Back
[00:14:42] <rue_house> abcminiuser, you working on a robot?
[00:15:03] <abcminiuser> rue_house, my thesis robot
[00:15:07] <abcminiuser> Uses my bluetooth stack
[00:15:12] <abcminiuser> It goes brum brum
[00:15:23] <abcminiuser> www.fourwalledcubicle.com/ExplorerBot.php
[00:15:28] <rue_house> abcminiuser, would I be right to presume you could use a hand with the hardware end?
[00:15:39] <abcminiuser> Too late :P
[00:15:46] <abcminiuser> I'm writing up the thesis on it now
[00:15:49] <rue_house> oh well, you did a nice job!
[00:16:08] <abcminiuser> It works, but it's not ideal :P
[00:16:10] <rue_house> looks like someone had a cnc at their disposal
[00:16:29] <abcminiuser> The H-Bridge I chose sucks up more power than brittany spears does...nevermind
[00:16:42] <rue_house> L298...
[00:16:45] <rue_house> 2.4V drop
[00:16:47] <abcminiuser> Our EE department got a cheap 3D printer
[00:16:58] <abcminiuser> Yup, it sucks
[00:16:58] <rue_house> aaaah
[00:17:15] <rue_house> there is a fet version you can substitute if your incomming power is more than 12V
[00:17:39] <rue_house> the part number is something like LM186xx
[00:18:19] <rue_house> tracked base looks off the shelf?
[00:18:27] <rue_house> but I havn't seen one like it before
[00:20:06] <rue_house> what are the two atmel modules?
[00:25:10] <abcminiuser> Yes, off the shelf, 6V motors and 7.2V battery
[00:25:15] <abcminiuser> Sensor boards
[00:25:19] <abcminiuser> Atmel Inertial One
[00:25:24] <abcminiuser> And the Atmel Pressure One
[00:26:56] <rue_house> hmm
[00:27:30] <rue_house> I found some good low voltage drivers, but I cant remember what they were rated for, might have been 6V abs max
[00:27:47] <rue_house> I remember I accidently gave one 12V and it did NOT live
[01:44:01] <Casper> so it look like canbus use pullup and pulldown, but only and only when transmitting
[01:44:18] <Casper> but they basically don't talk about it
[01:47:50] <Casper> so it mean I can go with 1 wire and it should be "as good"
[04:21:32] <h4x0r`> hey any Aussies here? need a recommendation on supacheaps Stanley MM
[04:45:03] <h4x0r`> peeps
[04:45:28] <h4x0r`> im grabbing a new avr tommorow, last time i tried i only got through lcd and basic pin functions, and dual usarts
[04:45:48] <h4x0r`> so without many other parts what is the next natural step, to take
[04:46:05] <h4x0r`> or what are some other basic functions i can run through using just a standard avr for example :)
[05:45:08] <inflex> lo there h4x0r`
[05:50:24] <h4x0r`> hey there inflex, hows the electronics coming along bud :)
[05:51:04] <h4x0r`> im just capping off in "fet theory" .. lol :)
[05:51:07] <inflex> h4x0r`: hoping the first PCBs might arrive tomorrow
[05:51:14] <inflex> oh, FET theory eh... which bits?
[05:51:36] <h4x0r`> n/p & c
[05:51:51] * inflex goes looking for a 3pole or 4-pole switch
[05:51:55] <h4x0r`> ahh
[05:51:59] <inflex> shame I can't find a 4-pole relay of compact dimensions
[05:52:06] <inflex> n/p & C ?
[05:52:14] <h4x0r`> n channel, p channel and cmos
[05:52:46] <h4x0r`> it all made sense when i realised cpu's are basically millions of transistors :p
[05:53:02] <inflex> ooooh okay
[05:53:04] <inflex> np
[05:53:21] <h4x0r`> its the drain part that gets me confused
[05:53:39] <inflex> how so?
[05:53:45] <h4x0r`> actually.. now its the gate part lol ok..
[05:54:20] <h4x0r`> so ive got a vcc going into source, then ive got the gate, then the drain is the next part of the circuit which i want on or off?
[05:54:34] <inflex> except Vcc should go in to -drain-
[05:54:46] <inflex> remember, FETs are labelled according to the electron flow model
[05:54:56] <h4x0r`> hmm
[05:55:07] <h4x0r`> yes.
[05:55:15] <h4x0r`> electrons and holes?
[05:55:20] <inflex> the gate basically is sort of like a push-button
[05:55:42] <h4x0r`> yep, if i tie it to ground then then circuit stays off
[05:56:04] <inflex> yep,and the opposite for P-ch
[05:56:13] <h4x0r`> and if i apply voltage then then the current is enabled.. for lack of a better word
[05:56:18] <h4x0r`> hmm
[05:56:34] <inflex> probably easier for now to just ignore P-ch
[05:56:42] <inflex> focus on N-ch, when you get to P-ch things are all opposite
[05:56:50] <h4x0r`> sweet
[05:56:56] <h4x0r`> off to jaycar tomoro :(
[05:57:07] <inflex> oh joy... what MOSFETs are you going to get?
[05:57:08] <h4x0r`> new mega162 = $30
[05:57:11] <h4x0r`> $20*
[05:57:12] <ziph> Going to buy a farting toy? :)
[05:57:15] <inflex> ouch - get it from Element14
[05:57:26] <h4x0r`> yea i prolly should order in lol
[05:57:45] <inflex> $8 from E14
[05:57:55] <inflex> so get 2 or so and get free delivery
[05:58:09] <inflex> oh, $6
[05:58:15] <inflex> ooohj, 162
[05:58:16] <inflex> not 168
[05:58:23] <h4x0r`> twin usarts
[05:58:36] <inflex> still only $12
[05:58:53] <inflex> http://au.element14.com/atmel/atmega162-16pu/ic-8bit-16k-flash-mcu-dip40-162/dp/9171169
[05:58:58] <inflex> Assuming you want DIP?
[05:59:10] <h4x0r`> so vcc goes to source, circuit runs off the drain when the gate is high?
