Back
[05:08:52] <aep> abcminiuser: ping
[05:09:00] <abcminiuser> Pong?
[05:09:26] <aep> abcminiuser: heya, seen my messages about 10 hours ago about AudioInput.c?
[05:09:42] <abcminiuser> Was out, left the machine idling
[05:09:48] <abcminiuser> You posted to the list IIRC tho
[05:09:52] <abcminiuser> Replied to one there today
[05:10:08] <aep> oh cool, thanks. i thought i'd get a notification oO
[05:11:25] <aep> oh. i did change the endpoint size acording to specs already. let me try changing the pid
[05:12:51] <aep> nope that did not help
[05:13:25] <aep> i also lowere sample frequency to 8000
[08:40:47] <theodor> "Hardware breakpoints can be combined to form data breakpoints" Does this mean that i can check for a larger size data in memory for comparison?
[11:14:25] <bss36504> Ni1njaneer: Whats up with your nick?
[11:20:29] <DanFrederiksen> atmel has no chips with digital comparators?
[11:20:44] <DanFrederiksen> instead of the much less useful analog comparators
[11:21:06] <DanFrederiksen> you set a digital value for comparison with external signal. like for current limiting
[11:21:22] <DanFrederiksen> seems like an overlooked feature
[11:21:29] <bss36504> much less useful? Analog comp could be used for digital comparator.
[11:21:44] <DanFrederiksen> with the use of a DA?
[11:22:08] <bss36504> no. Set the Aref pin to some Vcc-a little bit, boom, 1 bit digital comparator
[11:22:25] <bss36504> analog comparator ≠ADC
[11:23:03] <DanFrederiksen> That doesn't sound quite right
[11:23:36] <bss36504> which thing?
[11:25:42] <bss36504> what you're looking for is something like the PIC's latch thing. Just set up the analog comp correctly, and it will do almost the same thing. and again, the analog comp ≠the analog to digital converter
[11:41:17] <_abc_> Hello. What is the difference between say atmega32u2 and atmega32u4 or 6?
[11:46:58] <Casper> DanFrederiksen: what you want is pin change interrupt
[11:48:56] <bss36504> ^Yes, that
[11:50:03] <_abc_> Ooh an old acquaintance.
[12:17:02] <_abc_> What is the difference between say atmega32u2 and atmega32u4 or 6? Simply the usb engine revision?
[12:39:14] <Duality> _abc_: actually don't know but i am kinda curious too now :)
[12:51:06] <bss36504> not a whole lot happening in here lately...
[12:57:26] <_abc_> bss36504: you just broke the silence of the lurking contest.
[12:57:44] <bss36504> darn, I ruin everything :P
[12:58:02] <bss36504> but now we can talk, right guys!?
[12:59:12] <theodor> will ceramic 22pF 50V work with 30pF Crystal?
[12:59:35] <bss36504> maybe? try it.
[12:59:43] <bss36504> Crystals can be sensitive sometimes
[13:00:01] <bss36504> you need to speak nicely to it, and give it the right caps so it actually makes a good clock for you.
[13:00:12] <theodor> ok =)
[13:00:18] <bss36504> However, 30pf is outside the spec of both the crystal AND the MCU, so be warned.
[13:00:43] <theodor> what happens with my atmega??
[13:01:08] <bss36504> nothing bad per se. My guess is it wont clock properly
[13:01:17] <bss36504> and therfore, it will do exactly nothing.
[13:01:18] <theodor> are EVGA's cooler good?
[13:01:21] <theodor> could that be why my clock speed (led blink) became 4x slower?
[13:01:22] <theodor> *
[13:01:53] <theodor> i tought i burned my crystal but the microcontroller could be dead, is that what youre saying?
[13:02:34] <bss36504> I have no idea. Since I know nothing about the setup, I couldnt tell you. It's unlikely you burned out the crystal or the mcu, unless you did something really bad to them.
