Back
[02:32:05] <Nicholas`> Tom_itx: as you had assumed, backward compability does not work with Xilinx ISE, unfortunately.
[02:32:22] <Nicholas`> Looks like it's time for an XP vm
[02:32:39] <Nicholas`> The error message spat out from Xilinx is...
[02:32:41] <Nicholas`> "ERROR:ProjectMgmt:699 - The software version (13) is older than the version saved in the project file (14).
[02:32:54] <Nicholas`> Opening newer project with older software releases is not supported."
[03:38:46] <Tom_itx> so download 14
[03:47:25] <bezoka> hello
[04:38:42] <anton02> would you rather use a 6v or 12v DC motor to propel a 300 gram slow mobile RC car?
[05:47:47] <hetii> Hi;)
[06:05:24] <Lambda_Aurigae> morning hetii
[06:05:41] <Lambda_Aurigae> anton02, depends on what power supply I had available for said RC car.
[07:53:22] <hetii> Q: avrdude in version 6.0.1 have native support for linuxgpio or do I need compile it inside ?
[07:53:28] <hetii> hm
[08:15:50] <dufa> exit
[08:15:58] <dufa> uups
[09:15:57] <rue_more> Casper, ?
[09:21:42] <Casper> hi rue_more
[09:21:55] <Casper> do you do your own tax papers?
[09:22:22] <Casper> rue_more: who is elligible to that?
http://www.cra-arc.gc.ca/tx/ndvdls/tpcs/ncm-tx/rtrn/cmpltng/ddctns/lns409-485/440-eng.html
[09:22:37] <rue_more> tax forms use curcular references, I take it in
[09:22:55] <Casper> this tax software say I'm elligible, but I read the same thing in 2 different ways...
[09:23:08] <rue_more> quebec, its like canada but its not
[09:25:36] <rue_more> are you in quebec?
[09:25:56] <megal0maniac_afk> No
[09:27:27] <Casper> rue_more: yes
[10:16:46] <bezoka> hello
[10:16:57] <bezoka> is any website about avr tutorials etc?
[10:20:18] <megal0maniac_afk> Where the hell is beaky?
[10:25:49] <synic> this isn't a beaky channel, is it?
[10:26:47] <megal0maniac_afk> synic: It was, he hasn't been around in a while
[10:27:01] <Tom_itx> a good thing
[10:27:14] <synic> I thought he decided he hates avrs early and went the stm32 route
[10:27:24] <megal0maniac_afk> Last here on the 19th of Jan according to my logs
[10:27:46] <megal0maniac_afk> synic: Seems unlikely given how often he said "I love AVR"
[10:27:57] <megal0maniac_afk> He also said "I love SMPS"
[10:28:02] <megal0maniac_afk> And various other things
[10:28:03] <synic> but he says he loves stuff every 10 minutes
[10:30:08] <megal0maniac_afk> HAHAHA!
[10:30:27] <megal0maniac_afk> Okay I have to pastebin this
[10:30:51] <megal0maniac_afk> http://pastebin.com/fsSX6D4E
[10:31:29] <megal0maniac_afk> I removed other people mentioning it, but kept N1njaneer's comment because it seemed appropriate
[10:31:57] <synic> ok I'm going to do the same thing for #electronics, one sec
[10:33:32] <synic> wait, my logs are fugly
[10:33:34] <synic> https://gist.github.com/synic/ad4fa1441788dba7712f
[10:33:54] <megal0maniac_afk> Aw
[10:34:01] <megal0maniac_afk> That cuts deep
[10:34:23] <synic> https://gist.github.com/synic/ee9a33e9bf11eae0db63
[10:36:11] <megal0maniac_afk> That is incredible
[10:36:21] <megal0maniac_afk> I didn't even realise how bad it was, and I'm here all the time
[10:42:31] <megal0maniac_afk> My favourites: i love feet & ... no there are too many
[11:09:14] <megal0maniac_afk> i love relays they relay information for the good of humanity
[11:09:42] <synic> wow, I love feet
[11:09:45] <synic> that's awesome.
[11:56:02] <phinxy> does Solid State Relay get hot?