[05:59:26] <inflex> no, Vcc goes to drain
[05:59:27] <h4x0r`> yea dip man, dip 40.. verboardin for the moment
[05:59:31] <h4x0r`> Right .
[05:59:37] <h4x0r`> vcc -> drain, got it.
[05:59:37] <inflex> Vee to source
[05:59:43] <h4x0r`> just readin this wack source
[05:59:48] <h4x0r`> vee?
[05:59:55] <inflex> tie gate to source/gnd and it's off
[05:59:57] <h4x0r`> not familiar with the term
[06:00:03] <inflex> pull it to Vcc and it's on
[06:00:07] <h4x0r`> right
[06:00:11] <inflex> np, you get used to it - jus think of it as the source of electrons
[06:00:19] <h4x0r`> nods.
[06:00:24] <h4x0r`> :)
[06:00:38] * inflex actually has a lot more toruble with BJTs
[06:00:54] <h4x0r`> i recall theyre a little different
[06:01:05] <inflex> quite a bit, yes
[06:01:17] <inflex> at least with FETs you don't have to worry about gate currents
[06:01:24] <inflex> because there aren't any (significant)
[06:01:26] <h4x0r`> orly
[06:01:33] <h4x0r`> wow
[06:01:34] <inflex> it's just a voltage-controlled resistor really
[06:01:39] <h4x0r`> ahh
[06:01:45] <h4x0r`> sweet
[06:01:49] <inflex> (non-linear of course ;) )
[06:02:07] <inflex> but yeah, that's the essence of a MOSFET, it's just a voltage controlled resistor
[06:02:30] <h4x0r`> makes sense
[06:02:32] <inflex> (where's a BJT is more of a current controlled one)
[06:03:19] <inflex> (you know, the more current you pour through the base-emitter junction, the more flows from collector -> emitter
[06:03:34] <ziph> h4x0r`: Reading about the physics (at undergraduate level) is helpful.
[06:03:35] <h4x0r`> of course
[06:04:52] <h4x0r`> i had a $50 book about electronics from supacheap, lost it in a recent move, it explained the entire thing from electrons
[06:05:07] <h4x0r`> amazing book
[06:05:56] <h4x0r`> i think this video was confusing at first
[06:05:59] <h4x0r`> http://www.youtube.com/watch?v=Te5YYVZiOKs&feature=related
[06:06:34] <inflex> I still read the Art of Electronics on a daily basis
[06:06:45] <inflex> good book, well worth having a copy - eliminates the need for so many other books
[06:07:09] <h4x0r`> i might see if jaycar has a copy.. for less than $150 grand
[06:07:10] <h4x0r`> lol
[06:07:16] <inflex> I was lucky, got mine in south africa for $20, as they had the "3rd world country" type print edition
[06:08:33] <inflex> at the time I bought it, I didn't realise exactly what it was - I just noticed that it was written well and was fairly comprehensive
[06:09:43] <h4x0r`> cool
[06:10:02] * inflex needs to make a bigger lightbox... had to photoedit a lot of things on my LOM7 photo today
[06:10:08] <inflex> but the SCC1 mAh counters were easier
[06:10:53] <h4x0r`> lol
[06:11:00] <inflex> I think a box about 1m x 0.5m x 0.5m would be good
[06:11:02] <h4x0r`> has anyone here used the ocsope app for iphone?
[06:11:10] <h4x0r`> coz im thinkin omg.. not worthy.
[06:11:28] <h4x0r`> http://oscilloscopeapp.com/
[06:11:40] <h4x0r`> apprently the input is captured by the mic
[06:14:58] <h4x0r`> lol its on installous, cracked by bykte !
[06:18:36] <h4x0r`> son of a bitch it works :/
[06:56:50] <TwisteR> [13:52:58] <inflex> at least with FETs you don't have to worry about gate currents
[06:56:54] <TwisteR> ...until you don't have to deal with power mosfets with significant gate capacitance ;)
[06:58:42] <TwisteR> fast switching of such a mosfets is a real problem, as you have to supply enough current to charge to the gates... That current can be as big as several amperes...
[06:59:37] <TwisteR> to charge the gates*
[07:02:51] <inflex> I was meaning for the actual biasing :)
[07:10:12] <h4x0r`> caps?
[07:10:33] <h4x0r`> err, nvm..
[07:13:01] <h4x0r`> i presume someone here knows about scopes and waveforms, say i feed a 15v signal through a voltage divider, does the wave stay the same? havent done any wave theory yet
[07:13:27] <h4x0r`> coz theres a few oscope apps for iphone, if i make my own cable i can read waves
[07:13:51] <h4x0r`> but the input jack will max out at an input of 1.7v so i gotta run it through a divider, or a pot
[07:16:36] <TwisteR> yes, wave will stay the same
[07:16:52] <TwisteR> thing you are looking for is called attenuator
[07:17:32] <h4x0r`> hmm
[07:17:42] <h4x0r`> attenuator? what does it do?
[07:17:58] <h4x0r`> i presume it will regulate the voltage just like a voltage divider
[07:19:17] <TwisteR> the input part of the every o-scope, it allows to decrease the signal level with known ratio
[07:20:14] <TwisteR> yes, simple voltage divider is an partial case of attenuator
[07:24:46] <h4x0r`> TwisteR, thanks
[07:24:51] <h4x0r`> i checked them out..
[07:25:02] <h4x0r`> so ive got an iphone that has a nice lil oscope app on it
[07:25:20] <h4x0r`> it reads the wave via the mic input jack, which can only handle 1.7v
[07:25:27] <h4x0r`> so if i want to read a 5v signal
[07:25:34] <h4x0r`> ill need to use an attenuator?
[07:25:45] <h4x0r`> and adjustable one?
[07:25:50] <h4x0r`> coz hmm.
[07:26:43] <h4x0r`> this stuff is soo interesting.
[07:29:06] <TwisteR> be careful and doublecheck your divider before connecting to iphone
[07:30:22] <TwisteR> and don't forget to put some protective Zener diodes to protect input from occasional voltage spikes
[07:30:34] <h4x0r`> good call
[08:58:18] <spow> Hi. I want to make exercises using the A/D converter. I have a variable resistor 0-22k that I plan on putting between 5V and the appropriate pin of the µC. Could the µC get damaged if I turn the resistance down to 0ohm ?
