Back
[00:33:19] <mog> hi
[00:34:24] <mog> im having trouble unlocking bootloader flash lock on my atmega328p
[00:34:34] <mog> here is pastebin of avrdude ive tried
http://pastebin.com/BsDYpGHi
[00:34:41] <mog> it keeps reading fusebits as all 0
[00:37:22] <Casper> mog: you need to issue a chip erase...
[00:37:31] <mog> i tried that -e right?
[00:37:33] <Casper> else you can not touch the lock bits
[00:37:37] <Casper> I forgot the command
[00:37:58] <mog> -e is a chip erase according to man
[00:39:09] <mog> and as pastebin shows when i just issue -e it succeeds
[00:40:34] <Casper> I think you need to do it in one command...
[00:41:11] <Casper> try to swap the commanre
[00:41:17] <Casper> -e -U
[00:42:03] <mog> i tried that and that just fails like the other way
[00:42:45] <mog> ive tried slowing down rate to 19600 and still fails
[00:42:58] <mog> i have 3 of these chips i cant believe its all 3 broken
[00:44:46] <mog> the boards just have the 328p on them no oscillator could i need to set it to internal first or something?
[00:46:34] <Casper> if it's set for external crystal, then you need a clock
[00:47:27] <mog> all three of the chips keep showing fuse bits set to 0
[00:47:54] <mog> i cant reset that fuse bit without a clock?
[00:51:00] <inflex> very odd problem here
[00:51:21] <inflex> I'm picking up a tiny leakage current out of my LED from the AVR, just barely enough to fire the LED
[00:52:05] <Casper> mog: if the clock fuse is set to external clock/crystal, you need a clock. if it's set for internal RC, then you do not need
[00:52:06] <inflex> and the AVR is rebooting during current/voltage load transitions, even though it's running off a 3V regulator and the supply line is 5V... just isn't making a lot of sense :(
[00:53:34] <mog> it was never set and from the factory Casper its supposed to be set to internal
[00:54:18] <ziph> inflex: I warned you about sleeping with Murphy's wife.
[00:54:28] <ziph> inflex: Now look what's happening. ;)
[00:54:33] <Casper> mog: blank chip?
[00:54:39] <mog> from factory
[00:54:39] <mog> yes
[00:54:49] <inflex> ziph: bummer :( and I got the 4MBit flash memory working too :(
[00:55:00] <ziph> inflex: Which chip?
[00:55:04] <inflex> ziph: it's just odd... all I can think of is that there's protection diode leakage
[00:55:22] <ziph> inflex: Maybe they soldered it in the wrong way. ;)
[00:55:23] <Casper> mog: blank chip do not have lock bit set
[00:55:30] <Casper> they come completly empty
[00:55:35] <inflex> ziph: memory chip is a Microchip 4MBit SST25VF040B
[00:55:46] <Casper> if you can't read the lock fuse from a blank chip then you have problem with your programmer
[00:55:47] <mog> the ones i have used in past had fuse bits set not just all 0s
[00:56:02] <ziph> inflex: Ah, that's the one I used. Nice and cheap, but the amount of CPU you have to waste writing to it is a pain.
[00:56:20] <inflex> ziph: well, it works fine untill I plug the servo on... and then it works fine until I do a fast transition and it causes a reboot... BUT, I can stall the servo with my hands and it doesn't cause a reboot
[00:56:35] <inflex> ziph: yeah, esp ifyou're doing byte mode writing
[00:56:46] <inflex> ziph: though block-writing is okay
[00:56:59] <inflex> but yes... those 24-bit addresses are a bit long
[00:57:04] <ziph> inflex: Block writes are about the same except you don't send the address each time.
[00:57:37] <inflex> I find it interesting thatyou have to explicitly send the WREN each time
[00:57:50] <inflex> (same with the Atmel ones I had too)
[00:58:06] <ziph> The Dataflash let you DMA into the buffer though.
[00:58:08] <inflex> Anyhow, still a bit confounded as to where this leakage current is coming/going from
[00:58:19] <ziph> That way your MCU is only dealing with it once every 6ms or so.
[00:58:34] <inflex> well, I'm only pumping about 80 bytes/sec
[00:58:45] <ziph> Yeah, I wanted to do audio recording. ;)
[00:58:46] <Casper> mog: check the datasheet, 1 could be not programmed
[00:58:52] <inflex> ziph: yeah, that'll be a bit more intense :D
[00:58:55] <Casper> so be sure that you have the right one...
[00:58:58] <Casper> beside
[00:59:07] <Casper> not all bits are used, the unused one can read anything
[00:59:08] <inflex> ziph: I've noticed the "chip erase" works quick... amazingly quick
[01:00:10] <mog> Casper, i know i have tried two programmers im sure i did something wrong i just cant figure it
[02:53:30] <earthshine> Morning
[05:17:11] <amee2k> mmh, cyanoacrylate glue comes off with acetone, right?
