Back
[01:41:33] <budved> Hi all, does anyone know if it's possible for a 3-wire servo (ground, power, control) has the potential to cause an AVR chip to restart, or fire certain interrupts (such as spi_stc_vect)?
[01:47:16] <Kevin`> budved: servos use a moderate amount of power and can generate electrical noise. do you have bypass capacitors on the avr chip? do you have the brownout detector enabled?
[01:47:59] <budved> yeah i have some filter caps by the avr, but I don't have the brownout detector enabled (unless it's enabled by default?)
[01:48:24] <Casper> budved: a servo is known to cause enought noise to cause reset and badness in avr, this is why it's often suggested to power it from another regulator, or from no regulator at all
[01:48:46] <Kevin`> budved: it's not enabled by default
[01:49:02] <budved> okay, thanks for the insight guys
[01:49:21] <budved> I'll look into the brownout detector for now
[01:49:50] <Kevin`> you could also put some filtering on the power going to the servo
[01:50:19] <budved> Kevin`, thanks, I'll try that as well
[01:55:27] <budved> Kevin`, I'm having trouble reproducing the problem now that I added another filter cap near the servo, thanks for the suggestion
[02:12:24] <grummund> budved: check MCUSR register for cause of reset
[02:13:37] <grummund> code snippet here -
http://www.nongnu.org/avr-libc/user-manual/group__avr__watchdog.html
[02:14:17] <budved> grummund, alright thanks! I'll take a look
[03:43:13] <ferdna> would you prefer a computer based oscilloscope or a stand alone unit?
[03:44:43] <grummund> how kind of you to offer... ;)
[03:47:58] <ferdna> grummund, i have an old oscilloscope a tekronics... is yours if you want it... it doesnt work pay for shipping only.
[03:48:50] <grummund> hmm, does the computer based scope work?
[03:49:38] <ferdna> i dont have a computer based one... i was thinking on getting those chinese ones fro ~200
[03:51:14] <grummund> i'd go for a (working) bench scope, personally.
[03:52:00] <grummund> even 2nd hand off ebay, doesn't need to be all singing & dancing.
[03:52:44] <ferdna> grummund, why a bench scope?
[03:53:11] <grummund> personal preference i suppose, just wht i'm used to.
[03:53:54] <ferdna> i see
[04:21:42] <Amadiro> ferdna, most people swear on stand-alone units
[04:41:07] <ferdna> Amadiro, why is that?
[04:43:17] <Amadiro> ferdna, I don't know, I've never used a usb pc-based scope myself, but it appears most people have strong feelings against them.
[04:43:42] <ferdna> i see
[04:45:22] <OndraSter> I don't know, I don't trust them :P
[04:45:40] <OndraSter> anything PC requires usually their proprietary software
[04:57:53] <OndraSter> how the hell.. I have got uint16_t addr in which I calculate some address
[04:57:58] <OndraSter> how the hell do I read data from that address?
[04:58:10] <OndraSter> uint16_t* data = *addr does not work
[04:58:18] <OndraSter> whoops
[04:58:19] <OndraSter> I lied :D
[04:58:21] <OndraSter> typo
[04:58:22] <OndraSter> nvm
[04:59:31] <ferdna> well thanks
[04:59:36] <ferdna> good night.
[04:59:40] <OndraSter> gn
[05:03:18] <OndraSter> Corwin, high five for nobel prize!
[05:03:29] <Corwin> eh, what?
[05:03:36] <OndraSter> I have nobel prize
[05:03:38] <OndraSter> you have nobel prize
[05:03:46] <OndraSter> every EU member does
[05:03:46] <OndraSter> lmao
[05:03:49] <Corwin> oh, that
[05:04:37] <Amadiro> It's just a peace prize, though, nothing to get excited about
[05:04:45] <Amadiro> I anticipated getting one of those sooner or later by coincidence
[05:04:47] <OndraSter> *just*
[05:04:48] <OndraSter> lol
[05:04:53] <OndraSter> by an accident!
