lordpil's blog

omg panel pricing is sweet tho
like, it was maybe $80-$180 in tooling charges, than 10% or 15% that in per board cost
thats so much pcb!

Posted by renesis at 14:42 | permalink | 0 comments

i thought all my curves and 8mil drawn pours on the big panels was making their NC machine do nothing for longer than they liked
dud just didnt wanna do the gcode for a non matrixed single board panel
im like cmon, you no CNC ninja
like nothing happened until im like, um i have a CNC in my room i can cut it out myself
when they said 'factory says too hard' i thought it was like, a real problem
okay i guess ima take a nap and maybe when i wake up, i have new quote
under $180 shipped would be sweet
(shipping is $40, is for 2 each (their min) of two little panels)

Posted by renesis at 14:37 | permalink | 0 comments

dunno, but probably
dont always need both sides with modulated RF stuff
yeh russia and china are cheap
still new tubes even
kevtris: =O
how did you get it?
what you gonna do jack a tower
okay well china pcb fails at being on time, i can chill the fuck out
haha because of the board cutouts wtf

Posted by renesis at 14:32 | permalink | 0 comments

get some pipe hittin mofos with pliers to protect your relay
22:25 <@kevtris> I had a 1500 watt mexican telephone for awhile
wow it did an amp?
ha neat

Posted by renesis at 14:26 | permalink | 0 comments

so i guess summer is logic and RF
and strife
1:12pm Sunday (CST) - Time in Beijing, China
4 hours for my quote to happen
ha wow
antennas are neat
lots of people dont have cable
they watch daytime and local news
and sitcoms

Posted by renesis at 14:21 | permalink | 0 comments

check if theyre a publicly held company
and that would possibly be your answer
sad =(
hydra beasts
yeah, bets are off, cant sell themselves a nonevil anymore
when it goes digital broadcast
people will be like wtf?
i can rent or download my tv and movies ty
half the country goes WTF
and theyre all poor
and they all NEED tv to get thru the night and back at work next day
dude i dont think most people know
yeah sucks, hard
sometimes cuz family, sometimes cuz i dunno
yeah i cant deal with this passive media shit
kevtris: itd need to be semi organized
and actually fureal underground
and rolled out simultaneous
mobile, everywhere
itd be almost impossible
i say almost cuz im dullusional
i wanna do RF linear thing in summer

Posted by renesis at 14:16 | permalink | 0 comments

kevtris: its got multiplier builtins?
wtf! yayay
okay i need like two weeks for quartus and logic reorientation
shit looks too fun =(
wow so like maybe 51K of dual port?
and multiplier
do any have rf comm units?
ooo, pll
are they stable adjusting pll on the fly?
omg pwn

Posted by renesis at 14:00 | permalink | 0 comments

or you just wanna do it for sport
yeah i figured, heh
kevtris: pretty boards
everything worked out cool that design?
yeah, blank gate farm
any logic system has latency issues
thats how some of them sync up their stages
logic wouldnt work without spec'd delay
yesh but in real synchronous system, sometimes you still end up having to stagger gates
logic hurt sometimes =(
PC go faster
PC are mutant
wtf animals
also PC only really pwn at non realtime processing and media
kinda wtf for an embedded app
haha 924 wtf
yeah thats standard
FPGA farm = decryption

Posted by renesis at 13:55 | permalink | 0 comments

or compromised break freqs
haha fonts?
kevtris: you documenting any of this at some point?
or is secret
okay neat
like product or dev thing?
its your nes thinger, isnt it?
sid filtration, yeah?
you doing the switched analog filters?
or you figured out the dsp
wow thats prob a sexy board all assembled
kevtris: it was linear A and log DR ramping?
oh neato
i wanted to do that
because i did the log one, so linear should be simpler
put the pic in the corner
kevtris: they dont have em on open cores?

