<nicksydney_>
so normally the best way is apply flux first and then tin it right ?
<wpwrak_>
there's a third kind, "no-clean". you supposedly don't have to remove it though some experts say you really should anyway. i didn't find the one i tried to be very satisfying in terms of helping with soldering
<wpwrak_>
yes
<nicksydney_>
ok i will try on the test circuit that i have 2 on my board (on top and bottom) just to make sure :)
<wpwrak_>
looks pretty good
<nicksydney_>
i learn now that to save on etching combo i will use sharpie so that it will not be dissolved that way i can have my combo HCL / H2O2 longer :)
<nicksydney_>
the track i'm using 0.4mm
<nicksydney_>
as recommened by the expert here :)
<nicksydney_>
black and blue sharpie works other colour does not work
<nicksydney_>
think need to buy a better soldering iron
<wpwrak_>
the solder looks okay. not sure what you plan to do with paste, though. your process seems more compatible with solder in the form of wire
<nicksydney_>
the paste make it easier for me to solder i'm not that good in soldering with wire with small component...haven't got the hang of it :)
<wpwrak_>
what you do it that you put the solder on the tip of the iron. then you apply it with the iron.
<wpwrak_>
the flux in the solder will evaporate this way before you reach the component but you already have applied flux there, so it'll work
<nicksydney_>
will try it later on the test circuit :)
<wpwrak_>
with solder paste, you'd apply a tiny drop on each pad, then place the component, hold it down, and then use the iron. the big issue with this approach is that it's easy to smear solder paste around when you lose control of a component
<wpwrak_>
that unsoldered solder paste will still be conductive and when you apply heat to it or in its vicinity, it'll form solder bridges. so all in all, this is a messy process, especially as you move towards smaller components that tend to jump from your firm grip more often
porchao has joined #qi-hardware
<nicksydney_>
wpwrak_: true...need a steady hand
arielenter has quit [Quit: Leaving.]
porchaso0 has joined #qi-hardware
porchao has quit [Ping timeout: 260 seconds]
<nicksydney_>
wpwrak_: what osci you using ?
<wpwrak_>
osci ?
<nicksydney_>
oscilloscope
<nicksydney_>
:)
<wpwrak_>
ah, 'scope ;-) a dying Rigol DS1102CD
<nicksydney_>
neat....need to buy a new one i presume ? :)
<wpwrak_>
considering replacing it with a Rigol DS2072 or a Rigol DS1074Z
<wpwrak_>
yeah, without proper communication it's a lot less useful
<nicksydney_>
that's why i stick to software no investment just download
<nicksydney_>
hahahaha
fire has joined #qi-hardware
<wpwrak_>
the trick is that you can "unlock" 200 MHz mode in the 2072 :)
<nicksydney_>
yeah it is there in that post
<wpwrak_>
but yes, you have to put about USD 1000 on the table if you want a decent scope
<nicksydney_>
:)
<wpwrak_>
ah yes, eevblog is where you find all those things :)
<whitequark>
wpwrak_, nicksydney_: I've had horrible experience with water-soluble flux
<nicksydney_>
whitequark: what kind of bad experience ?
<whitequark>
the problem with it was that it's been marketed as both water-soluble and no-clean
<wpwrak_>
you can go a little cheaper with one of the 1000E/D series rigols. they're a bit better than the 1000C/CD series. not great but decent workhorses.
<nicksydney_>
wpwrak_: for me i can only spend max $150 for scope if ever i can get one....more than that is beyond my budget :(
<whitequark>
but in reality it was neither: I don't think I ever managed to completely remove it from any of my boards, either with water or various chemicals, AND the damn thing conducts well enough to account for inter-trace resistance of dozens of ohms
<whitequark>
also seems to be very hygroscopic
<wpwrak_>
nicksydney_: well, try 2nd hand 20 MHz analog scopes. i've heard that they made that sort before the war. the first punic war, to be precise ;-))
<whitequark>
it's called "flux TT", and hopefully it isn't produced or sold outside of Russia, because I feel for the poor soul who accidentally uses that
<wpwrak_>
whitequark: sounds great ;-)
<nicksydney_>
"very hygroscopic" ?
<whitequark>
attracts moisture from air
<wpwrak_>
"TT" = "Terribly Troublesome" ?