[11:56:13] <Tom_itx> yes
[11:56:20] <Tom_itx> if you overload it
[11:56:33] * Tom_itx introduces phinxy to heatsinks
[11:56:51] <phinxy> Do they generate heat when in both states?
[11:57:14] <Tom_itx> not when they're not under load
[12:24:33] <R0b0t1> Okay so double checking, I write a pin to be output and output low, but it stays high.
[12:24:42] <R0b0t1> Can I assume the device is damaged?
[12:24:47] <R0b0t1> No external pullups etc
[14:55:14] <bezoka>
[14:59:14] <Fleck> hi guys - I need a mistical circuit... voltages from 40V to 62V in input must show 0V to 5V in output (for avr - adc)... so that 51V would give me 2.5V and ADC read 512 for 10bit adc...
[15:09:26] <twnqx> misty indeed
[15:09:42] <Fleck> :)
[15:09:57] <twnqx> resistor divider + opamp for shifting, i would say
[15:10:34] <tzanger> yeah that's not mystical
[15:11:28] <Fleck> any principles for opamp shifting?
[15:11:49] <tzanger> scale the 22V down to 5V and subtract off whatever 40V is maybe
[15:13:10] <tzanger> principles? grab the forrest m mims engineer mini handbooks from your library or torrent
[15:13:17] <tzanger> one of them is op amp circuits
[15:13:34] <Fleck> :D
[15:14:28] <Fleck> http://www.scribd.com/doc/179043371/Forrest-M-Mims-III-Engineer-s-Mini-Notebook-en-Bookos-org ?
[15:14:30] <Fleck> this one?
[15:14:40] <tzanger> there are two operations you need to do... get 22V range to 0-5V range (that's a simple resistor scaling) and then subtract off the scaled 40V value so that your output will be 0-5
[15:15:02] <tzanger> that's environmental projects but yeah that series
[15:15:12] <tzanger> he also wrote "getting started in electronics" which was my bible growing up
[15:15:35] <tzanger> I had one copy that was covers missing, burned/torn pages, scribbling everywhere and then many years later I bought another copy which was pristine
[15:16:02] <tzanger> forrest m mims and robert grossblatt were probably my two childhood "heroes"
[15:16:43] <tzanger> don lancaster was too but a lot of what he wrote came off as written by some half crackpot who was obviously very smart but you weren't quite sure if he was conning you or not
[15:17:16] <tzanger> the art of electronics is what everyone tends to honk on about but I really did not like that book at all. way too math-y and theoretical
[15:17:18] <Fleck> https://kickass.to/forrest-mims-books-by-juribe610-t1409168.html << found it! :D
[15:18:25] <tzanger> those two/three guys though really got me started and on my way in this career. dad suplied the 30-in-one kit and the odd replacement part when I burned something out and I had a basement full of junk electronics to take apart and learn from
[15:20:59] <tzanger> http://www.forrestmims.com/
[15:29:41] <Fleck> tzanger: did not see opamp shifter in that book!
[15:31:12] <Fleck> http://im9.eu/picture/y32473
[15:33:26] <Fleck> which one is it tzanger? :D
[15:47:53] <w|zzy> http://www.fishpond.com.au/Books/Forrest-Mims-Engineers-Notebook-Forrest-M-Mims/9781878707031
[16:02:50] <xfebytes> does anyone know when they will begin sampling the SmartConnect Wi-Fi MCUs?
[16:08:08] <naquad> ermmm i have a very noob question. NRF24L01. can it work as both receiver and transmitter same time?
[16:10:05] <tzanger> basic noninverting amp will do scaling, and an op-amp subtractor will do the offset removal
[16:10:17] <tzanger> you can combine these into one as well
[16:10:39] <tzanger> oh it's right there
[16:10:41] <xfebytes> @naquad, they are transceivers. So yes, they can transmit and receive (not literally at the same time)
[16:10:43] <tzanger> difference amp
[16:11:04] <naquad> xfebytes, oh hell, so i've read datasheet correctly...
[16:11:32] <naquad> thanks for info
[16:12:26] <xfebytes> @naquad, no problem. If you do a quick google search you should be able to find tons of examples (both arduino and actual avrc)
[16:14:40] <naquad> i've found a library by Davide Gironi and from what i see he sets chip into one single mode
[16:14:43] <naquad> RX or TX
[16:14:52] <naquad> i expected to have duplex :(
[16:20:32] <xfebytes> @naqua, is it an nrf24L01 or an nrf24L01+?
