Back
[03:56:57] <pc_magas> goodmorning
[03:57:52] <TeknoJuce> http://3.bp.blogspot.com/_zqABT3suzXE/S_EwMNYzDhI/AAAAAAAAAW4/MHHY0U-xAhM/s1600/SD_M32_RTC.JPG
[03:58:42] <TeknoJuce> The 3.6v diodes there would you put the black line on the diodes connected to ground or to the lines
[04:02:08] <TeknoJuce> Guess the cathode should be connected to the lines
[04:02:16] <TeknoJuce> http://www.kpsec.freeuk.com/components/diode.htm
[04:04:58] <TeknoJuce> wonder why they chose to use PB1 instead of PB4 like everyone else
[04:06:28] <Kevin`> TeknoJuce: you should be limiting the current from the microcontroller to those diodes
[04:06:59] <Kevin`> why not just run the whole thing on 3.3v though?
[04:08:15] <TeknoJuce> I have 30 3.6 zener diodes I bought might as well use them for something :D
[04:09:30] <TeknoJuce> maybe the pololu stepper drivers run of 3.3v I didnt look
[04:11:29] <TeknoJuce> guess they do
http://www.pololu.com/catalog/product/1182
[04:21:39] <TeknoJuce> Kevin it says this: Schematic for ATmega8 is shown here (updated on 10 May 2010, SD series resistors are removed, as they were limiting the speed of SPI bus. 51k pullups are added on CMD/DAT lines. This gives better stability with different cards. Also, two 3.6v zeners are added to protect SD in case when the ISP programmer voltage levels are of 5v. these diodes are not required if your programmer has settings for 3.3v output)
[04:22:02] <ElMonkey> TeknoJuce, what are you building?
[04:22:12] <Kevin`> TeknoJuce: you have the avr'
[04:22:19] <Kevin`> TeknoJuce: you have the avr's power supply connected to 5v
[04:22:37] <TeknoJuce> Silly drawbot for something to do
[04:28:08] <TeknoJuce> Kevin I think he did that because he was running the max232 @ 5v as well but then he goes on to say that the max232 may operate at 3.3v just fine so no idea
[04:28:39] <elektrinis> 3232 can
[04:28:39] <TeknoJuce> what are you making ElMonkey
[04:28:40] <Kevin`> TeknoJuce: there are equivalents to the max232 that are designed to run on 3.3v or lower
[04:28:44] <elektrinis> 232 cant
[04:29:40] <TeknoJuce> ok then it makes more sense
[04:32:19] <TeknoJuce> these pololu drivers are driving me up the wall, guess I should buy stepper motors instead of savaging them.
[04:32:48] <TeknoJuce> I think they are stable and then hours latter they start jittering and then the driver fails
[04:37:05] <TeknoJuce> thing that erks me is that they are running ice cold yet still seem to trigger some kind of fail safe in them
[04:42:06] <ElMonkey> supply voltage ok?
[04:44:34] <TeknoJuce> 12v from psu seems stable even while it's going through its jitter death state
[04:46:18] <TeknoJuce> and then I have to leave it unplugged for some time and then it will finally start working again
[04:46:36] <ElMonkey> unplugged from everything?
[04:46:48] <TeknoJuce> I just unplug the psu
[04:47:00] <ElMonkey> ok
[04:47:18] <ElMonkey> are you sure the jitter isn't caused by bad input on the step/dir lines?
[04:47:43] <ElMonkey> any inputs left floating?
[04:48:14] <TeknoJuce> well I have two drivers plugged in one runs flawless but when the other starts to jitter it kills off the other stepper with it
[04:48:36] <TeknoJuce> so I cranked the motor that works better and it started going again
[04:48:57] <TeknoJuce> but the one that started to jitter wont until its unplugged for sometime
[04:49:34] <TeknoJuce> I changed the driver out to a different one and it did the same thing
[04:50:41] <TeknoJuce> I see data flowing fine on the scope from the uC even after it kanks out
[04:51:13] <ElMonkey> did you look at the power supply line with the scope, too?
[04:51:32] <TeknoJuce> when I changed up to the different second driver it now fails but the second driver will keep going with out me having to spin the shaft of it
[04:52:10] <ElMonkey> my hunch says supply noise issue
[04:52:13] <TeknoJuce> Yes, but what should I be looking for a drop in power??
[04:52:26] <TeknoJuce> I change the psu as well and this one acted the same as well
[04:52:28] <ElMonkey> tried putting a cap at the driver board's supply pins?
[04:52:58] <ElMonkey> the psu might be fine, noise might come from outside
[04:53:26] <TeknoJuce> have two caps one on the 5v 10uf and 12v 47uf as recommended by pololu
[04:53:38] <ElMonkey> hmm
[04:53:45] <ElMonkey> strange
[04:54:06] <ElMonkey> any high current lines running along logic?
[04:54:18] <TeknoJuce> really would like to know why one works totally fine
[04:55:46] <ElMonkey> you have other motors to test with?
[04:56:00] <TeknoJuce> I didnt want to try a forth encase this is damaging them some home
[04:56:04] <TeknoJuce> *how
[04:57:50] <TeknoJuce> I have 4 steppers 2 larger ones that I was first getting this symptom with nema 23 1.5amp 8ohm 12v
[04:58:12] <TeknoJuce> so I switched up to these smaller nema 17 ones that seemed to be more stable
[04:58:29] <TeknoJuce> thought my problems were solved then they started acting the same way as well
[04:59:05] <TeknoJuce> also swapped the motor from the driver that seems to be working well to the other one and it did the same thing.
