#avr Logs

Apr 17 2018

12:08 AM eszett: if trump is the madman, why do the other presidents as May and Macron play the same Rocket-Game?
04:30 AM nuxil: mornings
04:35 AM nuxil: so. i was thinking about my code https://pastebin.com/x2ZQb7pw and why it would do the "counter timer" like 15 sec faster than what it should do.
04:35 AM nuxil: I did not to consider the +-10% tolerance of the internal Rc. rechecking the math with some tolerance to see the values.
04:35 AM nuxil: +10% : 1100000 / 1024 / 97 -> 11,074 hz -> 0,090298 | 0,0902.. * 10 * 500 -> 451,49 seconds
04:35 AM nuxil: 0% : 1000000 / 1024 / 97 -> 10,067 hz -> 0,099328 | 0,0993.. * 10 * 500 -> 496,64 seconds
04:36 AM nuxil: -10% : 900000 / 1027 / 97 -> 9,0608 hz -> 0,1103644 | 0,1103.. * 10 * 500 -> 551,82 seconds
04:36 AM nuxil: man.. thats alot of seconds
04:37 AM nuxil: im really starting to hate the internal rc :p
04:39 AM Thrashbarg: ouch
04:41 AM nuxil: yea. not good
04:44 AM Thrashbarg: stick an antenna on it and sync it to mains noise :P
04:44 AM nuxil: :D
04:45 AM polprog: Thrashbarg: thats is not a stupid idea
04:45 AM nuxil: that might actually be more precise
04:45 AM polprog: iirc 50hz mains in very stable
04:45 AM Thrashbarg: yup
04:45 AM Thrashbarg: needs to be
04:45 AM polprog: or 49.999999hz in europe /s
04:45 AM nuxil: polprog, clocks in the old days used that trick.
04:45 AM Thrashbarg: hehe
04:45 AM polprog: i know ;)
04:45 AM Thrashbarg: and still do
04:45 AM Thrashbarg: like on appliances, ovens, etc
04:46 AM polprog: or slap a watch crystal on it
04:46 AM Thrashbarg: yer
04:46 AM nuxil: its a atiny13
04:46 AM polprog: oh
04:46 AM polprog: right
04:47 AM nuxil: might use the CLKI tho
04:47 AM nuxil: *can
04:55 AM day: polprog: except in europe. where it wasnt stable recently
04:56 AM nuxil: ?
04:56 AM nuxil: we never had a issue with mains freq here in norway.
04:56 AM nuxil: its always been 50hz and stable.
04:57 AM day: nuxil: dunno which countries were affected. it was a few month ago. there was a production shortage which caused some frequency issues. big enough to screw clocks over
05:00 AM day: nuxil: https://arstechnica.com/tech-policy/2018/03/ovens-across-europe-display-the-wrong-time-due-to-a-serbia-kosovo-grid-dispute/
05:00 AM day: last month actually
05:00 AM nuxil: hehe. ok
05:04 AM nohitzwork: commodore 64 uses mains 50hz for Time of Day clock input
05:04 AM Thrashbarg: yup
05:05 AM Thrashbarg: problem is for the TOD counter to work it needs 50/60Hz, so that makes it difficult to make a switch mode power supply for them
05:06 AM Thrashbarg: hell the damn thing uses a voltage doubler so it needs AC for that regardless
05:06 AM nohitzwork: http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/251138-2of2.gif its in the low middle
05:06 AM Thrashbarg: yeah
05:07 AM Thrashbarg: the CIAs have a TOD counter, but C90 there boosts the 9VAC up to something the 12V regulator can use, piggybacked off 5V
05:07 AM Thrashbarg: works fine with linear supplies
05:09 AM nuxil: so here is an idea. can i calculate the real F_cpu by sitting and timing my circuit with a stopwatch ? since i use the formula : f_timer = (f_cpu / prescalar / counter), or f_timer = f_cpu / (prescalar * counter)
05:09 AM nuxil: i can rearrange it so i get. f_cpu = t_timer * prescalar * counter. and knowing the values.
05:09 AM nuxil: example it uses 7 min 56 sec instead of 8 min 16 sec as it should, 7min 56 sec -> 476 secs.
05:09 AM nuxil: 476 / 10 / 500 -> 0.0952 sec per interrupt. 1 / 0.0952 = 10.50420.. Hz
05:09 AM nuxil: f_cpu = 10.50420 * 1024 * 97 -> 1043361 Hz.