[16:31:52] <FrankZappa> hey all. I don't know the vocabulary necessary to ask this question correctly. How would I go about creating a timer (for an ATmega328P) that has very high accuracy, and ticks of approx 2 milliseconds? Clock speed is set for me at 16 MHz. If you would like to reply with just a phrase I need to google, that would be great.
[16:34:43] <FrankZappa> hm, I've already found
http://www.seanet.com/~karllunt/interval.html
[16:41:59] <LoRez> you're going to need a clock that's some even multiple of 2ms
[16:42:21] <LoRez> (and hopefully divisible by some power of two to get to it)
[16:46:49] <FrankZappa> apparently not
[16:47:08] <FrankZappa> think I'll use ClocksPerSecond / MillisecondsPerSecond / Prescaler = InterruptsPerSecond
[16:47:40] <FrankZappa> using the last segment of a period as an offset or "trim" value seems to make sense
[16:47:54] <FrankZappa> unless someone here wants to tell me that's bad practice
[16:49:26] <FrankZappa> for a 20.000000MHz xtal this is (78 * 256 * 875) + (79 * 256 * 125) = 20000000 cycles
[16:51:16] <LoRez> that's not a timer, that's a timer plus a bunch of other shit.
[16:51:41] <FrankZappa> keeps better time
[17:05:28] <Tom_itx> Casper ping me when you're around
[17:38:15] <Fleck> tzanger:
http://yano.wasteonline.net/electronics/opamp/index.php?mode=1&vin=22&vout=5&rin=10000&rf= << I can't get REAL rezistor value there... you sure noninv opamp will do?
[17:41:19] <tzanger> Fleck: I misled you. you can't get a gain less than 1 with a non-inverting amplifier
[17:41:57] <Fleck> no problem, what should I use then?
[17:42:06] <tzanger> so you can either use an inverting amp which will work (then 5V = 40V and 0V = 62V, easy to flip in software) or use plain resistor division
[17:43:12] * Casper pings Tom_itx
[17:43:15] <Casper> I'm half there
[17:44:32] <Tom_itx> oh i'm fighting a virus on a pc
[17:44:45] <Tom_itx> i pulled the drive and am scanning it on another pc
[17:44:51] <Fleck> tzanger: hmm, how do I get two values? like... 0V (from 22V range) would be 5V and 22V will be 0V ?
[17:44:54] <Tom_itx> it won't let me boot at all
[17:45:27] <Tom_itx> takes over and pops up a splash screen warning
[17:45:32] <Fleck> tzanger: nah, I don't get it :D
[17:46:14] <tzanger> use the inverting amp solver
[17:46:44] <Fleck> Vin is what>?
[17:46:49] <Fleck> 22V ?
[17:46:59] <Fleck> http://yano.wasteonline.net/electronics/opamp/index.php?mode=0&vin=22&vout=5&rin=1000&rf= ?
[17:48:40] <Casper> Tom_itx: police virus?
[17:48:40] <tzanger> just use a resistor divider to get your 22:5 ratio and then subtract off the low end
[17:48:59] <Tom_itx> i dunno
[17:49:05] <tzanger> 40V scaled down by the same 4.4 factor will be about 9V
[17:50:06] <Tom_itx> pops up with a homeland security symbol on the page
[17:50:12] <Casper> Tom_itx: did you tried to boot in safemode with command prompt
[17:50:17] <Tom_itx> yeah
[17:50:18] <Casper> yeah that's the one I talk about
[17:50:19] <Tom_itx> no go
[17:50:38] <Tom_itx> what's the best way to get rid of it?
[17:50:48] <Tom_itx> right now i'm scannng the drive on another pc
[17:51:00] <Casper> I manually kill it
[17:51:09] <Tom_itx> how?
[17:51:35] <Casper> it usually hide in appdata/roaming
[17:51:45] <Casper> sometime in the user, sometime in the all user
[17:51:51] <Casper> which is programdata
[17:52:00] <Tom_itx> yeah, last time it was one user, this time it's all
[17:52:19] <Casper> it sometime hide in the startup menu too
[17:52:30] <Tom_itx> what to look for?