[05:18:01] <crazy_pete> it is supposed to
[05:18:09] <amee2k> good :)
[05:18:12] <crazy_pete> in my experience it doesn't work all that great
[05:18:31] <crazy_pete> and acetone will eat the hell out of most synthetic fabrics
[05:19:03] <amee2k> i need something to keep a heatsink on a board but bolting won't really work on this one
[05:19:24] <amee2k> but i sure want a way to get the heatsink off again
[05:23:04] <crazy_pete> oh acetone is not the answer
[05:23:15] <crazy_pete> acetone will remove UNCURED cyanoacrylate
[05:23:24] <crazy_pete> it will weaken it on an exposed surface
[05:23:36] <crazy_pete> but to get a heat sink off would probably require soaking for days
[05:23:59] <amee2k> eeew
[05:24:30] <amee2k> any other ideas for a glue that'll work well as just a thin layer and that i can get off again?
[05:26:26] <amee2k> thermal contact doesn't need to be stellar. i don't need to get rid of a lot of heat, but i need it steadily
[05:26:53] <crazy_pete> i thought they made special heat sink glues just for that
[05:27:37] <amee2k> hrm
[05:30:32] * Landon wonders... anyone here work at maxim by any chance?
[05:31:18] <amee2k> the porn magazine?
[05:31:19] <amee2k> :P
[05:32:10] <mrfrenzy> you want some "samples" Landon ? ;)
[05:32:22] * amee2k still recalls sitting mildly bored in a lecture and looking for some led drivers... idly typing maxim.com.... >_<
[05:32:48] <Landon> well
[05:32:50] <Landon> I want a job ;)
[05:33:03] <amee2k> how big are your boobs?
[05:33:11] <Landon> just trying to sift through potential BS on this opening
https://maximic.ats.hrsmart.com/cgi-bin/a/highlightjob.cgi?jobid=1845
[05:33:16] <amee2k> >_>
[05:33:52] <ziph> You could work in the supply chain department.
[05:34:00] <ziph> At Maxim that'd be the easiest job in the world.
[05:34:19] <Landon> why do you say that? :P
[05:34:41] <ziph> Because in contrast to other manufacturers Maxim doesn't actually have to sell their product.
[05:35:31] <crazy_pete> i used to work for Maxim, i found out there are at least 8 companies around the globe with that name
[05:36:05] <Landon> heh
[05:36:17] <crazy_pete> Yeah thats the maxim for whom i used to work
[05:36:35] <crazy_pete> those sort of companies are a bad deal in general. Honestly Maxim was the best of the lot
[05:36:53] <Landon> why's that?
[05:37:48] <crazy_pete> Well those computer headhunting firms are a nasty deal
[05:37:56] <pingec> I have a gsm modem with rs232 connector but datasheet says lines are +-4V I must not connect this straight to the PC right?
[05:38:02] <Landon> ah
[05:38:23] <crazy_pete> Some of those companies are just horrid. They recruit people from India and then literally enslave them
[05:38:31] <crazy_pete> Maxim actually treated people decently for the most part
[05:38:54] <Landon> well, not much other way to approach it, Kansas universities aren't a hot target for computer-y employers to visit
[05:39:16] <amee2k> pingec: the levels are a bit on the low side, iirc minimum swing is +/-3V or so. find the shortest cable you have and give it a shot
[05:39:36] <pingec> I won't fry anything?
[05:39:51] <amee2k> the 4V is probably 5V and a shitty charge pump to invert it
[05:40:36] <pingec> datasheet says 4V
[05:41:29] <amee2k> the line driver is going to take a volt off it
[05:41:49] <amee2k> anyway, sounds a bit low but otherwise normal to me
[05:42:00] <amee2k> what makes you think you're going to fry something?
[05:43:26] <pingec> doesnt the PC use +-12V ?
[05:43:38] <amee2k> depends
[05:43:54] <pingec> hmm ill try
[05:44:00] <amee2k> cheap shit ones use 0V/+5V and it works most of the time so noone says anything
[05:44:02] <pingec> if it doesnt work i need the max232 chip ??
[05:44:21] <ziph> pingec: Sure you aren't reading the minimum voltages?
[05:44:29] <crazy_pete> i thought by definition , RS-232 was 12V+
[05:44:36] <amee2k> if the modem claims RS232 compliance the inputs must be able to tolerate the 12V driver
[05:44:46] <crazy_pete> yes you need that level shifter for RS232
[05:45:01] <amee2k> from what i recall 232 can be anything from 3V to hmm 15V?
[05:45:09] <ziph> pingec: Stick a multimeter on the TX pin while it is idle.
[05:45:31] <crazy_pete> i am far from an expert amee2k but i thought a level shifter was REQUIRED for RS-232
[05:46:01] <amee2k> crazy_pete: you need it to convert RS232 level to whatever logic levels your circuit uses internally, yes
[05:46:01] <pingec> ok let me find a multimeter
[05:46:33] <crazy_pete> yes but isn't RS-232 defined as being 12-20V?
[05:46:39] <amee2k> in theory if your circuit used RS232 compatible levels internally you wouldn't even need the driver, but that would be a kinda ridiculous setup
[05:46:50] <pingec> http://www.elfa.spb.ru/uploads/tdpdf/CT63_eng_TDS.pdf
[05:46:55] <pingec> this is the thing
[05:46:57] <pingec> page 8
[05:47:03] <crazy_pete> it seems something is contradictory amee2k
[05:47:08] <pingec> can i connect it straight to the PC or not?
[05:47:36] <amee2k> "Valid signals are plus or minus 3 to 15 volts; the ±3 V range near zero volts is not a valid RS-232 level. The standard specifies a maximum open-circuit voltage of 25 volts."