05:09 AM nuxil: using a stopwatch i can react within 1sec if im real slow. 1 sec out of 476 sec is like 0.2% error.
05:09 AM nuxil: does this sound sound?
05:09 AM nuxil: ermn
05:10 AM Thrashbarg: anything stopping you from using a crystal? There are two crystal inputs on some ATmegas, one for CPU and one for a 32kHz RTC
05:10 AM nuxil: atiny13
05:10 AM Thrashbarg: I'm sure you're aware of that :P
05:10 AM Thrashbarg: yeah hmm
05:20 AM nuxil: im gonna try and time it a couple of times. then put in new f_cpu value to code and see if it works
05:20 AM Thrashbarg: there's an RC clock calibration register I believe
05:21 AM Thrashbarg: should be on the ATtiny
05:21 AM nuxil: user calibration
05:21 AM nuxil: yea
05:21 AM nuxil: wait what..
05:22 AM nuxil: a register.
05:22 AM Thrashbarg: to specify the RC frequency pull
05:22 AM Thrashbarg: in software
05:22 AM polprog: day: hence my joke
05:23 AM Thrashbarg: nuxil: have a look at the OSCCAL register
05:24 AM nuxil: i thought you calibrated it using temp and voltage..
05:24 AM nuxil: didnt know there was a register aswell
05:24 AM * polprog heats up nuxil's attiny with a hairdryer
05:25 AM Thrashbarg: heh
05:25 AM polprog: * clocking intensifies *
05:25 AM Thrashbarg: herhe
05:25 AM Thrashbarg: *hehe
05:26 AM polprog: it seems to have a CLKI input, not sure what that does
05:26 AM nuxil: i have no clock source to give it.
05:27 AM nuxil: i got a couple of crystals tho. not sure if they can be used
05:27 AM Thrashbarg: polprog: need to set fuses to make it a clock input, but it drives the processor
05:27 AM polprog: ah
05:27 AM polprog: maybe a cd4060 and a crystal? or a 4 pinned crystal generator?
05:28 AM polprog: http://www.hackersbench.com/Projects/1Hz/schematic.jpg
05:28 AM polprog: hmm hmm this looks cool
05:29 AM polprog: could make a small frequency box with it :)
05:29 AM Thrashbarg: hehe
05:29 AM polprog: ive got some watch crystals and some unknown condition crystals that i took out of boards and stuff
05:29 AM Thrashbarg: yeah
05:29 AM Thrashbarg: that crystal oscillator looks pretty sensible
05:30 AM polprog: im sure rue had a schem of one on his server
05:30 AM polprog: one that uses a cd4060 only
05:33 AM MrFahrenheit: when you google "halloumi without mint", and google spits in your face and the first 5 results are literally "halloumi with mint"
05:35 AM nuxil: Thrashbarg, can you explain how does this OSCCAL register work? i dont get it. it says i should (de|in)crement it only by 0x20 per step. but what is this "Typical Lowest Frequency
05:35 AM nuxil: with Respect to Nominal Frequency" and "Typical Highest Frequency
05:35 AM nuxil: with Respect to Nominal Frequency" ? with osccal value 0 its 50% and 100% and so on.
05:35 AM nuxil: erg.
05:35 AM theBear: and this is why we been using only duckduckgo the last few years... because it either stays an equal level of usefulness, OR improves in usefulness, over any given period of time
05:36 AM MrFahrenheit: google is REALLY bad lately
05:36 AM polprog: true
05:36 AM MrFahrenheit: always thinking it knows better than you
05:37 AM Thrashbarg: nuxil: so you'd have the nominal frequency is what it should run at, the lowest and highest frequencies are how much it can drift from that nominal frequency with the register setting
05:38 AM Thrashbarg: so if you set it to 0x00 it'll be as low as 50% of the nominal frequency, or as high as 100% the nominal frequency
05:38 AM Thrashbarg: the idea is to fiddle with it until it's set right
05:38 AM theBear: mmm it's true, and lately is stretched past a length of time best represented in years, cos 20+ months just sounds silly ;-)
05:39 AM Thrashbarg: and THANKS KiCad library person for specifying all the output pins on these 74xx IC's as inputs and making the DRC barf
05:41 AM Thrashbarg: argh even the 74LS93 has GND specified as an input... how'd they expect that to work
05:48 AM polprog: what i hate about 74xx lib on kicad is that they all have a hidden gnd and vcc pins. what if i need to connect them to another net?