[05:00:41] <ElMonkey> so its one particular driver board that fails?
[05:00:50] <ElMonkey> i am confused :)
[05:01:41] <TeknoJuce> i thought it was the board so I switched it to a 3rd driver and it thought it was fine then about 2 hours later had the same issue as the 2nd driver
[05:02:15] <TeknoJuce> but the 3rd driver this time wouldnt kill off the first driver that was working find like the second one did
[05:02:43] <TeknoJuce> I didnt want to try a forth driver encase something is damaging them
[05:04:11] <TeknoJuce> *fine
[05:04:52] <ElMonkey> you have the current limited ones, right?
[05:05:07] <TeknoJuce> yes they are set to a limit of 0.27
[05:05:09] <TeknoJuce> amps
[05:05:39] <ElMonkey> er, i mean short circuit protected
[05:05:58] <ElMonkey> 0.27 amps seems very little for the large motors
[05:06:31] <ElMonkey> jitter would make sense when they dont have enough torque and just get into steploss
[05:06:37] <TeknoJuce> yeah these are the smaller motors the pololu guys kept telling me to go lower and lower until it stops jittering
[05:06:41] <ElMonkey> why that would cause the driver to lock up though, i dont know
[05:06:54] <TeknoJuce> I started @ 0.7
[05:07:30] <ElMonkey> you have anything attached to the motors yet?
[05:07:34] <ElMonkey> or just free running?
[05:08:18] <TeknoJuce> the one that is running well has a belt and two gears on it the one that is free running
[05:08:36] <TeknoJuce> is not
[05:09:11] <ElMonkey> hmm, tried different speeds?
[05:09:25] <ElMonkey> it does make it seem like a harmonic oscillation issue on the motor
[05:09:48] <ElMonkey> though i dont have direct experience with the pololu drivers
[05:11:22] <TeknoJuce> like different microstep settings?
[05:12:04] <TeknoJuce> both drivers are setup identical is the strange factor
[05:12:24] <ElMonkey> you could try finer microsteps
[05:15:07] <TeknoJuce> thanks for your incite will bang my head off the desk some more after I get some sleep.
[05:19:05] <ElMonkey> yea, i hate weird stuff like this
[05:19:13] <ElMonkey> the other day i was baffled by the sprakfun serlcd
[05:19:32] <ElMonkey> turns out the thing is just buggy and hell and simply locks up when not waiting long enough after certain commands
[08:50:47] <rue_more> see, take a unipolar motor, apply some nice proper fets like a IRFZ44 and you wont have this problem
[08:51:04] <rue_more> is the driver chip on the polopu thing small than your small fingernail?
[08:51:36] <rue_more> if so, I'm gonna just guess its not good for more than 1A continious
[09:04:50] <specing> Is it normal for an ADC channel to affect results from an another channel, but not the other way around?
[09:05:18] <specing> I mean, going from 0 to ADC_MAX makes a difference of 50 on the other channel
[09:05:45] <specing> While 0 to ADC_MAX on the other channel only results in a difference of 5
[09:06:34] <rue_more> only if youhave your channel numbers mixed up AND you didn't ground out unused channels
[09:07:19] <rue_more> adjacent voltages will effect floating adc pins by about 5
[09:16:51] <rue_more> specing, ^^^^
[09:20:20] <rue_more> Tom_itx, how DO I change layers?
[09:20:26] <rue_more> I'v actaully never done that...
[09:20:56] <rue_more> aha context sensitive menus, got it
[09:21:48] <specing> Do I seriously need to ground them all, or can I turn them off somehow?
[09:21:57] <specing> I only use ADC 0 and 1
[09:22:15] <rue_more> you need to ground them so you know you have the right channel
[09:29:19] <izua> force them as outputs!
[09:29:29] <izua> powerusage++
[09:33:28] <specing> xD
[09:43:07] <OndraSter> what is the .avrsuo file created in project folder for?
[09:43:13] <OndraSter> I always delete it...
[09:43:25] <OndraSter> (or at least till now, had problems with it when SVNing it)
[10:12:08] <izua> don't svn it?
[10:12:38] <izua> i never svn files created by the IDE
[10:50:26] <OndraSter> izua, yeah
[10:50:30] <OndraSter> I added it to ignore
[10:50:32] <OndraSter> but what is it for...
[10:50:35] <OndraSter> nothing changes when I delete it
[10:50:37] <OndraSter> some kind of cache?
[10:50:53] <jadew> any idea why ISR code generated by gcc is so big?
[10:51:07] <OndraSter> disassemble it and take a look :)
[10:51:11] <jadew> an empty ISR routine will have 18 bytes
[10:51:15] <OndraSter> disassemble it and take a look :)
[10:51:33] <jadew> well, as far as I can tell it reads the sei register and writes it back in
[10:51:46] <specing> jadew: sei is not a register!!
[10:51:55] <jadew> well, the SREG
[10:52:10] <jadew> sorry
[10:52:20] <jadew> why would it do that?
[10:52:27] <specing> 18 bytes = 4 pushs + 4 pops + reti
[10:52:37] <specing> jadew: because it interrupts running code
[10:53:10] <jadew> isn't the avr doing that anyway? at the moment it calls the ISR?
[10:53:21] <specing> Nope
[10:53:38] <jadew> so what would happen if you wouldn't do that?
[10:54:14] <jadew> it doesn't make sense, if you are able to run code in that routine, it's pretty clear that you can't run code in the main loop
[10:54:49] <jadew> what am I missing?
