lordpil's blog


macegr: im doing reverse schottky and transzorb after 5x20 pcb fuse block next to 2p minifit jr

Posted by renesis at 18:39 | permalink | 0 comments


macegr: fuck a crossbar, fuse and diode
i already did the fuse block 3d model

Posted by renesis at 16:18 | permalink | 0 comments


i should make it a schottky and put a fuse
k
and something more badass than a 1n4001
i can do a to220 diode, feer
that will be fun to test
i can fuse right at the mean well current

Posted by renesis at 12:28 | permalink | 0 comments


timecop: check out this analol placement sex
https://i.imgur.com/QRNf4Wy.png
all that space i should prob put some sort of fuse on the suicide diode instead of depending on mean well protect mode
maybe a ptc
wait no a ptc is bad
because will extend voltage reversal

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


rab: i have digikey boxes like this
like, i didnt finish this project but now i have a lot of random prototype inventory
and sometimes i am like, oh shit i have had this for years, ive wanted to order this so long
im pretty sure rane has #1 graphic eq circuit
the multiple feedback constant Q shit

Posted by renesis at 10:36 | permalink | 0 comments


their class-b is supposedly fast enough that the crossover distortion isnt an issue, shrug

Posted by renesis at 10:15 | permalink | 0 comments


jero32: ^ like an audio specific dictionary
http://www.rane.com/note110.html
and this is about as close to a wiring standard as the industry has
and if you want to see amp schematics, QSC has them for a bunch of their amps
class AB, class B (yarly), class H, i think a class G (my fav, H is better efficiency tho)
probably a class-d amp now

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


woah wtf
rane pro audio ref now redirects to AES.org
i dont know how i feel about that
okay i guess
http://blog.rane.com/?p=558
all the links dead
http://www.aes.org/par/
yay
hmm where is the book
http://www.aes.org/par/guide/

Posted by renesis at 10:05 | permalink | 0 comments


rab: yeah i was so happy to see he still updates it!
theres a bunch of new content since i stopped reading as much, he has gotten new hosting tho
dude has never bothered with a domain name
rab: i basically referenced him for any circuits touching mains for years
linear supplies and stuff
jero32: tons of analog hackers have that site in common
beyond that there is nelson pass and linkwitz stuff, but really you just need esp articles and projects
you could pretty much rebuild most of the audio industry from zero based on his documentation

Posted by renesis at 10:00 | permalink | 0 comments


phillips is still pretty big?
isnt nxp phillips?
NXP said it was the fifth-largest non-memory semiconductor supplier in 2016, and the leading semiconductor supplier for the secure identification, automotive and digital networking industries.[4] The company was founded in 1953 as part of the electronics firm Philips, with manufacturing and development in Nijmegen, Netherlands.[5] Known then as Philips Semiconductors, the company was sold to a consort
ium of private equity investors in 2006, at which point the company's name was changed to NXP.[6]
to me, phillips = super mass production microcontrollers
i dunno i feel like that was their side scam
havent they always been huge component maker?
i figured its like fairchild

Posted by renesis at 09:55 | permalink | 0 comments


hes got an 80s tek on it, one with all the clear backlit buttons

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


you know there is probably some sort of web service that will ring your phone
rab: oh shit i think my friend got that
when we went to ham show

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


holy shit goldie did an inner city life remaster/remix single for diane charlemagnes funeral
theres a 2017 rebuild mix and a burial mix
wait it might be a guy named burial

Posted by renesis at 09:17 | permalink | 0 comments


i should recap that tektronix vector graphics DSO

Posted by renesis at 08:52 | permalink | 0 comments


so much temp burn in on my vizio

Posted by renesis at 08:13 | permalink | 0 comments


your keylogger just crashed
disguntled suicidal head devs
i like xfce setup kind of like windows
or *box

Posted by renesis at 07:38 | permalink | 0 comments


molex minifit stp file doesnt output from altium
everything is broken
okay not everything just the 3d export of one part
os/2 warp, yo

Posted by renesis at 07:20 | permalink | 0 comments


were trying but we have school and work
ATM machine?
?
i dont even remember that part
i just remember him jumping a dirtbike into the LA river
and governor pulling a shotgun out of the rose box
and sarah flipping out
but i mean, shes pretty much flipping out since the start of the first movie
we know that because why else would she drive a jeep to mexico at the end of the first movie
anyway i dont do RF and i dont know much about credit reader systems
both would get me into trouble, no doubt

Posted by renesis at 07:15 | permalink | 0 comments


oh nice, i was about to tell someone in cars to ban this fuck
so do the bots just sit there now
do the thing, bots
someone who joined the chan w/ MAGA in their nick
so im basically feeding the troll
i have no idea what you mean, take pics

Posted by renesis at 07:08 | permalink | 0 comments


https://youtu.be/kfwZH9qQJfg
cheater: you want acceleration?
you want the accel of y to be based on the x position?