[05:47:41] <ziph> pingec: Those are the minimum voltage levels, connect it to a PC.
[05:47:44] <amee2k> --
http://en.wikipedia.org/wiki/RS232#Voltage_levels
[05:48:01] <crazy_pete> the data sheet says "5-32V" so no problem
[05:48:15] <crazy_pete> lol i dunno that i would believe the sky was blue if wikipedia said it :)
[05:48:38] <amee2k> in practice its not very relevant... receivers should accept anything near ground and below as low level, and anything significantly above ground as high level
[05:49:02] <crazy_pete> but since your data sheet CLEARLY says 5-32V whats the matter?
[05:49:12] <amee2k> because lots of cheap equipment takes a dump on the standard and uses whatever they have. even straight 0V/5V TTL signals
[05:49:43] <pingec> crazy_pete where does it say 5-32?
[05:49:46] <ziph> He's off connecting his PC to it, why are you still arguing. :)
[05:49:51] <crazy_pete> hahah zak0
[05:49:54] <crazy_pete> er ziph :)
[05:50:22] <crazy_pete> gacck!
[05:50:59] <crazy_pete> i am sorry pingec i was looking at the wrong page , you're right , page 8 DOES say >4V <4V
[05:51:08] <crazy_pete> i was completely in error pingec
[05:51:16] <pingec> ok
[05:51:21] <ziph> 5-32 sounds more like the DC supply
[05:51:25] <crazy_pete> however on page 8 the data sheet does say >4V so i think
[05:51:29] <pingec> does that mean i should not connect it?
[05:51:35] <crazy_pete> yeah that is exactly what i was looking it (grins sheepishly)
[05:51:45] <crazy_pete> well it says >4V is a logic high
[05:52:01] <crazy_pete> and it says it is RS232 compliant so you should be very much ok
[05:52:13] <ziph> pingec: Yes, it is describing standard RS232. (Or something close enough to it).
[05:52:13] <crazy_pete> since it says GREATER than 4V not "Just 4V"
[05:52:23] <crazy_pete> so you are good to go, pingec :)
[05:52:46] <amee2k> 4V is sufficient for standard compliant RS232, and if they want to claim compliance the receiver must be able to handle appropriate voltage levels as well
[05:52:58] <amee2k> so long story short, it should work fine
[05:53:00] <amee2k> :)
[05:53:02] <ziph> Plus any designer who puts non RS232 level signals on a DE-9 is a retard and deserves to have his gear returned to him under warranty. ;)
[05:53:24] <pingec> ok
[05:53:33] <pingec> I dont have any experience in these things
[05:53:37] <crazy_pete> yeah it really should work fine pingec
[05:53:46] <crazy_pete> and yeah i was looking at the DC pinout :)
[05:53:52] <pingec> cool
[05:54:02] <amee2k> well, pretty much noone (except for hobbyists) still uses RS232 on consumer gear anymore
[05:54:27] <pingec> one more question
[05:54:36] <pingec> can i connect this connector directly to an avr uart?
[05:54:38] <amee2k> RS485 on a 4P4C connector would be kinda cool though
[05:54:56] <amee2k> pingec: nope, thats where you do need a line driver now
[05:55:13] <amee2k> USART has just normal TTL level outputs
[05:55:22] <ziph> pingec: No, an AVR UART will want 0V or 5V more or less exactly.
[05:55:41] <amee2k> if you need to ghetto rig it, chances are using the AVR driver as TX will work fine
[05:55:42] <pingec> so that's where i need a max232 ?
[05:55:55] <ziph> pingec: Also, the signals are inverted. 12V on RS232 is 0V on the AVR UART and -12V on RS232 is 5V on the AVR UART.
[05:56:09] <amee2k> and for the RX... 1k resistor and clamping diodes to ground and Vcc
[05:56:21] <ziph> The AVR will deal with a non-inverted signal?
[05:56:34] <pingec> sounds complicated
[05:56:51] <pingec> I'd better read some more
[05:56:53] <ziph> Or the MAX232 doesn't invert and I'm just confused with the bit mapping? :)
[05:56:56] <amee2k> ziph: i've done that once but i used a soft-uart for it
[05:57:05] <ziph> Ahh.
[05:57:15] <ziph> pingec: Just use a MAX232 or similar.
[05:57:19] <amee2k> i used the hardware uart to interface with another PC
[05:57:23] <pingec> ok
[05:57:37] <amee2k> max232 is nice but kinda expensive
[05:57:37] <ziph> pingec: It isn't worth trying to get the hacks working in my opinion.
[05:57:46] <ziph> The TI MAX232's are cheap.
[05:58:06] <amee2k> the hacks are only there if you can't get a hold of a driver, really
[05:58:11] <pingec> i dont care about price
[05:58:15] <pingec> I care about simplicity
[05:58:23] <pingec> One step at a time :D
[05:58:30] <amee2k> or if you are designing commercial stuff and the marketing department wants it cheaper
[05:58:42] <ziph> MAX3222 is cheap.
[05:58:45] <ziph> (From TI)
[05:58:58] <amee2k> pingec: the MAX232 is just the IC and a bunch of capacitors
[05:59:11] <amee2k> download the datasheet, it has an example schematic that works fine
[05:59:23] <pingec> ok i will
[05:59:25] <pingec> thank you all
[05:59:34] <ziph> MAX3222CDBR is $1.53 in singles from Digikey.