[09:00:29] <ziph> spow: If the uC pin is configured as an input then it is like a very high resistance resistor (e.g. 100's of k or several M's).
[09:00:53] <ziph> spow: If the uC pin is configured as an output then it is like a short to either 0V or 5V depending on the output value.
[09:01:36] <Tom_itx> i would add a small resistor between the wiper and the adc pin
[09:01:42] <Tom_itx> just as a precaution
[09:01:47] <Tom_itx> it should be ok though
[09:01:49] <spow> ziph: since it's to measure the input I guess it's fine the pin is shorted to vcc then
[09:02:14] <vectory> does the a/c convert current or voltage?
[09:02:19] <spow> ziph: actually, many thanks for the information I guess it also explains the tri-state logic
[09:02:22] <ziph> spow: A low value resistor can stop you from burning out the pin.
[09:02:26] <ziph> spow: e.g. 1k
[09:02:29] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/how_to/atmega168/mega168_adc_index.php
[09:02:36] <vectory> coulda looked that up, but i guess the question is simple enough
[09:02:43] <vectory> oh, ok i will read
[09:03:01] <Tom_itx> voltage
[09:03:10] <ziph> spow: Yeah, tri-state is either shorted to 0V, 5V or very high resistance (leakage).
[09:03:50] <vectory> then the varistor doesnt change voltage anyhow is it? you wnt to put the varistor in parallel with the ic i gues
[09:03:53] <vectory> +s
[09:04:22] <vectory> is it common to multiplex many lines on the adc pin?
[09:04:43] <spow> Tom_itx: ziph I found a 1k resistor, I'll add it serially. thanks :)
[12:36:48] <kobsu> Does anybody know how to use gnu mkdir from avr gnu-toolchain in Windows?
[12:37:25] <ftc> perhaps you might want to install sigwin?
[12:37:30] <ftc> er cygwin
[12:37:36] <kobsu> no i dont :)
[12:37:49] <kobsu> i have a virtual machine for that purpose
[12:38:19] <ftc> not sure where else you would get mkdir from though...
[12:38:41] <ftc> or is mkdir also a windows command?
[12:39:02] <ftc> ah yes, i am misremembering my windows commands, disregard me...
[12:39:40] <kobsu> yes, mkdir is also windows command
[12:39:55] <ftc> been a long time since I used a windows command line without cygwin
[12:40:34] <kobsu> if I change dir to gnu-toolchain and command there mkdir I'm able to use gnu-mkdir, but everywhere else it's windows mkdir although PATH is set correctly
[12:40:37] <mrfrenzy> specify path then
[12:40:51] <mrfrenzy> c:\gnu\mkdir.exe mynewfolder
[12:41:30] <kobsu> mrfrenzy: yes but then my Makefile won't work in linux :)
[12:42:17] <ftc> hmm, that sounds messy trying to make a make file cross platform...
[12:42:21] <mrfrenzy> If you make your makefile depend on a customized(Broken) windows environment it won't work in many windows installations
[12:43:03] <ftc> I seem to recall that cygwin allows you to use linux style paths with windows
[12:43:18] <ftc> but again with the customization thing
[13:01:00] <RikusW> !thislog
[13:01:01] <tobbor> This one:
http://rueshouse.dyndns.org:82/~ircjunk/irclogs/html/%23avr-2011-11-13.html
[13:01:25] <RikusW> Forbidden You don't have permission to access /~ircjunk/irclogs/html/#avr-2011-11-13.html on this server.
[13:01:27] <RikusW> ?!
[13:02:22] <RikusW> rue: did you change permissions by accident or on purpose ?
[13:05:40] <rue_house> oooh we had a big reboot, 1 min
[13:07:45] <rue_house> !thslog
[13:07:52] <rue_house> !thslog
[13:07:53] <RikusW> better :)
[13:07:56] <rue_house> aha!
[13:07:59] <rue_house> !thislog
[13:08:00] <tobbor> This one:
http://rueshouse.dyndns.org:82/~ircjunk/irclogs/html/%23avr-2011-11-13.html
[13:08:42] <rue_house> thanks, there are a lot of things I have to fix after a major power outage, I forget some
[13:09:52] <RikusW> how can a reboot cause loss of permissions ?
[13:10:10] <rue_house> loss of mounted filesystem
[13:10:16] <RikusW> ah
[13:10:26] <RikusW> why not put it in /etc/fstab ?
[13:10:40] <rue_house> hmm
[13:12:20] <RikusW> I even put my external hd in there with noauto, so I don't have to type everyting
[13:12:27] <RikusW> +h
[13:12:42] <RikusW> so how many files are there in html/ ? 3000 ?
[13:29:32] <RikusW> doublebeta: get old broken motherboard and PCI cards and practice some smd soldering on that
[13:35:08] <rue_house> oh sorry
[13:35:20] <rue_house> I can have them counted
[13:36:16] <rue_house> about 11000
[13:36:38] <rue_house> RikusW, ^
[13:36:46] <Tom_itx> 11k what?
[13:37:25] <rue_house> files
[13:53:58] <RikusW> thats a lot :O
[13:54:51] <vectory> idd
[13:55:06] <vectory> wonder whos gonna read all of them
[13:55:35] <vectory> but its sure noce to know they are there for when you just might need em
[13:55:47] <vectory> typically this is after you decide to delete it
[14:06:18] <RikusW> This is weird, I actually saw -> (22:01:46) RikusW left the room (quit: Ping timeout: 252 seconds).
[14:06:23] <RikusW> how on earth ?!
[14:06:45] <Tom_itx> you have been cloned
[14:07:53] <Casper> RikusW: probably that the uplink was broken but not the downlink and that the ircd have a small bug
[14:08:04] <Casper> that sent you a last packet before closing the connection
[14:08:08] <RikusW> heh
[14:39:49] * Casper goes datasheet hunting on digikey
[14:40:50] <Steffanx> Have fun
[14:43:02] <Casper> yeah
[14:43:09] <Casper> it's driving me crazy
[14:43:20] <Casper> it's a simple thing, but I can't find what I seek...