[10:56:28] <specing> imagine this: (I hope you know what these instruction do, else look them up)
[10:56:33] <specing> tst r0
[10:56:39] <specing> -ISR fires
[10:56:53] <specing> brne label
[10:56:55] <OndraSter> guys, when I am using PB7 as output, which is also OC2, when I set the PB7 low, will it fire the interrupt even when the OC2 output function is disabled?
[10:56:59] <specing> What would that do?
[10:57:16] <jadew> I don't know what brne label is, is that a jmp?
[10:57:23] <specing> LOOK IT UP
[10:57:57] <jadew> ok, jne
[10:58:46] <jadew> yeah, makes sense now
[10:58:58] <jadew> would have hoped the avr would do that for you
[10:59:32] <jadew> in this case the 8 prescaler is useless
[10:59:44] <jadew> when using interrupts
[11:00:41] <specing> 8 prescaler?
[11:01:32] <jadew> yeah, you set up a timer with a clk/8 prescaler and you have an interrupt firing up, if you have nother one fireing up after this one, it won't get ran
[11:02:12] <jadew> consider having an interupt on OVERFLOW and a CTC one for 1
[11:02:36] <jadew> the call to the overflow interrupt will screw up the other one
[11:04:04] <ziph> The timer interrupts should be edge triggered, which means they'll sit around waiting for you to handle them.
[11:04:34] <jadew> well, I want to generate some pwm
[11:04:39] <jadew> soft pwm
[11:04:45] <jadew> this thing ruins it
[11:05:08] <ziph> You want to generate soft PWM every 8 clock cycles? :)
[11:05:13] <jadew> yeah
[11:05:18] <jadew> it actually works really
[11:05:30] <jadew> the only problem is the big ISR routine
[11:05:43] <ziph> Isn't the interrupt latency 5 or so?
[11:05:50] <specing> I don't think using an ISR for that is appropriate
[11:05:57] <specing> ziph: 4
[11:06:02] <jadew> if it didn't have all that push and pop, it would work
[11:06:33] <ziph> How are you calculating values that fast?
[11:06:53] <jadew> I don't
[11:07:10] <ziph> What signal are you generating then?
[11:07:22] <jadew> I have an overflow interrupt which sets the line to 1, and a CTC interrupt which sets the line to 0
[11:07:41] <jadew> so I simply set OCR1A to the desired value
[11:07:47] <specing> Oh
[11:07:57] <ziph> Why aren't you using the PWM logic?
[11:07:58] <Kevin`> jadew: why can't you use the hardware functions for that?
[11:07:58] <specing> So write your ISRs in ASM
[11:08:12] <jadew> Kevin`, not enough resolution
[11:08:28] <specing> jadew: owerflow: sbi port, bit; ctc: cbi port, bit
[11:08:41] <jadew> specing, can you do it from whithin c?
[11:08:42] <specing> two instruction isr :D
[11:08:46] <specing> jadew: ofcourse
[11:08:48] <jadew> specing, I know
[11:09:03] <Kevin`> jadew: so what additional calculation are you doing with it?
[11:09:03] <ziph> How would the PWM resolution be less than the timer resolution?
[11:09:09] <jadew> Kevin`, none
[11:09:13] <specing> ISR(CTC_vect) __attribute__((naked));
[11:09:20] <specing> ISR(CTC_vect) {
[11:09:22] <Kevin`> jadew: it sounds like you are just duplicating the hardware function
[11:09:28] <specing> asm volatile (
[11:09:46] <specing> "cbi %0, %1\n\t"
[11:09:48] <jadew> Kevin`, no, the hardware function won't support two pwm's at 10bit resolution (not on this mcu)
[11:09:53] <specing> "reti \n\t"
[11:10:10] <ziph> Just write the ISR in a GAS file if you want to do that, having it naked and inline is an unnecessary complication. :)
[11:10:21] <jadew> so I'm using two OCR1A and OCR1B to generate two PWM's on the single 16bit timer I have
[11:10:35] <jadew> specing, thanks for the info
[11:10:37] <specing> : : "I"(_SFR_IO_ADDR(port)), "I"(bit) :);
[11:10:40] <Kevin`> jadew: there's no pins associated with those two output compares?
[11:10:45] <specing> }
[11:10:59] <izua> oc1a and oc1b
[11:11:25] <jadew> Kevin`, there are, but for some reason that's not working when using it in CTC mode with the top value defined by ICR
[11:13:04] <izua> did you set DDR to output manually?
[11:13:15] <jadew> yeah
[11:13:46] <ziph> Is there an #avr that isn't mainly people asking how to do perverse things? ;)
[11:15:07] <jadew> gonna try it again with the output pins, that would be the desired solution really
[11:15:33] <ziph> The MSP430 has really nice PWM support.
[11:16:13] <ziph> Some of the ARM Cortex M3's can do 4 or more 32 bit PWM's at 120MHz too.
[11:16:56] <jadew> nice
[11:17:17] <jadew> but I'm using pwm just cuz I'm cheap and I don't want to buy a DAC
[11:17:35] <jadew> actually, I would, but I just got a delivery and I would hate to pay for the shipping again, just for those
[11:17:37] <ziph> You don't have half a dozen resistors?
[11:18:13] <jadew> I think pwm will give you better accuracy
[11:18:30] <jadew> and it doesn't waste 8 pins :)
[11:18:42] <ziph> :)
[11:19:17] <ziph> If I recall correctly MSP430's have nice DAC's too, even on the low end chips.