Posted by renesis at 13:50 | permalink | 0 comments

okay last quote gerbers sent
ourpcb offline
timecop: make them come online
blackmoon: our parts room tech at school was like a total watchdog with parts
always talking shit cuz id bend up his pins
by the end, i could walk in there and grab whatever part i wanted
just me and a few others, heh
yeah totally
blackmoon: my filters are prob gonna be least 10 diff medium value resistors and caps =(
if im like ninja about part selection
if im not, maybe 20 or 30 values

Posted by renesis at 13:45 | permalink | 0 comments

i cant afford fast fedex and rush
im getting these no silk
hehe, prototype sex

Posted by renesis at 13:30 | permalink | 0 comments

final quote time
i got it down to $250, shipped, last time
but for 2x what i needed, because 2 panel min
chopped the panels in half
blackmoon: do you know about international fedex?
wait nono

Posted by renesis at 13:25 | permalink | 0 comments

macegr panel method is fail
(silk layer)
maybe its cheaper without silk
im not gonna include those gerbers, its all fucked up anyway

Posted by renesis at 13:03 | permalink | 0 comments

i want pics of eagle saying the board was done with one ratwire left
i almost even left a comment
my sister brought her canned goods cabinet over from old apartment
omg heavy ass milk crate
ragu for years
two panels minimum
and i want two CS systems
not 4
and headphone amps to maybe even recover cost
haha ima make em do a 3rd quote
watch i get another email from susan 'our engineer says he cant finish your job'
okay but yeah, i can make the boards like half size
techsmurf: /usr/local/oldroot/old/backup/2003.02.root/home/ren/backup/usr/local/root/
and yes thats an exageration, but not rly by much =(

Posted by renesis at 12:23 | permalink | 0 comments

haha peemask!
i dont think ive ever seen it
i bet they have one package of yellow they never used
i got my all and parts express stuff
and a quote for the qauranteed late for maker faire panels
but i get get 2 day rush for for like $600 extra
$150 per panel
$50 for 5 day
im like, 5 day rush maybe 4 days ago...
yeah rev1b =D
for the maker one
i want to kit these so i want test/demo boards
and like 20 headphone amp boards
if those dont work i will maybe stop doing electronics forever
just maybe, tho (cuz i really like teh electrons, yes)
dude did you see that pic?
i think there might have been a trace stub
anyway, i found it offensive he said eagle failed him

Posted by renesis at 12:18 | permalink | 0 comments

For solder mask color: you can chose Black, red, blue,yellow for your production order.

Posted by renesis at 12:09 | permalink | 0 comments

kinda sucks, susan doesnt talk to me anymore
just the tech =(

Posted by renesis at 12:02 | permalink | 0 comments

okay, John says he looking at the new panels now
im all out of ziplock snack baggies
i think TS, but theyre very big cans
i dont remember the small can high volt ones (i dont do much design right on mains)
are you?
TS and i think TU are the panasonic snap in series caps
the ones like 2" diameter
backwards, charged up?
thatd be an amazing anti personel device
blackmoon: found use for those niobiums

Posted by renesis at 11:55 | permalink | 0 comments

and that itd be cheaper
i got a quote within 20min
so it was like $30
er, $300
for two pretty big panels, like 2x the curiously strong systems and maybe 40 headphone amp boards
so i cut headphone amp boards to maybe 12
also it was for qty2 on each panel
tho that only saves me $40
tooling cost was most of it, per board after that was maybe 10% of the tooling cost
nice invoices =D
but i mean, thats cool for future orders
this is just prototype/demo stuff, i dont want alot
and $40 is like, alot of parts
but panel 2 was like $80 tooling
compared to like $160 for panel one
so size seems to be big factor
so i cut down size on both boards maybe 30%
yeah i gotta go hang with exexgf
she been cleaning my living room while i been working

Posted by renesis at 11:49 | permalink | 0 comments

yeah no prob
cant imagine everyone wants to spend the free time comparing datasheet esr parameters
just me =\
blackmoon: did you read about china pcb?
arent those fucked up expensive?
and low volts
(mad farads tho, if i remember right)
hehe yeah
ill find something to do with those at some point
rf amp or something
blackmoon: ha no, thats timecops issue
so it wasnt the photoplotter NC
it was the routing/vscore NC
its either to small to do full panels, end to end, or the guy was just like, oh man wtf this isnt like when we tile them
like, when they tile single boards, the routing is just straight across
my shit would have a bit of gcode h4x
anyway, at some point it became clear i had a CNC in my bedroom
and he was like, oh we can do a panel, and you can cutout?
im like, yeh
hes like, ok old panels are fine