[05:59:44] <pingec> I hope to get this working soon :)
[05:59:51] <amee2k> you're welcome :)
[05:59:53] <ziph> pingec: Good luck with it.
[05:59:54] <amee2k> ziph: nice :)
[06:00:07] <ziph> pingec: By the way, you might want to organise to be able to reset the modem from software.
[06:00:23] <ziph> pingec: In my 10 years of dealing with GSM modems they will all lock up given enough time.
[06:00:28] <amee2k> "AT Z" doesn't work?
[06:00:41] <pingec> ok
[06:00:46] <ziph> You can have the AT interface die on some of them.
[06:00:54] <amee2k> oww
[06:01:05] <amee2k> welcome to china
[06:01:07] <ziph> There's just some standard industry rule that they're not allowed to be reliable.
[06:01:08] <pingec> So if I reset it once in a while periodically should be fine?
[06:01:09] <amee2k> thats industry quality for ya
[06:01:21] <ziph> amee2k: The ones installed on $3000 motorola units die too. :)
[06:01:23] <pingec> lol
[06:01:32] <crazy_pete> are they also made in China? :)
[06:01:39] <ziph> pingec: If your software can't get a response after a while then reset it.
[06:01:52] <ziph> pingec: or if dialing fails or AT commands fail for a while.
[06:01:58] <amee2k> i think these days the list what isn't made in china is /much/ shorter than what is
[06:01:58] <pingec> ok
[06:02:07] <pingec> will take into consideration
[06:02:23] <crazy_pete> yeah almost everything is made in china and is a total fail
[06:02:28] <ziph> crazy_pete: Probably, but in cases like that Motorola also sends people over to China to stick people's nuts in a vice if things don't get done properly.
[06:02:36] <amee2k> hire a nig... afroamerican with a huge club
[06:03:04] <amee2k> that modem won't dare to freeze up... and it'll try only once
[06:03:04] <crazy_pete> people around where i live are actually learning to make things themselves again like in the old days simply because western capitalists working with china have sabotaged the worlds manufacturing ability
[06:03:38] <ziph> Sign.
[06:03:39] <ziph> Sigh.
[06:03:48] <crazy_pete> what happened ziph ?
[06:03:49] <amee2k> crazy_pete: how do you think i got into electronics... being able to fix stuff has made my life much easier on occasion
[06:03:52] <ziph> Must remember not to Apple-W on the wrong screen.
[06:04:01] <crazy_pete> indeed amee2k that is why i learnt to work on cars
[06:04:06] <crazy_pete> lol ziph
[06:04:52] <amee2k> ziph: is that along the lines of "xchat sucks if you're used to irssi?
[06:05:13] <ziph> The problem is that I have two monitors on this PC.
[06:05:37] <ziph> And with Windows in a VM I can stare at one and think I'm focused on that when I'm not.
[06:05:48] <amee2k> the person who thought it is a good idea to make ctrl+w do something random undesirable on GUIs deserves to lose his fertility over this
[06:06:08] <ziph> Apple-W closes windows.
[06:06:17] <mrfrenzy> as does ctrl+w
[06:06:28] <amee2k> except when its a terminal...
[06:06:41] <ziph> Yeah, that's a bit nasty.
[06:06:52] <ziph> At least on a Mac ctrl does nothing. ;)
[06:06:56] <crazy_pete> roflmao
[06:07:04] <crazy_pete> i had to try ctrl-w to see what it did
[06:07:09] <crazy_pete> i see why you sighed, ziph :)
[06:07:20] <amee2k> especially since GUI people can't even make up their mind what close window should be... ctrl+w, ctrl+x, ctrl+q, wtf??
[06:07:44] <ziph> It's very consistent on the Mac, Apple-W for a window and Apple-Q for an app
[06:09:17] <amee2k> only because the marketing department doesn't know about it :P
[06:10:22] <jacekowski> amee2k: alt+f4
[06:10:33] <crazy_pete> yeah there were CUA standards for that
[06:10:34] <amee2k> mmh right, i missed that one
[06:10:43] <crazy_pete> then MS blew them away as soon as they got them accepted :)
[06:10:52] <crazy_pete> then the Gnome project had to blow them away too :)
[06:11:07] <jacekowski> tbh alt+f4 is working on a lot of platforms
[06:11:11] <jacekowski> it worked like that on amiga
[06:11:13] <jacekowski> windows
[06:11:16] <jacekowski> linux with kde
[06:11:39] <amee2k> crazy_pete: the problem isn't that there are no standards, the problem is that there are either too many of them, or people don't even know they exist
[06:11:56] <amee2k> so they just duplicate the behaviour of their favourite program without thinking about it
[06:14:58] <crazy_pete> well CUA did start out as a standard
[06:15:19] <crazy_pete> and apple did follow it IIRC (i never used apple, but apple users claimed apple was all CUA compliant)
[06:30:52] <Tom_itx> amee2k use silicon
[06:33:14] <Tom_itx> Landon, try TI in Dallas and go from there
[06:34:14] <amee2k> isn't silicone a pretty poor thermal conductor unless its made for that kind of application?