[14:43:34] <Casper> can bus... how does the transceiver work internally?!?
[14:43:58] <Casper> all what I find show completly incomplete one
[14:44:36] <Casper> the "best" I found so far is: open collector style but no pullup (and reverse for the canh)
[14:44:47] <Casper> in other words, it just can not work
[14:47:02] <Casper> I'm starting to wonder if the waveform that we see everywhere are wrong... another datasheet with only the open collector and nothing to bring the line high
[14:50:00] <Casper> TI: open collector
[14:50:38] <Casper> NXP: open collector with 2x 25k to a 1/2vcc suply
[14:51:34] <Casper> microchip: is the same but no value
[14:51:36] <Casper> so...
[14:51:47] <Casper> it mean it's not true differential
[14:52:18] <vectory> maybe you can find some scientific papers on it
[14:52:21] <vectory> somewhere
[14:52:40] <Casper> can bus documentation isn't great, seriously...
[14:52:51] <Casper> it's not something that hobyist would regularly use
[14:53:28] <Casper> I can see that it's slightly better than 1 wire, but not by much
[15:09:46] <TwisteR> Casper: you can ask at #open-bldc here at freenode, CAN bus is their favorite bus, IIRC
[15:26:59] <Casper> TwisteR: asking
[15:29:14] <TwisteR> Casper: wait a while until someone (esden, probably) will answer
[18:49:40] <abcminiuser> \me Can't tell if netsplit, or everyone's just quiet today
[18:49:56] <Tom_itx> quiet
[18:50:00] * abcminiuser rather
[18:50:12] <Tom_itx> make some noise
[18:50:26] * Valen made some noise
[18:50:30] <Valen> but then everybody left
[18:51:11] <Tom_itx> i want a harmonic drive but can't find one in the budget
[18:51:22] <Valen> want something more affordable ;->
[18:51:43] <Valen> where would i find some HL1606 IC's?
[18:51:51] <Tom_itx> what is it?
[18:52:21] <Tom_itx> led driver?
[18:52:27] <Valen> for a sane price lol
[18:52:29] <Valen> yeah
[18:52:33] <Valen> http://www.ebay.com.au/sch/i.html?_from=R40&_trksid=p3907.m570.l1313&_nkw=HL1606&_sacat=See-All-Categories
[18:53:08] * abcminiuser is CONVERSATION CATALYST MAN
[18:53:47] <Valen> idunno you could be considered consumed by the conversation
[18:53:58] <Valen> thus your a reagent not a CATALYST
[18:54:09] <Valen> caps cos i couldnt be assed working out how to spell it
[18:55:05] <abcminiuser> Aww
[18:57:27] <Landon> CONVERSATION REAGENT MAN sounds like someone who wants to sell me secondhand conversations about the weather
[18:57:56] <Valen> if your not part of the solution your part of the precipitate
[18:58:37] <grummund> lol. /me adds that one to the quotes list
[18:58:38] <inflex> hey Valen
[18:58:52] <inflex> Valen: your glass/paper photo-setup worked nicely for me
[19:00:02] <inflex> http://nqrc.com/images/PLD-LOM7.jpg and
http://nqrc.com/images/PLD-SCC1-minv.jpg were done with it
[19:01:49] <Valen> glass paper photo?
[19:07:13] <vectory> lcd in the second pic looks rather neat
[19:08:01] <Landon> those are impresssive pictures
[19:08:15] <Valen> does, i'm just not sure what your thanking me for lol
[19:13:56] <inflex> Valen: I thought you told me you took photos of a board you did using a light box with paper under a sheet of glass
[19:14:00] <inflex> hmmm... wonder who it was then
[19:15:58] <inflex> maybe it was ziph ... or some other aussie
[19:16:53] * inflex starts the long search through his logs
[19:16:56] <inflex> (using grep)
[19:20:19] <inflex> so sad that I have almost a gigabyte of chat logs
[19:20:45] * Casper wonders how much he have...
[19:20:45] <Tom_itx> not much mention of a light box
[19:20:50] <Landon> 744M irclogs/
[19:20:56] <Landon> you win :(
[19:21:08] <Casper> inflex: 1.9G of log chat here!
[19:21:20] <inflex> damn, Casper wins
[19:21:44] <Tom_itx> some of mine go back to 1999
[19:21:50] <inflex> Aha! it was from Ziph!
[19:22:00] <inflex> <ziph> Nup, just a single flash. The board was sitting on a pane of glass with A4 paper underneath and the flash was pointed towards the large white ceiling. :)
[19:22:07] <inflex> but I see now it was a PM
[19:22:29] <inflex> Anyhow, I was really pleased with the results
[19:22:45] <inflex> Sure there was some photo-editing required, but it was still very quick
[19:23:11] <inflex> the hardest was with the LOM7 picture, because of the casing being beige and the white backgruond - but GIMP's intelligent tracer worked fine
[19:23:19] <Tom_itx> use white paper on a couple sides and it will help eliminate the shadows
[19:23:40] <inflex> Tom_itx: it was in a white styrofoam box
[19:24:15] <inflex> I put the box on its side, paper and glass on the bottom (side), then just using the built-in flash on the camera
[19:27:04] <inflex> btw Valen - redesigning the dual-lipo part of my system again :
[19:27:08] <inflex> (for the headphone amp)
[19:47:31] <h4x0r`> hey, when using filtering caps on xtal, does the voltage rating matter?
[19:47:39] <h4x0r`> I can only find 200v 22pf caps lol
[19:51:53] <inflex> O_o
[19:52:19] <inflex> the voltage of the xtal would only be at most 2x the supply... I think
[19:52:25] <inflex> so, in short, no :)
[19:53:55] <rue_shop> yea, dont use caps on your crystal rated for less than 5V
[19:53:57] <LikWidCHH> inflex whats that little circut you made?
[19:54:21] <rue_shop> if you do, they will explode and fly off your pcb
[19:54:23] <rue_shop> :)
[19:54:25] <inflex> LikWidCHH: what? which one?
