] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[16:20] <AlexBreton> to get publicity =)
[16:21] <AlexBreton> can someone remind me of how far away the guy in Ireland was who tracked out balloon?
[16:23] <AlexBreton> was it 550km?
[16:24] <jonsowman> 55k5m
[16:24] <AlexBreton> 555?
[16:24] <jonsowman> *555km
[16:24] <AlexBreton> thanks
[16:24] <jonsowman> yes, i can't type
[16:24] <jonsowman> sorry
[16:24] <AlexBreton> Radiometrix were chuffed when I told them
[16:24] <jonsowman> :)
[16:24] <AlexBreton> precise town where this ham lives?
[16:25] <jonsowman> hmm not sure
[16:25] <jonsowman> MI6VIM
[16:25] <jonsowman> Philip Heron
[16:25] <jonsowman> 32 Milburn Park, Cookstown
[16:25] <jonsowman> Co. Tyrone, BT80 8HQ
[16:25] <jonsowman> Northern Ireland
[16:25] <jonsowman> fsphil on here :)
[16:26] <AlexBreton> nice
[16:27] <jonsowman> i like his callsign
[16:27] <SpeedEvil> AlexBreton: It'll be in the next datasheet as 'typical range' :)
[16:27] <AlexBreton> hehe
[16:27] <jonsowman> haha probably
[16:27] <AlexBreton> also mentioning that someone in Merseyside got S6
[16:28] jasonb (~jasonb@adsl-66-124-73-250.dsl.sntc01.pacbell.net) left irc: Ping timeout: 260 seconds
[16:30] rharrison (~rharrison@gateway.hgf.com) left irc: Disconnected by services
[16:30] rharrison_ (~rharrison@gateway.hgf.com) joined #highaltitude.
[16:33] <AlexBreton> all done
[16:34] <AlexBreton> asked for an NiM2
[16:34] <AlexBreton> dan said we need one
[16:34] <jonsowman> for alien 2?
[16:35] <jonsowman> i was talking to him wrt uplink the other day
[16:47] <AlexBreton> yeah
[16:47] <AlexBreton> apparently we need an uplink
[16:47] <AlexBreton> I'm in no position to argue
[16:48] <jonsowman> its a nice technical exercise and can come in handy
[16:52] GW8RAK (~chatzilla@host86-134-241-131.range86-134.btcentralplus.com) left irc: Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]
[16:52] jasonb (~jasonb@m350536d0.tmodns.net) joined #highaltitude.
[16:57] Futurity (~bill@212.44.16.158) left irc:
[16:59] rharrison_ (~rharrison@gateway.hgf.com) left irc: Quit: Leaving
[16:59] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 240 seconds
[17:01] <AlexBreton> we plan to sell prints at the upcoming school spring fair
[17:01] <AlexBreton> or rather take orders
[17:01] <AlexBreton> hopefully we will be in the newspaper before then so it attracts attention
[17:07] sbasuita (ssb@unaffiliated/sbasuita) joined #highaltitude.
[17:07] jcoxon (~jcoxon@92.40.113.219.sub.mbb.three.co.uk) joined #highaltitude.
[17:10] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[17:12] <jcoxon> afternoon
[17:33] <sbasuita> http://southgatearc.org/news/may2010/alien_1_success.htm :D
[17:33] jasonb (~jasonb@m350536d0.tmodns.net) left irc: Ping timeout: 268 seconds
[17:35] <Randomskk> sbasuita: excellent photos, good job
[17:35] <Randomskk> I'd like to do a dual balloon launch to get a photo of another balloon at altitude
[17:35] <Randomskk> be awesome to see your payload with the earth as a backdrop :P
[17:36] <jcoxon> Randomskk, the skys the limit!
[17:36] <jcoxon> 'scuse the pun
[17:36] <jcoxon> :-p
[17:36] <Randomskk> :P
[17:36] <Randomskk> keeping them together and having the camera actually aim right would be quite tricky though I imagine
[17:36] <jcoxon> i think the alien guys are publicity lovers
[17:37] <Randomskk> otoh a double payload, one higher up the string than the other...
[17:37] <earthshine> That's going to be very tricky to get them to stay together i think
[17:37] <Randomskk> not as good though
[17:37] <Randomskk> earthshine: yea
[17:37] <russss> swarming balloons!