Posted by renesis at 11:44 | permalink | 0 comments

18:37 < renesis> NHG = small, FC = very good, FM = best ESR
cheaper shit is extremely low ripple
for high load, youd have to buy more to do it right, so its not really worth it in most instances
and i tend to buy extra, so i want what i have around to be 'general purpose'
meaning i dont have to baby them and calc out/sim ripple voltage and current for almost everything i would use an electrolytic for
yeah $25 is alot of caps, even the FM ones
youre not buying values near the extreme high end of any family
so they should be super expensive
shouldnt shouldnt shouldnt

Posted by renesis at 10:55 | permalink | 0 comments

measure your caps
dont pray, just do it right
theres a can diamater and a pin spacing
height isnt really an issue

Posted by renesis at 10:50 | permalink | 0 comments

NHG = small, FC = very good, FM = best ESR
thats pretty much the 3 series of panasonic i use
i design with
for years i been buying the United Chemi-Con equivelents
about 50% cost
yeah i dont really know, i could tell you for sure without knowing kinda what circuit its in and the load
yeah, look at can sizes in a voltage range
for FM and FC
if FC come in 2x the value at the same can diameter, get FC at the high value
if not, get FM
at the value or above the original

Posted by renesis at 10:40 | permalink | 0 comments

i should ask what kind of machine they use for CNC routing and vcuts
possibly they will accept gcode i write for routing my board (then they only have to review it, prob just load it into the machine and see if it throws limit errors)
tho i guess then its only a few minutes less work for me than doing it at home
just setup of the panel
timecop: so maybe with super rush, they actually show on time

Posted by renesis at 10:17 | permalink | 0 comments

okay hes sending quote now
he said its much cheaper like this
i wonder if they write the gcode to do routes by hand
oh damn
im like ignoring exexgf

Posted by renesis at 10:09 | permalink | 0 comments

so i guess the photoplotter NC isnt the prob
its maybe just the limits on the routing nc, shrug
so,we can just make them in panels without V-cut or CNC, you rout it by yourself?
old panels are ok

Posted by renesis at 10:04 | permalink | 0 comments

timecop: he said its difficult to vcut or cnc
too many boards on panel
he a cnc dunce, basically
i would have if they told me they suck at cnc
i asked if he wants me to make it so they line up
so its just cuts across and down
timecop: hahaha
i think he has maybe the same prob i have in my bedroom
the cnc they use to cutout might not have the axis length to cut whole boards
he wants it in 4 or 5 diff smaller panels
hes checking the files again, i asked if about half size on the Y is ok

Posted by renesis at 09:59 | permalink | 0 comments

i told them id pay up front
i told them i do that too
because the told me i could panel it
so it did
they told me to panel is smaller
less designs and smaller panels
so i did
ive done 3 panels for them
yeah but its maybe how the fills are drawn and the data on the file
um, its not that
dude, its exporting gerbers fine
their trolling or their machine is old
when you "panel" a single file on an NC it does do the data for everything drawn
it draws, move to the next origin, and repeats

Posted by renesis at 09:54 | permalink | 0 comments

i dunno cuz you said you were going to sleep last i remember
who is Season Sun?
Our engineer told me we cannot finish your job,sorry.
Best Regards,
Season Sun
so i asked why
but yeah i think i just overloaded their NC or something
for the tools?
fuck i dunno
prob monday or tuesday
yeah they shipped yesterday

Posted by renesis at 09:49 | permalink | 0 comments

14:01 < DX^> renesis: omg did you get endmills
shipped yesterday
im prob gonna do curiously strong rev 1b
for home brew milling
i think maybe 4 or 5 vias need to be moved
arent you sleeping?