[06:36:02] <Tom_itx> what about epoxy?
[06:36:43] <amee2k> i want to have a way to get it off again
[06:36:54] <amee2k> epoxy is a huge pain in the ass for that
[06:37:12] <Tom_itx> ask inflex what he uses
[06:37:17] <Tom_itx> not here right now
[06:37:37] <amee2k> i'm tempted to just try regular thermal grease and superglue the corners down
[06:41:53] <Tom_itx> superglue doesn't flex much
[06:42:43] <Tom_itx> and if the board is twisted or something it may pop loose taking whatever it's attached to with it
[06:42:46] <amee2k> i was going to try a *thin* film of superglue over the entire area, but not sure if that is practical to get off again
[06:43:29] <amee2k> there is nothing attached directly to the heatsink, the board has some SMT parts that get hot
[06:44:03] <amee2k> and heatsinking SMT parts is kinda impossible, so i'm now using a huge ground plane like heat spreader and heatsink the board instead
[06:44:33] <amee2k> but the plane isn't electrically neutral so i can't put the heatsink directly on it
[06:45:30] <RikusW> amee2k: qfn does have a heatsink pad.....
[06:45:55] <amee2k> if nothing else works, i'll have to redesign the board so i can bolt it to the heatsink :/
[06:46:14] <amee2k> RikusW: but its kinda useless because you can't bolt it to a heatsink like a TO220
[06:46:19] <Tom_itx> it's not necessarily a heatsink pad but rather a pad to keep the chip from flexing on the board so the tiny pads don't get ripped off
[06:46:29] <Tom_itx> in some cases it is a heatsink
[06:46:29] <amee2k> all the heat goes into the board, and then you're stuck with the same problem again
[06:46:40] <Tom_itx> but i think all qfn have a center pad regardless
[06:47:56] <Tom_itx> glue a butterfly to the top of it
[06:47:57] <Tom_itx> :)
[06:48:06] <amee2k> ??
[06:49:15] <Tom_itx> you should know better than to take my advise when i first wake up
[06:49:25] <amee2k> ...
[06:50:11] <RikusW> mosquite might work better :-P
[06:50:48] <amee2k> torturing a mosquito certainly has a higher satisfaction factor
[06:50:58] <amee2k> butterflies are kinda cool
[06:51:36] <RikusW> thought so too ;)
[06:52:07] <keenerd> I've heatsinked SMT parts before. Cut some copper flashing into strips and glue them on with thermal cement.
[07:25:42] <RikusW> grummund: you're there ?
[07:32:49] <Steffanx> Are you there RikusW ?
[07:33:13] <RikusW> Of course ;)
[07:33:32] * RikusW need a stk600 protocol dump for jtag
[07:33:48] <RikusW> stk600 protocol appnote is vague
[08:16:42] <scuzzy> hey RikusW
[08:16:43] <scuzzy> how you doing?
[08:16:50] <scuzzy> Guess what?
[08:18:39] <Steffanx> No guessing today scuzzy
[08:18:48] <scuzzy> apparently
[08:18:55] <scuzzy> I guess I just wanna know if he's listening
[08:20:01] <RikusW> hi scuzzy
[08:20:01] <scuzzy> or... reading rather
[08:20:04] <scuzzy> yo!
[08:20:10] <scuzzy> guess what?
[08:20:13] <scuzzy> I got a letter from SARS
[08:20:14] <RikusW> what ?
[08:20:17] <scuzzy> saying I owe them R21k
[08:20:17] <RikusW> and ?
[08:20:22] <scuzzy> teh suck
[08:20:25] <RikusW> ugh, really ?
[08:20:37] <scuzzy> THEN, I got a letter from UK Revenue service (I used to work in the UK)
[08:20:40] <scuzzy> £3700
[08:20:44] <scuzzy> even more teh suck!
[08:21:01] <scuzzy> then I realised, that actually, they owed me the £3700
[08:21:05] <RikusW> seriously ?!
[08:21:12] <scuzzy> it too me a while to get off the floor and stop have heart failure
[08:21:18] <scuzzy> and read the letter properly
[08:21:48] <scuzzy> so, like, UK Revenue is paying SA revenue... sort of
[08:21:50] <scuzzy> funny stuff
[08:22:00] <RikusW> hmm
[08:22:03] <RikusW> lucky for you
[08:22:10] <scuzzy> yeah
[08:22:14] <scuzzy> dodged a bullet there
[08:22:39] <RikusW> Just finished soldering a t10 to a breakout
[08:22:47] <scuzzy> how'd that go?
[08:22:50] <Steffanx> What's SARS scuzzy ?
[08:22:51] <scuzzy> did you make the breakout?
[08:22:58] <RikusW> more like a 2x3 pinheader on a little piece of pcb
[08:22:59] <scuzzy> South African Revenue Service
[08:23:15] <Steffanx> I only 'know' SARS as in Severe Acute Respiratory Syndrome
[08:23:25] <scuzzy> Steffanx: yeah, this is wrose
[08:23:26] <scuzzy> worse
[08:23:34] <Steffanx> Worse?
[08:23:36] <scuzzy> sort of the same
[08:23:38] <RikusW> pin 2+5 fits exactly between the header pins
[08:23:39] <scuzzy> SARS and SARS
[08:24:05] <RikusW> added little wires for the other pins
[08:24:14] <scuzzy> RikusW: you gonna get PDI working?
