highaltitude.log.20080125

[00:28] natrium42 (n=alexei@bas3-kitchener06-1096650452.dsl.bell.ca) joined #highaltitude.
[00:54] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[00:54] flowolf (n=flowolf@87.13.187.42) left irc: Remote closed the connection
[02:16] fnoble_ (n=fnoble@fn217.quns.cam.ac.uk) left irc: Read error: 110 (Connection timed out)
[04:35] smealum (n=smealum@82.243.132.64) left irc: Read error: 113 (No route to host)
[05:35] GhostChe (n=che@c-24-91-163-92.hsd1.ma.comcast.net) left irc: Read error: 110 (Connection timed out)
[05:36] GhostChe (n=che@c-24-91-163-92.hsd1.ma.comcast.net) joined #highaltitude.
[07:30] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[08:09] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[11:38] flowolf (n=flowolf@87.13.187.42) joined #highaltitude.
[11:48] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[13:31] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[14:59] prpplague (n=dave@mail.americanmicrosystems.com) joined #highaltitude.
[15:01] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[15:32] Laurenceb (n=Laurence@fwnat-pub-1.physics.ox.ac.uk) joined #highaltitude.
[15:32] <Laurenceb> hi all
[15:32] Action: Laurenceb has assembled the radio and spi thermopile amplifier boards
[15:50] borism (n=boris@195-50-199-202-dsl.krw.estpak.ee) joined #highaltitude.
[16:01] <Laurenceb> edmoore: hows it going?
[16:24] flowolf (n=flowolf@87.13.187.42) left irc: "Leaving"
[17:02] Laurenceb (n=Laurence@fwnat-pub-1.physics.ox.ac.uk) left irc: Read error: 104 (Connection reset by peer)
[19:37] GhostChe (n=che@c-24-91-163-92.hsd1.ma.comcast.net) left irc: Read error: 110 (Connection timed out)
[19:38] GhostChe (n=che@c-24-91-163-92.hsd1.ma.comcast.net) joined #highaltitude.
[19:51] Hiena (n=Hiena@81.93.195.181.datatrans.hu) joined #highaltitude.
[19:51] <Hiena> Good evening!
[20:58] smealum (n=smealum@82.243.132.64) joined #highaltitude.
[21:13] Laurenceb (n=laurence@dhcp38-235.sthughs.ox.ac.uk) joined #highaltitude.
[21:13] <Laurenceb> hi all
[21:14] <Laurenceb> TSSOP and 0603 parts isnt too hard :P
[21:14] <Laurenceb> I've assembled the thermopile amplifier without any issues
[21:15] <Hiena> Yeah. I'm dealing with these bastards every day.
[21:15] <Laurenceb> need to find a supplier that can go smaller than batchpcb :D
[21:16] <Hiena> Smaller?
[21:16] <Laurenceb> http://wiki.ukhas.org.uk/projects:ukhas_glider_project:components
[21:16] <Laurenceb> smaller is better :D
[21:17] <Hiena> Ah. are you using the Paparazzi's system?
[21:17] <Laurenceb> no
[21:17] <Laurenceb> but its paparazzi inspired
[21:17] <Laurenceb> Its simpler, and avr based
[21:17] <Laurenceb> but designed with high altitudes in mind
[21:19] <Laurenceb> http://wiki.ukhas.org.uk/projects:ukhas_glider_project:jobs
[21:19] <Hiena> I see. Currently i'm working on a camera based horizont finding and measuring system. But it's not suited for a high altitude projects.
[21:20] <natrium42> Hiena, on fpga?
[21:21] <Laurenceb> Hiena: you work with fpgas ?
[21:22] <Hiena> Nope. It's a land based system. The camera mounted on the plane with the transmitter and the code running on the ground base.
[21:22] <natrium42> ah, ok
[21:22] <Hiena> At work, i'm soldering everykind of parts from the 200 pin BGAs to the 1210 size resistors.
[21:23] <Laurenceb> soldering BGA ?!
[21:23] <natrium42> you use ovens?
[21:23] <Hiena> I can insert 700 component/day.
[21:23] <Hiena> Nope. I'm using an ELSA BGA rework station.