[13:04:36] <theodor> ill come back when ive tried to switch out the crystal and caps.
[13:51:42] <spec__> Hi all. I'm trying to read/verify the flash contents of a ATmega64 via JTAG using the AVR dragon. I'm using avrdude in interactive mode to control it.
[13:52:12] <spec__> It 'succeeds', but returns only 64 bytes, not 64K bytes.
[13:52:29] <spec__> Reading the fuse bits returns 0xFF, so there is no flash read protection.
[13:59:06] <jerkey> spec__ i know some programmers can't handle the larger chips because of the protocol they use, but i'm not sure where the dragon fits on that scale
[13:59:37] <bss36504> People should just stop using the Dragon, IMO. JTAGICE3 >> Dragon
[14:01:11] <jerkey> spec__ without knowing any real answers for you I would first confirm that your dragon is compatible with that chip.
[14:06:18] <DanFrederiksen> Casper, is that a trigger on a certain analog level?
[14:06:39] <DanFrederiksen> say above 3V
[14:06:53] <Casper> pin change interrupt is whenever the logical status on it change
[14:07:09] <Casper> if you want analog, use the analog comparator in the avr
[14:07:15] <DanFrederiksen> but that's not what I want though
[14:07:28] <DanFrederiksen> I want a digitally set analog comparator
[14:08:45] <DanFrederiksen> value threshold trigger
[14:09:13] <bss36504> you can use the AVCC pin, or the bandgap as a reference for the analog comp. If you want more resolution, you will need to supply the AVCC pin, or use the ADC and do the comparison in software
[14:09:53] <bss36504> It would be very rare to have the feature you want, since it would suck up a lot of die space to make an accurate power supply to act as a reference just for one peripheral.
[14:10:04] <Roklobsta> isn't dragon limited to 32k?
[14:11:03] <bss36504> DanFrederiksen: What I mean is, there is nothing else on the chip that would ever need more than whatever the logic supply voltage is, and if it does need another voltage, it's supplied externally (ie, the AVCC pin, or the UVCC pin on USB parts)
[14:11:12] <Tom_itx> Roklobsta, not anymore
[14:12:04] <eatyourguitar> has anyone tried the ATXMEGA8E5-AU
[14:12:06] <eatyourguitar> ?
[14:12:10] <Roklobsta> ah the limit was lifted. i had to give my dragon back. stuck with mkII-CN
[14:12:27] <bss36504> Roklobsta: You should get a JTAGICE3 :)
[14:12:33] <bss36504> you deserve it
[14:12:46] <Tom_itx> what's special about it?
[14:12:48] <DanFrederiksen> bss36504, it wouldn't take die space
[14:13:04] <Roklobsta> Tom_itx: it isn't made by you? ;)
[14:13:23] <Tom_itx> that's harsh
[14:13:30] <Roklobsta> hey there was a ;)
[14:13:38] <bss36504> DanFrederiksen: It absolutely would. Power supplies suck space like nothing else on a chip.
[14:13:56] <bss36504> Roklobsta, Tom_itx: Shots fired! :)
[14:14:32] <bss36504> DanFrederiksen: Either way, its not a feature. So you'll have to use the methods I described or find a new brand, since no Atmel part has the feature you want.
[14:14:54] <Roklobsta> mrs would like me to make a nixie clock controller - obviously i'll use avr.
[14:16:04] <DanFrederiksen> bss36504, it's not a power supply
[14:16:29] <DanFrederiksen> and I can always do it with a voltage divider
[14:16:44] <bss36504> DanFrederiksen: It's a reference. Sure. Either way, it would be big, and totally unnecessary since you can get the feature you want with an ADC and like 3 lines of code.
[14:16:47] <DanFrederiksen> but it's an obvious function that should be in the chip instead
[14:17:01] <bss36504> I respectfully disagree.