[08:24:14] <RikusW> still tiny...
[08:24:23] <scuzzy> take a pic
[08:24:25] <RikusW> TPI
[08:24:38] <scuzzy> ok
[08:24:50] <scuzzy> nice one
[08:25:02] <scuzzy> Steffanx: it was supposed to be a joke
[08:25:20] <Steffanx> Oh, haha :P
[08:26:31] <RikusW> scuzzy: thats a triple pun ;)
[08:29:25] <scuzzy> RikusW: hehehe
[08:35:45] <vectory> scuzzy: i lol'ed
[08:35:58] <scuzzy> in rl?
[08:36:04] <vectory> ye
[08:36:19] <vectory> i feel funny today anyway
[08:36:32] <vectory> might even start my profect -_-
[08:36:37] <vectory> project*
[08:38:40] <scuzzy> what is your project?
[08:38:51] <vectory> getting started with avr
[08:39:05] <vectory> building a small led matrix
[08:39:09] <vectory> on breadboard
[08:39:31] <vectory> to move it to perf board later and have a nce xmas present
[08:40:02] <vectory> xmas is all about blinky lights anyway
[08:40:56] <vectory> bbl
[08:41:06] <scuzzy> sweet
[08:44:03] <grummund> RikusW
[08:44:17] <RikusW> hi grummund
[08:44:44] <grummund> do i need to undox the stk600 ?
[08:44:49] <grummund> *unbox
[08:45:08] <RikusW> you could do it whenever its convenient for you
[08:45:25] <RikusW> you'll need to have a jtag avr
[08:45:40] <RikusW> and do some flash/fuses r/w on it
[08:45:52] <RikusW> I'm not in too much of a hurry ;)
[08:46:11] <RikusW> but will appreciate it, if I can get that sometime
[08:46:57] <grummund> how i've got an xmega xplain card
[08:47:09] <RikusW> needs to be avr8
[08:47:22] <RikusW> but you can send the xmega dump too
[08:47:23] <amee2k> okay, new question. i'm looking for a 2-conductor connector (polarized or not isn't important) that can carry mains voltage up to 10 amps, but under no conditions whatsoever must be confused with a power inlet connector and connected straight to a wall outlet. any ideas what type would do?
[08:47:30] <RikusW> that will be used later as well :)
[08:47:35] <grummund> which also has an at90usb (so it has 2 JTAG connectors)
[08:47:57] <inflex> Right, power is back ... that was a fun 3 hours
[08:48:01] <RikusW> at90usb is fine
[08:48:10] <grummund> and a atmega2560 card for the stk600, which has never been used
[08:48:18] <amee2k> preferably it shouldn't be much larger than an IEC inlet connector. the live part is the cable so the male connector should go on the case
[08:48:25] <grummund> xmega is not avr8 ?
[08:48:28] <RikusW> will work too
[08:48:47] <RikusW> grummund: it use PDI and stk600 xprog protocol
[08:49:12] <grummund> oh, and a at90usbkey
[08:49:57] <grummund> hmm, xmega has both pdi and jtag
[08:50:26] <RikusW> xmega jtag != atmega jtag....
[08:50:36] <RikusW> stk600 use different commands
[08:54:09] <grummund> atmega2560 daughter board hooked up and avrstudio reads the device sig. ok.
[08:54:56] <RikusW> logging is still on ?
[08:55:18] <grummund> no idea. what do you need?
[08:55:27] <RikusW> read and write some innocent fuse
[08:55:46] <RikusW> and close programming dialog
[08:56:06] <RikusW> you remember last time with the clock settings you turned on logging ?
[08:56:30] <grummund> no i have the memory of a fish
[08:56:45] <Steffanx> How good is the memory of a fish?
[08:57:13] <grummund> better than mine ;)
[08:57:28] <RikusW> HKEY_CURRENT_USER\Software\Atmel\AVRTools\STK500\LogFilePath
[08:57:36] <RikusW> you set a filename there
[08:58:51] <RikusW> I want that file
[08:59:05] <RikusW> after you close the programming dialog
[08:59:11] <RikusW> the log is saved there
[09:00:29] <grummund> http://www.dailymail.co.uk/sciencetech/article-1106884/Three-second-memory-myth-Fish-remember-months.html
[09:03:43] <grummund> atMega2560 clear/set CKDIV8 -
http://pastebin.com/kkhLqjiY
[09:05:30] <soul-d> woulnd that be obvious if you look at those trained fish :P
[09:05:59] <soul-d> http://www.youtube.com/watch?v=H3LZtfslbRA
[09:07:40] <RikusW> thanks grummund
[09:08:29] <grummund> np
[09:09:29] <RikusW> was that fuse hex value 0x62 ?
[09:09:52] <grummund> um, the dialog box is closed now ;-/
[09:09:59] <RikusW> 0x62 0x99 0xFF
[09:10:01] <RikusW> nvm
[09:10:38] <RikusW> will check against my current stk600 jtag code
[09:10:48] <RikusW> guess I'll have to modify it a bit
[09:11:23] <grummund> CKDIV8 is bit-7 in lofuse
[09:12:41] <RikusW> yup
[09:12:57] <RikusW> just entered that values in my Qt app
[09:13:08] <RikusW> ff 99 62 -> ff99 e2
[09:13:51] <RikusW> thanks to you I now have some exact protocol "docs" :)
[09:14:31] <RikusW> grummund: you have any other atmel programmers around ?