[05:04:57] <Steffann> It's a devaluation of a nobel price ...
[05:05:12] <OndraSter> nobel prize hasn't had value for quite some time
[05:05:17] <Amadiro> OndraSter, that's how they are given out -- some guy throwing darts at a long list
[05:05:18] <Steffann> EU and peace.. EU => money => peace :S
[05:05:25] <vsync_> :SSss
[05:05:27] <OndraSter> Amadiro, hehe
[05:05:31] <Amadiro> This time the dart arrow just shattered and they couldn't be arsed to buy a new one
[05:05:44] <OndraSter> not enough money!
[05:06:10] <Amadiro> OndraSter, the other nobel prizes are still fairly prestigious, though
[05:06:20] <OndraSter> if they were I would have one!
[05:06:24] <OndraSter> :P
[05:06:34] <Amadiro> I'd probably be more aiming for a fields medal or abel prize, though...
[05:07:12] <Amadiro> Kind-of depressing for mathematicians between 40 and 60 or so -- fields medal is for people under 40, abel prize is never given to anyone under 60
[05:07:20] <OndraSter> heh
[05:07:32] <Richard_Cavell> How's everyone going?
[05:07:32] <OndraSter> and MY INTERNET IS STILL DOWN
[05:07:42] <Corwin> *your* internet ?
[05:07:45] <OndraSter> yes
[05:07:47] <Richard_Cavell> I've almost talked this cute girl at work into going out with me
[05:07:47] <OndraSter> the one in my home
[05:07:55] <OndraSter> Richard_Cavell, was her name AVR?
[05:07:58] <Richard_Cavell> no
[05:08:02] <OndraSter> PIC ?!
[05:08:09] <Amadiro> the heresy!
[05:08:12] <Richard_Cavell> Ellen
[05:08:24] <Amadiro> Ellen? that ain't no ISA I've ever heard off!
[05:08:55] <Steffann> Why people from aussieland always talk about their love life in publicly logged channels?
[05:08:58] <tld> I'm not sure I'm into the whole "dating a cow-orker"-thing.
[05:09:19] <tld> Steffann: because the nearest neighbor is 200km away?
[05:09:26] <Steffann> Could be :)
[05:09:27] <OndraSter> LOL
[05:09:28] <OndraSter> good one
[05:09:36] <Amadiro> Or too much vegemite blurrs your better judgement
[05:09:44] <Amadiro> That stuff is potent
[05:10:08] <tld> I just ordered 30awg kynar, against better judgement.
[05:10:08] <Steffann> Amadiro had a nice conversation with abcminiuser? :P
[05:10:22] * tld will bbl
[05:10:23] <Amadiro> Steffann, about vegemite? nop
[05:11:03] <Steffann> No, but perhaps you noticed some blurryness
[05:15:34] <RikusW> its called Marmite over here ;)
[05:21:11] <creep> Richard_Cavell<< you mean how many different girls we fuck every week?
[05:21:51] <creep> i heard AIDS is from africa
[05:21:55] <RikusW> Steffann: which other aussies are you referring to ? :-P
[05:24:40] <Steffann> Aussies all around the web
[05:24:46] <Steffann> *over
[05:25:19] <Richard_Cavell> creep: I didn't ask you that
[05:25:30] <Richard_Cavell> I just said, triumphantly, that there's this cute girl at work and I asked her out
[05:25:37] <Richard_Cavell> And she's a bit shy but she's keen
[05:27:29] <RikusW> Richard_Cavell: do you know what a burr cell and schistocyte is ?