[19:54:36] * inflex has gazillions of circuits :(
[19:54:42] <LikWidCHH> lcd with red/black wires comming out of it
[19:54:52] <LikWidCHH> that like a battery draw gauge or something?
[19:54:58] <LikWidCHH> current..
[19:55:42] <vectory_> 2nd of the two posted photos
[20:01:12] <inflex> LikWidCHH: aaah yes, that's a commercial product I'm mkaing
[20:01:40] <inflex> LikWidCHH: it's a mAh counter, peak current tracker, voltage tracker and run-time tracker
[20:01:46] <inflex> LikWidCHH: used for model aircraft
[20:04:10] <LikWidCHH> Oh ok, so im guessing it weighs 10-20 pounds? :)
[20:04:49] <inflex> hah... the unit weighs about 15g
[20:05:01] <LikWidCHH> I was like 2-3 oz off..
[20:05:15] <LikWidCHH> thats pretty cool looking
[20:05:19] <inflex> http://nqrc.com/?vp=PLD-SCC1
[20:05:59] <LikWidCHH> funny stuff! is that literally about 5 dollers worth of components ? :)
[20:06:17] <inflex> no, quite a bit more actually
[20:06:36] <LikWidCHH> whats on the back side of it?
[20:07:26] <inflex> there's several items on the board which push the price up quite a bit
[20:08:47] <inflex> and of course I'm making a profit - since I run a business, not a charity.
[20:09:25] <inflex> there's also the painstaking hours of firmware development and spawning of genius ideas to get around limitations
[20:12:04] <Landon> inflex: what's the going rate on genius ideas?
[20:12:22] <LikWidCHH> inflex do you do lots of airplane stuff? or RC stuff in general?
[20:12:33] <abcminiuser> Hrmph, is a 30nS short a long period of time?
[20:12:40] <abcminiuser> Relatively speaking, that is
[20:13:19] <LikWidCHH> to the speed of light its a long time, relative to a human its not measureable... I hope ive answered your question and or made it worse
[20:13:47] <inflex> Landon: a billion dollars per idea! mwhwhahahaaa
[20:14:17] <inflex> LikWidCHH: R/C in general - though I do a lot of projects outside of R/C as well, just that R/C was (until recently) a fairly good market.
[20:15:14] <abcminiuser> LikWidCHH, *sigh*
[20:15:29] <LikWidCHH> well i have this idea and since you do RC stuff it peaked my interest... basicly what the idea is to make a product that converts rc pwm into like i2c/spi/serial out for robotics
[20:15:45] <LikWidCHH> abcminiuser its terriably fast amount of time
[20:16:33] <abcminiuser> Indeed - I had an issue with my robot, where switching the direction of the motor would cause high current spikes
[20:16:53] <abcminiuser> I'm trying to figure out if the 30ns short that occurs during the switch over would cause it
[20:17:30] <LikWidCHH> well its the magnetic field thats causing that
[20:18:06] <LikWidCHH> atleast thats my guess
[20:18:44] <inflex> LikWidCHH: not a difficult task, you should be able to do that purely with bitbashing
[20:19:18] <inflex> LikWidCHH: I'd go SPI/serial though, I2C would be a bit painful for the polling part as an I2C slave on smaller AVRs
[20:19:28] <inflex> but SPI/serial on something as tiny as the T10 would work
[20:19:39] <LikWidCHH> well I have zero skill in C
[20:19:48] <inflex> well, no time like now to learn :)
[20:20:09] <LikWidCHH> yeah thats kinda what im thinking, the idea would be to make some device that would support 7 rc channels
[20:20:27] <LikWidCHH> but if the chip was large enough as many channels as you throw at it really
[20:20:57] <inflex> can get tricky with multiple channels, because modern RXs can be .... unpredictable... in their scheduling of the servo pulses out the pins
[20:21:16] <inflex> in the "old" days, the pulses came out sequentially, eg, you'd never get two pulses out of the RX concurrently
[20:21:21] <LikWidCHH> well I wouldnt say its unpredictable... I have an oscilliscope and have measured it
[20:21:36] <inflex> but with new RXs some put them out sequentially, some concurrently...
[20:21:48] <inflex> some are split so you have sequential and concurrent
[20:22:02] <LikWidCHH> yeah I have a reciever that channels 6 and 7 overlap
[20:22:10] <LikWidCHH> one sec
[20:22:20] <inflex> that makes it a major PITA to deal with
[20:22:49] <LikWidCHH> http://www.lynxmotion.net/download/file.php?id=2418&mode=view
[20:23:25] <LikWidCHH> thats what my reciever does
[20:23:53] <LikWidCHH> so two passes would be able to capture the 7 channels worth of data which really isnt a big deal if you think about it
[20:24:10] <inflex> none of those overlap
[20:24:17] <inflex> that's a purely sequential output
[20:24:30] <inflex> just that 6 & 7 come out before 1.
[20:24:59] <LikWidCHH> http://www.lynxmotion.net/download/file.php?id=2417
[20:25:12] <LikWidCHH> thats funny! this guys looks like it overlaps
[20:25:28] <inflex> okay, now -that- one does
[20:25:51] <inflex> anyhow, I'll leave that to you to work out.
[20:25:54] <LikWidCHH> wonder which one is right the first photo I did, the second was with a scalie logic kit..
[20:26:03] <LikWidCHH> not mine.
[20:26:58] <LikWidCHH> Anyway... Im wondering if a product already exists do to what I want? I mean people use rc stuff for those quad copters
[20:28:06] <inflex> check with the DIY-Electronics section on RCGroups
[20:31:51] <LikWidCHH> but the general idea with reading pwm is starting a timer when the signal goes high and stopping it when the signal drops, doing some division and spitting out a value right?
[20:33:40] <inflex> yes, use a 1 or 8MHz clock and you'll have a neat value
[20:35:54] <LikWidCHH> if I used an 8mhz clock then 1mS would give me a value of 8000 correct? depending on how stable the clock is right.... ovbiously if im looking for a range of lets say 0-500 doing division will clean that irratic clock out if there is one right?