[17:53:05] <Casper> random character file, most probably hidden, 193k or 233k
[17:53:22] <Casper> also, a skype.something file
[17:53:42] <Casper> you may want to do a date search too
[17:53:55] <Tom_itx> yeah
[17:54:14] <Tom_itx> it was scanned yesterday so what happened was overnight
[17:54:21] <Tom_itx> at least i think
[17:54:43] <Tom_itx> it appeared when the email was opened
[17:54:50] <Casper> yeah...
[17:54:51] <Tom_itx> not read but just started
[17:55:00] <Casper> if you want, I can help you later tonight
[17:55:03] <Casper> I'm at work
[17:55:06] <Tom_itx> np
[17:55:11] <Tom_itx> i'll try a few things
[17:55:39] <Casper> hiren's boot cd is also usefull
[17:55:53] <Casper> it have a remote registry editor
[17:56:01] <Tom_itx> where do you get that?
[17:56:04] <Tom_itx> i'll google..
[17:56:21] <Tom_itx> i have a boot cd with a bunch of tools on it but it's pretty old
[17:56:58] <Casper> http://www.hirensbootcd.org/files/Hirens.BootCD.15.2.zip
[17:57:37] <Casper> I usually ends up checking the registry and find the path there
[17:58:54] <Tom_itx> malware has detected 5 things so far
[18:00:29] <Casper> mbam?
[18:01:20] <Tom_itx> yeah
[18:01:58] <Casper> what I used to clean it is combofix, but beware, it disable autorun
[18:02:06] <Casper> but I have a fix that I made to fix it
[18:02:31] <Tom_itx> i tried that but it started on the wrong drive
[18:02:52] <Casper> it scan all the drives... which is an issue
[18:02:55] <Tom_itx> i was tryin to run it on the other pc as the D drive
[18:03:02] <Tom_itx> i aborted it
[18:03:11] <Casper> it is not made to clean another install
[18:03:26] <Tom_itx> i had no way to boot into this one
[18:03:38] <Fleck> tzanger: that means I have to create constant 14.09V for Vin2 for difference amplifier and Vin2 will change with that 4.4 factor? so that Vin1 is 11.59V then difference would be 2.5V ?
[18:03:52] <Casper> you tried safemode and also the one with command prompt?
[18:03:58] <Tom_itx> yes
[18:03:58] <Fleck> err
[18:04:08] <Tom_itx> cmd promt just kept rebooting
[18:04:11] <Fleck> tzanger: that means I have to create constant 14.09V for Vin2 for difference amplifier and Vin1 will change with that 4.4 factor? so that if Vin1 is 11.59V then difference would be 2.5V ?
[18:04:24] <Casper> you got the heavy duty version
[18:04:30] <Casper> that one is fun
[18:04:33] <Tom_itx> gee
[18:04:46] <Tom_itx> well any tips are appreciated
[18:05:05] <Tom_itx> i can pull files and wipe the drive but i'd rather not
[18:08:54] <Tom_itx> just jumped to 42
[18:10:56] <Casper> I put some files on my usb key, they may come handy for you
[18:11:08] <Tom_itx> i have one as well
[18:11:19] <Tom_itx> with a few like rkill etc
[18:11:51] <Casper> but not safemode fixer :D
[18:11:53] <Tom_itx> finished the scan, i may try putting it back in the laptop
[18:13:12] <Tom_itx> is that in the zip you linked?
[18:13:23] <Casper> no
[18:13:34] <Casper> bbl, time to go home
[18:13:38] <Tom_itx> k
[18:39:25] <tzanger> Fleck:
https://www.circuitlab.com/circuit/75395g/scalediff/
[18:39:53] <tzanger> play with the scaling resistor networks R1/R2 and R8/R9 but that's the right idea
[18:40:48] <tzanger> the sim is a little goofy, but it works. R4/R5/R7 are just highish value resistors (all the same value, 10k would probably work fine too) and R6 represents the input impedance of the ADC. It's not critical
[18:41:24] <tzanger> pretty much any op amp you plug in there will work just fine to drive an ADC. you will notice some clipping at the rails even with rail-to-rail op-amps but it shouldn't be anything you can't handle
[18:47:50] <tzanger> 680 ohms is a little lwo for practical shunt resistance so multiply the scaling resitors by 10, 22k and 6k8 should be plenty and if you stick with 68k-100k for the input/feedback resistors for the opamp you shouldn't cause any weirdness with loading
[18:48:36] <tzanger> in theory you could dick around with teh feedback network, get rid of the scaling resistors and subtract 40V from the input and then scale it in the op-amp, but do you want to find and use high voltage op amps?