[17:37] <Randomskk> russ has a good point
[17:37] <Randomskk> 99 luftballoons :P
[17:37] <earthshine> maybe we shoudl wait till there are about 5 people waiting to launch then do them all at once
[17:37] <jcoxon> well in theory with te same ascent rate they should stick together
[17:37] <russss> need an adjustable sail, or something ;)
[17:38] <jcoxon> then with cutdown at set altitude
[17:38] <jcoxon> you should be able to hold them together for nearly all the flight
[17:39] <Randomskk> easy recovery too
[17:39] <Randomskk> they'd need the same parachutes
[17:39] <Randomskk> that said
[17:39] <Randomskk> just do like 5 identical payloads
[17:39] <Randomskk> same components
[17:39] <Randomskk> stay together like a charm
[17:39] <jcoxon> what would be interesting would be do identical launches without a cutdown
[17:39] <Randomskk> not sure of the utility
[17:39] <jcoxon> and see what sort of difference the balloons have
[17:39] <jcoxon> between them
[17:53] <russss> launch two balloons tied together
[17:54] <russss> and then detach them at altitude
[17:54] <Randomskk> ooh, could work
[17:58] <jcoxon> this summer (well my summer so June/July) we should have a few launch days for people to come along to
[17:58] <jcoxon> perhaps with some experiment flights etc
[17:59] <jcoxon> to test out various ideas that have cropped up
[17:59] jasonb (~jasonb@dsl027-180-244.sfo1.dsl.speakeasy.net) joined #highaltitude.
[18:04] <earthshine> good idea James
[18:04] <jcoxon> to go onto my 'after exams' list
[18:10] AlexBreton (Alexander@client-80-5-40-26.cht-bng-014.adsl.virginmedia.net) left #highaltitude.
[18:22] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 240 seconds
[18:33] JamesLeeds (~chatzilla@5ac30595.bb.sky.com) joined #highaltitude.
[18:42] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[18:45] jcoxon (~jcoxon@92.40.113.219.sub.mbb.three.co.uk) left irc: Quit: Leaving
[18:46] Jasperw (~jasperw@skeleton2.london.iofc.org) left irc: Quit: Leaving.
[18:57] DanielRichman (~DanielRic@unaffiliated/danielrichman) joined #highaltitude.
[19:05] AlexBreton (~Alexander@client-80-5-40-26.cht-bng-014.adsl.virginmedia.net) joined #highaltitude.
[19:06] <AlexBreton> guys, what do you think about having 3 cameras on the payload in a way that the fields of view can be put together to make a 180 degree or more panorama? synchronised shutters obviously
[19:06] <SpeedEvil> well...
[19:07] <SpeedEvil> Do you really care?
[19:07] <AlexBreton> about...
[19:07] <SpeedEvil> Is the image going to be moving fast enough that three pics taken over 10s will ahve motion artifacts
[19:08] <AlexBreton> what do you mean?
[19:10] <SpeedEvil> Consider a 360 degree panorama.
[19:10] <SpeedEvil> 360*360
[19:10] <SpeedEvil> If you make this out of simultaneos pictures - it all 'just works'.
[19:11] <SpeedEvil> If you try to make it by taking pictures several seconds apart and stitching them - it will still 'just work' - if the scene is stationary
[19:11] <SpeedEvil> try to fake it
[19:12] <SpeedEvil> And the scene is only - from a HAB - not stationary when fairly close to the ground
[19:12] <DanielRichman> AlexBreton, calm down. We've got funding now but that doesn't mean we just need more stuff
[19:13] <SpeedEvil> Indeed. You could buy beer with the money instead.
[19:13] <DanielRichman> ON a similar note: people that have sent up video cameras: worth it? Personally I don't think so; tried stopmotioning the pictures we had and it was too shaky anyway; would look exactly the same if we took a video and sped it up
[19:14] <SpeedEvil> I'm not sure about that
[19:14] <AlexBreton> SpeedEvil: the shutters would be synced to go off at the same time
[19:14] <SpeedEvil> http://cgi.ebay.co.uk/MINI-DV-Spy-Camera-Key-Chain-Video-DV-Camcorder-720x480-/150436036471?cmd=ViewItem&pt=UK_CCTV&hash=item2306afbf77
[19:14] <SpeedEvil> 4 of the above could be fun
[19:15] <SpeedEvil> AlexBreton: yes - I mean - if you have only one camera, and are relying on it swinging randomly while taking pictures to do your panorame
[19:15] <SpeedEvil> hence the 10s delay
[19:15] <AlexBreton> SpeedEvil: that's why I suggested three at different angles
[19:16] <AlexBreton> so you have simultaneous pictures
[19:16] <SpeedEvil> AlexBreton: yes - I'm just attempting to point out that the advantage is quite limited.
[19:16] <SpeedEvil> It makes the stitching a lot better.
[19:16] <SpeedEvil> And it will improve panoramas close to the ground
[19:16] <SpeedEvil> makes the stitching a lot easier rather
[19:16] <AlexBreton> well there you go
[19:16] <SpeedEvil> not a lot better
[19:17] <SpeedEvil> But do you care about panoramas close to the ground?