[11:21:10] <jadew> they look interresting
[11:22:27] <ziph> Yup.
[11:22:50] <ziph> And the official programmers are cheap and the GCC build excellent.
[11:23:36] <jadew> eh, will stick with avr's for now
[11:24:23] <devilsadvocate> avrs can make do with free programmers :P
[11:24:37] <devilsadvocate> the msp still needs you to actually buy something
[11:25:25] <ziph> Unlike AVR's that are distributed for free by Social Security? :)
[11:28:57] <ziph> https://estore.ti.com/MSP-EXP430G2-MSP430-LaunchPad-Value-Line-Development-kit-P2031.aspx
[11:29:38] <ziph> There you go, MSP430 board with a full emulator for less than you can buy a shitty AVR and hack up some ghetto parallel port SPI read/write only programmer.
[11:29:55] <ziph> ($4.30)
[11:32:11] <ziph> They're 16 bit processors too, so no silly huge globs of code to do 16 or 32 bit arithmetic.
[11:32:13] <voodoofish430> the launchpad is pretty nice inthat respect..
[11:33:04] <ziph> I'm assuming you can rig that up to program/debug other MSP430's, but I haven't checked. ;)
[11:34:52] <voodoofish430> you can up to a certain memory limit on the ti related programmers. With the mspgcc compiler, you aren't memory limited. Most ifnot all of the valueline procs can be used.
[11:35:20] <voodoofish430> programmers being the dev environment...not the hw programmer itself.
[11:36:04] <specing> I have a launchpad in my drawer, it has been there for half a year now and I still haven't touched it
[11:36:22] <specing> fucking school leaves me no time to hack :(
[11:36:38] <ziph> High School or Uni?
[11:36:38] <voodoofish430> the latest version actually comes with some nice chips.
[11:37:02] <voodoofish430> it usses the 2553 chip and a 24xx series both 20pin....
[11:37:04] <specing> high school
[11:37:10] <ziph> You're doing it wrong then. :)
[11:37:45] <ziph> You should have loads of free time.
[11:38:06] <voodoofish430> unless he's taking the ap classes for college
[11:40:09] <ziph> Does that reduce your time/load in collage?
[11:40:14] <specing> Not sure what AP classes are
[11:40:26] <specing> But I am doing Cisco CCNA
[11:40:30] <voodoofish430> advanced placement
[11:40:36] <ziph> They're Uni level classes that US High School kids can do.
[11:41:01] * specing is not in that patent-infested country
[11:41:14] <voodoofish430> in highschool?
[11:41:59] <Steffanx> specing is on the good side of the globe :)
[11:43:24] <voodoofish430> alrighty, time to head to work....adios for now.
[12:07:46] <ziph> I just realized that by some cosmic coincidence all my clients are on holiday tomorrow.
[12:08:04] <ziph> Three day weekend time.
[12:08:06] <Steffanx> Partyyy
[12:14:50] <dirty_d> any of you guys use eagle?
[12:14:54] <OndraSter> I do
[12:14:55] <dirty_d> does it not come with any components?
[12:14:59] <OndraSter> it does
[12:15:03] <OndraSter> just enable (load) the libraries
[12:15:04] <dirty_d> hmm
[12:15:12] <dirty_d> where do you do that
[12:15:21] <OndraSter> libraries -> use all?
[12:15:30] <dirty_d> let me try that
[12:16:38] <vectory> you can also import more libraries, if need be, e.g. for use with VHDL software
[12:16:55] <vectory> whats more common in this chan, vhdl or verilog?
[12:17:19] <specing> A mixture of none
[12:17:32] <vectory> i read vhdl/verilog is a eu/us thing, but where does .au stand#?
[12:17:43] <vectory> specing: not for avr, true :)
[12:18:19] <dirty_d> OndraSter, that did the trick
[12:19:13] <OndraSter> I am sure you know it: you bring some new piece of hardware home
[12:19:19] <OndraSter> and no matter if there is still warranty, you just HAVE to open it up
[12:19:20] <OndraSter> lol
[12:19:25] <dirty_d> dont have hte xmega i need, but im sure that must be around somewhere
[12:19:39] <specing> vectory: Anyway, I would like to start hacking on FPGAs/CPLDs/whatever but I do not have time or equipment to do it
[12:19:43] <OndraSter> there is AVR library for Eagle
[12:19:48] <OndraSter> though but not sure if it has xmegas
[12:19:52] <OndraSter> *-but
[12:20:00] <dirty_d> i can just make it myself pretty easily right?>
[12:20:08] <vectory> use regular qtfp
[12:20:11] <dirty_d> just copy one and change the names of pins etc?
[12:20:33] <dirty_d> this one might actually already be the same pins
[12:20:44] <vectory> whatever package u need you can edit pinlables to suit your needs
[12:21:35] <vectory> thats what i tried, although the fileformat was rather crude
[12:22:05] <vectory> having to position lables "manually" iirc, sometimes appearing outside the chip etc
[12:22:27] <OndraSter> use the library editor :)
[12:22:36] <dirty_d> yea
[12:22:45] <OndraSter> btw, I have just noticed... There is support for AT91SAM9263EK inside Riff Box...
[12:22:54] <OndraSter> maybe they could add libs for other Atmel ARM MCUs... :)
[12:23:00] <OndraSter> into the app
[12:23:35] <ziph> vectory: You're picking a HDL?
[12:24:27] <vectory> nope, just curious. we started using vhdl in a course just when i flunked out.