[09:15:04] <grummund> only a mkii clone based on abcminiuser's code
[09:15:32] <RikusW> grummund:
http://sites.google.com/site/megau2s/home/supporting-software
[09:15:39] <RikusW> you can get RavrProg there
[09:15:50] <RikusW> abc mkii works with it
[09:16:03] <RikusW> you'll need libusb and qt dev files in Linux
[09:16:18] <RikusW> havent tried compiling on XP yet
[09:17:39] <RikusW> will fix the stk600 jtag code a bit later
[09:18:14] <Steffanx> When you're ready for a beta release RikusW :)
[09:18:18] <Steffanx> ?
[09:18:49] <RikusW> that one is alpha ;)
[09:18:57] <RikusW> ISP and JTAG works
[09:19:05] <RikusW> except for stk600
[09:19:13] <RikusW> isp on stk600 might work
[09:27:15] <grummund> wirelessly is not a word?
[09:28:00] <grummund> if you connect two devices by wireless means, are they not wirelessly connected?
[09:28:10] <RikusW> isn't it ?!
[09:28:19] <grummund> LibreOffice seems not to think so.
[09:29:06] <RikusW> maybe its just technophobic ;)
[09:31:41] <vectory> maybe form of the adverb is the same as the adjective
[09:32:14] <vectory> fastly doesnt exits eutehr, but aspell is a crap spellchecker
[09:32:33] <vectory> tho, it would help my typing here ^^
[09:46:47] <inflex> I've yet to find a decent spell checking engine on *nix systems
[09:48:42] <_Shurik_> Chromium!
[09:50:15] <RikusW> yay, t10 breakout + adapter to abcminiuser's mkii works :)
[09:52:04] <inflex> course it does *tsk*
[09:52:10] * inflex has two of them from Tom :)
[09:52:32] <RikusW> inflex: I loaded it onto one of my 32u2 boards
[09:52:53] <RikusW> its the adapter wiring and breakout I'm talking about
[09:53:36] <RikusW> inflex: you're using linux ?
[09:53:45] <inflex> RikusW: yes
[09:53:59] <RikusW> look at the link I pasted above
[09:54:09] <RikusW> it will work with the mkii's ISP mode
[09:54:14] <RikusW> tested it
[09:54:37] <RikusW> download the RavrProg tbz file
[09:57:10] * RikusW is going for a walk, its 17:55 here
[10:28:52] <Landon> Tom_itx: just saw the recommendation, why TI in particular?
[12:11:35] <Tom_itx> Landon because they're close by
[12:27:59] <Landon> Tom_itx: yeah, I've been focusing on texas electronics companies for a while
[12:28:12] <Landon> and a couple of other midwest-y ones like motorola
[12:36:18] <mapee> hi
[12:36:18] <tobbor> mapee! like, totally tell us about the project!
[12:36:58] <mapee> pardon?
[12:43:20] <Steffanx> just ignore that bot mapee
[12:44:20] <mapee> ah, ok :)
[12:46:45] <Tom_itx> anybody ever try replacing the power connector on a laptop?
[12:48:12] <Tom_itx> gotta figure out if i can get one from digikey
[13:19:33] <vectory> CANNUCK
[13:19:45] <vectory> :(
[13:20:14] <vectory> so, whats the best way to clock an avr for use as rtc?
[13:20:16] <Steffanx> CANUCK
[13:20:17] <tobbor> SOMEONE MAKE HIM SHUT UP! or I'll cry.
[13:20:37] <Steffanx> 32.XXXkHz crystal?
[13:20:44] <vectory> got that
[13:20:57] <vectory> does it need to be filtered or resonated somehow
[13:21:08] <vectory> i see setups with 2 22pF caps
[13:22:06] <mrfrenzy> Tom_itx: you can often find them on ebay
[13:30:06] <Tom_itx> so i see
[14:18:20] <buzzsaw> I have a little brother that has been asking out mom for a robotics kit and shows interest in programming such. Is there a learning program that would work with teens or should I point her to the arduino platform?
[14:19:14] <buzzsaw> He is 13-14 and we live on the opposite side of the country so I cant sit down with him and show him things from scratch like I learned.
[14:20:15] <amews_aj> Maybe he could start with something a bit less low-level ?
[14:20:51] <buzzsaw> any ideas or suggestions?
[14:20:56] <RikusW> maybe learn C/C++ programming on a pc first
[14:21:04] <amews_aj> PICAXE ? Easy to get started, and has quite a few kits available..
[14:21:08] <RikusW> without any microcontrollers involved
[14:21:29] <keenerd> Mindstorms if you are made of money.
[14:21:46] <amews_aj> http://www.picaxe.com/Hardware/Robot-Kits/
[14:22:05] <buzzsaw> RikusW if he is anything like I was when he was younger I think he would get the hang of things more if there was something flashy ;-)
[14:23:05] <buzzsaw> but i guess c/c++ could be fun.
[14:23:11] <amews_aj> PICAXE is programmed in BASIC, and is well-documented. Still a concept of registers and such.