[19:17] <AlexBreton> it's impossible to do a proper panorama at high altitude with 1 camera
[19:17] <AlexBreton> as you would on the ground
[19:17] <SpeedEvil> When it's 3* the weignt, meaning a heavier balloon, ...
[19:17] <SpeedEvil> Why do you think it's impossible?
[19:17] <AlexBreton> it moves around too much
[19:17] <AlexBreton> on the ground you always use a tripod
[19:18] <SpeedEvil> Look at some of the high atltitude shots.
[19:18] <SpeedEvil> Even though - yes - it's moving around a lot.
[19:18] <SpeedEvil> There is not usually noticable motion blur - along one or another axis.
[19:18] <SpeedEvil> Simply as the air is quite blurry.
[19:19] <AlexBreton> has anyone made a proper panorama at high altitude yet?
[19:19] <AlexBreton> something that looks recognisable
[19:19] <SpeedEvil> And softens all of the things you might otherwise see blur in.
[19:19] <SpeedEvil> yes
[19:19] <AlexBreton> I know sbasuita is trying now
[19:19] <DanielRichman> AlexBreton, yes we showed you at lunch
[19:19] <AlexBreton> it's not a proper panorama - I mean like t^2's DanielRichman
[19:19] <SpeedEvil> http://www.nivnac.co.uk/blog1.php
[19:19] <DanielRichman> AlexBreton, that's how it's made
[19:20] <SpeedEvil> http://www.nivnac.co.uk/files/HAPSD_NOVA8/HAPS-D_pan1_fudged_polar_half.jpg
[19:21] <AlexBreton> that's horizons stitched up to make a circle
[19:21] <SpeedEvil> no, it's not
[19:21] <AlexBreton> SpeedEvil: http://www.nivnac.co.uk/files/HAPSD_NOVA8/HAPS-D_pan1_quarter.jpg thinking more like this
[19:21] <AlexBreton> but without all the fudging
[19:21] <SpeedEvil> It's an interpolation onto a sphere, and a mapping to a rectangle
[19:22] <AlexBreton> SpeedEvil: it's not recognisably the earth though
[19:22] <SpeedEvil> Sure it is.
[19:22] <AlexBreton> it looks like a sphere sure, but not like a satellite photo
[19:22] <AlexBreton> for obvious reasons
[19:23] <SpeedEvil> The image I posted accurately reflects what you would see if you were at the location of the balloon, looking at the earth through a fisheye lens
[19:23] Upu (~Upu@ubn.upuaut.net) left irc: Ping timeout: 248 seconds
[19:24] Upu (~Upu@ubn.upuaut.net) joined #highaltitude.
[19:25] <AlexBreton> a circular fisheye
[19:25] <AlexBreton> I see
[19:25] <AlexBreton> I prefer the horizontal panorama
[19:25] <SpeedEvil> The panorama mapping is no more real.
[19:26] Simon-MPFH (~simon@phantom.mpfh.co.uk) left irc: Read error: Connection reset by peer
[19:55] <rjharrison> ping DanielRichman
[19:55] <DanielRichman> rjharrison, hi
[19:56] <rjharrison> Quick question on uC's
[19:56] <rjharrison> I see you have managed to log during tx's so i'm guessing your interrupt based
[19:56] <DanielRichman> fire, but you owe me one apostrophe
[19:56] <DanielRichman> yeah totally interrupt based
[19:57] <DanielRichman> bit of a pain to code :P but everything runs simultaneously
[19:57] <rjharrison> So how do you make sure the pin switches at the right time for the tx to work
[19:57] <rjharrison> ie it isn't busy doing something else
[19:57] <DanielRichman> I kept all the interrupt routines short
[19:57] <DanielRichman> short enough that if it were busy when TIMERx (can't remember which it was) did fire
[19:57] <DanielRichman> it wouldn't matter
[19:58] <rjharrison> Cool, what was your frequency for the uC
[19:58] <DanielRichman> 16MHz
[19:58] <rjharrison> Ok i'm only at the lofty height of 8 atm
[19:59] <DanielRichman> It should be enough :)
[19:59] <DanielRichman> I didn't investigate extensively but it might be possible to time all your different interrupts, work out worse-case, and then prove that it cannot go wrong
[19:59] <rjharrison> I may have a go at moving to interrupt driven routines
[20:00] <rjharrison> I like the it worked in test should work in practice aproach
[20:00] <rjharrison> nay new images?
[20:00] <rjharrison> any
[20:00] <DanielRichman> I don't think so
[20:01] <rjharrison> still working on the pano?