<nicksydney_>
wpwrak_: there aren't that many of those analog scopes available locally here and if they are available you bet they will be selling more tha $200
<whitequark>
wpwrak_: perhaps it's a reference to TT pistol, as the preferential way to end your suffering
<nicksydney_>
whitequark: so TT flux is suicide pretty much :)
<whitequark>
yeah, it's absolutely terrible
<whitequark>
and it's not just my hands :p
<nicksydney_>
whitequark: thanks...i keep that in mind
<wpwrak_>
whitequark: ever tried ultrasound cleaning ? that may even be able to defeat some of the nastier fluxed
<whitequark>
wpwrak_: I've subjected those TT-stained boards to more than four ultrasonic baths, while rinsing them with water, acetone and alcohol in between
<wpwrak_>
for my "water soluble", i just leave it some 10 minutes in the cleaner. with demineralized water. (though "water soluble" really refers to hot water under pressure)
<whitequark>
I don't know in whose corrupted imagination is that thing soluble in ANYTHING
<nicksydney_>
whitequark: $380 is WAY beyond me :(
<nicksydney_>
need to save up for few months first i suppose
<wpwrak_>
yeah, the DS1052E should be a decent choice if you can "unlock" the 100 MHz. else it's a bit slow if you care about things like checking signal integrity
freespace has joined #qi-hardware
<whitequark>
wpwrak_: dsoquad seems to have mostly same technical specs as 1052e, but I'm somewhat wary of trusting that thing
<nicksydney_>
"checking signal integrity" in design like communication and things like that or also important if designing board like anelok, beaglebone and such ?
<wpwrak_>
nicksydney_: it's for those "what the hell is wrong ?" moments :)
<nicksydney_>
like the one problem you had with the OLED being short circuit ?
<wpwrak_>
nicksydney_: still very slow. that means that you can't properly handle signals above ~20 MHz. well, they don't seem to specify the analog bandwidth. maybe that's already the limit anyway.
<wpwrak_>
the OLED power drop is something even a slow scope would catch. but for example glitches on a digital line or heavily distorted signals need something better
<nicksydney_>
so if an MCU is running at 25MHz it won't be able to measure it ..is my understanding correct ?
<wpwrak_>
(DAQ32) 32 kSa/s divided by the number of channels ? why not just use a multimeter and switch channels manually ? ;-)
<whitequark>
*g*
<nicksydney_>
wpwrak_: like the way you put it bluntly ;)
mth_ has joined #qi-hardware
<wpwrak_>
if the cpu outputs 25 MHz signals, you could get aliasing and such. also, if the analog bandwidth is in that area, you'd get attenuation
mth has quit [Write error: Broken pipe]
<wpwrak_>
analog bw = the frequency at which the signal is attenuated by 3 dB
<rz2k>
never buy those dso_quad - they are horrible and slow. buy an 10-20 year old digital scope from hp-agilent/tektronix/lecroy and have a good hacking. one of my pals has 1991 year HP-Agilent logic analyzer system, it still can measure what a typical hardware hacker would ever need.
zear_ has joined #qi-hardware
<rz2k>
it runs some weird dos and screen saw better days, but with last calibration around 2000-s it is still pretty accurate
<nicksydney_>
rz2k: that's way OVER my budget ..unfortunately
<whitequark>
a CRT!
<wpwrak_>
for a modern scope you want: at least 60 MHz analog BW, preferably >= 100. sample rate at least 5x the BW, preferably higher. memory, 1 MSa or more. the more the merrier. digital connectivity for readout and also for automated setup.
<nicksydney_>
Dual channel-channel bandwidth 60MHZ-100MHZ; • The real time sampling rate is 1GSas;
<whitequark>
>Floppy Drive: Yes
<whitequark>
hehe
lilvinz has quit [*.net *.split]
zear has quit [*.net *.split]
sanderr has quit [*.net *.split]
<wpwrak_>
so the entry-levels rigols just fit that profile. oh, and you want to avoid the lesser china crap, like owon, hantek, siglent, and such
<wpwrak_>
they often have "nice" specs but basically everybody who has one curses it
<nicksydney_>
wpwrak_: good tip....looks like just focus on either older model or Rigol
<whitequark>
that lecroy scope is rather impressive btw, for its cost
<nicksydney_>
what really makes a scope so very expensive ? is is the components inside it or the amount of time it takes to develop the fw ?
sanderr has joined #qi-hardware
<rz2k>
nicksydney_: low product quantity (compare scope to your coffemachine), extremely hard to make - custom made hybrid ICs on the frontend, lots of RAM and FPGAs on the backend to aquire that megasamples (that is literally million samples taken from the frontend per second to the backend and then shown to you, or stored)
<wpwrak_>
"Probes: No" hehe :)
<rz2k>
so yes, before Rigol did their 50Mhz low price scope on Altera FPGAs and some cheapy frontend, they were $1K+ and no lower for digital ones
<nicksydney_>
rz2k: but isn't the price of FPGA, RAMs are now very cheap compared to how it was few years ago ?