[14:17:10] <DanFrederiksen> you are respectfully wrong
[14:17:19] <Tom_itx> saiz who?
[14:17:41] <DanFrederiksen> God
[14:17:55] <Tom_itx> which one?
[14:18:13] <DanFrederiksen> there is only one
[14:18:25] <DanFrederiksen> same as there is only one truth
[14:20:12] <bss36504> DanFrederiksen: Find one other person on here who says that a "Digitally-controlled-reference analog comparator" is "an obvious function that should be in the chip instead [of using the ADC/Analog comp w/ external reference]" and I'll eat my words.
[14:20:55] <Tom_itx> bss36504 don't get him started. it's what got him banned in robotics once
[14:21:26] <Tom_itx> or twice
[14:21:50] <bss36504> Should I care? I don't appreciate being made to feel like I don't know what I'm talking about when I do, and a number of people would back me up on this IRC.
[14:29:09] <Duality> hi I am compiling my code with avr-g++ but in my .h file I have to add a include .cpp at the end of the code. why is this? else it won't work.
[14:29:43] <bss36504> That sounds wonky as hell. Can you paste the code up? pastebin.com
[14:31:09] <Duality> http://pastebin.com/L9iAMxBP If i don't do the include it gives me alot of undefined references
[14:33:31] <bss36504> did you try including the DmDisplay.h file in the DmDisplay.cpp file?
[14:33:43] <bss36504> This...is an odd way of doing things, for sure.
[14:34:20] <bss36504> a #include tells the preprocessor to literally copy and paste all the junk in the include source into the destination.
[14:34:45] <Duality> this is what the .cpp file looks like
[14:34:48] <Duality> http://pastebin.com/4fDib10A
[14:35:26] <Duality> yes, and I don't know why I have to do it like this, I normaly do it the way you say
[14:35:46] <bss36504> What exactly is the problem? Is it at runtime, or is there a compilation error?
[14:37:39] <Duality> well when i call functions from the class if I do it the normal way, it gives me the errors that the functions i call are a undefined reference
[14:38:06] <Duality> when I make a .elf file
[14:40:19] <Duality> like for example GameOfLife.cpp:(.text._GLOBAL__I_Font+0xe): undefined reference to `DmDisplay::DmDisplay()'
[14:42:41] <bss36504> I assume you include DmDisplay in GameOfLife.cpp as well?
[14:43:45] <Duality> bss36504: yes
[14:45:56] <Duality> bss36504: my .cpp file looks like this:
http://pastebin.com/qzC16m10
[14:47:26] <bss36504> Well thats odd. Youre compiling this at the command line or in AS6?
[14:48:09] <Duality> command line
[14:48:45] <bss36504> C++ isnt my best language, but in your constructor call of DmDisplay, should it be DmDisplay lcd(); ?
[14:49:25] <Duality> bss36504: hmm could be, I have done it this way lots of times before
[14:50:03] <Duality> command i use for compiling.
http://pastebin.com/TVeMMAvA
[14:51:06] <Duality> bss36504: if i do lcd(); it gives me GameOfLife.cpp:23:6: error: request for member ‘lcdChar’ in ‘lcd’, which is of non-class type ‘DmDisplay()’
[14:55:06] <bss36504> well isnt avr-gcc NOT avr-g++?
[14:57:57] <Duality> bss36504: well actually when you call avr-gcc with c++ sources it automaticly uses avr-g++ ( i read )
[14:58:46] <Duality> and changing avr-gcc into avr-g++ produces the exact same output.
[15:00:01] <bss36504> The more you know haha. Ok...I'm assuming you pass all of the cpp files into the compiler?
[15:00:19] * bss36504 is grasping at straws now...
[15:00:20] <theodor> For a low pass filter for AVcc does it matter if the capacitor (1uF) is rated for 50V?
[15:00:34] <theodor> its quite large
[15:00:36] <bss36504> theodor: nope, you're food
[15:00:38] <bss36504> good*
[15:02:05] <bss36504> theodor: The voltage on a cap is the max rated voltage of the cap, not what it *will* charge to.