[20:01] <DanielRichman> I think sbasuita is
[20:01] <DanielRichman> rjharrison, if you want to log whilst txing then I guess you only need to put the tx on an interrupt
[20:01] <DanielRichman> (and make sure that the shared variables are volatile_
[20:01] <rjharrison> So do you have a counter or a pulse for the tx
[20:01] <DanielRichman> to trigger the interrupt? I think I used TIMER1
[20:02] <DanielRichman> setup to clear on timer match
[20:02] <DanielRichman> and set the interrupt flag
[20:02] <rjharrison> Then fill a buffer to be sent
[20:02] <DanielRichman> yeah; stuff a buffer then start the timer
[20:02] <rjharrison> is it a circular buffer or do you refill when empty?
[20:04] <DanielRichman> I didn't use a buffer because I used an equally overcomplicated (and in hindsight, a bad idea) system of generating one character of the message at a time
[20:04] <DanielRichman> but a refill-when-empty might be easier since you'll be sending data in large blocks, strings
[20:04] <DanielRichman> ie. it's easy to split it up
[20:05] <DanielRichman> your strings arn't longer than ~90bytes? you should have that much ram, you can snprintf into it
[20:05] <rjharrison> Yep I think grab the next sentence when buffer empty
[20:06] <rjharrison> I like the idea of haveing time to do other stuff whilst tx'ing
[20:06] <DanielRichman> be wary of race conditions
[20:06] <rjharrison> explain a bit futher
[20:06] <rjharrison> further
[20:07] <DanielRichman> the compiler (and indeed, more advanced processors) are free to reschedule and rearrange your relatively high-level C to make it faster. To that extent, if you write " var1 = 3; var2 = 4;" on the desktop there's no guarantee that it will be executed in that order.
[20:07] <DanielRichman> similarly the compiler for the atmega may choose not to perform certain instructions (ie. keep that variable in a register)
[20:07] <DanielRichman> for example if you write in main()
[20:08] <DanielRichman> var1 = 4; do_something_else(); if (var1 == 4) ...
[20:08] <DanielRichman> and an interrupt fires and modifies var1 whilst it's doing something else, it's possible that the compiler won't reload the new value; it will have cached it in a register
[20:08] <rjharrison> ahh
[20:08] <rjharrison> Is there a way to make it behave
[20:08] <DanielRichman> so to solve that problem you have the "volatile" keyword
[20:08] <rjharrison> ahh cool
[20:09] <DanielRichman> furthermore, if you write
[20:09] <rjharrison> Yep I do use that in the uart function
[20:09] <DanielRichman> snprintf(my_tx_buffer, 90, "blah blah blah"); lock_my_tx_buffer = 1
[20:09] <DanielRichman> (bad example)
[20:10] <DanielRichman> IF, rjharrison, you don't do any sort of locking, and you write
[20:10] <DanielRichman> snprintf(my_tx_buffer, 90, "%s blah blah blah");
[20:10] <DanielRichman> and an interrupt fires in the middle of the snprintf function
[20:10] <DanielRichman> then it may well start transmitting a dodgy half string
[20:10] <DanielRichman> (though in this example that's unlikely)
[20:10] <rjharrison> Ahh good point
[20:10] <DanielRichman> rjharrison, also be aware of the sei() and cli() instructions
[20:11] <rjharrison> I see what you mean so I need a lock on the tx function
[20:11] <DanielRichman> which disable and enable the execution of interrupts
[20:11] <rjharrison> and it's locked when writing to the string
[20:11] <DanielRichman> they don't discard any interrupts that would have executed; they "pause" and wait for sei() before then executing them
[20:12] <rjharrison> Ok cool I did wonder about that before
[20:12] <rjharrison> So they just queue up
[20:13] <rjharrison> Dare I sak where you learnt all this stuff. When I last did GCSE's this wasn't in the syllabus
[20:14] <DanielRichman> the majority of the specific details are buried very deep in the atmega datasheets
[20:14] <rjharrison> does your tx vary slightly if another intterupt fires or is in progress but it's by such a small amount it makes no difference
[20:15] <rjharrison> Or does your tx interrupt take priority
[20:15] <DanielRichman> rjharrison, if one interrupt is executing (avrlibc documentation) then interrupts are disabled... I think avr libc wraps your interrupt code with a call to cli() and then sei() (a feature you can disable). THis is to prevent deadlocking inside an interrupt
[20:16] <DanielRichman> ie. you can't interrupt an interrupt unless you specifically set it up to do so
[20:16] <DanielRichman> rjharrison, I never tried measuring the practical effect on the tx
[20:19] Jasperw (~jasperw@78-86-9-131.zone2.bethere.co.uk) joined #highaltitude.
[20:19] <rjharrison> so that said if you were in the process of writing a char to the SD card when you were ment to be flipping the tx from a 0 to a 1 there would be (albeit a slight) delay flipping the tx
[20:19] <DanielRichman> correct.