[12:25:02] <ziph> Use Verilog then, you won't have the bad memories. ;)
[12:25:26] <vectory> but i didnt edit anything, it was about 74' chips, layed out in eagle and with timing diagramms in mentorgraphics, kinda cool
[12:27:12] <vectory> edited version of eagle, even. with a button to start modelsim simulation, making use of vhdl components in eagle, all very confusing to me %-)
[12:27:36] <ziph> That's weird.
[12:28:38] <vectory> i have i all in a virtual machine, if you wanna download 11 gb :)
[12:28:52] <ziph> Not really.
[12:29:12] <vectory> never worked on my pc. appearantly vm ware on ubuntu isnt as good as on win xp
[12:30:27] <ziph> I don't know why they wouldn't just get you to write VHDL/Verilog.
[12:33:16] <vectory> ziph: woulda been the next step, builing a oscilator with cpld as controller
[12:33:38] <ziph> A PLL Synthesizer?
[12:34:12] <vectory> have no idea, really
[12:34:22] <vectory> probs why i flunked, too xD
[12:34:32] <vectory> s/why/because/
[12:36:02] <dirty_d> hmm, no QFN24 packages?
[12:36:46] <dirty_d> oh, i guess its just not searching for that
[13:02:13] <Tom_itx> eagle also has a ref-packages lib that has alot of footprints to make lib parts
[13:03:28] <Tom_itx> hi abcminiuser
[13:05:26] <Fleck> anyone has some idea - why
http://bbs.dianyuan.com/bbs/u/54/1252531182223253.pdf << this chip does not generate, reading datasheet and cant get it - what can make it to stop generate on OP1 and OP2
[13:05:38] <abcminiuser> Yahoyoyoyoyoyoyoy
[13:05:46] <abcminiuser> AS5.1 is out people
[13:05:51] <OndraSter> final?
[13:05:55] <abcminiuser> Go download and cut down on my support tickets
[13:05:55] <Tom_itx> gawd help us
[13:05:56] <abcminiuser> Yep
[13:05:57] <OndraSter> I installed beta like 2 days ago
[13:05:59] <OndraSter> lol
[13:06:13] <OndraSter> direct download link so we don't have to mess on the website? :P
[13:06:24] <OndraSter> and register as fu.k@you.com?
[13:06:27] <Tom_itx> abcminiuser, is it different than the beta i got?
[13:06:28] <OndraSter> :P
[13:06:47] <abcminiuser> Tom_itx, very
[13:06:56] <abcminiuser> It's like 200 builds later or something
[13:07:09] <abcminiuser> This one actually (kinda) works in places it didn't work before
[13:07:29] <OndraSter> hmm
[13:07:37] <OndraSter> <OndraSter> direct download link so we don't have to mess on the website? :P
[13:07:59] <specing> Heh, moved halfway across the globe to work in support ;D
[13:08:39] <dirty_d> if this part im adding to the library has NC pins, would it be best to just leave those out of the sybol?
[13:08:45] <dirty_d> symbol
[13:10:28] <OndraSter> hmm
[13:10:37] <OndraSter> too bad that website search on atmel.com is giving me "page not found" errors :P
[13:11:25] <OndraSter> just as this
[13:11:25] <OndraSter> http://www.atmel.com/PFResults.aspx#(data:(category:'34864',type:!(3)),sc:2)
[13:11:29] <OndraSter> only 5.0 and 5.1 beta
[13:11:34] <OndraSter> abcminiuser, help us to get the new studio :P
[13:12:42] <abcminiuser> OndraSter, one sec
[13:13:33] <specing> I bet abcminiuser is specialised in supporting #avr
[13:13:38] <abcminiuser> Full Installer (+.NET, +VS Shell):
http://www.atmel.com/Images/as5installer-stable-5.1.208-full.exe
[13:13:53] <abcminiuser> Smaller Installer:
http://www.atmel.com/Images/as5installer-stable-5.1.208-small.exe
[13:13:59] <OndraSter> thanks
[13:14:08] <abcminiuser> specing, I'm a customer support engineer at the moment
[13:14:11] <abcminiuser> Oh wait, sorry
[13:14:17] <ziph> abcminiuser: Heya, better day today? :)
[13:14:19] <abcminiuser> Dear Customer, thank you for your enquiry
[13:14:28] <specing> ...
[13:14:38] <abcminiuser> Above you will find the latest direct links to Atmel AVRStudio 5.1, the latest version of Atmel AVR Studio
[13:14:43] <OndraSter> :D
[13:15:03] <abcminiuser> If you have any additional questions, please respond directly to this ticket and I will assist you further
[13:15:05] <Tom_itx> great tech support there!
[13:15:16] <abcminiuser> Herp Derpington, Derp Support Tech
[13:15:16] <specing> I feel sorry for abcminiuser :[
[13:15:24] <Tom_itx> i don't
[13:15:32] <Tom_itx> he begged to move to Norway
[13:15:35] <Tom_itx> :)
[13:15:43] <abcminiuser> specing, half my time is replicating the customer's elaborate setup so I can click the button they asked about and report back
[13:15:49] <Tom_itx> although i'd much rather work in the lab
[13:16:11] <abcminiuser> "Does the XYZ device support signature row programming?" *I check, get back to them* "Yes."
[13:16:45] <ziph> abcminiuser: How'd you end up in support?
[13:18:01] <abcminiuser> ziph, all apps engineers start there to train them up for a few weeks
[13:18:11] <ziph> abcminiuser: Ahh, ok.