[21:23] <Laurenceb> ok
[21:23] Action: natrium42 will need to have a small BGA soldered soonish
[21:24] <Laurenceb> cool
[21:24] <natrium42> have been planning to try using my hot air reflow station...
[21:24] <natrium42> never done BGA with it
[21:24] <Hiena> Also, i'm flip-BGA expert.
[21:24] <Laurenceb> the space instruments lab where I'm working is so outdated
[21:24] <Laurenceb> the use dip for everything
[21:25] <Hiena> Because it's a right way.
[21:25] <Laurenceb> well, when it comes to flight hardware, they use bare die
[21:25] <Laurenceb> and mount straight onto the board
[21:25] <Hiena> The dip is much more reliable as the SMT or BGA componenets.
[21:25] <natrium42> do they epoxy the board afterwards?
[21:25] <Laurenceb> no
[21:25] <natrium42> but what if the components come loose?
[21:26] <Laurenceb> epoxy could expand or contract by a different factor
[21:26] <Laurenceb> they are glued on from below
[21:26] <Laurenceb> with epoxy
[21:26] <natrium42> ah
[21:26] <natrium42> interesting
[21:26] <Laurenceb> but not epoxy encased
[21:26] <Hiena> The bonding is much stronger as the soldering.
[21:26] <Laurenceb> the gold bond wires are exposed
[21:27] <prpplague> Hiena: you're doing flip-chip work?
[21:27] <Laurenceb> I was talking to the guy who made the beagle 2 wind sensor earlier
[21:28] <prpplague> Laurenceb: cool
[21:28] <Hiena> Yup. The biggest was 64 pin. Now i thinking an 217 pin run. ;)
[21:28] <Laurenceb> he's working on a smaller one for exomars
[21:28] <Laurenceb> its about 2mm*1mm
[21:28] <prpplague> Hiena: interesting, i did some research on using flip chip on some of products
[21:28] <Hiena> Laurenceb, neat.
[21:28] <prpplague> Hiena: but we don't have the volume as of yet
[21:29] <prpplague> Hiena: most of what i looked at you needed around 20k a year to make it economical
[21:30] <Laurenceb> basically a quartz cylinder with 1 micron titanium coating etched to form tracks
[21:30] <Laurenceb> these heat due to current flow
[21:30] <Laurenceb> and the resistance changes
[21:30] <Laurenceb> if you measure the current required to keep the temperature constant
[21:31] <Laurenceb> and compare different sides, you can calculate windspeed and direction
[21:31] <Hiena> prpplague, the flip-BGA is much easier for the prototypes and the small workshops.
[21:32] <prpplague> Hiena: yea, we have a pick-n-place machine
[21:33] <prpplague> Hiena: we do about 10k per year, maybe when we get to 20k we'll start doing flip-chip stuff
[21:33] <prpplague> Hiena: http://www.elinux.org/Hammer_Board
[21:34] <Hiena> I think we talking different things. The flip-BGA is when you glue the BGA to the board positioned the piuns upward, and wired one by ones.
[21:35] <prpplague> Hiena: right, and flip-chip is similiar but done with bare silicon and then epoxed over
[21:35] <Hiena> Wow. Hom much is one Hammer chip?
[21:36] <prpplague> Hiena: $159USD in single quantity or $249 for a kit
[21:36] <Laurenceb> not bad
[21:37] <Laurenceb> I've just started experimenting with embedded linux
[21:37] <prpplague> Laurenceb: yea its a little higher than we want, we are working on some larger oem deals to help bring down the single qty price
[21:37] <Hiena> Ugh...It's out of my budget.
[21:37] <Laurenceb> SSH is a nice feature
[21:37] <prpplague> Hiena: whats your budget range?
[21:37] <Laurenceb> I could leave my payload at the lab and work on it from home :P
[21:38] <prpplague> Laurenceb: indeed
[21:38] <prpplague> Laurenceb: i did a complete board bringup remotely once via ssh
[21:39] <Laurenceb> say would anyone be interested in some of the thermopile amplifier boards?
[21:39] <Laurenceb> I might have 4 or so going spare
[21:39] <Hiena> I guess somewhere your lunch money. ;) Ususaly, 75USD/half year.