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


i am not the math geek, tho
i think hes saying
what is i

Posted by renesis at 06:45 | permalink | 0 comments


jero32: heh, w/ diffs the variables are other equations
and yeah, math w/ circuits is why opamps are called opamps
yeah fuck that
that frenchy d means problems
yeah thats what i said, the frenchy d
i think its actually specific to partial diffs
d/dx is how reg derivatives are usually tagged
anyway i think the diff is dealing with single value variables vs dealing with dynamic systems as variables

Posted by renesis at 06:40 | permalink | 0 comments


https://www.digikey.com/product-detail/en/on-shore-technology-inc/WMS-080Z/ED2680-ND/2175337
fucked machined pins and its going to non-stock soon, but thats sexy

Posted by renesis at 06:20 | permalink | 0 comments


i really miss cat
i cannot afford cat
almost is too much not free
and cost me money when it gets bit by a racoon
it doesnt really work like that
that sounds really boring
ive been playing wasteland 2
its better than wasteland 1
yeah i wouldnt know that because most the timers i played it i didnt know i had to find the instruction manual to read that

Posted by renesis at 05:54 | permalink | 0 comments


that sucks
like a normal comedian
yeah he has that little magnet dealy that covers it
and it works
heh
i need to coffee and do stuff
ha

Posted by renesis at 05:49 | permalink | 0 comments


have you done steady state ac analysis?
like, where you calc the resistance of the caps at a specific freq, then do normal resistive circuit analysis?
their two stage circuit is really easy to analyze like that, because it just turns into an easy opamp circuit and a couple dividers
yeah but you dont need to fuck with phase and all that
like, the feedback loop turns into a buffered divider, so its basically 2 voltage dividers after some simplification
verne?
that sucks
he was legit is a really cool guy

Posted by renesis at 05:44 | permalink | 0 comments


two common single amp designs, and their more ideal dual amp solution
yeah
its a funky curve, it levels out in the middle so its like a shelving EQ and a HPF or LPF
depending which direction youre going
http://www.ti.com/lit/an/snoa586d/snoa586d.pdf
they just buffer their first stage with a 2nd amp
so like, you could impliment it as a single amp design if you have opamps following it doing other stuff

Posted by renesis at 05:39 | permalink | 0 comments


my personal experience, its like hey batteries arent you supposed to have sploded like 10 dents ago?
and this mod will warn you on charge mismatch but it doesnt seem to do much about it
pretty sure ive gotten a cell below 3v
blackmoon: so an issue seems to be flat top contact vs insulation from the ground edge on the flat top side
like, they will ship with this little insulator ring under the heatshrink
so some stuff wont make contact past the offset
so you can take the wrap off, but in some mods youre shorting cells on each other and the chassis
the wrap is also fucked up thin and a lot of the mod cases are metal, so you have to rewrap a lot
when i had a mech mod my batteries would short internal a lot
theres a natsemi(ti) doc about that

Posted by renesis at 05:34 | permalink | 0 comments


i kind of feel like you assume all responsibility once you slap them into your mod
the application is inherently unsafe
youre shorting lithium ion batteries next to your face
yeah
if this goes anywhere, LG will just say thats not a safe application for our products, and its the distributor or consumers responsibility
amazon will just point at the reseller, because i doubt they sold these direct
the reseller will probably just point at all the warnings on forums and mod manufacturer websites
in general, samsung/LG 18650, and probably the fakes that mostly meet spec, seem a lot safer than expected