[13:18:46] <ziph> And then you'll be off making toys to show off AVR's?
[13:19:14] <abcminiuser> Well then we rotate, 5 or so weeks developing ASF, a week on support
[13:19:54] <Steffanx> Yay
[13:20:04] <Steffanx> And when you screw up you get 2 weeks?
[13:20:14] <ziph> Hah.
[13:20:39] <abcminiuser> Hehe
[13:20:47] <abcminiuser> Just quit emailing support
[13:20:55] <abcminiuser> Or email them with "You're all awesome"
[13:21:10] <Steffanx> I will
[13:21:11] <Steffanx> Email?
[13:21:20] <ziph> Does everyone get supported equally?
[13:21:24] <Steffanx> support+abcminiuser@atmel.com ?
[13:21:43] <specing> dean.the.supporter@atmel.com?
[13:23:13] <Casper> abcminiuser: is there a way to recover from the signature corruption bug?
[13:23:39] <Steffanx> Casper corrupted a signature?
[13:24:22] <abcminiuser> ziph, we have priority support and regular support
[13:24:33] <abcminiuser> Priority (key customers, distributors, etc.) are 24 hour
[13:24:38] <abcminiuser> All others are 72 hour
[13:25:01] <Casper> Steffanx: yeah somehow the 3 bytes signature get corrupted sometime for unknown reason
[13:25:04] <Steffanx> How is it corrupted Casper? I wonder because I corrupted 2 m16's with my own jtag-ish programmer
[13:25:06] <ziph> Ahh, right.
[13:25:18] <Steffanx> They now all read as 0xFF
[13:25:28] <Casper> it happened to me once, couln't find a real fix but the "just force the programmer"...
[13:25:57] <Steffanx> Yeah, but what do you mean with 'corrupted'?
[13:26:12] <ziph> ...They aren't hardwired on the IC? :)
[13:26:21] <Casper> Steffanx: the bytes get fucked up and return something else
[13:26:29] <Steffanx> Who knows ziph
[13:26:41] <OndraSter> you burnt it!
[13:26:45] <OndraSter> like I almost burnt the SRAM :P
[13:26:58] <Steffanx> Aren't the calibration bytes in the same region?
[13:27:00] <OndraSter> still have no idea what was that about... works just fine
[13:27:07] <Casper> so instead of reading for example 84 d4 6e you can get ff ff ff or 91 68 93, so the programmer think you have specified the wrong part and error out
[13:27:10] <abcminiuser> ziph, no, they're variable
[13:27:27] <abcminiuser> Just not normally alterable by the user :S
[13:27:37] <ziph> :)
[13:27:51] <Casper> WOW WHAT A MIRACLE!!!!!!! I succeded to burn a dvd!
[13:28:01] <Steffanx> Free cake or beer for everyone?
[13:28:18] <specing> DVD? I though people stopped using them in the 2000s
[13:28:47] <Casper> sadly, still required sometime :/
[13:29:23] <OndraSter> abcminiuser, my devilish brain says "you make 10 same chips and depending on how much flash/ram works, you sell it as specified device" :P
[13:29:26] <specing> windoze?
[13:29:42] <Steffanx> Just change the signature and sell them OndraSter :P
[13:29:46] <OndraSter> exactly
[13:29:55] <abcminiuser> OndraSter, that's an interesting idea
[13:29:57] <OndraSter> lock half of the flash and sell them as mega8 instead mega16
[13:30:06] <abcminiuser> So's that one
[13:30:12] <Steffanx> So get some m8's and sell them as 328p or something like that
[13:30:18] <OndraSter> and is it just idea or reality on megas? :P
[13:30:25] <OndraSter> Steffanx, o_O
[13:30:26] <OndraSter> lol
[13:30:28] <OndraSter> great idea
[13:31:06] <OndraSter> this technique is common on CPUs, GPUs in PCs etc
[13:31:14] <OndraSter> (the one with locking, not renaming lol)
[13:31:46] <dirty_d> OndraSter, i created a new library, and created a symbold for an IMU which is QFN44, do you know how to pull in an existing QFN44 package into my library?
[13:31:54] <OndraSter> no
[13:31:59] <OndraSter> sadly no
[13:32:08] <dirty_d> hmm, maybe copy/paste
[13:33:06] <Kevin`> dirty_d: a single chip imu? what's this?
[13:36:24] <dirty_d> that worked
[13:36:34] <dirty_d> Kevin`, yup, an mpu-6050
[13:36:46] <dirty_d> 6 axis gyro/accelerometer on a single die
[13:36:55] <dirty_d> it wasnt cheap and it was hard to find
[13:36:57] <dirty_d> $15
[13:38:23] <Kevin`> 'hard to find' and 'qfn' probably make it not worth it vs seperate chips, but thanks. I'm gonna need something like that for a project soon
[13:39:18] <dirty_d> yea
[13:39:37] <dirty_d> its nice to already have the rotation and accelration axes aligned though
[13:39:46] <dirty_d> its 16 bit too, most other ones ive seen are 10
[13:39:56] <dirty_d> what are you building?
[13:40:44] <Kevin`> one of those rc flying machines
[13:42:54] <dirty_d> Kevin`, me too
[13:43:00] <dirty_d> a very big one
[13:43:11] <dirty_d> like able to lift 20 pounds big
[13:47:49] <specing> Niiiice, now translate that to kg
[13:48:02] <dirty_d> lets call it 8
[13:49:05] <specing> That is some serious lifting
[13:49:14] <dirty_d> well that wasnt that bad, i guess ill just make a new xmega part to match even though this one has the same pins
[13:49:20] <dirty_d> specing, yea
[13:49:21] <specing> Is it an aeroplane or a chopper?