<wpwrak_>
nicksydney_: ah yes, atten ... more china crap ;-)
<nicksydney_>
rz2k: ahhh
<wpwrak_>
looks like a dumbed down rigol. be assured that the dials and such will wear out rather quickly :)
mth__ has joined #qi-hardware
<nicksydney_>
saw this post ...."I would like to encourage everyone interested in used equipment to check out Agilent's premium equipment store on eBay. They offer almost new, slightly used equipment incl. cal certificates at huge discounts of up to 70% of list price. Examples are the Agilent Arb 33522 that has a list price of 3,200 US$ is offered for 2,100 US$ and they accept offers below their asking price. The equipment is
<nicksydney_>
shipped directly out of Agilent's factory in Malaysia (at no cost for the buyer). You have to be somewhat patient and watch the store ~ once per week to find equipment that fits your need.
<nicksydney_>
Most of the equipment is still very expensive, but if you are patient and with some luck you may find what you need, more or less brand new at 70% discount.....
<wpwrak_>
(cheap ram) that's why you can now get scopes with 140 MSa :) (ridol DS4000 series. alas, serious budget-busters)
valhalla has quit [Read error: No route to host]
<wpwrak_>
agilent is good but unfortunately a bit pricy
<rz2k>
and as I said, things like scopes or good multimeters dont wear out - you could buy that 500mhz lecroy in 1995 and you can still use it now, just calibrate it and you are good to go.
<wpwrak_>
they also have somewhat low specs on memory and such. and a lot of nice things are optional.
mth_ has quit [*.net *.split]
<wpwrak_>
well, eventually something will get them ...
<nicksydney_>
seems like no way out except save some more for me :)
<rz2k>
buy unlocked rigol on eevblog forum
<rz2k>
there were a series of rigols that you could unlock from 50mhz to 100mhz without problems
<rz2k>
only uart needed
<wpwrak_>
or just a code in some cases
<wpwrak_>
(DS1000Z, DS2000, DS4000)
<nicksydney_>
will troll around the forum and see if i can pick one uip
<rz2k>
I bet rigol themselves said what and where is used, they have much more money from selling the units to every hardware hacker around the world because of the keygen than from selling those unlockable options
<rz2k>
I have one around, it has EMI shield consisting of a little piece of copper scotch tape above one of the DACs, should I mention that this scope is horribly noisy and slow?
<nicksydney_>
reading through the forums i concur that it would be ideal to stay away from non-Rigol brand
<nicksydney_>
seems like lots of humans are having problem with non-Rigol :)
<nicksydney_>
wpwrak_: regarding USB....if the datasheet from MCU does not specify resistor value for D+ and D- ... what is the best way to estimate it ?
<whitequark>
nicksydney_: terminating resistor or pull-up?
<nicksydney_>
terminating i guess
<whitequark>
afaik that should match the cabling. it's specified in USB spec
<nicksydney_>
pull up normally used for D- if i correctly read the schematics
<whitequark>
it depends on speed. D+ for low-speed, D- for high-speed (or vice versa, never could remember that)
<whitequark>
so, terminating resistors for USB are 68 ohm
<nicksydney_>
so pull up resistor if in D+ indicate low speed while on D- it's high speed ? is that how i read it ?
<whitequark>
let me recheck
<whitequark>
yeah I mixed it up. pull-up on D- is low-speed, D+ is full-speed
<whitequark>
high-speed is full-speed plus additional negotiation
<whitequark>
whoever named those modes must have deliberately been confusing
<nicksydney_>
cool thanks whitequark
<whitequark>
low-speed is 1.5MBps, full-speed is 12Mbps, high-speed is 480Mbps
<wpwrak_>
(flir) what, no cat walking into the image on that nice big empty floor ?