Posted by renesis at 09:44 | permalink | 0 comments

timecop: okay you add the pot in later?
wtf email, why does it say not done
it doesnt matter =)
okay tho, ill look at it and hook it up in a bit
i need like, feeds and liquid
exexgf woke up
meh, no time, energy
haha shes interviewing for some acting thing
because shes all super tiny looks kinda like a kid
makeup would do it
so yeah, she could play a teenager, they dont have to fuck with all that kid shit
timecop: okay night ill look it over see if i spot nonsane

Posted by renesis at 02:43 | permalink | 0 comments

ac-130u: id prob go with the salt lake CC one
but id have to be in there and asking shit and seeing machine
their course outline just reads better, its by hours and focuses
the other school..
UVS, they have typical courses
but notice a few of the courses say on sufficient demand
this either means the class is active when they have a student
or it means they wont have the class unless enough students take it same time
which can be fucked
it means you might not finish for awhile
and it means if people leave, you might have to drop classes third way thru or something
ask how long a typical student takes to finish the program, full time student and working
thats just how shit rolls in a community college
tech departments are a money drain
they *probably* wont have a class for one student
but its tech, it might not matter

Posted by renesis at 02:38 | permalink | 0 comments

we got some stuff in the school shop
italian shit
had symbols
for the main gearing
i dont remember any of the other goofy symbols, but i think the dials were like tenth mm or something
i think we maybe had metric threads on a conversion table plaque
on the logans
but maybe im tripping im remembering another table
the valley in the 70s
and 60s
fuck that i dont wanna know
they machines fuel rods and and inspected the spent ones at santa susana
supposedly had fires

Posted by renesis at 02:33 | permalink | 0 comments

yeah but most big shops take damn near anything to keep the machines spinning
im way more an electronics tech than a machinist
im okay at setup, but im impatient, so i suck at everything kinda
thats why avrfreak used to crack me up
im like, you dont have a machine, and im not a very good machinist
sure i can but out .001 parts
after scrapping 2
fact i was called scrappy for two years cuz of me on the manual machines =\
close, close, close AW FUCK TO FAR WTF
yeah .005 was pretty easy
on the bridgeports and logan mills and lathes
which was weird cuz they were WAAAAAAAAAAAy sloppy
but they were huge, and we had micrometers
and if we tried, we could get shit dead on with .001 calipers
40 year old tank machines, heh
i never machine metric
i hated gd&t
its not an inch standard anymore
officially its metric
one of our instructors will not teach is because the spec is mm specific
french machine =)

Posted by renesis at 02:28 | permalink | 0 comments

in academics, they care about you, they want you to learn and fuck up a little, just not get hurt
at a job, they dont give a fuck
youre safety is almost def not their priority, if theyre turning out mad parts
thats good
also, if the teachers dont stress it constantly, id be worried
ha, scary
yeah im afraid of them
i dont even wanna work in a production shop
prototype or research is diff
almost as dangerous because possibly rushed
but having like 100 machines and machinists walking around everywhere
and you cant here and maybe welding flashes going off
like, that shit isnt safe no matter what, if you aint on your toes
def cool work, but job shops are rush rush, or not making money, usually
itd have to be almost full CNC with a couple manual support machines and very low on grinding and welding for me to consider it
CNC with shields and modern controllers dont scare me so much

Posted by renesis at 02:23 | permalink | 0 comments

yeah TC
thats a bad rheostat config pot =)
haha, its cool thats just the symbolism they used
nono, joking
you dont leave them open, when you make them
you short them
else its a place for noise to get in
looking at the other
first def has better page
this is like typical catalog stuff, heh
ac-130u: does salt lake CC have CNC?
ok yeah
its prob just in the 200 hours of machine operation
you need to go to both school and check out their shops, too
thats the most important thing, picking a school or a job
like, when you interview for a shop job, ask to see the machines and ask questions about the machines and what tooling
ok neat
but yeah, youre getting paid to put your life on the line
or youre paying to put your life on the line, in a school
so just stay away if you see the machines and feel uncomfortable about their condition

Posted by renesis at 02:18 | permalink | 0 comments

prob make sense when parts were 3lbs each and a "pin" was prob $.05 cents of metal
$.05 but you a whole gourmet meal back then!
thas what old people always sayin
okay lemme check the other ac
and its called "industrial Print Reading" in my program
misleading name
they know, they didnt pick it
fundamentals of geometric dimensioning and tolerancing
is the one we have, im pretty sure, ill check for the author, sec
timecop: single turn box trimmer or screw disc trimmer
short one pin to the center/adj/wiper
its 3 pin out?
thats what you want
its two pot outputs?
and you can adjust the pot?
wait, top or bottom, you can flip it with code?