05:48 AM MrFahrenheit: you can unhide them iirc
05:48 AM polprog: yeah but a separate power block for multi part ics would be way better
05:49 AM polprog: im talking about opamps too
05:49 AM MrFahrenheit: multipart blocks are odd, there's no clean way to handle those in schematics
05:49 AM polprog: my schem is all covered in +V and -V labels cos every amp has them
05:49 AM polprog: good point
05:49 AM MrFahrenheit: having 2 pin blocks everywhere just looks odd
05:49 AM Haohmaru: polprog you can do that.. if you make your own symbol
05:50 AM polprog: i know
05:50 AM polprog: bbl
05:51 AM Haohmaru: MrFahrenheit it looks odd in eagle schematics i've seen, because iirc it doesn't indicate who's pins those are
05:51 AM Haohmaru: in kicad it's a unit, so it has the refdes (IC14) and value (LM358N) together with it
05:52 AM Thrashbarg: polprog: I make a copy of the library and edit the components I need as I go ... done it before :/
05:52 AM Haohmaru: i've already made a few multi-unit symbols like that, opamps, logic gates..
05:52 AM polprog: hmm
05:53 AM polprog: good idea
05:53 AM Haohmaru: a better idea is to just clone the symbol you need into your own library, modify it there
06:07 AM Haohmaru: crap, i need some preprocessor dark magic glue
06:08 AM Haohmaru: i need to glue together USARTC0 with _DRE_vect to form USARTC0_DRE_vect, but USARTC0 comes from a preprocessor definition
06:42 AM polprog: reference says iw would be #define USART_VCT(name) ##name_DRE_VECT
06:42 AM polprog: i think
06:42 AM polprog: it would be *
06:43 AM polprog: and then youd do USART_VCT(SOMETHING)
06:44 AM cehteh: eh ##foo is pretty pointless
06:44 AM cehteh: sure it isnt name##_DRE_VECT
06:45 AM Haohmaru: yes, i managed to concatenate USARTC0 and _DRE_vect together
06:45 AM Haohmaru: into a macro
06:45 AM cehteh: and when you want to evaluate the 'name' part then you need some poxying
06:45 AM Haohmaru: no
06:46 AM Haohmaru: i want to use it in the ISR(here)
06:46 AM Haohmaru: #define WUB = concat(WHICH_USART,_DRE_vect)
06:46 AM Haohmaru: ISR(WUB)
06:47 AM cehteh: i mean when you have something like #define foo bar #define concat(a,b) a##b then call concat(foo,foo)
06:47 AM Haohmaru: what i get is a warning that i am probably incorrectly defining a vector with a name WUB
06:47 AM cehteh: that gives foofoo not barbar
06:48 AM Haohmaru: so i need the concatenated thing (which is held in WUB) to be evaluated
06:48 AM cehteh: #define concat(a,b) concat_proxy(a,b) #define concat_proxy(a,b) a##b
06:49 AM cehteh: just notice, maybe unrelated for you now
06:51 AM cehteh: for ISR's iirc gcc does some magic, one needs to dig deeper into that to figure the details about the warnings out
06:51 AM Haohmaru: it's clear..
06:51 AM Haohmaru: WUB doesn't evaluate
06:51 AM Haohmaru: because "ISR()" itself is a macro
06:52 AM Haohmaru: the final result is an extern "C" declaration of a void __vector61() function with interrupt attributes
06:53 AM Haohmaru: my broken macro turns into an extern "C" declaration of a void WUB()
06:53 AM Haohmaru: and avrgcc knows (either hardcoded, or from some linker scripts or whatever) that this is probably not valid
06:54 AM cehteh: yes something along that
06:55 AM Haohmaru: the io.h header for my chip declares USARTC0_DRE_vect as __vector61 (again thru some macro functions)
06:55 AM cehteh: does gcc still uses .spec files somewhere to drive some details?
06:55 AM Haohmaru: *shrug*
06:55 AM cehteh: dunno much about the details nowadays
07:51 AM Haohmaru: i blame Emil
07:51 AM * Haohmaru runs around waving hangs irresponsibly
07:51 AM Haohmaru: * hands
08:10 AM polprog: hang responsibly
08:11 AM * Haohmaru zaps polprog with 3.3 volts
08:11 AM Haohmaru: >:(
08:12 AM polprog: what amperage?