[18:52:21] <Valen> Ihigh voltage opamps are hard to find :-<
[18:52:34] <Valen> I would like a 100V amplifier
[18:53:58] <tzanger> well it's high common mode really not high output but you still need to feed the opamp > 80V here if you want to go that route
[18:54:23] <tzanger> still, for the cost of 4 1% resistors it's hardly worth trying to save.
[18:56:00] <Valen> lol crap, found a high voltage amp, I was hoping to use in place of 144 piezo drivers
[18:56:30] <Valen> its the same price as the piezo driver that comes with a 200 volt boost converter and a swath of other stuff
[18:57:35] <Valen> and sorting by supply voltage is *really* hard when its "1v to 18 volt"
[19:32:20] <Casper> Tom_itx: got success?
[19:33:15] <Tom_itx> it boots
[19:33:23] <Tom_itx> running mbam again
[19:33:46] <Tom_itx> there was a file i can't seem to find though
[19:33:58] <Tom_itx> c2s_service.exe
[19:34:05] <Tom_itx> in some skype subdir
[19:36:44] <Casper> try combofix too
[19:37:22] <Tom_itx> it keeps crashing
[19:38:19] <Tom_itx> what's tdsskiller?
[19:39:01] <Casper> rootkit killer
[19:39:12] <Casper> from kaspersky, do run it too
[19:39:37] <Casper> https://www.dropbox.com/s/4zpqky7f2mr4w4j/autorun%20fix.reg <=== that's my autorun fix for combofix
[19:40:53] <tzanger> Tom_itx: caught something?
[19:41:04] <Tom_itx> one of the laptops did
[19:41:10] <Tom_itx> not mine
[19:43:54] <Tom_itx> Casper, add all that back to the registry?
[19:44:01] <Casper> yup
[19:44:16] <Casper> that was actually fun to hunt down
[19:44:39] <Casper> dump registry on fresh windows, run combofix, dump, compare (thanks linux!)
[19:49:00] <rue_more> yay! my resistors arrived today
[19:49:16] <rue_more> now I have the 1k and 2k resistors I need for my 64 bit DAC
[19:49:23] <Casper> rue_more: the answer to the tax thing: I am elligible to it! :D
[19:49:30] <Casper> did you read the description? :D
[19:49:55] <rue_more> lots of times, but it was written in typical governish
[19:50:55] <Casper> it's a med tax refund
[19:51:07] <Casper> we pay both to canada and quebec
[19:51:16] <Casper> so canada refund part of that double pay...
[19:51:42] <rue_more> so if I used drugs in the first plae I might be abel to get something back?
[19:52:25] <rue_more> and my rail to rail amps arrived
[19:53:51] <Casper> I need to get some
[19:54:07] <rue_more> check out MCP6002
[19:54:55] <rue_more> I got 50 for $16
[20:19:44] <Casper> what speed and supply voltage? can it work single rail? is it also rail to rail output or just input?
[20:33:00] <rue_more> these are input and output
[20:33:06] <rue_more> I think they are just 1Mhz
[20:33:12] <rue_more> dual op-amp
[20:33:16] <rue_more> single rail
[20:33:42] <rue_more> http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en010434
[20:34:16] <rue_more> er low voltage, thats what I forgot
[20:34:32] <rue_more> mm, kinda wanted something 36V capable
[22:50:19] <anton02> is it safe to connect the txd and rxd of a TTL serial module directly to the rxd and txd of an amtega328? Will it draw too much current and damage the chip? If so, do you use a transisstor to prevent damage?