[21:40] <Laurenceb> you'll need an AD8574 and AD7923 to make it work
[21:40] <prpplague> Hiena: hehe, not my lunch money, but yea, i can understand your price range
[21:42] <prpplague> Hiena: we do give away 2 or 3 kits a month to people doing interesting projects and or documentation on the wiki pages
[21:45] <Laurenceb> strangely my ngw100 isnt talking to my avr :(
[21:45] <Laurenceb> it gets loads of nonsense back
[21:46] <Laurenceb> do you have to take care to time serial comms correctly under linux?
[21:47] <prpplague> Laurenceb: no, standard rs-232?
[21:47] <Laurenceb> as in if the avr starts to transmit too soon, will it screw the whole transmission
[21:47] <Laurenceb> 3.3v rs232
[21:47] <Hiena> prpplague, i see. I'll show it to the guys at the university, and aquire a budget. Will you ship it oversea?
[21:48] <prpplague> Laurenceb: no, standard rs-232 is usually handled very well, might double check that you have the baud rates set properly
[21:48] <Laurenceb> yes
[21:48] <Laurenceb> or... perhaps the avr is resetting itself
[21:48] <prpplague> Laurenceb: thats possible too
[21:48] <Laurenceb> I'll check the pull up off reset
[21:48] <prpplague> Laurenceb: rs-232 under linux is pretty darn robust, so i'd suspect something on the settings wise
[21:49] <Laurenceb> and the crystal, I swapped that for a rather odd old one from the cupboard
[21:49] <prpplague> Hiena: yes, we ship overseas and are RoHS compliant
[21:49] <prpplague> Laurenceb: OH
[21:49] <Laurenceb> maybe the avr dosnt like it
[21:49] <prpplague> Laurenceb: the rs-232 baud rates are pll's of the main clock
[21:49] <Laurenceb> sure, this is avr crystal
[21:49] <Laurenceb> not ngw100 crystal
[21:50] <prpplague> Laurenceb: right
[21:50] <prpplague> Laurenceb: the avr's i've used, have pll calculations for the baud rate
[21:50] <prpplague> Laurenceb: based on the main xtal value
[21:50] Dave2 (i=Dave2@freenode/staff/dave2) left #highaltitude ("Leaving").
[21:51] <prpplague> Laurenceb: what xtal freq are you using?
[21:51] <Laurenceb> 10MHz
[21:51] <prpplague> Laurenceb: which avr are you using?
[21:52] <Laurenceb> mega168
[21:52] Action: prpplague grabs the datasheet
[21:52] <Hiena> prpplague, i think the ROHS is the last thing what they cares about at the uni. Maybe around november, they will cleans up the budget, buy some chips.
[21:52] <Laurenceb> I'm using the pyserial extension to talk to the port from python
[21:53] <prpplague> Hiena: great, if you have any questions, we discuss the hammer and related stuff over on #edev
[21:54] <Hiena> Laurenceb, did you checked the output with the serial terminal?
[21:54] <Laurenceb> no
[21:54] <Laurenceb> theres no terminal on the ngw I think
[21:55] <Laurenceb> unless I cross compile
[21:55] <prpplague> Laurenceb: the code you are running on the atmega168, did you calculate the correct baud rate for the 10mhz xtal?
[21:55] <Hiena> I means connecting the serial lines to a PC and check it.
[21:56] <Laurenceb> prpplague: I've tested the code using a pc
[21:56] <Laurenceb> but not with the present design for the board
[21:56] <prpplague> Laurenceb: ahh
[21:56] <prpplague> Laurenceb: are you sure you are using the same xtal freq as before?
[21:56] <Laurenceb> I'm not
[21:57] <prpplague> Laurenceb: what were you using before?
[21:57] <Laurenceb> but I changed the config file
[21:57] <Laurenceb> from 16MHz to 10MHz
[21:57] <Laurenceb> so winavr whould have changed everything
[21:57] <prpplague> Laurenceb: hmm, yea the first thing i'd check is your UBRR value
[21:58] <Laurenceb> ok
[21:58] <prpplague> Laurenceb: what baud rate are you using?