[20:20] <DanielRichman> rjharrison, there's one exception I'd like to point out
[20:20] <DanielRichman> rjharrison, the temperature sensors (1 wire) are not interrupt based. They do all their business with interrupts disabled; because the timings have to be quite precise
[20:21] <rjharrison> Yay
[20:21] <rjharrison> So how do you do that and tx strings
[20:21] <rjharrison> it bust be at the end of a sentence
[20:22] <DanielRichman> I made sure that the routines don't last longer than 20ms
[20:22] <rjharrison> must
[20:22] <DanielRichman> then I do the temperature sensor stuff after flipping a tx bit
[20:22] <rjharrison> so you set the sensor for a read
[20:22] <rjharrison> then you go back to interrupts
[20:22] <DanielRichman> They're triggered by the same timer: timer1 fires 50 times a second; and every 50th hit a "1 second" function is executed, after flipping the bit
[20:23] <DanielRichman> every 60th execution of *this* function we set the sensors to read; then 1 second after we retrieve the reading
[20:23] <DanielRichman> ie. take temperature every 60 seconds
[20:23] <rjharrison> Oh ok I guess that's as about as often as I do
[20:24] Jasperw (~jasperw@78-86-9-131.zone2.bethere.co.uk) left irc: Ping timeout: 276 seconds
[20:24] Jasperw1 (~jasperw@78-86-9-131.zone2.bethere.co.uk) joined #highaltitude.
[20:24] <DanielRichman> rjharrison, also there's some code to try and synchronise the interrupts that are longer with the gaps between GPS sentences. I never actually tried to see if it has effect, but the theory was that 4800 baud might be effected if something a bit heavier was running
[20:24] <DanielRichman> ie. all temperature and sms code is executed in these gaps
[20:25] <rjharrison> Cool I'll think before embarking on this
[20:25] <DanielRichman> rjharrison, http://code.google.com/p/alien-flightcomputer/source/browse/#svn/trunk/alien1/atmega162/final
[20:25] <rjharrison> The lazy person would just off load the tx to a smaller cpu
[20:26] <DanielRichman> doing so has other benefits
[20:26] <DanielRichman> it allows you in the future to use that very other-cpu attached to something like an ARM
[20:26] <DanielRichman> I'm tempted to do that when I get my hands on a NiM2 ( AlexBreton !! )
[20:27] <rjharrison> I think there will be an arm in Icarus IV
[20:28] <rjharrison> They are nice chips and V fast compared to Atmegas
[20:28] <rjharrison> I think there is RTOS for them too
[20:28] <DanielRichman> rjharrison, and best of all they do proper multitasking
[20:28] <DanielRichman> no messing with interrupts
[20:29] <rjharrison> Ohh nice. But avr's are cool for cutting your teeth and unstanding basic uC thinking
[20:29] <junderwood> Not sure whether the AVRs you are using have DMA but if so, it's a great way of driving your radio transmit in the background
[20:32] <junderwood> Hmm. Seems not. Shame
[20:32] <DanielRichman> rjharrison, there's quite a large range of arms though. I think that we'll be using avr until we need heavy lifting
[20:33] <DanielRichman> I should measure the %CPU used on our atmega actually
[20:37] <rjharrison> Sorry kid distraction
[20:51] fsphil (~phil@2001:470:1f09:483:21f:c6ff:fe44:b25b) joined #highaltitude.
[21:01] <rjharrison> all in bed now
[21:02] <rjharrison> DanielRichman other than trying to calc %age used what else can you do
[21:02] <rjharrison> I guess you could inc a counter over time and measure
[21:03] <DanielRichman> you could test, at the end of every interrupt, if you've been keeping the tx interrupt waiting
[21:03] <DanielRichman> or you could get an oscilloscope out
[21:03] <DanielRichman> but you'd have to artificially manufacture the worst case scenario
[21:03] <rjharrison> Yep
[21:05] <rjharrison> scoping for a bad pattern sounds like time consuming work
[21:05] <rjharrison> I guess this is where sims come in
[21:05] <rjharrison> simulators
[21:05] <rjharrison> Where you have enough processing around the sim to see whats going on
[21:06] <DanielRichman> you could just attach two atmegas together, have the second one use its own clock to monitor how true to 50baud it remains
[21:11] <sbasuita> Definitely don't need three cameras btw
[21:11] <sbasuita> One side and one down at most
[21:11] <rjharrison> Oh I guess the scope would be good enough to monitor tx by just flicking between 0 and 1 each time
[21:12] <rjharrison> sbasuita, why not do what I have done for icarus III and use a servo
[21:12] <DanielRichman> sbasuita, 360 degrees of freedom robot arm what
[21:12] <sbasuita> i made a quick panorama of 26 photos
[21:12] <sbasuita> looks good
[21:13] <rjharrison> I get up down and horizon + some funky moves
[21:13] <rjharrison> lol
[21:13] <rjharrison> I like the robot arm
[21:13] <rjharrison> might be a bit heavy
[21:13] <Upu> sounds like alot of weight
[21:13] <Upu> snap
[21:13] <rjharrison> so are we going to see a lunch II from alien?
