highaltitude.log.20091002

[00:02] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[00:11] <Laurenceb> http://www.liquidware.com/shop/show/QCT/QuadCore+Tower+Kit
[00:11] Action: Laurenceb contemplates as a form of suicide
[00:12] <Randomskk> but it's just four arduinos. :|
[00:12] <Randomskk> oh, actually five.
[00:12] <Laurenceb> lol
[00:12] sbasuita (n=sbasuita@unaffiliated/drebellion) left irc: Remote closed the connection
[00:12] <Randomskk> I wonder how you program them :P
[00:12] <Laurenceb> Randomskk: are you using analogue cmparitor for your brushless controller?
[00:13] <Randomskk> not in the current design, but I've not programmed the breadboard version yet so don't know if it'd be needed
[00:13] <Randomskk> I assume for monitoring the voltage on the shunt resistor?
[00:14] <Laurenceb> for the brushless phase detection
[00:14] <Randomskk> oh, right. you'd need three inputs though surely?
[00:14] <Randomskk> it's just got each phase going through a potential divider and low pass filter to an ADC input
[00:34] <Laurenceb> ah that solution
[00:34] <Laurenceb> there is another way to do it
[00:34] <Laurenceb> you have a resistor network and use the analogue comparitor
[00:35] <Randomskk> atmel note I was reading used the analogue comparitor for comparing the current to a fixed value to very quickly stop in overcurrent
[00:35] <Laurenceb> then you set some output pins to balance the resistor network dependant on the H bridge outputs so you can use the comparitor for phase detection
[00:35] <Randomskk> I guess either method works
[00:35] <Randomskk> I see
[00:35] <Randomskk> I can see how it'd work, will look into it
[00:36] <Randomskk> don't know if there's any advantage to using ADCs besides the fast response time
[00:36] <Laurenceb> yeah overcurrent protection is a nice feature
[00:36] <Randomskk> well and interrupt triggered "voltage at midpoint"
[00:36] <Laurenceb> most of the RC stuff doesnt have it
[00:36] <Randomskk> can do overcurrent in software too, just reading the voltage on the shunt using an ADC
[00:36] <Randomskk> but that limits reaction time to one control loop
[00:36] <Laurenceb> yeah
[00:36] <Randomskk> personally I think that's acceptable
[00:37] <Laurenceb> funny story - SSTL seem to think brushless ESCs are too hard
[00:37] <Laurenceb> so they use off the shelf ASICs
[00:37] <Randomskk> it also does stall detection and attempted restart, based on a watchdog timer that is reset every commutation
[00:37] <Laurenceb> neat
[00:37] <Randomskk> most of that is just straight from atmel's provided source, but I'll be hacking it up a fair bit to make it compile under gcc-avr and also adding my own stuff
[00:38] <Randomskk> not got the time to do that atm though, still working on these engineering questions and getting other stuff ready in a general panic before saturday
[00:38] <Laurenceb> heh
[00:38] <Randomskk> considering adding a temperature monitor for the mosfet heatsinks, too, it'd be cool to see that on the ground computer
[00:38] <Randomskk> it can already do voltage, current and rpm over i2c
[00:39] <Randomskk> temperature as well would be fun, if only because I want a cool screen with a wireframe 3d model showing the attitude with boxes over each ESC showing vital stats etc :P
[00:39] <Laurenceb> hehe
[00:39] <Laurenceb> have you considered a central ESC for all 4 motors?
[00:39] <Laurenceb> guess it increases EMI
[00:40] <Laurenceb> - interference
[00:40] <Randomskk> not really thought about it too much - seems like you'd still need 3 outputs per motor
[00:40] <Randomskk> I don't know if the motors could be driven off the same outputs
[00:40] <Laurenceb> no
[00:40] <Laurenceb> but it would be a single PCB
[00:40] <Laurenceb> and maybe a single uC
[00:40] <Randomskk> this way already uses every pin and timer on the atmega, so I'd need either a more powerful mcu or more hardware
[00:40] <Laurenceb> maybe even use the ARM
[00:41] <Laurenceb> ah
[00:41] <Randomskk> but even an stm32 would be hard pressed to do 4 of them based on its timers and peripherals
[00:41] <Randomskk> could be done, but tbh four small, cheap PCBs that can be mounted on the struts far from my sensors seems like a better idea
[00:41] <Laurenceb> yes
[00:41] <Randomskk> also if I want to go to eight rotors it's easier this way too
[00:41] <Laurenceb> what the...