<wpwrak_>
nicksydney_: resistors depend on the chip. some have built-in resistors, some have some resistors build-in and others not, and there are variations in the termination resistors as well
qi-bot has quit [Ping timeout: 245 seconds]
<larsc>
this definitly reassures your trust in the kernel code: /* AB: FIXME!!! This is definitely nonsense */
porchao has quit [Quit: Leaving...]
porchao has joined #qi-hardware
dos1 has quit [Read error: Operation timed out]
dos1 has joined #qi-hardware
dos1 has quit [Ping timeout: 265 seconds]
qi-bot has joined #qi-hardware
rz2k has quit []
<whitequark>
nicksydney_: on the second though my advice about resistor values was flawed, ignore it
panda|x201 has quit [Ping timeout: 252 seconds]
Guest88517 has quit [Ping timeout: 264 seconds]
wej has quit [Ping timeout: 252 seconds]
wej has joined #qi-hardware
<nicksydney_>
wpwrak_: you use the yellow paper for the toner transfer correct ?
<wpwrak_>
no, i use a "glossy" printer paper
<nicksydney_>
ahh ok ... thought you are using the same yellow paper that i just got today
<wpwrak_>
naw, always had good luck with easily sourced printer paper :)
<wpwrak_>
haven't seen this one yet, no. there are others and also some nasty chemicals. but just tinning with the soldering iron accomplishes the same and is quick and easy.
pcercuei has joined #qi-hardware
<whitequark>
nicksydney_: wpwrak_ is right. I heard it's possible to tin pcbs with rose's alloy, citric acid and water, but apparently the resulting coating is suboptimal
<whitequark>
so while it looks very neat, I wouldn't really recommend that.
<DocScrutinizer05>
\o/ 500246471 a) 597pcs 6.50EUR/1 14days shipping b) 1000pcs 2.55USD/1 shipping 3-5days c) 1000pcs 1.55USD/1 shipping 1 week \o/
wej has quit [Ping timeout: 245 seconds]
<wpwrak_>
now, quick, let's buy them all before DocScrutinizer05 does ;-)
<DocScrutinizer05>
good luck! you'll not even receive the offers before we receive our components
<DocScrutinizer05>
this are true offers mailed to us, so no hoax
<DocScrutinizer05>
and you can't find those in the intarweb
<nicksydney_>
DocScrutinizer05: what component is that for ?
<DocScrutinizer05>
Neo900 display connector
<DocScrutinizer05>
the damn unobtainium obsolete component
<DocScrutinizer05>
we got three *real* sources now
* DocScrutinizer05
does a dance of joy
<nicksydney_>
can you show me url with the image curious want to see how it looks like
<nicksydney_>
so looks like there are money to made if we can hoard component and keep it for few years :)
* nicksydney_
managed to find on how to create multiple copies inside PcbNew so as not to waste copper .. create 2 design in 1 copper..wohoo
<wpwrak_>
whoa. massive thunderstorm. a c't some 1.5 from the window got drenched :)
<DocScrutinizer05>
LOL
<nicksydney_>
DocScrutinizer05: it looks like a clipped connector
<nicksydney_>
i can see the mechanical drawing from datasheet
<nicksydney_>
wpwrak_: pop up the vodka and enjoy the day :)
<wpwrak_>
hmm yes, a nice icy vodka from the freezer would be nice :)
<wpwrak_>
the ice comes free from the sky at the moment
<nicksydney_>
all works out just fine when you trust life to nature :)
<nicksydney_>
played around with kicad and realised that you can create a connection in PcbNew even if it does not exist in the schematics..awesome !
pcercuei has quit [Quit: brb]
<wpwrak_>
just turn off DRC and pcbnew will let you do anything you want, no matter how evil ;-)
<nicksydney_>
how to turn it off
<DocScrutinizer05>
wpwrak_: don't take that ice, don't help them to discard their toxic waste and fallout
pcercuei has joined #qi-hardware
wej has joined #qi-hardware
<DocScrutinizer05>
now for that nasty common-anode vs common-catode RGB LED issue. Will cost me another day
<nicksydney_>
DocScrutinizer05: new chinese phone startup http://oneplus.net/ ... they are joining forces with Cyanogen (commercial entity of Cyanogenmod)
<DocScrutinizer05>
mhm
<wpwrak_>
nicksydney_: click on the little bug icon at the top of the left icon bar
<nicksydney_>
wpwrak_: there is no option inside that to turn off the DRC
<nicksydney_>
DocScrutinizer05: the so called 'CEO' of that company is ex OPPO manager :)
<nicksydney_>
out-of-the-box Cynogenmod version of Android phone :)
<DocScrutinizer05>
I'm not much interested in any flavor of andridiot
<nicksydney_>
andridiot :)
<nicksydney_>
wpwrak_: sorry misunderstood you ... found the 'bug' icon
<DocScrutinizer05>
I'm just enjoying the joy of dealing with a fsckdup Qualcom phone that is based on a android HAL, though they put some new OS (sailfish) on top of it
<DocScrutinizer05>
causing nausea
<nicksydney_>
miss the good ol' nokia days...where computer is a computer and phone is for sms and voice...life was simpler
lekernel has joined #qi-hardware
mstevens has joined #qi-hardware
<mstevens>
is anything interesting going on in the free hardware world?
mth__ is now known as mth
<DocScrutinizer05>
wpwrak_: do you still have a mail account on openmoko.org?