08:12 AM Haohmaru: i won't tell you
08:12 AM * polprog dies just in case
08:12 AM Haohmaru: crap
08:12 AM * Haohmaru hides polprog under the rug
08:22 AM * polprog moves rag, opens trapdoor
08:25 AM Haohmaru: >:(
08:25 AM * Haohmaru puts a grand piano on top
09:23 AM dStruct: polprog: I finally got all the kinks worked out, and it works perfectly, very stable clock and logic signals
09:23 AM polprog: good job :)
09:24 AM dStruct: polprog: well you helped quite a bit, so I say thank you kindly :D
09:24 AM polprog: does it generate the data by itself or it has to measure stuff or get it gr
09:24 AM polprog: from a computer*?
09:24 AM polprog: hah thanks :P
09:25 AM dStruct: polprog: so I'm taking in 20-bits of 10vdc strobing binary coded decimal, and converting that into an array (still TBD, needs level shifting down to 5v) and then serializes it and streams it out based on the 10khz clock
09:25 AM dStruct: polprog: it's actually coming in from an I/O system being driven by an Encore/SEL/Gould 32/67 mainframe lol
09:25 AM polprog: is it a paralell input or serial?
09:26 AM dStruct: parallel and it's going to have to go through a shift reg
09:26 AM polprog: wow that mainframe is this kind of a fridge-style one
09:26 AM polprog: hehe.
09:26 AM dStruct: polprog: so I need to do a few things, level shift down to 5, build a shift reg capable of doing 20-bits, and also build protection diode/level shift up to 12v on the output side
09:29 AM polprog: i think you can use a max232 for the level shifting
09:29 AM polprog: if the high voltage side is paralell
09:29 AM dStruct: polprog: it's about 5.5ft tall weighs about a billion pounds, and draws enough power to light a small city heh
09:29 AM dStruct: the incoming BCD bits are +/-10vdc but in this case they would only ever be 0 or +10
09:29 AM polprog: single max232 has 2 inputs and two outputs so i think the strobe signal can be converted as well (if i'm imagining this correctly)
09:29 AM dStruct: I was thinking a 74hc series possibly, 8-bit, x3 of them
09:29 AM dStruct: now if I could find a shift reg that would accept 10-15vdc on the input side and deal with the voltage shifting that would be ideal but I don't think it exists
09:31 AM polprog: yeah but the 10v data, is it serial or paralell? why do you need a shift reg?
09:32 AM dStruct: it's parallel, so I have 20 outputs from an I/O system feeding all 20 into an analog card currently, where it's shifting and NAND gating it to make this signal conversion the old school way
09:33 AM polprog: i see
09:33 AM polprog: i remember there was some kind of a single mosfet logic level limiter
09:34 AM dStruct: so currently 20-bits come in, shift down to 5v, bit shifted into single serial output, and then lots of NAND gating to build the sync and then all 3 are combined analog clock, sync, and this data line to make the final output, then it all gets shifted up to 12v
09:34 AM polprog: whoa wait a sec
09:35 AM polprog: is it some kind of a recognized protocol? so i could look at a pdf of some kind
09:35 AM polprog: i mean interface not protocol
09:35 AM dStruct: polprog: http://www.cems.uwe.ac.uk/~a2-lenz/n-gunton/worksheets/ARINC-429.pdf look for ARINC 568
09:36 AM polprog: ah this
09:38 AM dStruct: it's a little confusing but it shares some stuff from 429, but uses the 568 wiring and signalling format
09:43 AM dStruct: the input side and shift register stuff is pretty straight forward, however the output side I'm trying to figure out the best way to get clean fast 12v switching without losing voltage or drawing too much current, not to mention preventing reverse current, I have a couple 2N2222 which might work
10:17 AM nuxil: Thrashbarg, i saw a video about the osccal register, he had setup his fuses so that he had a the clock output. hooked that up to his scope. read the current osccal value and changed it a litte. he could then tweek it to get it rather precise. now. how is this suppose to be done with atiny13 since it doesnt have a clock output. except the sclk for spi
10:18 AM rue_bed: dStruct ULN2803
10:29 AM dStruct: rue_bed: thanks for the suggestion, that looks good
11:08 AM MrFahrenheit: halloumi is almost done, needs to cook for a bit now, then it's getting salted
11:13 AM Thrashbarg: awesome
11:14 AM polprog: nice
12:47 PM MrFahrenheit: halloumi done and fried up https://www.dropbox.com/s/a90vua02v6siasv/2018-04-17%2019.24.53.jpg?dl=0
12:48 PM Thrashbarg: mmm excellent
12:48 PM Thrashbarg: got lemons?