[15:03:13] <theodor> bss36504, i see thanks
[15:04:45] <sinclair> hi everyone
[15:04:50] <bss36504> hey there
[15:05:55] <sinclair> i'm having problems compiling a simple blink_led program, the build fails with "'doskey' is not recognized as an internal or external command, operable program or batch file." running on Windows 8.1, Atmel Studio 6.1 and a JTAGICE3
[15:07:05] <bss36504> Well thats a new one on me. Does AS6 support Win8?
[15:08:01] <sinclair> worked before i fried my atmega324p, selected wrong internal clock ... here is the full output ->
http://pastebin.com/fYD9P0ec
[15:08:14] <sinclair> (this is a new atmega324p btw)
[15:08:33] <bss36504> right, but device problems shouldnt affect the build...
[15:09:03] <bss36504> Do a find in files for doskey
[15:09:33] <sinclair> nothing found
[15:10:16] <bss36504> What does the timeline of failure look like? It worked, then what, then it didnt.
[15:12:43] <sinclair> it worked, flashed blink_led successfully, changed internal clock unable to enter debugging mode, ordered new one, replaced it, opened same project, rebuild, build fails
[15:12:53] <sinclair> i've tried the repair option of atmel studio
[15:13:07] <sinclair> maybe i should uninstall and try to clean up and reinstall
[15:13:53] <bss36504> ok thats odd. A quick google of doskey returns that it's some sort of command line shell extension that microsoft makes, but I see no reason why AS would need it. Did you try the ol' restart yet?
[15:14:14] <sinclair> yeah several times
[15:14:16] <Duality> bss36504: how ? how do i pass all the cpp files into the compiler ?
[15:16:56] <bss36504> sinclair: I guess you could try a fresh install. Wish I could help more. toolchain issues are not my strong point.
[15:18:06] <bss36504> Duality: I think google will provide a better answer about how to pass files into gcc. Pretty sure the general format is avr-gcc [flags] -o outfileName sourceFiles?
[15:18:11] <bss36504> not sure though.
[15:18:45] <bss36504> I use my IDEs as a crutch for my lack of interest in defining commands manually :) Maybe try this in AS6
[15:21:39] <sinclair> uhm
[15:21:58] <sinclair> i noticed something... it tries to generate .elf files now, shouldn't it be .hex?
[15:22:48] <bss36504> You can set that stuff in the project settings, but either one should be programmable to the chip i believe.
[15:23:53] <sinclair> ok
[15:34:03] <Duality> bss36504: yes :)
[15:34:33] <Duality> Someday i will understand how to make this working :P I hope it won't take me to long.
[15:35:37] <bss36504> Duality: I'm a fan of AS6, and it hasnt let me down yet. That's where I'm comfortable :)
[15:35:37] <bss36504> \
[15:48:52] <Duality> bss36504: I can't use AS^
[15:48:56] <Duality> AS6 *
[15:51:24] <bss36504> Duality: Why not?
[15:52:26] <Duality> bss36504: AS6 is windows only ?
[15:52:41] <bss36504> yes
[15:52:46] <bss36504> Virtual machines man
[15:52:47] <Duality> that is why :)
[15:52:58] <bss36504> I have a Mac. I just run AS6 in a VM.
[15:53:15] <bss36504> You can download winXP from all over the internet too.
[15:55:04] <tzanger> I also have a mac. I use the command line. I have a strong dislike for IDEs
[15:56:43] <bss36504> I like IDEs for various reasons: No need to muck around with the command line (not that I cant, I just prefer not to), code completion, good text formatting, etc.
[15:57:06] <bss36504> I get more productivity using an IDE versus a text editor and a command line, but that's just me.