Posted by renesis at 05:29 | permalink | 0 comments


http://www.ni.com/cms/images/devzone/tut/GRAS_Sensors.JPG
thats basically what the spendy stuff is
is he sueing the box mod company?
its not if theyre legit batteries
im pretty sure mine are fake samsung and ive been using them almost two years
was basically shorting them across .5R coils
maybe
mine have dents
from when i dropped my mod and the battery terminals punch the batteries
thing is, he cant prove he didnt tear the battery wrap

Posted by renesis at 05:24 | permalink | 0 comments


only because you asked someone who lived this stuff every day for 6 years
you can make a measurement mic
http://sound.whsites.net/project93.htm
http://sound.whsites.net/index2.html
this guy really smart, read all the things
all about analog audio stuff
simplest is just sticking one of those panasonic electrets in a metal tube from hardware store with some epoxy

Posted by renesis at 05:19 | permalink | 0 comments


for up to like 100-200 Hz
they look so good its like you faked them
room doesnt matter
jero32: minidsp people have a pretty popular usb measurement mic
if not, you can just get a behringer measurement mic
theyre good enough, mostly the cheap measurement mics just have a shitty noise floor
but theyre flat
i think its like $30, the usb thing is closer to $100
roomeq wizard pretty good for diy measurements, can make eq files for a lot of diff hardware

Posted by renesis at 05:10 | permalink | 0 comments


measurements stand up to almost anything
dsp genelecs defy reality
like, measure so flat im like, did it just measure the fuck anechoic chamber and correct for it?
for design you measure them using near field or ground plane
which is basically pro low frequency measurement technique
for vented stuff, you do summed nearfield, you basically add them in proportion to their radiating area
for mid and highs, you probably want to try and find a good open location and do a lot of measurements and average
finam measurements are done in a room to check for craziness
but you cant do any kind of design measurement in a room
you voice with room response
but voicing with passive xovers is kind of a joke
you dont, much
cheapie: it will also determine how much cut you get in other spots
er, _champi0n ^
anyway you can do nearfield measurements literally anywhere

Posted by renesis at 05:05 | permalink | 0 comments


minidsp, you buy little $5 or $10 software modules, and you can do full parametric EQ and crossovers
industry doesnt trust consumers with parametric EQ because a lot of studio techs dunno how to use them
right thats like, 70s tech
home AV reciver is kind of like, zombie tech
home users never got real speakers
they went from dinosaur speakers to little shitty active all in ones
but most of those all in ones, when not driven to the point that the active EQ and bass processing kick in, perform better than a lot of 'audiophile' home speakers
no it is
nope
the prototypes are usually a compromise, and the spread in production is usually huge
done right, the parts are incredibily expensive
so its almost never done right
and even done right, you cant correct for speaker issues
its just the wrong way to do it
its the cheapest way
im really not _champi0n
yes jero32
you can load FIR filters and get whatever response you want
which is how its done with pro speakers
like, active genelecs are heroic

Posted by renesis at 05:00 | permalink | 0 comments


actually i guess that is what youre saying, which is nuts
all i see is passive xover with like, amateur mistakes
like, parallel adjacent caps and coils
some of the single filter boards look nice
_champi0n: i get no active xover boards in a search, but minidsp in a box pops up
which ive already suggested to jero
if you meant blank passive xover boards, i already told jero why i think thats a mistake considering his background and the end result
it depends
theyre usually really shit
and sometimes hard configged to xover tops and subs
and i wouldnt trust a home AV reciver directly connected to tweeters
their user interfaces are very crackish

Posted by renesis at 04:55 | permalink | 0 comments


https://i.imgur.com/migBxlR.jpg
maybe not first finished project, shrug
pretty sure i laid that shit out in autocad
i dont fuck around with that shit!
i think i did a total shit parametric EQ before that
with random all electronics parts on cheap vector board
it worked but it sounded shit
im pretty sure that was after a lot of elliot sound product reading, and lots of lurking on gainclone amp threads
yes ofc
so that was maybe 06 or 07
yeah some of those boards are okay
salen key filters are pretty off the shelf
a lot of shit like that at parts express is like, china shit from the 80s
there is laminates involved
thats like saying all pcb are the same

Posted by renesis at 04:50 | permalink | 0 comments


with an active xover, matching driver levels is trivial
imo, moving xover frequencies is easier
with DSP crossovers, youre pretty much unstoppable in terms of xover and correction
minidsp is maybe $100
you can get AD sigma studio boards for less
and you dont even really program them, its all click config and block diagrams
was my first finished electronics, can confirm, was fun
https://i.imgur.com/TagfknH.jpg