[21:58] Action: Hiena heads back to the XFLR5
[21:58] <Laurenceb> 9600
[21:58] <Laurenceb> I'll try some simple code to say while(1){printf("hello world \r\n");}
[21:59] <prpplague> Laurenceb: your UBRR should be 64 with a 10mhz xtal
[21:59] <Laurenceb> and see if anything comes out
[21:59] <Laurenceb> ok thanx
[21:59] <Laurenceb> but its all handled by my #define statements
[21:59] <prpplague> and 103 for 16mhz
[21:59] <prpplague> Laurenceb: yea, _it should be_
[22:00] <prpplague> Laurenceb: but it something to double check
[22:00] <Laurenceb> my gps works ok, so I know pyserial works
[22:01] <prpplague> pyserial is running on the avr32 board?
[22:01] <Laurenceb> yes
[22:01] Action: prpplague isn't familiar with pyserial
[22:01] <Laurenceb> its a python extension
[22:15] <Laurenceb> lol http://www.youtube.com/watch?v=stPqr6B_zEY
[22:16] <Laurenceb> genius
[22:16] <prpplague> work safe?
[22:17] <Laurenceb> lol
[22:17] <Laurenceb> maybe
[22:17] <Laurenceb> its on youtube
[22:17] <Laurenceb> its a parody of "chavs"
[22:18] <prpplague> sorry not familiar with that
[22:18] <Laurenceb> Uk thing
[22:19] <prpplague> yea reading about it now
[22:19] <Hiena> Ehhhh...
[22:20] <Hiena> Please, rather link some high alt. MIG-25 video with a cheesey synth. music.
[22:20] <Laurenceb> lol sorry
[22:21] <prpplague> hehe
[22:21] <Laurenceb> in which case you would like: http://pripyat.com/
[22:27] <Laurenceb> i love the logo
[22:28] <Hiena> No thanks. I'm not browsing such pictures, when i could do almost same ones within one hour driving. Same buildings, less radiation. Welcome to East-Europe...
[22:29] <Laurenceb> lol
[22:29] <Laurenceb> this is interesting http://gallery.pripyat.com/browseimages.php?s=b33cbe1b753c90af1a38edb329f51dbb&c=5
[22:33] <Laurenceb> wow http://gallery.pripyat.com/files//279-P101004.jpg
[22:33] <Laurenceb> wouldnt catch me up there
[22:38] GhostChe (n=che@c-24-91-163-92.hsd1.ma.comcast.net) left irc: Read error: 110 (Connection timed out)
[22:38] <Hiena> Laurenceb, what kind of glider will be drop?
[22:38] <Laurenceb> with the thermopile guidance?
[22:38] <Laurenceb> probably a flying wing
[22:39] GhostChe (n=che@c-24-91-163-92.hsd1.ma.comcast.net) joined #highaltitude.
[22:40] <Hiena> It will be autostable?
[22:41] <Laurenceb> no
[22:41] <Laurenceb> pid control using the thermopiles
[22:41] <Hiena> Well, that case, it will be interesting drop.
[22:42] <Laurenceb> http://wiki.ukhas.org.uk/projects:ukhas_glider_project:code
[22:52] prpplague (n=dave@mail.americanmicrosystems.com) left irc: "Leaving"
[22:53] <Hiena> Hmmmm... I rather try a delta wing configuration with a high V-angle settings. I think, it has a fat chance to start a spin at the high altitude, if it's not autostable.
[22:53] <Hiena> The FMX5 proved that.
[22:55] <Laurenceb> yes, maybe a more stsble design would be best
[22:56] <Laurenceb> but the electronics could be used on lots of different designs
[22:56] <Laurenceb> bbl
[22:57] <Hiena> The problem is at HA, the boundary layer separation happens much easier. If your control loop not smooth enough, you willenter to spin immedietly.
[22:58] <Hiena> Also, if it start at the ailerons and the plane is a flying wing, that means immediately control loss.
[23:09] Hiena (n=Hiena@81.93.195.181.datatrans.hu) left irc: "-=Alkoholmentes-sör és repülõgép-szimulátor, biztos út a guminõ felé=-"
[23:52] <Laurenceb> interesting
[23:52] <Laurenceb> I hadn't looked at the fluid dynamics, but that makes sense
[00:00] --- Sat Jan 26 2008