[15:57:37] <tzanger> heh, I find text/code completion and things like autoindent etc more of a pain than anything, and I really like being able to ssh in, screen -r and continue owrking away from the actual hardware
[15:57:54] <tzanger> yep it's a very personal thing, some people love 'em, some hate 'em
[15:59:07] <bss36504> Have you ever tried Netbeans as an IDE? I think it is just lightyears ahead of most other IDEs. Stable, fast, excellent multi-(programming)language support, cross platform, highly customizable. Currently I'm using it for java and C development, and I've had no problems.
[15:59:50] <bss36504> And actually, its running on my work's linux servers, so I just open my VNC session and go.
[15:59:58] <tzanger> yep, tried eclipse, netbeans, vs-based ones, emacs based ones... they're all nasty
[16:00:12] <bss36504> eclipse leaves a bad taste in my mouth.
[16:00:21] <tzanger> VNC is a very poor substitute for screen or even RDP
[16:00:49] <bss36504> meh, idk. Its what we all use. seems to work ok. It's plenty fast over a LAN
[16:02:30] <tzanger> over a LAN, sure
[16:02:41] <tzanger> RDP as much as I hate MS technologies, works very, very well
[16:05:23] <bss36504> alright, I've just been staring at my code for like 40 minutes. Time to go home. g'night all.
[16:29:58] <theodor> in my experience over network windows built in remote desktop is really good
[16:30:18] <theodor> +1
[16:52:08] <catphish> i just compiled a very simple program and uploaded it to an arduino mega 2560, it doesn't appear to work, and i'm hoping for some pointers because i have no idea how to go about debugging it:
http://pastebin.com/wemYU4b9 /
http://pastebin.com/CHqfmDYk
[16:52:19] <Duality> it was a choice of mine not to use windows ever again :)
[16:53:06] <catphish> i would have expected PORTB|128 to have an affect on the arduino LED
[16:57:49] <Duality> the code shows 0xFF
[16:58:49] <catphish> yes, when 128 didn't work, i thought i'd try flooding the whole port
[16:58:51] <Duality> it should toggle all 8 pins
[16:59:09] <Duality> are you sure the led is on that port ?
[17:00:34] <catphish> all the documentaton i can find says its on PB7
[17:03:01] <catphish> the schematic is here:
http://arduino.cc/en/uploads/Main/arduino-mega2560_R3-sch.pdf
[17:07:21] <Duality> hmm yea but it's connected through a opamp, what if you tried connecting a led to any of the other pins on the port ?
[17:07:28] <Duality> should work though
[17:09:32] <Duality> atleast I don't see any reason why it shouldn't
[17:09:45] <catphish> i tried the same thing on PORTD, which should flash the LEDs designed to monitor serial io activity, no affect there either
[17:11:45] <catphish> is there anything i missed that might prevent that code from actually starting? looks like it programmed successfully
[17:19:56] <catphish> i think i need to look at the fuses / oscillator settings
[17:52:44] <phinxy> Is there 2-row breadboards you can use for ground/vcc?
[17:52:50] <catphish> it appears i've accidentally set my atmega fuses to use the internal oscilator and now it won't respond to my programmer at all, not sure what it needs to reset it :(
[17:53:04] <Epsilon-Auriga> phinxy: yes.
[17:53:22] <catphish> lots of breadboards have special rows at the edges for that
[17:53:25] <Tom_itx> if you set to internal osc it should work
[17:53:30] <Epsilon-Auriga> catphish: if it is set to internal oscillator it should be running at least 1MHz one would think.
[17:53:42] <Epsilon-Auriga> phinxy: let me find the ones I use.
[17:53:45] <Tom_itx> lower your spi frequency
[17:53:52] <jerkey> catphish if should respond if set to the internal oscillator. perhaps you set it to use some other oscillator source.