12:49 PM MrFahrenheit: life gave me lemons, yes
12:49 PM Thrashbarg: halloumi + lemons I mean
12:49 PM Thrashbarg: juice
12:50 PM MrFahrenheit: I didn't use lemons, it was garlic + cream chicken and fried mushrooms
12:50 PM Thrashbarg: nice
12:50 PM MrFahrenheit: this was my first time trying halloumi, actually
12:50 PM Thrashbarg: and?
12:50 PM MrFahrenheit: gotta say, I expected better
12:50 PM Thrashbarg: aw
12:51 PM MrFahrenheit: people built it up too much, it couldn't live up to my expectations
12:51 PM Thrashbarg: I see
12:51 PM Thrashbarg: now I'm hungry bah
12:51 PM * Thrashbarg rummages
12:52 PM MrFahrenheit: I like cheeses with a stronger taste, so it was very underwhelming for me
12:52 PM MrFahrenheit: maybe if I added lipase and brined it for a month it would be more flavourful
12:52 PM polprog: looks like pumped up pierogies. mmmm
12:53 PM polprog: this analog devices filter 101 is pretty nice, maybe up to the sallen key filter where he explains it simply but then he jumps into all those funky filter types and higher maths that i have absolutely no idea about
12:53 PM polprog: but it looks like a sallen key filter is a nice thing
12:54 PM MrFahrenheit: sorry I didn't work on my lib today, polprog
12:54 PM polprog: oh
12:54 PM polprog: well i didnt work on my mixer today as well ;P - got other stuff to do
12:55 PM polprog: i wish i would actually.. ill email another seller for the blueprints of a chassis i like
12:55 PM polprog: so i can order it (cheaper - 20pln vs 60 pln / piece) and make a PCB for my mixer so it fits
12:55 PM MrFahrenheit: nuxil, convince polprog to get a 3d printer
12:56 PM polprog: no i dont like them
12:56 PM MrFahrenheit: any case you want...
12:56 PM polprog: i found a nice box-and-lid enclosure, aluminum, i will just replace the lid with my PCB faceplate
12:56 PM polprog: already drilled and labeled
12:57 PM polprog: maybe gonna spraypaint the aluminum chassis
12:57 PM polprog: ive got 4 colors of spray,
12:57 PM MrFahrenheit: I recently got one of those tiny aluminium enclosures, the extruded ones - very small
12:57 PM MrFahrenheit: was thinking of using it to make a diy ovenised crystal chamber
12:58 PM polprog: glossy white, glossy black and construction site spray in highlighter pink color
12:58 PM polprog: ooo nice
12:59 PM polprog: ovened oscillator :D
12:59 PM MrFahrenheit: ocxo
12:59 PM polprog: so i was thinking about painting the mixer case in some funky way
12:59 PM polprog: i could really use a black glossy PCB and maybe add a white strip on it on the silk screen and then continue that strip by painting it on with spray on the box itself
01:00 PM MrFahrenheit: can't find info on best practices for them though, like what kind of temperature control is good enough, should the control circuit also be in the oven or not...
01:00 PM polprog: what do you think
01:00 PM MrFahrenheit: check out the subreddit on diy stomp boxes
01:00 PM MrFahrenheit: they do stuff like that all the time, but it's a process
01:01 PM MrFahrenheit: https://www.reddit.com/r/diypedals/
01:02 PM polprog: they look cool :D
01:02 PM polprog: some of them like Zero Cool's laptop
01:02 PM MrFahrenheit: I think most of them are etched
01:04 PM polprog: im was going to use sideways mounted pots, just screwed onto the faceplate, and connect them with wires to headers on the PCB
01:04 PM polprog: maybe add some washers on the inside to distribute the force
01:05 PM polprog: i need to replace the on/off switches, i could even make them as momentary and use flipflops to toggle them
01:05 PM polprog: but i wanted to have tactile ones that you see they are either pushed in or out
01:05 PM MrFahrenheit: use toggle switches
01:05 PM polprog: yeah
01:06 PM polprog: the problem is that those i've got are pretty huge (but they have builtin backlight)
01:06 PM MrFahrenheit: show me
01:06 PM polprog: just a sec
01:07 PM MrFahrenheit: never seen ones with a light
01:09 PM polprog: https://puu.sh/A5azQ/f99ae326d4.jpg switches
01:09 PM polprog: and this is the general panel layout https://puu.sh/A5azw/bcd08fbf70.jpg
01:10 PM polprog: of the parts are upside down because they didnt want to stand still
01:10 PM polprog: and the rightmost pot will be the same as the two, just needed a prop.