[22:50:42] <Casper> it is safe and actually the proper way
[22:51:26] <Casper> however, if you have code bug it can be an issue, a low value resistor can be used (like 220 ohms) to protect for such case
[22:51:53] <Casper> the bug would be: set the pin as output when it should be an input
[23:06:35] <anton02> Casper: so serial communucation devices can draw a fair bit of current on the txd and rxd?
[23:13:55] <Casper> no
[23:14:03] <Casper> basically zero current
[23:14:09] <Casper> unless you fuck up :D
[23:14:38] <Casper> an input is basically nothing
[23:14:53] <Casper> the input of an AVR is a few Mohms
[23:15:50] <Casper> your serial dongle may have a "many kohms" pulldown on it's input, typical is 10k or 22k, or even higher
[23:16:23] <Casper> that resistor is used to make sure that the input is at a known level if not connected, else it may switch at the nearest AM radio station frequency
[23:16:45] <Casper> the issue is really if you have 2 outputs connected together
[23:16:51] <anton02> Casper: if it's almost no current, then why would you want to use a transistor?
[23:16:59] <Casper> an output may only have a few ohms of impedance
[23:17:32] <Casper> so 2 of them connected together may allow lots of current to pass, more than it can handle, basically causing a short
[23:17:40] <Casper> and I didn't said a transistor, but a resistor
[23:19:15] <anton02> i see. Though i dont really see why lots of current would just happen to want to cross if the devices dont draw it.
[23:20:24] <Casper> the issue as I said if TX-TX
[23:21:35] <Casper> let's say a total resistance of 10 ohms... one is at 0V, the other at 5V, 0.5A would flow... but that can only happend if you screw up in your avr setup/code
[23:22:18] <Thrashbarg> even then you can't really source or sink 500mA from an AVR output pin :P
[23:23:24] <Casper> no, but enought to possibly damage something
[23:23:36] <Casper> maybe not short term, but long term it can
[23:23:38] <Thrashbarg> I thought the limit was 20mA or so
[23:23:40] <Thrashbarg> yeah
[23:24:02] <anton02> what do you mean long term
[23:24:04] <Casper> 40mA maximum absolute
[23:24:07] <Thrashbarg> ok
[23:24:11] <anton02> current is an instantaneous value
[23:24:20] <Casper> so 20mA for the 50% safety
[23:24:28] <Casper> but the avr can provide more
[23:24:48] <Casper> anton02: if you excede the current capability, it cause overheating
[23:25:00] <Thrashbarg> or blow the output pin in question
[23:25:13] <Casper> hence the possibly not immediate damage but can after a while (as heat build up)
[23:26:30] <anton02> how about from the battery to the VCC of the microchip? Should that also have a resistor?
[23:26:56] <Thrashbarg> no
[23:27:14] <anton02> low impedance wires though
[23:27:35] <Casper> supply pins should never be current limited
[23:27:40] <Casper> it will cause malfunction
[23:27:51] <Thrashbarg> you want as much current *available* to the AVR as possible so it can use it when it needs to
[23:28:10] <Casper> they are designed to be direct, in fact, you should put a capacitor there to make sure that all the resistance and trace impedance are "killed"
[23:28:26] <Valen> I have put a resistor in a micro power feed line
[23:28:28] <anton02> so I guess that particular pin of the chip can handle the current.
[23:28:41] <Valen> was going into a car so I put a 1 ohm resistor and diode in the power line
[23:28:48] <Valen> and a massive cap after it all
[23:29:17] <Valen> so the massive voltage spikes wouldn't make it into the chip
[23:29:23] <Valen> seems to have worked so far ;->
[23:29:26] <Thrashbarg> good
[23:29:35] <anton02> can using big caps make up for bad soldering
[23:30:14] <Valen> (literally this thing flashes an LED and its got a cap the size of my thumb in it) junk drawer engineering ftw
[23:30:22] <anton02> also, how big of a cap should i use for a microchips vcc?
[23:30:23] <Thrashbarg> heh
[23:31:53] <Valen> depends on lots of things
[23:32:15] <Casper> anton02: rules of thumb: you never current limit except some exception. The main exception is led and transistor base
[23:32:26] <Casper> anton02: cap to not fix bad soldering<
[23:32:35] <anton02> in soldering something i think may be true, but im not completely sure is that solder moves onto a wire more readily if you first heat the wire up. would a hot wire OR a cold wire + flux be more accepting of solder?