[13:49:26] <dirty_d> quadcopter
[13:49:36] <dirty_d> im having trouble finding props
[13:49:54] <dirty_d> im looking for something around 18"x6" props
[13:49:56] <specing> with each one having to lift 3 kg, that is going to be hard indeed
[13:50:01] <dirty_d> nah
[13:50:14] <dirty_d> tahts not that much
[13:50:27] <dirty_d> rc stuff is very cheap
[13:50:37] <dirty_d> even the really powerful stuff
[13:50:58] <dirty_d> the motors will probably be around 1500W
[13:51:01] <dirty_d> each
[13:51:22] <specing> Wow
[13:51:30] <specing> How are you going to power them?
[13:51:35] <dirty_d> lipo
[13:51:44] <specing> That is not going to last long >_>
[13:51:48] <dirty_d> the motors are about $30 each, same for batteries
[13:51:55] <dirty_d> not if its lifiting 20 pounds
[13:52:07] <dirty_d> but if its just flying with no load probably like 20 minutes
[13:53:14] <dirty_d> it would be cool to use like gasoline engines for the main power and electric motors to maintain stability or soemthing
[13:53:47] <specing> I was thinking of building a plane though
[13:54:08] <specing> helicopters consume too much power
[13:54:17] <dirty_d> yea
[13:54:32] <dirty_d> i had a plane but you need too much room to use it
[13:54:40] <dirty_d> i fly my other rc heli in the house
[13:57:47] <specing> How accurately can you pilot these things though?
[13:59:36] <specing> But before I can do that, I have to learn RC
[13:59:54] <specing> Which translates to loads of electronics
[14:01:23] <Kevin`> i'm planning on building a tiltrotor plane, so it flys longer
[14:02:03] <Kevin`> specing: bah, just use radio modules designed by someone else. radio is black magic =p
[14:11:24] <dirty_d> specing, pretty accurate
[14:12:10] <dirty_d> im just interfacing from the rc reciever
[14:12:17] <dirty_d> i dont have to deal with the actual radio stuff
[14:12:34] <dirty_d> its just pulse widths that you work with
[14:13:25] <dirty_d> the hardest thing for me so far has just been balancing accuracy and speed in the calculations for stability control
[14:14:05] <dirty_d> its pretty tedious
[14:14:35] <dirty_d> its all integer based and scaled by powers of two to maintain precision while being fast
[14:15:04] <dirty_d> and making sure now to overflow
[14:15:06] <dirty_d> not
[14:19:37] <ureif> the ATmega16U2 and the ATmega8U2 are identical except for memory size ?
[14:19:54] <Kevin`> i'm using digital radio transceivers so I can get telemetry back and send more complex control information. the downside of course is that it's a whole seperate project to make that work
[14:20:07] <Kevin`> right now i'm just setting up two of the radios as a link between computers
[14:20:35] <specing> How much does such equipment cost? And what is the range of it?
[14:21:04] <Kevin`> both figures vary. $3 and a few km?
[14:21:51] <Kevin`> (more like $10 for a few km. the ones i'm playing with now are either $3 or $5 though, I forget)
[14:22:16] <Kevin`> you can spend as much as you want for various features or high level interface apis
[14:23:13] <Kevin`> i'd be done already if I used those xbee modules, for example
[14:24:56] <dirty_d> what kinda radios are you using?
[14:25:10] <dirty_d> specing, im using a 24GHz rc tx and rx
[14:25:16] <dirty_d> i think range is at least 1km
[14:25:23] <dirty_d> 2.4GHz
[14:26:20] <Kevin`> dirty_d: this is the one i'm fiddling with now:
http://www.hoperf.com/pro/rf/fsk/rfm22b.htm
[14:27:13] <OndraSter> dirty_d, let me come with bluetotoh/wifi/nrf24l01 :)
[14:27:25] <OndraSter> and your 2.4GHz range will worse :P
[14:28:49] <dirty_d> thats cool
[14:29:03] <dirty_d> lol
[14:29:13] <Kevin`> why is 2.4ghz so popular for rc stuff? I wouldn't expect it's performance to be very good compared with the ~27mhz stuff
[14:29:25] <OndraSter> it isn't licensed
[14:29:33] <dirty_d> you dont have to worry about interferance
[14:29:33] <Kevin`> or is that 72mhz
[14:29:35] <Kevin`> whatever
[14:29:43] <dirty_d> it frequency hops or something
[14:29:54] <Kevin`> OndraSter: there's a few frequency bands for rc stuff
[14:30:00] <OndraSter> oh
[14:30:01] <OndraSter> hmm
[14:30:11] <dirty_d> i dont think anyone uses 27 and 75MHz anymore
[14:30:18] <dirty_d> its all 2.4Ghz now
[14:30:28] <Kevin`> wacky.
[14:30:41] <dirty_d> you dont need to match crystals or anything
[14:30:51] <dirty_d> you "bind" to a reciever
[14:30:55] <dirty_d> based an a UID
[14:30:58] <Kevin`> I can see that being a big advantage for meets
[14:31:01] <dirty_d> yea
[14:31:07] <Kevin`> but.. there's no reason you couldn't so something similar in other bands
[14:31:13] <dirty_d> i guess
[14:31:21] <dirty_d> maybe the data rate isnt fast enough?