[17:53:52] <Tom_itx> must be 1/4 the crystal freq
[17:53:54] <Tom_itx> or less
[17:54:10] <phinxy> Epsilon-Auriga so kind of you :)
[17:54:15] <catphish> that's what i thought, i set (copy and paste accident) -p m2560 -c usbasp -e -U lfuse:w:0x42:m -U hfuse:w:0x99:m -U efuse:w:0xff:m
[17:54:37] <jerkey> http://www.engbedded.com/fusecalc
[17:54:46] <Tom_itx> yeah check yourself.
[17:54:55] <Epsilon-Auriga> http://www.circuitspecialists.com/solderless-breadboard-wb-102.html these are the ones I use and the strips down both sides are removable.
[17:55:06] <catphish> they're its defaults for atmega2560
[17:55:08] <Epsilon-Auriga> you have to cut the paper backing but they separate easily after that.
[17:56:04] <Epsilon-Auriga> http://www.futurlec.com/Breadboards.shtml I also have some of these....on this page there are some just power strip boards.
[17:56:18] <phinxy> Epsilon-Auriga, oh. i have the same one didnt know you could remove the sides!
[17:56:20] <Epsilon-Auriga> warning on futurlec....shipping is 1 to 6 weeks...usually on the order of 4 weeks.
[17:56:31] <phinxy> i switched to another better quality one so thats why
[17:57:14] <Epsilon-Auriga> digikey also sells them.
[17:57:25] <Epsilon-Auriga> different brands an whatnot.
[17:58:51] <Epsilon-Auriga> http://www.digikey.com/product-search/en?site=us&lang=en&vendor=19&Keywords=solderless%20breadboards%20-I&stock=01
[17:59:12] <Epsilon-Auriga> http://www.digikey.com/product-detail/en/923277-I/923277-ND/3808
[17:59:56] <Epsilon-Auriga> but those are very very pricy.
[18:02:01] <catphish> looks like the fuses wrote successfully, then the board stopped responding, right now i can only assume the internal clock isn't working as it should, i assume there's no way i could have completely disabled ISP programming is there?
[18:05:19] <Epsilon-Auriga> catphish: yes, depending on the chip, it is possible.
[18:15:09] <catphish> Epsilon-Auriga: thanks, i fixed it by slowing the programmer bitrte, at 1MHz, it couldn't respond fast enough, very happy i didnt kill it :)
[18:16:52] <catphish> Duality: thanks for your help also, after disabling the bootloader flag, my program loads, since it was programmed at 0x0000
[18:36:33] <clixxIO_> Hi DrLuke
[18:37:15] <clixxIO_> everybody..
[18:37:25] <Duality> catphish: ok no problem, hadn't thought about the bootloader though, I always program directly :)
[18:38:00] <catphish> i'm working on an arduino, so it was configured with a bootloader previously
[18:38:07] <Tom_itx> you can load the program but until you change the boot bits it won't know where to start execution
[18:38:33] <catphish> i'm not totally sure i understand why the bootloader doesn't just sit at 0x0000, then start a program later in memory
[18:39:01] <Epsilon-Auriga> also, wasn't someone trying to use the arduino bootloader to change fuses with avrdude the other day?
[18:39:11] <Epsilon-Auriga> catphish: because.
[18:39:22] <Tom_itx> i believe it's because your reset vector sits there
[18:39:34] <Tom_itx> look at a pdf of your memory map
[18:40:33] <Epsilon-Auriga> and don't forget that programs running from bootloader section of memory have the ability to change the rest of the flash....not sure what that has to do with why it doesn't start at 0000 but,,,
[18:40:57] <Tom_itx> pics start at 0x03
[18:41:01] <Tom_itx> go figure
[18:41:23] <catphish> actually if a program uses absolute memory locations, it would need to be at 0x0000, or at least know where its going to be
[18:41:37] <Tom_itx> yup
[18:41:46] <Tom_itx> you set that in the makefile i believe
[18:42:27] <Tom_itx> same for eeprom data
[18:42:38] <Tom_itx> needs to know where to load it
[18:43:34] <catphish> sure, i probably won't be using a bootloader any more, i see no need with ISP programming
[18:43:42] <Tom_itx> nope
[18:43:43] <phinxy> should the cap polarity for AVcc have ground against the pin? power -> resistor -> cap -> AVcc-pin
[18:43:58] <Tom_itx> for a ceramic cap?