[14:23:24] <keenerd> Ugh, Mindstorms is still +$200. Thought it would have come down by now.
[14:23:43] <buzzsaw> lol surly not made of money :-)
[14:23:45] <RikusW> buzzsaw: something flashy will definitly be added incentive ;)
[14:24:36] <buzzsaw> I thought about teaching him a 3x3x3 led cube then go to a 3x3x3 rgb :-)
[14:25:01] <specing> buzzsaw: msps are cheap
[14:25:06] <specing> buzzsaw: launchpads
[14:25:16] <buzzsaw> msps?
[14:25:33] <specing> MSP430
[14:25:38] <specing> by texas instuments
[14:25:47] <buzzsaw> ah
[14:28:18] <buzzsaw> sweet, i will look at them and figure one out
[14:28:47] <buzzsaw> besides who needs a robot kit when you can just plug a bunch of crap into a mc ;-)
[15:41:38] <devcoder> what pcb cad software does everyone use
[15:41:58] <devcoder> i am using eagle at the moment but the board size limitation is getting annoying
[15:42:56] <devcoder> i've heard about kicad, seems pretty full featured, any thoughts on it?
[15:44:43] <keenerd> Works pretty nice. Better than the expensive Orcad software I used at a dayjob.
[15:49:13] <devcoder> how is the parts bin, quite extensive
[15:52:12] <karlp> I use eagel, but feel guilty and wish I used kicad...
[15:53:01] <keenerd> devcoder: Not as extensive as some programs, but the footprint editor is so slick. It look me just as long to draw a new footprint in Kicad as it did to look one up in Orcad.
[15:56:41] <karlp> footprints in kicad are text based right? you allegedly script create them from the datasheet "recommended lands" rather than having to draw them right?
[16:11:36] <jakllsch> not sure about kicad, but they are in "PCB"
[16:15:52] <devcoder> sounding good
[16:16:38] <devcoder> I like eagle, but the board size is sooo limiting. that is my only gripe with it
[16:17:29] <devcoder> need to make a board 7" long, 4" max allowed unless you get the professional. even freemium doesn't do it
[16:23:29] <ys0> oi
[21:58:14] <inflex> anyone have recommendations for sensor input isolation/protection into the AVR ADC's from car sensors?
[22:00:04] <ziph> inflex: You want to protect from overvoltage transients?
[22:00:26] <ziph> inflex: Or do you want galvanic isolation?
[22:01:59] <inflex> ziph: just wondering what's really required
[22:02:17] <inflex> all I know is that the sensors are currently 0~5V output... but I wonder...
[22:02:27] <inflex> atm I've got the usual zener protection
[22:05:04] <ziph> You could look at TVS diodes, but they're just faster.
[22:09:16] <Kevin`> faster can be made to not matter very much if you filter the signal to make it slower
[22:12:23] <ziph> Yeah, a capacitor plus the usual 2kV on the IC is sometimes all you need to get 15kV ratings.
[22:13:18] <ziph> From ESD, anyhow. continuous overvoltage is obviously a different story.
[22:13:27] <inflex> hrmm... most sensors are single-wire on a car aren't they?
[22:13:33] <inflex> eg, they all rely on the common/chassis ground
[22:15:35] <ziph> inflex: You couldn't get what you need from the CAN bus could you?
[22:16:08] <inflex> not in this case from what I know.... the person is monitoring 3rd party sensors
[23:15:58] <rue_house> canbus? stilll using canbus? use modbus!
[23:17:13] <Casper> inflex: got 66% efficiency with the smps from 31 to 5V 800mA :D
[23:17:23] <rue_house> cool
[23:17:26] <rue_house> what controller?
[23:17:29] <Casper> time to move to a better IC I guess now
[23:17:56] <Casper> sg3524
[23:18:05] <Casper> half bridge forward
[23:18:10] <rue_house> never heard of it
[23:18:11] <Casper> with salvaged parts
[23:18:18] <rue_house> forward converter!?
[23:18:22] <rue_house> ickgh
[23:18:25] <Casper> yup
[23:18:28] <Casper> half bridge too
[23:18:42] <rue_house> fet or bipolar?
[23:18:48] <Casper> bjt
[23:19:16] <rue_house> hmm, 31 is between the fet vs bipolar line
[23:19:21] <rue_house> might do better with a fet
[23:19:34] <rue_house> so synchronous rectification then?
[23:19:39] <Casper> no
[23:19:41] <Tom_itx> rue_house
[23:19:43] <Tom_itx> mail?
[23:19:57] <rue_house> oh, thougt thats what you mean by half bridge
[23:20:02] <rue_house> Tom_itx, brb
[23:20:07] <Casper> rue_house:
http://img100.imageshack.us/img100/2780/smpforward.png
[23:20:43] <Casper> tomorrow I'll do an higher current test and will check what heat up
[23:20:58] <Casper> tonight I made a short, so one of the transistor went boiling hot
[23:21:00] <Tom_itx> found 3 plugs priced from $25 to $15 to $2.80
[23:21:04] <Tom_itx> same plug
[23:28:51] <rue_house> ah its a TL494 knockoff
[23:57:09] <rue_house> and that not a forward, thats a pushpull
[23:57:36] <rue_house> forward is a flyback where they use the wrong part of the cycle for output