Posted by renesis at 02:13 | permalink | 0 comments

200 hours of lathe/mill and its got all the little things you wanna learn too
thats kinda similar to me program
sculptor_: architects, if them anymore
but to them, its symbolic i think
thats like your fureal hardcopy (i seen contractor fools on jobs looking at big unscrolled prints)
one half a pot is a rheostat
variable resistor
youre shorting out part of the pot
well, i would use a 40R
and put that on top
a 100R or 50R pot
short out one side, thatll determine if up or down will make it more or less resistive
which side you short the wiper too, that is
then you set at 0 resistance for max bridgt
or full resistance (as determined in firmware) for least bright
its like a series adjustable resistor like that
you can get actual rheostats
but its typically an old control thing, or audio stuff
really, theres never a need, you can turn any pot into a rheostat, and set which direction raises the value

Posted by renesis at 02:08 | permalink | 0 comments

oh, its not gd&t at my school either
explain in a minute
because its drafting skill
its drafting and inspection
so its like in the domain of CAD departments and machining departments
well, kinda
its more for inspection
well no
not in machining, its prob all digital and print paper
but architects prob still do
im willing to bet architects still use blueprint for final output
the book we have is writtin by one of the guys on the board who did the spec
good shit
its fucked up tho if you go thru it page by page tho
hurts brain
ac-130u: salt lake CC
that one looks cool, the hours split on the specific course work

Posted by renesis at 02:03 | permalink | 0 comments

and its maybe rejecting or sitting there for 3min prereading the code
im saying its a bit harder on the machine
the curves plus the 8mil fills might just be making it take a long time for the machine to preprocess the code
yeah these are big tho
i dont think the designs would be a prob one at a time
because then they just like, do one set of code and move to a diff workspace coordinate, if the machine or tech is smart
yeah because extremely high density and batched
im asking
im not sure
im guessing
but i predicted itd break their machine after dong the gerber export
i was only half joking

Posted by renesis at 01:58 | permalink | 0 comments

Our engineer told me we cannot finish your job, sorry.
timecop: ourpcb fails
read previous line
im asking
im asking
it might just be too much for the machine because the pours are drawn with .008 traces
they asked for smaller ones, and i did how they said
inches fool
4 or 5
all my shit DRC at 7mil
8mil really
thats not what i mean
stu didnt get it either
a) curves
b) the POUR FILLS are .008 traces
well, its an NC machine
the photoplotter
yes it is
what do you think gerbers are, timecop?
its gcode
rs274 is gcode spec for CNC machines
rs274x is gcode for photoplotters
i can read that shit =)
instead of tools, it has shape descriptions
and it uses them like a pen
so whats happening probably is that the curves and tons of .008 traces for the polys is loading up the NC machine hardware

Posted by renesis at 01:53 | permalink | 0 comments

kevtris: you get the SD card all figured out?
and the SDHC thing too?
yeah i didnt remember the end
and cmon like a week ago?
timecop: wtf
do the pot low side it should be ok
timecop: tiny soic8
is what it looks like
whats .65mm in real units?
like .025?
yeah its those tiny soic packages im always like WTF ARE THOSE CALLED
sot238 wtf
eck0|wrk: hi

Posted by renesis at 01:47 | permalink | 0 comments

ima call truck buddy and buy bins and more clother drawer tuppers, and trashcan
cant do that on bike =\
alot of crazy people are quite sane
they like, decided they dont like society
some of them are like genius and shit
fuckin weird shit
09:23 * Daveman goes off to make a motivational poster... chinks - leading the industry in failed espionage
09:23 * Daveman goes off to make a motivational poster... chinks - leading the industry in failed espionage
09:23 * Daveman goes off to make a motivational poster... chinks - leading the industry in failed espionage
wrong trackball clicker
im all looking at my firefox like WHY WONT THE PAGE SHOW UP