Posted by renesis at 04:45 | permalink | 0 comments


if you go passive, its quite likely the actual speaker response will be pretty shit
yes
or tweeter and midwoofer, but its basically a big one and a little one
tweeters have a lot of characteristics related to resonance that dont necessarily show up in plots
like, how they ring, waterfall decay plots are actually pretty useful but this isnt published much
so in general, metal tends to be higher sensitivity, but tends to ring more
and high sensitivity really only means you save money on amps, because tweeters are almost always already too sensitive for the paired woofer
soft tweeters arent as sensitive, may distort a bit more, but dont ring as much
theyre damped
theyre prob still more sensitive than the paired woofer
so with a passive xover, you really need an L-pad attenuator in most designs

Posted by renesis at 04:40 | permalink | 0 comments


the book also covers stuffing, diff types and the intended effect to factor into your enclosure volume calculations
theres also tons of overlaid plots
and enough of the math is there for you to build a spreadsheet and start playing with driver datasheets
my guess is a little less reviewed
probably the safest thing to do is look into full range drivers
like 3-6" things, and build a sealed system for those
like, bunch of little boxes, with unequal XYZ dimensions
and then build a vented sub
next step in complexity is 2-ways instead of full range

Posted by renesis at 04:35 | permalink | 0 comments


do you have loudspeaker design handbook?
by vance dickason
DIY budget is pretty high compared to what mass production speakers cost
get loudspeaker design handbook
sure, that
first section is thielle smalls for sealed enclosures
which is simple and *will* get you the response you design for
you can measure near field and see it easily
section after goes into vented designs
wtf sell it
download it
its pretty much community property at this point
ive paid for it before, pretend i lent you my copy
anyway, professional vented designs almost never hit target first pass
unless you know what direction to fudge the calcs

Posted by renesis at 04:30 | permalink | 0 comments


so chinese

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


we will fight over abortion and rights for guys in dresses, and continue to get large scale fucked
aerith_gx: hi
what in the fuck
the chuwi booted to some rendered 3d battery charge graphic
now it booted to windows 10, about 3 or 4 sec to login screen

Posted by renesis at 02:58 | permalink | 0 comments


jero32: in our system i was making >$60k a year with one of the better health insurance options, and could barely find doctors, and was flooded with a lot of paperwork i didnt understand every time i tried to use it
like, mfkrs why are you asking me to justify this, i go to the dr i do what they says
i would argue that you system has nothing to do with ours then =(
the people: yo these private insurance companies are really a problem for almost all of us
the dem leadership: hey republicans have this plan where we all use private health insurance
i dont think that
i think you have lower copays if any and better coverage and better outcomes
were setup to make profit
making profit is basically fucking people over more if possible
like, fucking ourselves over is built into a market laws
its not going to go away

Posted by renesis at 02:53 | permalink | 0 comments


if that guy ran out of money for hiv meds, lakers fans would pay for them
california has medi-cal, pretty sure it covers a lot of that stuff
i was more afraid of using medical coverage when i was using private insurance

Posted by renesis at 02:47 | permalink | 0 comments


those have happened
jero32: pretty sure magic johnson is still alive
right was basically agreeing
also it relates to spyberts point

Posted by renesis at 02:42 | permalink | 0 comments


we need like a 3rd senator for each state that has a stem education

Posted by renesis at 02:36 | permalink | 0 comments


These data may underreport HIV diagnoses among transgender people because of challenges in accurately identifying and reporting gender identity in HIV surveillance.
so theres an assumption there that finding more transgender people would lead to a higher ratio
it may also overreport, in terms of HIV relative to the entire population
thats just the first line
Includes Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia.
NY, CA, and WA are not there

Posted by renesis at 02:30 | permalink | 0 comments


he hates the tweeters, so does my departed acoustic engineer fren
partsexpress.com projects and cabinets
join freenode:#redditaudio
and ask nyt
or you can join #a/v here but its a bit dead
yes nyt
net geek, does a ton of speaker hacking
yes
he is big shit on home theater audio forums, builds some pretty big systems
not really an EE but pretty smart about systems and can hack together a decent passice xover
estimates by who? and yeah that doesnt really surprise me
hardcore data
when is the study dated?
well, it was
its american jero, and were kind of having a moment, in terms of government and science