[21:13] <rjharrison> Launch
[21:14] <sbasuita> definitely
[21:14] <sbasuita> after exams
[21:14] <sbasuita> with video
[21:14] <rjharrison> You'll have all summer to play
[21:14] <sbasuita> :)
[21:14] <rjharrison> Video is quite easy
[21:14] <Upu> I'm going to crack on after the wedding, got my break out board on the way back from that Olimex place
[21:15] <DanielRichman> rjharrison, is video worth it?
[21:15] <Upu> Thoughs on a bullet camera with a fish eye lens on a boom ?
[21:15] <rjharrison> DanielRichman, It helps fill up the 16 GB in the SD card
[21:15] <SpeedEvil> I think one servo will get you most of it
[21:16] <SpeedEvil> Just the ability to pan the camera way up, or way down
[21:16] <SpeedEvil> the balloon spins plenty for the other axis
[21:16] <DanielRichman> but but... I always wanted an excuse to build a robot arm
[21:16] <rjharrison> It depends really I think pics are best for wall and pano's but vid is nice for some events
[21:16] <rjharrison> Launch, burst, landing
[21:16] <DanielRichman> rjharrison, I see
[21:17] <Upu> I'd like to put a camera on a boom with a fish eye so you can see lot in situ
[21:17] <DanielRichman> rjharrison, we're probably going to see a joint launch of the alien1 flight computer & camera with a second camera + servo and an "alien2" payload
[21:17] <sbasuita> like to get a dedicated video camera
[21:17] <DanielRichman> but it's not confirmed yet
[21:17] <DanielRichman> sbasuita, AlexBreton, want to make a deal :P ?
[21:17] <rjharrison> I would like to see the effect of alt on the payload
[21:17] <rjharrison> If you think that we use closed cell foams I would like to see the amount of expansion / deformation you get
[21:18] <DanielRichman> sbasuita, AlexBreton: here's the deal. You two do whatever you (REASONABLY) want with the cameras if you leave me alone to mess with uplink & downlink dominoex fun
[21:18] <SpeedEvil> DanielRichman: you could also use the robot arm for a cutdown
[21:18] <DanielRichman> scissors in the arm, yes!
[21:19] <DanielRichman> It might drop the camera though
[21:19] <rjharrison> DanielRichman nice one on the domino ex attempt
[21:19] <rjharrison> I fancied that
[21:19] <SpeedEvil> DanielRichman: teach it juggling.
[21:19] <rjharrison> but there is only som much time in ones life
[21:19] <AlexBreton> honda asimo style
[21:20] <DanielRichman> rjharrison, I've got all the components lined up after a masssive maxim samples order. I'm the head of R&D at alien tch which is producing 50k units, apparently
[21:20] <sbasuita> DanielRichman: is dominoex the best mfsk mode?
[21:20] <DanielRichman> sbasuita, it's pretty good. Might try some olivia
[21:20] <sbasuita> DanielRichman: olivia is ascii though iirc
[21:20] <rjharrison> DominoX has a stong hab following
[21:21] <DanielRichman> SpeedEvil, I meant to ask you; as a reference voltage for a DAC, the Maxim dedicated "reference" chips require a large Vin which is a pain to provide. Is a voltage-buffered/op amp follower buffered potential dividor good enough to provide the 2.5V?
[21:21] <SpeedEvil> maybe
[21:21] <DanielRichman> what would you do?
[21:21] <SpeedEvil> You know about rail-rail opamps?
[21:22] <SpeedEvil> And polarity inversion?
[21:22] <fsphil> how tolerant is olivia or dominoex against frequency shifting?
[21:22] <DanielRichman> SpeedEvil, to a certain extent
[21:22] <DanielRichman> sbasuita, probably going to have 5 modes, interchangable by uplink: rtty50, rtty300, dominoex, olivia and midi. It will fall back to default 50rtty if no uplink comms are received
[21:22] <SpeedEvil> DanielRichman: If you've got a rail-rail opamp - then yes - a buffer will just work
[21:22] <SpeedEvil> DanielRichman: there is the seperate issue
[21:22] <rjharrison> Are you using the C11111 thing for uplink?
[21:23] <SpeedEvil> DanielRichman: Does the ntx2 frequency depend on the voltage of the control input.