[14:31:33] <dirty_d> 2.4GHz is faster for digital stuff right
[14:31:49] <Kevin`> well there's more bandwidth available, but using it means shorter range
[14:32:00] <Kevin`> so I wouldn't expect the radios to be very high data rate
[14:32:43] <dirty_d> it might be
[14:36:36] <specing> Kevin`: Where did you buy it from?
[14:36:47] <dirty_d> i guess the range of mine is about 1.5km
[14:37:03] <dirty_d> but you can get booster antennas etc
[14:37:16] <dirty_d> i dont think i want to fly that far anyway, i wouldnt be able to see it lol
[14:37:24] <dirty_d> i wanna get a fpv camera and stuff though someday
[14:38:18] <Kevin`> specing: direct, but depending on which one you want, there are a few companies who stock them
[15:08:10] <Tom_itx> abcminiuser, did you see the online class MIT is offering?
[15:08:17] <Tom_itx> http://www2.electronicproducts.com/MIT_announces_first_free_online_course-article-fajb_free_MIT_course_feb2012-html.aspx
[15:14:00] <Tom_itx> abcminiuser, does that 5.1 small install overwrite the beta or do we need to uninstall the beta first?
[15:14:14] <abcminiuser> Tom_itx, no, haven't seen it
[15:14:20] <abcminiuser> Uninstall if possible
[15:14:29] <Tom_itx> ok
[15:14:30] <abcminiuser> I did inplace upgrades, but migth as well start fresh
[15:14:57] * specing passes his BFG9000 to Tom_itx
[15:14:57] <Tom_itx> the MIT thing looks neat for someone wanting those courses
[15:15:12] <specing> I want those curses but lack the time
[15:15:35] <specing> courses*
[15:15:36] <specing> meh
[17:55:11] <Jan-> hihi avr peeps :)
[17:58:44] <Casper> Jan-: hi ya
[18:02:09] <dirty_d> hi
[18:02:09] <tobbor> Hello dirty_d
[18:12:02] * Jan- puts her feet up on the table
[18:12:03] <Jan-> *thump*
[18:13:10] * specing puts his fingers up on the keyboard
[18:21:13] * Roamin puts the weed in the paper.
[18:22:01] <Tom_itx> has anybody used FLIP under linux?
[18:22:15] <Jan-> Of course not, it's under linux
[18:22:28] <Jan-> What people have done with FLIP under linux is spend 300 years trying to make it work, then gave up.
[18:22:30] * specing smacks Jan- with a large trout
[18:22:31] <Tom_itx> then i wouldn't be speaking to you
[18:23:27] <Tom_itx> there is actually a linux release of it is why i asked
[18:23:42] <Jan-> there's linux releases of lots of things
[18:23:56] <Roamin> i haven't, but that linux release is 4.5 years old, and x86 only it seems
[18:24:15] <Tom_itx> yeah it looks that way
[18:24:29] <Tom_itx> DFU seems the choice for that platform
[18:24:46] <specing> I've used DFU...once
[18:24:51] <specing> upon a time...
[18:25:05] <Roamin> i use DFU all the time
[18:26:18] <Tom_itx> well, i just updated mine and noticed a ver for linux
[18:26:42] <Tom_itx> more device support for the usb enabled xmegas looks like about it
[18:26:52] <Tom_itx> dfu may be lagging there
[18:37:05] <dirty_d> well there is is!
http://cisweb.bristolcc.edu/~klabs/quadcopter_sch.jpg
[18:37:29] <dirty_d> see anything immediately alarming?
[18:38:41] <specing> No crystal?
[18:40:19] <dirty_d> nah. xmega has 32mhz internal
[18:42:05] <specing> hmm who do the xmegas have to have that *$%^(@#$ pdi?!
[18:42:16] <specing> why*
[18:42:29] <Casper> why not?
[18:43:10] <specing> I'd love if they had regular ISP
[18:43:12] * Jan- splutters
[18:43:20] <Jan-> How on earth does an american end up with the name Wierzbowski?
[18:43:32] <dirty_d> specing, why?
[18:43:40] <dirty_d> dont have a programmer that supports it?
[18:43:46] <dirty_d> get tom's
[18:43:47] <specing> too lazy to build one
[18:43:51] <dirty_d> its $30
[18:44:01] <specing> actually, no time to do it
[18:44:08] <specing> And I like doing stuff myself
[18:45:06] <dirty_d> uhhg, i hate board layout
[18:45:17] <specing> iLike :D
[18:51:20] <dirty_d> its likea birds nest
[19:08:37] <Tom_itx> specing, it's less now
[19:09:00] * Jan- slithers slowly down in the chair until her feet hit the coffee table
[19:09:01] <Jan-> Bleargh
[19:09:03] * Jan- is exhausted
[19:09:10] <Tom_itx> tmi
[21:53:39] <Tom_itx> !thislog
[21:53:44] <Tom_itx> !thislog
[21:53:44] <tobbor> This one:
http://rueshouse.dyndns.org:82/~ircjunk/irclogs/html/%23avr-2012-02-17.html
[22:28:17] <CapnKernel> I wonder if brentbxr^ got any interest...
[23:28:33] <Casper> hi Richard_Cavell
[23:28:43] <Richard_Cavell> hi Casper
[23:28:47] <Richard_Cavell> I've been doing some electronics, mate
[23:28:49] <Richard_Cavell> analogue stuff
[23:28:54] <Richard_Cavell> Building up to doing some AVR work
[23:28:57] <Casper> :D