<lekernel>
mstevens, yes, migen, misoc, mixxeo
<lekernel>
and the soon to be announced physics application of the first two
<lekernel>
mstevens, and yosys
<mstevens>
lekernel: hmm, I can see how those might be cool if you could design hardware
<wpwrak_>
DocScrutinizer05: mail to openmoko gets forwarded to almesberger.net
<wpwrak_>
the winter in the US must be really bad. last evening, i ordered from fedex about an hours before the deadline and it only shipped today. first time ever that they don't do it on the same day.
<DocScrutinizer05>
wpwrak_: that's fine. I however wonder why my TLS-ESMTP fails
<DocScrutinizer05>
can't send anymore, receive fine
Luke-Jr has quit [Read error: Connection reset by peer]
Luke-Jr has joined #qi-hardware
Luke-Jr has quit [Read error: Connection reset by peer]
Luke-Jr has joined #qi-hardware
<DocScrutinizer05>
roh: ping
<DocScrutinizer05>
roh: I could use some help with failing TLS-SMTP on openmoko.org MX
<DocScrutinizer05>
I am a bit lost with my wireshark analysis of what's going on
jekhor_ has quit [Ping timeout: 276 seconds]
porchao has quit [Ping timeout: 260 seconds]
porchao has joined #qi-hardware
<DocScrutinizer05>
wpwrak_: number 4 offer
<DocScrutinizer05>
3.60EUR, also 597 ;-)
FDCX__ has quit [Ping timeout: 245 seconds]
<wpwrak_>
heh, that's the same stock then ;-)
FDCX has joined #qi-hardware
arielenter has joined #qi-hardware
<roh>
DocScrutinizer05: failing? i wonder that its still up
<DocScrutinizer05>
yeah, looks like
<roh>
my guess: invalid certs...
<roh>
validity time ran out or so
<DocScrutinizer05>
yep, my guess as well, but that's a server side issue, no?
<DocScrutinizer05>
wpwrak_: what did you do? you talked to Mr Brill? :-D
<DocScrutinizer05>
wpwrak_: when you suggest a camera I should get, you should also provide a exact spec you would suggest, since I have no idea how much this thing would cost
Luke-Jr has quit [Read error: Connection reset by peer]
<wpwrak_>
hehe, i did ;-)
<wpwrak_>
both, actually :)
mstevens has left #qi-hardware [#qi-hardware]
<wpwrak_>
the one i'm suggesting is actually fairly cheap, only some EUR 200 i think. the Canon ELPH 330 HS / IXUS 255 HS
<wpwrak_>
(canon have crazy naming)
<wpwrak_>
wos monster thunderstorm
<wpwrak_>
wow. at 2 pm. didn't see that one coming.
jekhor_ has joined #qi-hardware
<whitequark>
oh, fuck. apparently I need to sign a contract that I won't import over 1kEUR and over 31 kg per calendar month.
<whitequark>
gah, I hate customs so much
<whitequark>
good thing though, the cnc machine is being packed!
<wpwrak_>
nicksydney_: no idea ;-) you keep on finding weird devices nobody has ever used for such things yet :)
<nicksydney_>
technically if i connect a micro USB connector for pin D+, D- and VBUS to a resistor and LED and when it is plug into the computer the LED will light up .. correct ? http://ctrlv.in/280364
<nicksydney_>
wpwrak_: borderline between creativity and weirdo is very thin :)
<nicksydney_>
thought of making some micro usb pcb as i have few of those to learn how to solder them :)
Luke-Jr has quit [Excess Flood]
<wpwrak_>
don't use D+/D- for that :) and you can't control ID, the 5th signal.
<wpwrak_>
one LED between VBUS and GND is fine. anything else needs a bit more intelligence.
<wpwrak_>
e.g., an MKL2x... :)
Luke-Jr has joined #qi-hardware
<nicksydney_>
need to read up on D+ and D-..thanks wpwrak_