[21:23] <DanielRichman> SpeedEvil, I have http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1267 a MAX494 and 492
[21:23] <DanielRichman> SpeedEvil, yes
[21:23] <SpeedEvil> DanielRichman: Or does it depend on the voltage of the control input as a fraction of the supply voltage
[21:23] <DanielRichman> rjharrison, nah a NiM2, an ADC and an Atmega
[21:23] <DanielRichman> then on the ground the biggest transmitter I can get my hands on
[21:23] <DanielRichman> by then hopefully I should be able to tx 50W
[21:24] futurity (~anonymous@cpc7-cmbg15-2-0-cust14.know.cable.virginmedia.com) joined #highaltitude.
[21:24] <AlexBreton> DanielRichman: what will we be uplinking?
[21:24] <DanielRichman> SpeedEvil, hmm I think the NTX2 has internal voltage regulation
[21:24] <DanielRichman> no wait...
[21:24] <DanielRichman> I'm not sure. I'd have to try it
[21:25] <SpeedEvil> DanielRichman: you need - ideally - your reference for the DAC to drift the same as the reference for the ntx2
[21:25] <DanielRichman> so if I power the ntx2 from the circuits main 5v regulator
[21:25] <DanielRichman> then divide that 5v into 2; 2.5V for the DAC...
[21:25] <SpeedEvil> yes.
[21:25] <futurity> Hi, does anyone know if the Alien team have uploaded any images yet from their launch?
[21:25] <DanielRichman> perfect.
[21:26] <sbasuita> futurity: http://flickr.com/alienhab/sets
[21:26] <futurity> Many thanks
[21:26] <DanielRichman> SpeedEvil, thank you again :)
[21:26] <SpeedEvil> np
[21:27] <DanielRichman> I'll later put my breadboard in a freezer & replace the regulator with something less stable for some stress testing
[21:31] <futurity> some great photos :)
[21:32] <futurity> ttfn
[21:32] futurity (~anonymous@cpc7-cmbg15-2-0-cust14.know.cable.virginmedia.com) left irc: Quit: futurity
[21:32] <fsphil> the bad weather might have worked to your advantage, the clouds looks amazing from above
[21:33] <sbasuita> yeah there's a great variety
[21:34] <SpeedEvil> http://www.foddy.net/Athletics.html - completely off-topic. Well - maybe not - robotic legs
[21:37] <DanielRichman> fsphil, I made your packaging changes work on mingw
[21:38] <DanielRichman> also rjharrison I think the crash is fixed.
[21:38] <DanielRichman> (I'm not 100% sure)
[21:38] <fsphil> DanielRichman, I fixed a crash earlier -- same one? (updating hab interface when not in hab mode?)
[21:38] <DanielRichman> fsphil, that's my suspicion
[21:39] <fsphil> aye .. someone here mentioned it during the alien flight
[21:39] <DanielRichman> I mentioned ability to reproduce a bug rjharrison mentioned the day before the flight
[21:39] <DanielRichman> good you caught it
[21:39] <fsphil> What do you guys think of changing the executable name to dl-fldigi?
[21:39] <DanielRichman> anyway about to upload a new windows binary
[21:39] <DanielRichman> fsphil, I merged that in; it's a good idea allows people to have both installed
[21:39] <fsphil> I recorded the alien signal so was able to test it out a bit
[21:40] <DanielRichman> yes didn't you set a record?
[21:40] <fsphil> wow, you're way ahead of me mate lol
[21:40] <fsphil> I think so, 555km
[21:40] <DanielRichman> sbasuita, btw the RADARC guys asked me to do a presentation :X
[21:40] <DanielRichman> sbasuita, I said aprez exams
[21:41] <fsphil> that's a good part of the way towards me making a package for Fedora, and I believe someone else is going to try submitting it to debian
[21:42] <sbasuita> DanielRichman: on the thing in general or just the electronics?
[21:42] <sbasuita> DanielRichman: also who to
[21:42] <DanielRichman> the general radarc meet peeps. they get a crowd of ~30
[21:42] <DanielRichman> sbasuita, of the thing in general but including electronics I imagine
[21:42] <sbasuita> fsphil: yeah when the new github one gets polished up i will be working with a debian developer to do that
[21:43] <sbasuita> DanielRichman: cool
[21:46] jcoxon (~jcoxon@92.40.239.216.sub.mbb.three.co.uk) joined #highaltitude.
[21:47] <jcoxon> evening
[21:47] <Upu> Evening
[21:48] <jcoxon> ping DanielRichman rjharrison, fsphil
[21:48] <DanielRichman> sup jcoxon
[21:49] <jcoxon> dl-fldigi meeting?
[21:49] <DanielRichman> about to post a new dl-fldigi binary to my windows downloads section
[21:50] <jcoxon> okay
[21:50] <jcoxon> guess i missed the others
[21:51] <DanielRichman> fsphil is was here ~9 mins ago; rjh ~30
[21:51] SpeedEvil1 (~user@tor/regular/SpeedEvil) joined #highaltitude.