01:10 PM polprog: its two tracks stereo mixer so it uses dual gang pots
01:11 PM MrFahrenheit: I didn't mean those, I meant these https://www.cordlessrenovations.com/wp-content/uploads/2013/06/SPDT-Mini-Toggle-Switch-1.jpg
01:11 PM polprog: those lever ones?
01:11 PM polprog: nah
01:11 PM polprog: or maybe....
01:11 PM MrFahrenheit: they're called toggle switches
01:11 PM polprog: ikr
01:12 PM polprog: most mixing consoles ive seen use pushbuttons
01:12 PM MrFahrenheit: toggle ftw
01:12 PM polprog: i guess its an idea
01:12 PM MrFahrenheit: makes it seem like you're operating an apollo lander
01:12 PM polprog: hmm true
01:12 PM polprog: i must make sure to make two pieces
01:13 PM Tom_L: nasa approved
01:13 PM MrFahrenheit: you can even get ones with a cap
01:13 PM polprog: i know, i think i have some
01:14 PM MrFahrenheit: do you use them to launch missiles?
01:16 PM Tom_L: naw, that takes a key
01:16 PM polprog: no, it takes two keys
01:16 PM polprog: actually i have a big red button and a key switch
01:17 PM polprog: and beat one chassis up
01:17 PM polprog: wait
01:17 PM polprog: i wanted to say i should make two pieces and beat one chassis up before its assembled so it looks older
01:21 PM polprog: https://puu.sh/A5b2t/6978c9eb68.jpg
01:23 PM polprog: i should add a key hole labeled "COUNTRY MUSIC ENABLE"
01:23 PM polprog: and some danger signs around it
01:23 PM MrFahrenheit: https://potato.io/
01:24 PM polprog: and an AVR that would play "Cotton Eye Joe" with the DAC, and inject it into master
01:25 PM polprog: https://www.youtube.com/watch?v=uK_7AqH1VGQ [this is actually cool]
01:33 PM nohitzzz: what is this
01:33 PM nohitzzz: is this yours
01:33 PM nohitzzz: <polprog>https://puu.sh/A5b2t/6978c9eb68.jpg
01:35 PM polprog: yeah
01:35 PM polprog: junk drawer dump
01:35 PM nohitzzz: ok
02:07 PM MrFahrenheit: ugh, my 3d printer tracking stopped 2 days ago and the last 2 updates are useless... "Delivered by air transport"
02:08 PM MrFahrenheit: delivered where, and to whom? :@
02:28 PM polprog: nohitzzz: why did you ask?
02:52 PM polprog: https://www.youtube.com/watch?v=C8KGMa6m8qE
02:53 PM polprog: hehe attiny
02:53 PM polprog: 5:30
03:00 PM nohitzzz: no reason
03:03 PM nohitzzz: but you are building some sort of audio device?
03:06 PM polprog: yeah
03:06 PM polprog: a small audio mixer
03:06 PM polprog: just to get into the world of analog voodoo
03:07 PM MrFahrenheit: that unboxing
03:08 PM polprog: ave is awesome. no bullshit
03:08 PM MrFahrenheit: in small doses, yes
03:08 PM polprog: i need to get that bench mat he has
03:09 PM polprog: something like that
03:09 PM MrFahrenheit: get an antistatic one
03:12 PM polprog: yeah probbaly
03:12 PM polprog: i wonderhow expensive they are
03:12 PM MrFahrenheit: I got one that's about a2 sized
03:13 PM MrFahrenheit: never get around to connect it to ground though
03:22 PM polprog: i should get a cordless drill instead
03:22 PM polprog: to tired of taking out the wired skill drill
03:22 PM polprog: maybe the only plus of the skill drill is that it's 1200W
03:40 PM MrFahrenheit: I think the wiki page on aluminium electrolytic caps was written by a capacitor manufacturer
03:40 PM MrFahrenheit: Aluminum electrolytic capacitors with non-solid electrolytes have, in terms of quality, a relatively negative public image. This is contrary to industrial experience, where electrolytic capacitors are considered to be reliable components if used within their specified specifications during the calculated lifetime. The negative public image might be, among other reasons, because failed electrolytic capacitors in devices are easily and
03:40 PM MrFahrenheit: immediately visible.[62] This is exceptional and not the case with other electronic components.
04:21 PM polprog: goodnight
04:23 PM * Tom_L checks the sun still in the sky..
11:03 PM polprog: morning
11:27 PM day__ is now known as day