[18:44:01] <Tom_itx> non polarized
[18:44:13] <phinxy> its not ceramic its a 1uF polarised
[18:44:32] <phinxy> )|
[18:44:46] <Tom_itx> i don't recall wiring it that way
[18:45:02] <Tom_itx> why do you need all that for the analog vcc?
[18:45:06] <Tom_itx> tie it to your source
[18:45:12] <Tom_itx> put a cap to the pin and gnd
[18:45:21] <Tom_itx> and
[18:45:29] <Tom_itx> i wouldn't use a electrolytic there anyway
[18:45:32] <Tom_itx> they're noisy
[18:46:01] <Tom_itx> i think the pdf shows how it should be
[18:46:07] <Tom_itx> i don't recall from memory
[18:46:34] <phinxy> Im following a post where he sounds very knowledgeable
[18:46:38] <phinxy> its basicly
http://i.stack.imgur.com/UEot5.png
[18:46:58] <phinxy> but instead of a inductors, theres a resistor
[18:47:01] <Tom_itx> The ADC has a separate analog supply voltage pin, AVCC. AVCC must not differ more than ±
[18:47:01] <Tom_itx> 0.3V from VCC. See the paragraph “ADC Noise Canceler” on page 251 on how to connect this
[18:47:01] <Tom_itx> pin.
[18:47:11] <phinxy> yse he mention that
[18:47:28] <Tom_itx> i get my info straight from the data sheets generally
[18:47:28] <phinxy> " The datasheet says AVCC shouldn't be lower than VCC - 0.3 V, but I couldn't find how much current AVCC uses. That won't be much, say 10 µA maximum."
[18:47:37] <Tom_itx> don't trust everything you hear
[18:47:45] <phinxy> but he post charts :D
[18:47:49] <Tom_itx> don't worry about the current
[18:47:54] <Tom_itx> just connect it to 5v
[18:53:06] <Tom_itx> i see they're using the resistor as part of the noise canceling
[18:53:25] <Tom_itx> what chip is this?
[18:53:31] <Tom_itx> i'm looking at a 168 data sheet
[18:53:38] <Tom_itx> they all should be pretty much the same
[18:54:18] <Tom_itx> oh wait... what i was looking at wasn't for avcc
[18:54:23] <Tom_itx> ignore that
[18:55:02] <Tom_itx> they use a 10µH choke to vcc for the avcc with a cap to gnd
[18:55:21] <Tom_itx> there is a diagram for it in the data sheet
[18:55:50] <Tom_itx> also run a GND plane under the adc pins for better noise immunity
[18:56:26] <Epsilon-Auriga> why not wrap the whole thing in a piece of aluminum foil and ground that?
[18:56:37] <Epsilon-Auriga> will keep the aliens out anyhow.
[18:56:52] <Tom_itx> that's not gonna separate the digital noise from the analog circuit
[18:57:16] <phinxy> http://electronics.stackexchange.com/questions/36167/understanding-the-avcc-pin-wiring-on-arduinoleonardo-low-pass-filter
[18:57:26] <phinxy> check out the second post. third is informative too
[18:58:34] <Tom_itx> that's basically what the data sheet says
[18:58:47] <Tom_itx> i just prefer to get my info first hand
[18:59:22] <phinxy> i guess the only way to know which type of filter is best is with an oscilloscope :\
[18:59:35] <phinxy> screw it. directly to ground it is
[18:59:37] <Tom_itx> they both show a choke
[18:59:49] <Tom_itx> it would be good to run it thru the choke
[19:00:04] <Tom_itx> to gnd??