[21:54] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 260 seconds
[21:54] <DanielRichman> jcoxon, http://github.com/danielrichman/dl-fldigi-windows-build/downloads
[21:54] SpeedEvil (1000@tor/regular/SpeedEvil) left irc: Ping timeout: 260 seconds
[21:57] Jasperw1 (~jasperw@78-86-9-131.zone2.bethere.co.uk) left irc: Quit: Leaving.
[21:58] <fsphil> back
[21:59] Hiena (~Hiena@81.93.195.181.datatrans.hu) joined #highaltitude.
[22:19] superkuh (hukrepus@unaffiliated/superkuh) joined #highaltitude.
[22:19] JamesLeeds (~chatzilla@5ac30595.bb.sky.com) left irc: Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]
[22:25] <earthshine> evening
[22:27] Nick change: SpeedEvil1 -> SpeedEvil
[22:30] <Hiena> Hi all!
[22:30] <earthshine> \o/
[22:30] <Hiena> I need some fast worst case temperature data for 3000m.
[22:31] <jonsowman> -10C
[22:31] <jonsowman> *very* rough guess
[22:32] <Hiena> The guys at the uni, want to use stock microservos, and i fear it's out of the spec.
[22:32] <jonsowman> what is their spec range
[22:32] <Hiena> 0-40
[22:33] <SpeedEvil> 3000m?
[22:33] <SpeedEvil> oh
[22:33] <jonsowman> you'll probably get away with that
[22:33] <SpeedEvil> 6C/5km IIRC
[22:33] <jonsowman> http://www.wolframalpha.com/input/?i=temperature+at+3km+altitude
[22:33] <SpeedEvil> or is that 6c/m
[22:33] <SpeedEvil> it will _totally_ depend where you are
[22:33] <SpeedEvil> this is for a flying thing?
[22:33] <Hiena> Yeah, but how about the winter conditions?
[22:34] <SpeedEvil> where in the world are you?
[22:34] <Hiena> Middle europe.
[22:34] <SpeedEvil> 3km in winter over kenya is gonna be not cold.
[22:34] <SpeedEvil> -35 is probably pessimistic
[22:34] <SpeedEvil> has this got to fly all the time?
[22:35] <Hiena> Well, just forget it. I'll kick some balls for the better specifications.
[22:35] <sbasuita> panorama is looking very promising
[22:35] <sbasuita> running 120 image hugin now
[22:35] <SpeedEvil> http://weather.unisys.com/model/details.html see that hie
[22:36] <SpeedEvil> 850 millibars is about rightish
[22:53] <sbasuita> DanielRichman: for a2 let's do sstv with the camera on a servo controlled by uplink
[22:53] <DanielRichman> yes
[22:53] <DanielRichman> sbasuita, I'd love to
[22:53] <jcoxon> :-D
[22:53] juxta (fourtytwo@203.122.193.94) joined #highaltitude.
[22:54] <sbasuita> DanielRichman: it will be interesting to see what sort of entries that specialist science competition that longstaff mentioned gets
[22:58] Upu (~Upu@ubn.upuaut.net) left irc:
[23:05] Hiena (~Hiena@81.93.195.181.datatrans.hu) left irc: Quit: -=Got bored from the net. Gone blowing up things.=-
[23:05] junderwood (~John@adsl.jcu.me.uk) left irc: Quit: Leaving
[23:10] DanielRichman (~DanielRic@unaffiliated/danielrichman) left irc: Quit: Leaving
[23:10] smealum (~smealum@82.243.132.64) left irc: Ping timeout: 276 seconds
[23:11] jcoxon (~jcoxon@92.40.239.216.sub.mbb.three.co.uk) left irc: Quit: Leaving
[23:13] Laurenceb (~laurence@host86-136-234-181.range86-136.btcentralplus.com) joined #highaltitude.
[23:15] SpeedEvil (~user@tor/regular/SpeedEvil) left irc: Read error: Operation timed out
[23:29] smealum (~smealum@82.243.132.64) joined #highaltitude.
[23:34] N900evil (~Speedevil@tor/regular/SpeedEvil) joined #highaltitude.
[23:43] Cleo (~Cleo@unaffiliated/cleo) left irc: Ping timeout: 265 seconds
[23:45] SpeedEvil (~user@tor/regular/SpeedEvil) joined #highaltitude.
[23:48] N900evil (~Speedevil@tor/regular/SpeedEvil) left irc: Ping timeout: 240 seconds
[23:48] Cleo (~Cleo@unaffiliated/cleo) joined #highaltitude.
[00:00] --- Wed May 5 2010