[20:37:15] <inflex> at 8MHz, just take the clock count and divide by 8 (which is fast and easy), that'll give you the time in microseconds
[20:37:23] <inflex> use an xtal to drive the uC
[20:37:29] <inflex> and your time will be essentially perfect
[20:37:53] <LikWidCHH> do I need to have an external crystal?
[20:38:03] <inflex> if you don't then you'll be +/-10% of the value
[20:38:12] <inflex> if you're dealing with servo timings then you should use an xtal/resonator
[20:38:15] <LikWidCHH> Oh i see! depending on the tide or moon phase right?
[20:39:10] <LikWidCHH> so in theory the circuit should really be two devices, the avr and a xtal
[20:40:16] <inflex> you're goingto have a lot more parts than just those two
[20:40:29] <inflex> maybe go get yourself an arduino kit
[20:40:49] <LikWidCHH> oh I was thinking that I could just use a single raw avr chip
[20:41:08] <inflex> hahahahaaaa no.
[20:41:18] <LikWidCHH> well one sec
[20:42:01] <Casper> inflex: btw, can bus isn't trully differential. basically canl is open collector, canh is... open emitter? both have a resistor of 25k to a VCC/2 reference
[20:42:06] <LikWidCHH> http://www.lynxmotion.net/viewtopic.php?f=8&t=7633 4th photo down
[20:42:18] <Casper> so both line never cross the other voltage
[20:42:25] <Casper> more like mirrored
[20:43:14] <inflex> LikWidCHH: they're cheating in various ways - it's a hackjob that won't be reliable
[20:43:18] <LikWidCHH> inflex that link above is the thing I made from this site
http://jreise.de/PPM/R617FS.html google translate if your curious about it
[20:43:54] <LikWidCHH> ok so when you say not reliable are you speaking the output can fluxuate +- 10% ? from your comment eairler
[20:44:23] <Casper> internal RC is +/-10% from nominal frequency
[20:44:40] <Casper> it can vary from chip to chip, voltage, temperature and aging
[20:45:09] <Casper> it is also influenced by electrical noise, due to them being varying voltage
[20:45:30] <Casper> xtal is easy to use, one xtal + 2 capacitors
[20:45:41] <LikWidCHH> so that 1.5ms pulsewidth signal can jump around quite a bit? Honestly with as small as my robot thingy is I don't think its going to cause a problem at the scale its at.. now if it were 10 times bigger I can see for that level of accuracy
[20:45:45] <inflex> or a resonator, only 3 pins (X1, X2, GND)
[20:46:04] <inflex> LikWidCHH: you'd be surprised, a 10% drift will make a notable difference
[20:46:14] <inflex> even a 0.5% drift causes problems with servos
[20:46:25] <Casper> 1.5ms could be 1.35
[20:46:36] <Casper> as it could be 1.65
[20:46:37] <LikWidCHH> i see! well I'll keep that in mind about the xtal / resonator
[20:46:47] <inflex> you can try that simple circuit - but believe me, while it may "work", it won't be reliable.
[20:46:50] <LikWidCHH> i have never seen it be that off
[20:47:08] <LikWidCHH> inflex well it does work, you can see my oscilliscope post below
[20:47:16] <Casper> btw, if you use a crystal, take some extra time to select the right frequency
[20:47:28] <Landon> can you tune AVR clocks?
[20:47:49] <Casper> for example, 16MHz is a very bad choice if you do serial data transfert, like with uart
[20:47:49] * Landon remembers being able to do that with the freescale chips in uc class, but hasn't pursued it with AVR projects
[20:47:52] <Casper> Landon: yes and no
[20:47:54] <LikWidCHH> honestly if I could write code to make an led blink right now Id be happy
[20:47:54] <inflex> *sigh* I'm not going to waste time on this. LikWidCHH go ahead, build it try it - no harm there - but I've been down this path a thousand ttimes in R/C - I'm only trying to give you solid advice based on experience - but it would seem yet again, people think they can avoid it.
[20:48:10] <Casper> Landon: you can tune, but you need to tune each chip, and you still have jitters and all
[20:48:21] <inflex> and temp drift
[20:48:23] <LikWidCHH> inflex I'm not arguing with you at all!! this is new stuff for me
[20:48:30] <Landon> so yeah, that's what I Was thinking it might be
[20:49:01] <LikWidCHH> I just want to know why / how / good bad habbits exct... looking at the signal with my dsoquad it looks pretty noisy
[20:49:27] <Casper> Landon: you can tune up I think -50% to +100%, in non-linear way, with an 8 bits setting... so still quite a big change for each increment
[20:49:49] <ripthejacker> Hey got recommended to come here for project help..anyone will to help :(
[20:49:51] <Casper> LikWidCHH: basically, if you need precise timing, use a crystal (of apropriate speed)
[20:50:00] <LikWidCHH> what do you guys usually code your avr with? c or something else
[20:50:11] <Casper> if it do not need precision, then internal clock work
[20:50:14] <Casper> in C
[20:50:25] <Casper> you could go with ASM, but I don't recommand it
[20:50:39] <Casper> and AVOID TO USE THE STUPID ARDUINO LIBRARY
[20:50:48] <Landon> heh
[20:50:48] <Casper> however you can use their boards
[20:50:50] <abcminiuser> "Avoid to use"?
[20:50:52] <ripthejacker> lol
[20:50:54] <LikWidCHH> good deal casper, as for output format is there one thats least stressful ? honestly since rc is concerned since it runs at 50hz would I be able to get that many updates per second?
[20:51:02] <Landon> please avoid to use the stupid arduino library, abcminiuser
[20:51:13] <abcminiuser> Sure thing Landon :P
[20:51:18] <abcminiuser> I shall do the needful in this case
[20:51:27] <Landon> much appreciations
[20:51:33] <inflex> Is very much happy fish programming, yes.
[20:51:34] <Casper> LikWidCHH: ???
[20:51:41] <Casper> the servo is 50Hz, not the avr
[20:51:54] <LikWidCHH> yeah.... one sec ill explain
[20:52:14] <Casper> ideally, you would like to have one crystal value that allow your timer to run at 50Hz
[20:52:58] <inflex> thing is, TX's don't stick with 50Hz
[20:53:05] <inflex> some are slower (like JR), some are faster
[20:53:07] <ripthejacker> what would be the best approach to output voltage waveforms from an attiny/atmega in real time?