[19:00:09] <Tom_itx> avcc goes to 5v
[19:00:49] <phinxy> Who said anything else?
[19:00:49] <Tom_itx> avcc--choke---5v
[19:00:56] <Duality> lol nice motto in the topic
[19:01:09] <Tom_itx> rue must have done that
[19:02:07] <Duality> it's a good motto, it's just my ears hurt when i read it :)
[19:02:19] <Tom_itx> yeah blame rue for that
[19:02:59] <Epsilon-Auriga> I always encourage people...specially those playing with explosives and high voltage.
[19:03:18] <Tom_itx> at the same time even
[19:03:29] <Epsilon-Auriga> of cours
[19:03:33] <Epsilon-Auriga> of course!
[19:03:53] <Epsilon-Auriga> I even encourage them to put it on video, live feed if at all possible..
[19:04:15] * phinxy shakes head
[19:04:45] <Tom_itx> umm that first pic looks wrong to me at first glance
[19:05:10] <Tom_itx> i dunno what MH2029-300Y is though
[19:05:22] <Tom_itx> i assume a choke since it's labelled L1
[19:06:19] <Tom_itx> no it's right
[19:06:38] <Tom_itx> i was seeing the AGND next to it and thinking that was the label for that wire
[19:07:23] <phinxy> I find different shematics everywhere
[19:07:36] <Tom_itx> you shouldn't find much variation
[19:07:46] <Tom_itx> that's pretty clear how it should be
[19:08:16] <Tom_itx> if you're just messin around you can tie avcc to 5v directly
[19:08:24] <Tom_itx> otherwise add the filter
[19:08:31] <phinxy> http://d32zx1or0t1x0y.cloudfront.net/2011/02/adconverter_04_lrg.jpg
[19:08:34] <phinxy> this is what ive done
[19:08:49] <phinxy> except aref is directly to ground with another cap
[19:09:22] <phinxy> but then i found this:
http://i39.tinypic.com/2i2c0ux.jpg
[19:10:17] <phinxy> maybe that other cap is something different
[19:10:26] <phinxy> down right
[19:13:56] <clixxIO_> I think what they mean is that any filtering is better than none
[19:15:06] <phinxy> How would you guys test if the filter is doing its job? theres no easy way is it?
[19:16:40] <clixxIO_> hey Tom_otx: question
[19:17:05] <clixxIO_> I have a signal which is 2v, I want to detect that with adc on interrupts
[19:17:37] <clixxIO_> is that realistic, or do I need to pull it up to 5v and use it as a digital io?
[19:18:03] <clixxIO_> it's a pulse, from a car distributor. actually it's 1v
[19:18:03] <Epsilon-Auriga> analog comparitor is your friend.
[19:18:21] <clixxIO_> inside the avr?
[19:18:28] <Epsilon-Auriga> yup.
[19:18:36] <clixxIO_> ah - I see
[19:18:50] <clixxIO_> ok - I will try go learn that
[19:19:47] <clixxIO_> by the sounds of it, that would suite my needs
[19:20:09] <clixxIO_> thanks
[19:25:15] <Tom_itx> yeah you can set the trigger voltage on a comparator
[19:26:01] <clixxIO_> Hi Tom, I've quite liked your programmer
[19:26:16] <clixxIO_> works perfectly with my attiny boards
[19:26:32] <Tom_itx> glad you like it
[19:27:19] <Tom_itx> phinxy, i'm not so sure why you keep searching the web for circuits when it's spelled out in the data sheet
[19:27:30] <Tom_itx> not that the ones you found are wrong...
[19:27:35] <clixxIO_> https://plus.google.com/109366813998920635083/posts/LKHYdpLWbfb
[19:28:18] <phinxy> Tom_itx, i searched for 'low pass' and got 3 results so i went to google
[21:38:09] <DrLuke> hi ClixxIO