[00:41] <Laurenceb> as if 4 wasnt bad enough :P
[00:41] <Randomskk> most importantly, the wires coming out of the motors are pretty short
[00:42] <Randomskk> :P
[00:42] <Laurenceb> I'm actually interested in a dual motor controller with one micro
[00:42] <Randomskk> I do want to get the main flight controller all on one PCB
[00:42] <Randomskk> which is why I really, really hope the ST gyros will prove good enough
[00:42] <Randomskk> no daughterboards = wooo
[00:42] <Randomskk> they do them in a reasonable range of sensitivities and they all provide a 4x amplified output so it seems like they could cope
[00:43] <Laurenceb> I think the ST gyros should be fine
[00:43] <Randomskk> just wish I had another month of holiday really
[00:43] <Randomskk> then I could get this thing flying and hopefully fairly stable before going up
[00:43] <Laurenceb> people have made quadcopters with standard heli gyros
[00:43] <Randomskk> true
[00:43] <Randomskk> will compare them to the AD ones anyway
[00:44] <Randomskk> shame no one makes a 6DOF accel+gyro IC yet
[00:44] <Laurenceb> gtg, cya
[00:44] <Randomskk> better yet, one with internal 12bit ADCs on SPI. /dream
[00:44] Laurenceb (n=laurence@host81-154-154-237.range81-154.btcentralplus.com) left irc: Remote closed the connection
[00:44] <Randomskk> seeya
[01:04] Action: SpeedEvil sighs.
[02:41] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 110 (Connection timed out)
[03:08] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[04:09] Nick change: xatm092 -> xatm092|bed
[04:11] gordonjcp (n=gordonjc@symmetria.fi) got netsplit.
[04:11] ybit (i=ybit@unaffiliated/ybit) got netsplit.
[04:11] bittwist (n=a@unaffiliated/bittwist) got netsplit.
[04:11] jasonb (n=jasonb@dsl027-180-244.sfo1.dsl.speakeasy.net) got netsplit.
[04:11] smealum (n=smealum@smea.servebeer.com) got netsplit.
[04:11] herabot (n=herabot@78.144.200.73) got netsplit.
[04:13] jasonb (n=jasonb@dsl027-180-244.sfo1.dsl.speakeasy.net) returned to #highaltitude.
[04:13] gordonjcp (n=gordonjc@symmetria.fi) returned to #highaltitude.
[04:13] ybit (i=ybit@unaffiliated/ybit) returned to #highaltitude.
[04:13] bittwist (n=a@unaffiliated/bittwist) returned to #highaltitude.
[04:13] herabot (n=herabot@78.144.200.73) returned to #highaltitude.
[04:13] smealum (n=smealum@smea.servebeer.com) returned to #highaltitude.
[04:20] juxta (i=fourtytw@122-49-156-209.ip.adam.com.au) joined #highaltitude.
[04:45] jasonb (n=jasonb@dsl027-180-244.sfo1.dsl.speakeasy.net) left irc: Read error: 110 (Connection timed out)
[04:49] natrium42 (n=natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Read error: 110 (Connection timed out)
[05:21] jasonb (n=jasonb@m480436d0.tmodns.net) joined #highaltitude.
[05:28] darknesslord_ (n=darkness@187.146.33.243) joined #highaltitude.
[05:38] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[05:54] jasonb (n=jasonb@m480436d0.tmodns.net) left irc: Read error: 113 (No route to host)
[05:57] darknesslord_ (n=darkness@187.146.33.243) left irc: "2nibbles!!"
[06:36] DanielRichman (n=daniel@unaffiliated/danielrichman) joined #highaltitude.
[06:37] DanielRichman (n=daniel@unaffiliated/danielrichman) left irc: Client Quit
[06:54] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: Read error: 54 (Connection reset by peer)
[06:54] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[06:56] jasonb (n=jasonb@adsl-66-124-73-250.dsl.sntc01.pacbell.net) joined #highaltitude.
[07:18] natrium42 (n=natrium4@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[08:19] bt42 (n=a@unaffiliated/bittwist) joined #highaltitude.
[08:20] akawaka (n=akawaka@icculus.org) left irc: Remote closed the connection
[08:20] akawaka (n=akawaka@icculus.org) joined #highaltitude.
[08:37] natrium42 (n=natrium4@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left #highaltitude.
[08:37] bittwist (n=a@unaffiliated/bittwist) left irc: Read error: 113 (No route to host)
[08:42] junderwood (n=chatzill@87.194.214.222) joined #highaltitude.
[09:55] RocketBoy (n=Steve@217.47.75.8) joined #highaltitude.
[11:16] SpeedEvil (n=user@tor/regular/SpeedEvil) left irc: "Leaving."
[11:34] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[11:57] SpeedEvil (n=user@tor/regular/SpeedEvil) joined #highaltitude.
[12:07] Laurenceb (n=laurence@host81-154-154-237.range81-154.btcentralplus.com) joined #highaltitude.
[12:07] <Laurenceb> hi
[12:07] <Cooleo> hi
[12:22] RobertB (n=robert@p4FC13E3A.dip.t-dialin.net) joined #highaltitude.
[12:40] Nick change: xatm092|bed -> xatm092
[12:59] xatm092 (i=xatm092@138-38-213-239.resnet.bath.ac.uk) left irc:
[13:11] Cooleo (n=Mike@host86-132-27-37.range86-132.btcentralplus.com) left irc: Read error: 113 (No route to host)