Posted by renesis at 01:42 | permalink | 0 comments

i hate how they call their shit different
i need to learn the jdec numbers
jadec? jedec? the acronym numbers
they black chromed it timecop
thats pretty sweey
T = trippy
timecop: sot-23-6
check sizes
sometimes they just kinda scale packages up and down to the next common pitch
fuck that
thats a shitt trashcan
you know youre gonna put trash in that thing, right?
its gonna get dirty and youre gonna want a new one eventually
i have a pile of 200 cigs
on a 3" ashtray

Posted by renesis at 01:37 | permalink | 0 comments

you need about 10mA min im guessing
45mA nice
blackmoon: yeah you do =\
at 10mA
thats prob cool for actual backlight use
like, in the dark
timecop: hehe yeah LED kinda peak then fade
its diff depending on LED
it really should be a datasheet parameter
burn time decay:
luminance peak at burning
because like, image shorting out an array of 1000 LEDs
you get a nice flash and fadeout
one time use, but damn that would be a sweet effect
omg you can prob do it with like 10V
shit pot
this is a dev thing, yeah?
well, then id go like, some resistors and fets or cheap pot
yeah chinese Vf
typical: 1.5V max: 5V

Posted by renesis at 01:32 | permalink | 0 comments

they do more than anyone else
so in total numbers, of course theyre gonna fail more
timecop: vcc and io are same?
timecop: no the vcc rails and the io
LED +A LED Anode
LED -K LED Cathode
also, what you described sounds about what happens when you dead short an LED with a regulated supply
sounds about right
yeah youre dropping like no voltage

Posted by renesis at 01:27 | permalink | 0 comments

haha, shitty
did you do the pinout upside down?
oh fuck i remember that
thats why its called K and A
i think its just breakouts on the led, most LCD
so you need to current resistor it
yeah it might be there already
yeah check for the actual resistor
haha, kids
daveman: no we <3 them
for as long as youre gonna live
09:23 * Daveman goes off to make a motivational poster... chinks - leading the industry in failed espionage
prob not, but funny anyway

Posted by renesis at 01:22 | permalink | 0 comments

09:13 < Blackmoon> 'man got some good dirt for ya. pure black gold man, just try a sniff'
haha sniff
haha i bet farmer people do that, too =\
just to change contrast
alternatively, use a pot to test good level
then DMM the pot
to see what the resistance is, to see the divider ratio
then just buy similar resistors
its be fixed contrast unless youre switching in and out resistors (fets) or using a digital pot
you just need like 3 resistors and fets for 8 values
well, 7, you prob dont even wanna use 000
timecop: you have pots to test with?

Posted by renesis at 01:17 | permalink | 0 comments

but yeah, their soil mix seems to be like: dirt, leaves, bugs
well i just use the soil in the pots
from the tea bushes
because they came in huge pots, and i ran out of bagged cactus soil
i thought my soil baggie was bigger

Posted by renesis at 01:12 | permalink | 0 comments

its not in the leaves
its on the leaves
its not the leaves
stupid bug families
well, 5W two weeks, hostile buggies, and they about the same
i didnt expect a big jump
ive done drawer CFL plants before at 15W
they dont grow that great
but they grow
so yeah, want to try cleaner soil
the CFL plants didnt grow either
tho no shrooms in there
and i dont thing those bugs dotted the leaves
well, they got watered the same
which wasnt super often
yeah they popped up in a day
and died
about 1" tall, then spread to 1" wide
then died
its from a landscaping nursury
its like literally miles of plants, under the power lines
so they planting literally thousands of plants a week

Posted by renesis at 01:07 | permalink | 0 comments

and i know its clean
thats possible, but the plants are doing okay
it might be because its a bit warm
well i have a shop down the street
theyll help me out
im sure they have something for that
an algae isnt dangerous at all
when i was a kid, my dad wouldnt pay pool dude
so our pool would get swampy
literally, like 6" visibility
but yeah, wed refill the pool
and then id swim around and hit all the algae
im sure other stuff can get rid of it tho
like with a pumice stone (my pool tool, didnt like it when it was slimy when i pushed off bottom and walls)
but thats whats pissing me off
the bugs came with the dirt
there were there like day one
fuck if i know
atmel is good docs
olimex is error docs
charles: no way these are good
they dotted like whole leaves
yeh =(
but bugs are growing
and plants arent dying
so i mean, 5W sustained it
and the tea bushes seem like whatever
they kinda like minitrees
yeah but im pretty sure 2 weeks no light would kill most plants
i almost did a no light grow
to control that
blackmoon: white i think