Posted by renesis at 02:25 | permalink | 0 comments


rab: that shouldnt matter too much
like, you are not super sensitive to phase down low, so its just going to be boosted and cut in different spots
like, walk around and find what fits, heh
i feel so shit about selling those g1 krk
they were def QA winners, even tho they seemed years apart
they measured on top of each other
like, they were maybe +/-1.5 in the middle, but the two were like half that from speaker to speaker
the shade of yellow didnt even match, i could not believe
the EE i worked with there who sold them to me was like fuck off, you buy whatever speakers you want when youre done

Posted by renesis at 02:20 | permalink | 0 comments


like 90% of the designs ive seen
you should prob read elliot sound products pages, too
no but its like building a high power switching amplifier
its not complicated but sourcing parts can be drama and they tent to burn up more in r&d
but you get a decent module and theyre pretty reliable, low drama
AB you can just buy a couple of $60 boxes if you want
with RCA in and banana out
AB amps are mostly big opamps now
up to like 100W
rab: yeah thats not going to translate anything like what the producers did
is it centered in the room?
that actually helps a ton, ive never had a room that would work for surround or pinpoint stereo imaging so ive never put any time into it

Posted by renesis at 02:15 | permalink | 0 comments


trying to get low end from a bunch of midwoofers while also getting clear mids
yeah
i mean, no one really has a good 5.1 setup
making certain types of good ones is trivial
full range driver, sealed box, simple and you will get exactly your target response
and very well behaved low end, even if you dont get the extension
making transducers is mostly about glues and shims
everything else models pretty well
have you looked at parts express parts and projects?
i think you should make a 2 way active speaker with active crossovers
you should skip passive crossovers
youre young, theres no reason for you to waste time on that trash
then either buy some little AB amplifiers or do a little chipamp PCB
actually
you could buy a minidsp
theyre not cheap but theyre not expensive and they can do a lot
yeah its usually salen key filters

Posted by renesis at 02:10 | permalink | 0 comments


so producers maybe tilt some of the eq for that expectation
but generally, shit sounds sex on matched speakers
which makes sense
rab: like, generally rears and centers are almost afterthoughts in design
because putting matched speakers in a rear or center is wasted 90% of the time
anyway soundbars just create reverb
beam steering doesnt work on small systems, you need 90ft line arrays
and wall bouncing isnt going to work unless you live in a perfect box with no furniture
like, pinpoint soundstage becomes like suggestive puffclouds of location
i watched bunch of phd mfkrs get off on how good their dsp shit is, because good feedback from listeners, only to find out later the algorithm was wrong and the output was semi random
also the low end challenges are high

Posted by renesis at 02:05 | permalink | 0 comments


it should downmix fine
anyway
i would maybe get a decent stereo setup
see if you can find a used 5.1 or 7.1 receiver
and maybe buy some decent little full range speakers for it
thats kind of the affordable, flexible route
thats usually what a downmix is
it should work
unless its mastered shit and they didnt do any sort of level correction and shit is clipped
i could see that happening, heh
almost every movie has a stereo track
oh, no idea
ive only worked on the other way
like, why is this shit stereo on the 5.1 config
thats not really true
movies are mixed on monitor speakers
and really good surround systems are matched speakers
no one spends money for really good surround systems

Posted by renesis at 02:00 | permalink | 0 comments


jero32: do you have money?
if you dont have money you shouldnt be considering a soundbar, really
theyre a huge engineering challenge, if you want decent audio
like, if you *only* have TV speaker, a *good* soundbar will be better
well, the shit i worked on would be worth it, but its like $600 i think
like, if youre not paying $300-$600 for a soundbar, chances are its pretty shit
also, DSP in these things is a coin toss
sometimes its accurate, sometimes its not, sometimes its not but people like the results, because reverby
soundbar might not fix that at all
it sounds like you have a shit middle channel, honestly
in a lot of mixes, most dialog goes to center
so if your center isnt good, dialog will sound shit
like, you are most sensitive in this region of audio spectrum
and people are sensitive to distortion
like, frequency response preferences are all over the place
distortion preferences are not
people prefer below 1% THD

Posted by renesis at 01:55 | 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