[23:32:45] <Casper> anton02: cap on the VCC of the chip should be 0.01-0.1uF ceramic
[23:33:18] <anton02> what if the chip has a dedicated button battery? should i still use a cap?
[23:34:00] <Casper> anton02: yes, every chip should have it's own capacitor, one per VCC pin
[23:34:30] <anton02> Casper: just out of curiousity, how come?
[23:34:31] <Valen> button battery is just for clock and nvram as a rule
[23:34:32] <Casper> for soldering, always heat the pin AND the pad at the same time, apply the solder on the other side of the wire
[23:34:45] <Valen> everything needs to be hot to solder
[23:35:26] <Valen> if you have something big and something small heat the big thing up, wet the iron with a bit of solder first then heat the big thing then touch it to the small thing and add a bit more solder
[23:35:28] <Casper> anton02: the avr is made of mosfet, a mosfet gate is basically a capacitor, when you try to charge that gate it basically drain an huge current for a tiny time
[23:35:48] <Casper> that small capacitor on the VCC pin ensure that the voltage do not drop too much, which would cause a malfunction
[23:36:23] <anton02> but why would the voltage ever drop? Because batteries are imperfect?
[23:36:42] <Casper> that, and wire/trace impedance
[23:38:01] <Valen> when the chip clocks the current draw can be very high
[23:38:05] <Valen> in the amps range
[23:38:11] <Valen> but on average its low
[23:38:58] <anton02> so im guessing i should put the cap close to the chips vcc? that way wire impedance between cap and chip is reduced
[23:39:07] <Valen> yes
[23:39:09] <Casper> impedance = resistance + inductance... inductance is an inductor, what is an inductor? generally a coiled piece of wire. why coiled? because it concentrate the magnetic field that is generated by the wire, causing it to get "magnified" more. Any wire cause a magnetic field as soon as current flow. That cause a difficulty to change the current flowing
[23:39:25] <Casper> capacitors resist the change in voltage, inductor resist change in current
[23:39:41] <Valen> in the old days with socketed chips it wasn't uncommon to have the decoupling cap actually under the chip itself soldered directly to the socket
[23:40:07] <Casper> and yes, cap should be as close as possible to the chip, which is another reason why smd caps are more and more popular, even for thru-hole chip
[23:40:13] <Casper> they can be put between 2 pins
[23:40:30] <Valen> they have better charaistics as well, lower ESR and inductance
[23:41:04] <Casper> Valen: fortunatelly, today's caps are better, so no need to be as close, and chips are getting better
[23:41:37] <Casper> heck, intel next generation of cpu is said to have build in caps for the supply, with build in inductors and power mosfet
[23:41:48] <Valen> todays chips run so much faster its still an issue
[23:41:50] <Casper> basically, it will be the return of "12V cpu" :D
[23:42:03] <Valen> thats interesting, I do like that
[23:42:08] <Casper> yeah
[23:42:24] <Casper> they claim an huge increase in efficiency
[23:42:30] <Casper> in the order of like 30%
[23:42:37] <Valen> the cpus today are on a carrier board with a bunch of caps on it
[23:42:52] <Valen> wouldn't have expected that
[23:43:13] <Casper> since there will be many independant smps, they can fine tune the voltage of each, and even shutdown those that ain't needed
[23:43:48] <Casper> like if the os use only a single core, it can literally powerdown the other cores
[23:44:14] <Valen> they do that now
[23:44:18] <Valen> thats just on/off
[23:44:40] <Valen> http://en.wikipedia.org/wiki/Haswell_(microarchitecture)
[23:44:44] <Valen> Fully integrated voltage regulator (FIVR), thereby moving some of the components from motherboard onto the CPU.[41][42][43]
[23:45:44] <Valen> http://hothardware.com/News/Haswell-Takes-A-Major-Step-Forward-Integrates-Voltage-Regulator/
[23:46:17] <Casper> yeah read about that one
[23:46:23] <Casper> but the board using them still have regulators
[23:46:32] <Casper> so it's not fully implemented
[23:46:50] <Casper> but I'm expecting the next gen to not have reg at all on the board