Posted by renesis at 01:02 | permalink | 0 comments

yeah thats when electromagnetics couple, charles
i found it
somehow ended up behind the baggie of single port RCA jack
my electromechanicals are splattered all over the carpet
damn i want my pell grant check already
blow it on drawer bins and couple tool boxes
switching to hydro soon
fuck bugs
dirty soild from tree farm
mulch stuff
except nothing dead yet
nothing is dying
but nothing is totally growing either
yeah theyre not drying
theyre not thriving either
i just loosened root balls before planting
yeah the clones arent dying
but they got shrooms
stupid dead stuff in the dirt
so yeah, like two people suggested hydro
ha, both of them do it themselves, and said its not hard

Posted by renesis at 00:57 | permalink | 0 comments

like, electrolytics wont see the high freq spikes much
i think because the esr
but yeah typically you want .1uF across very big caps
big caps are for psu ripple, little caps are for EMI
man FUCK
so like 100 baggies of parts fell off my bed onto the floor
i think my weed is there too
you dont need 50MHz xtal to drive to 100MHz =)
you can do it with PLL probably

Posted by renesis at 00:52 | permalink | 0 comments

unless they got some insane caps, but if theyre blown up, its unlikely
FM is best
but it comes in lower values
like, check the can size
and if you can get FM in 2x the value, same size, check ripple on both
er, FC in 2x the value, sry
id get the FC if they ripple is close
because with the larger value, youll have less ripple
which is better for the psu and its prob relatively the same load on the cap
the atmel data?
just do what it says
thats pretty much what i would suggest for the parts around the uC
why you dont have 4.7?
you can use 1uF's in parallel, just add up farads in parallel
hehe, over time is maybe just cheaper to get the 4.7 uF
thats for which?
oh, yeah do it
its like decoupling for your decoupling cap

Posted by renesis at 00:47 | permalink | 0 comments

timecop: looks like they go in with voltage at s ingle point
and then just mad bypass at the other pins
yeah FM cans are like the best in that size
ESR wise, on a mobo you want that
its no such a big deal i think in the psu but dunno
because you can get FC in 2x the size
so there would be like 1/2 the ripple
so you dont need as much ripple current rating
what value are they?
if you can get FM in the same value, size, get those
if not, FC probably ok
stu: yes
mostly because of the gloss tho
oh thats chrome
stu: that style in flat black would be cool
oh thats nicer, stu
thats alot tho
no psu, either?
workbench: yeah i think all those values come in FM series
at 35V or 50V, its 1000uF max
so at those voltages, you should be covered

Posted by renesis at 00:42 | permalink | 0 comments

FM come in about half the value max, per voltage
okay yeah those are usually like FC/FM size cans
they dont use bigass snapins
okay yeah for mobo id go FM
timecop: haha, yeah with all the cap values and stuff
timecop: haha
is 1.8 io legal?
timecop: weird
wtf is asia-x
if theres SMT caps, you can get better esr than FM i think
its called VC or something so
i think it says in FC datasheet what the upgrade was

Posted by renesis at 00:37 | permalink | 0 comments

wow, so weird
my exexgf's mom dropped of my exexgf last night
her moms didnt have money
dude all the shit on my bed/shelf is on the floor
kinda sucks
like i didnt 'put' it on the floor
i think she went to a dudes house a few blocks away
like mom like daughter, or something
you done with schema?
oh well yeah i need to like do nothing today
i might go nut i do one more pcb
but yeah ill do that shit
FC or FM or chemicon equivs
if you need more, the big TS series cans
or matve TU, T something
but FC in parallel is very, very good esr
and high values
yeah two FC in parallel is prob better esr/ripple than FM
and you get way higher values

Posted by renesis at 00:32 | permalink | 0 comments

Top | Add to Technorati Favorites

© 2007 lordpil.   XHTML 1.0! CSS! Site design by GNAA  Blog Engine by pbx | MULTI2 | ian hanschen | lolwat