[20:53:09] <LikWidCHH> since the reciever puts out 50updates per second, would I be able to get that many reads on all 7 channels and then have the avr output all 7 values on a single pin, in "XYZ" fromat? someone said spi/serial for that... honestly serial would be the eaysist for me to interpert on the next device that needs to read that
[20:53:17] <inflex> depends on how many channels they're trying to cram in
[20:54:01] <Casper> ripthejacker: what frequency and waveform wanted?
[20:54:04] <abcminiuser> "Do the needful" sounds like an awesome dance
[20:54:31] <inflex> abcminiuser: probably a lot of hip-thrusting in that one
[20:54:45] * Landon was thinking similar
[20:55:32] <abcminiuser> I'm TOTALLY starting that craze then
[20:56:00] <ripthejacker> well basically im trying ot mak ea circuit break. make microcontroller constant check voltage and current. and at anytime i want to see the voltage/curren toutput, be able to see it. frequency i guess 4hz ( voltage point plotted at every 1/4 of a second)
[20:56:05] <inflex> Do the Bender... oh, yeah... oh... yeah...
[20:56:32] <LikWidCHH> ripthejacker they have a premade circuit for that
[20:56:57] <ripthejacker> lol i know...but i woul dlike to make my own from scratch, for a school proj as well
[20:57:06] <ripthejacker> goign to biuld my own SMPS aswell :)
[20:57:16] <ripthejacker> *build
[20:57:18] <inflex> JFC.... can anyone type in here?
[20:57:19] <LikWidCHH> not sure what that is..
[20:57:21] <inflex> properly!
[20:57:37] <LikWidCHH> spell check doesnt work on irc, it only works on pidgin and firefox and chrome
[20:57:43] <inflex> gawd, it's like I'm in #electronics all over again
[20:57:49] <inflex> IRC spell check works fine in Xchat
[20:57:53] <ripthejacker> switch mode power supply but yeah, trouble deciding between attiny and atmega
[20:58:05] <LikWidCHH> hah
[21:00:10] <LikWidCHH> Inflex thanks for your help... I'm going to get my learn on
[21:06:39] <abcminiuser> &^*&^%$^%#^%$#@$#@$#!@$%
[21:06:53] <abcminiuser> The &^%$&^%$^&$#^%$ing cat just ran off the tiles onto the carpet and yacked up everywhere
[21:07:23] <Tom_itx> gotta love pets
[21:09:23] <inflex> yep, probably hairballs at this time of the year
[21:11:18] <abcminiuser> But running off the nice tiles and onto the carpet?
[21:11:20] <abcminiuser> Damn kitty
[21:12:05] <inflex> it's absorbent - so it doesn't run all over the place :D
[21:12:18] <inflex> oh ffs... I wish my replacement PCBs for my LOM would arrive
[21:12:40] <inflex> until they do I feel like I can't get out there and push/promote the LOM for more sales
[21:13:01] <inflex> It's a bit silly really, since the thermal drift right now is no worse than 0.05%... but it still burns me
[21:13:12] <inflex> nothing must move!
[21:13:37] * inflex suspects he's going to end up using the 4-wire TO220 10R resistors just so he can ignore the thermals with impunity
[21:13:47] <Tom_itx> inflex, is your hobby site still open?
[21:13:52] <inflex> Tom_itx: yes
[21:15:58] <Casper> ripthejacker: smps may happend soon to me too, planning some big projects, will need to make 3 smps... that will be annoying :D
[21:18:35] <ripthejacker> Casper : lol o wow. haha lucky me im only doing one :p .
[21:25:19] <Casper> ripthejacker: I will need to do a 19->13.65V input current limited to 3.4A. then I need to do a 10-16V to 48V 2A, then some 40-55 to 5V 2A :D
[21:28:06] <ripthejacker> Casper : lol nice. I have to do 120 to 15V/5V.
[21:28:31] <Casper> yuck... better have it isolated, and I hope it's not your first one
[21:28:56] <ripthejacker> Definetly will be isolated :p and it is my first time but theres a team of us.
[21:30:35] <Casper> good luck, better wear safety google when you power it up
[21:36:28] <vectory> Casper: have you found infos on the can bus?
[21:39:26] <vectory> no?
[21:39:42] <vectory> thats how i feel all the time with electronics -_-
[21:43:16] <inflex> Casper: CAN bus sounds odd
[21:44:09] <inflex> wonder what design criterion made them go the way they did, rather than things like RS422 or such
[21:51:11] <Casper> multi master with collision avoidance
[21:51:29] <Casper> they can't use push pull driver, because that would corrupt the data of the other master
[21:51:44] <Casper> so what they did is: the device with the lowest ID have priority
[21:51:54] <Casper> and the device ID is the first thing that get sent
[21:52:03] <Casper> bit 0 get priority
[21:52:10] <Casper> since it's basically open collector
[21:52:36] <Casper> but what I wonder is why they didn't used a pullup to VCC and pulldown to gnd instead of VCC/2
[22:07:17] <inflex> should have just made it ethernet
[22:10:43] <karlp> it's been around about as long...
[22:10:50] <karlp> can bus is what, early-mid 80s?
[22:12:31] <Valen> inflex: where are you going to sell your loms?
[22:12:44] <vectory> karlp: ye
[22:13:05] <vectory> the beginnings
[22:13:05] <Valen> ethernet is non deterministic
[22:13:17] <Valen> you want determanistic stuff running your ABS ;->
[22:18:41] <Casper> I wonder what would happend if you get in a collision at 200mph and the abs deploy... would it make it worse? would it trigger after you hit the dash? or don't deploy? :D
[22:19:27] <Valen> abs = anti brake skidding
[22:19:36] <Valen> stops the wheels from locking up
[22:21:14] <Casper> err
[22:21:17] <Casper> I mean airbag
[22:21:23] * Casper thinks he's tired
[22:23:20] <Casper> sure thing is... the acid is well stired in the batts
[22:23:29] <Casper> time to wait until it stop to gas :D
[22:33:48] <Valen> if the airbag goes off and you have a body part in its deployment space, that body part will rapidly leave that space
[22:34:46] <inflex> Valen: not sure precisely - but they'll be availalbe on my site of course
[22:34:57] <inflex> Valen: but I'm looking to push them at the EV people... anyone dealing with 100A+
[22:37:08] <ziph> abcminiuser: 30ns is fairly slow if you're talking about rise times.
[22:38:59] <abcminiuser> Yeah, I was wondering about it
[22:39:34] <abcminiuser> I need a good 5ms dead time to prevent system reboots however, never figured out why
[22:39:45] <abcminiuser> Err, USB bus reboots, I mean
[22:40:02] <ziph> Before switching directions on the motors?
[22:41:55] <abcminiuser> Yup
[22:42:02] <abcminiuser> In the driver, I check if the direction is changing
[22:42:20] <abcminiuser> If it is, I disable PWM, wait 5ms, change the direction and re-enable PWM
[22:42:40] <abcminiuser> https://code.google.com/p/bluetooth-explorerbot/source/browse/trunk/Code/Drivers/Motors.c
[22:45:12] <inflex> ziph: see I thanked you for your help with the photography?"
[22:45:29] <ziph> inflex: Yeap. What did you have to edit out?
[22:45:42] <inflex> ziph: just the various non-white bits :)
[22:45:58] <inflex> but it was quick enough using the GIMP outline tracer
[22:46:24] <inflex> my glass was less than clean and the foam-box was quite dirty too
[22:46:42] <ziph> With small things and a flash you can aim you can get the background to +3 stop white while still exposing the object well. Probably far more difficult with that big box though. :)
[22:46:44] <inflex> also the glass was double-thick (2x3mm) which made it a bit different again
[22:47:02] <inflex> ziph: fixed "flip up" flash on this thing
[22:47:15] <inflex> ziph: Fujifilm S5700
[22:47:35] <ziph> Yeah, that's excellent for a flip up.
[22:48:19] <inflex> http://nqrc.com/images/PLD-LOM7.jpg <=- obviously I put in the fake blue blend on that
[22:50:05] <inflex> Anyhow, it works well for now,and for 99.9% of the things I make/sell, it's perfectly good
[22:50:18] <inflex> the LOM7 is the largest physical thing I've had to photograph
[22:51:08] <ziph> I did a house recently.
[22:51:13] <ziph> Couldn't get glass under that. ;)
[22:51:40] <ziph> It was a pain though, I had to take some photographs at 5 AM and others at 6 PM. :)
[22:51:42] <inflex> true... but at least you have sunshine... too bad it moves
[22:51:49] <inflex> yeah, golden-hours
[22:52:04] <ziph> Well, 5 AM was just to get some light on the front of the house.
[22:52:57] <ziph> And it has brown brick walls and no external lighting, so it is fairly close to being the proverbial black bear in a black cave. :)
[22:53:05] <ziph> (dark cave)
[22:53:49] <inflex> hah, glad it's you,not me.
[22:54:00] <inflex> time, for me to go and do my postal run for the day - talk to you all later.
[22:59:48] <inflex> ziph: anyhow, thanks for the tips
[22:59:59] <inflex> I feel the pictures are a lot more "pro
[23:03:09] <ziph> inflex:
http://www.amazon.com/Light-Science-Introduction-Photographic-Lighting/dp/0240802756 is an excellent book on photographing different materials.
[23:04:10] <ziph> inflex: It's completely different from the usual book-full-of-pictures-with-random-tips that usually get published. :)
[23:11:18] <doublebeta> rue_shop: please don't tell me you mounted a non-UNIX filesystem on a server?
[23:15:09] <Landon> doublebeta: you aren't living until you run linux off of an NTFS filesystem
[23:16:26] <doublebeta> Nooo ty
[23:16:36] <doublebeta> You aren't living AFTER doing that
[23:36:08] <inflex> hah
[23:42:59] <inflex> ziph: I think I'll need to sell a lot more to buy more books :(
[23:43:12] <ziph> inflex: :)
[23:43:25] <ziph> inflex: Been ages since I've been able to buy books too.
[23:44:48] <inflex> last book I bought was a jQuery pocket guide book by OReilly.
[23:45:01] <ziph> Did you end up writing any jQuery?
[23:45:12] <inflex> yes,I have written some, but there's a lot more to come
[23:45:43] <inflex> if nothing else jQuery makes a very nice / easy layer to access parts of your page
[23:46:14] <ziph> inflex: Did I ever show you the store site I did?
[23:46:52] <inflex> not sure - try me and it might jog my memory
[23:46:55] <ziph> inflex: The one where you can buy soap bars via a jQuery set up? :)
[23:47:44] <inflex> ziph: nice clean site... no pun intended
[23:47:58] <ziph> Heh. :)
[23:48:17] <inflex> interesting way indeed to do the quantities
[23:48:22] <inflex> saves people typign
[23:48:26] <ziph> Yup.
[23:48:40] <ziph> Plus you can play with the quantities and see how it changes the total price easily.
[23:48:46] <inflex> yep
[23:48:49] <inflex> nice
[23:48:54] <inflex> didyou invent that?
[23:48:57] <ziph> That gets registered in the database immediately too, so you can see people changing things as they go. :)
[23:49:41] <inflex> that does remind me - I need to update the NQRC store a bit, add some more 'features'
[23:50:01] <inflex> but as per usual with sites that "organically" grow, it's getting a bit twisted and ugly behind the folliage
[23:50:54] <ziph> I was going to suggest you remove the pages for items you no longer stock, but it looks like you did that recently. :)
[23:51:50] <inflex> yep
[23:52:02] <inflex> the model kits will be filled up again soon
[23:52:04] <inflex> have a box of them coming over
[23:54:11] <ziph> Are there any local markets near you?
[23:55:20] <inflex> There are, yes
[23:55:51] <inflex> Though the wrong sort of market for what I want - those people are covered with the local toy-store