[05:27:39] <Richard_Cavell> nup
[05:27:50] <RikusW> o nvm then ;)
[05:28:06] <RikusW> seems hematology isn't your speciality
[05:28:31] * RikusW have been reading a lot of med stuff recently
[05:29:09] <RikusW> Richard_Cavell: basically the presence of such RBCs are due to high BUN
[05:29:50] <Richard_Cavell> ok
[05:29:57] <Richard_Cavell> no haematology is not my specialty
[05:31:43] <RikusW> I do have a microscope with 4 10 and 40x objectives and a 10 and 16x eyepiece
[05:31:53] <RikusW> and plenty of time to mess around
[05:32:02] <creep> 40x s not too much
[05:32:13] <RikusW> 40x10 = 400
[05:32:17] <creep> ah
[05:32:26] <RikusW> or 640
[05:32:41] * RikusW wants a 100x oil immersion objective now :)
[05:32:48] <RikusW> 1000x !
[05:32:51] <OndraSter> wow
[05:32:54] <OndraSter> YOU CAN NEARLY SEE ATOMS!
[05:33:01] <RikusW> not even close
[05:33:05] <creep> inspect the tail of the sperms
[05:33:15] <RikusW> yeah
[05:33:26] <OndraSter> RikusW, when I write something in caps it is usually very ironic
[05:33:35] <RikusW> or watch them swim around too :-P
[05:33:49] <Corwin> ehm... am i really in #avr?
[05:33:50] <creep> and make video and post to youtube
[05:33:57] <Richard_Cavell> Corwin: Yes
[05:34:20] <OndraSter> RikusW, you can dissect an old AVR MCU and make high high high resolution photos of it
[05:34:22] <OndraSter> of the die
[05:34:25] <RikusW> Corwin: this is now the Animal Veterinary Research channel :-P
[05:34:28] <OndraSter> and reverse engineer it :P
[05:34:32] <Corwin> :D
[05:34:47] <creep> animal planet
[05:34:58] <creep> discovery channel
[05:35:23] <RikusW> OndraSter: I have photographed some UVEPROMS
[05:35:30] <OndraSter> that is not AVR
[05:35:58] <RikusW> though the 40x lens cant focus on the die, it needs to be <1mm from the surface to focus
[05:36:30] <RikusW> so 160x is the best I can do
[05:36:37] <creep> that is the thickness of the cd disc
[05:36:46] <RikusW> less
[05:36:58] <creep> dvdv disc is half that then
[05:38:27] <creep> this is why dvd is better, data surface is sandwiched between 2 equally thick polycarbonate sheets
[05:42:14] <RikusW> yeah, scratching a CD on top is devastating...
[05:42:36] <creep> just write on it with a pen
[05:48:57] <creep> can you examine mushroom spores with it ? or 300nm particles/viruses?
[05:49:43] <vsync_> can you examine aids with it
[05:49:47] <RikusW> spores easily
[05:50:05] <GuShH> vsync_: yeah just hand him a sample of your blood
[05:50:15] <vsync_> that's mean
[05:50:21] <RikusW> creep: um sized particles are easily visible
[05:50:22] <GuShH> naw
[05:50:40] * GuShH turns vsync off
[05:50:57] <RikusW> creep: you'll need an electron microscope for viruses
[05:51:08] <RikusW> or do a PCR test
[05:51:17] <RikusW> aka DNA test
[05:52:48] <vsync_> so, makes me wonder, you're an MD by profession?
[05:52:52] <vsync_> or?
[05:53:23] <OndraSter> he is The House.
[05:54:07] <RikusW> creep: using a 2Mp cellphone camera and 640x its about 4 pixels / um
[05:54:36] <RikusW> I have a micrometer slide too, with 10um markings
[05:55:43] <specing> Anyone here already has the stellaris launchpad?
[05:56:18] <vsync_> RikusW?
[06:00:19] <creep> an M8 threaded rod driven by a 1.8 deg/step stepper motor gives 6.25um/fullstep
[06:00:40] <RikusW> and how much backlash ?
[06:00:47] <creep> 0
[06:01:01] <RikusW> can't be 0
[06:01:04] <creep> use a backlash free screw
[06:01:40] <creep> there will probably be linearity errors due to dust and greases and machining errors
[06:02:03] <vsync_> sometimes it is hard trying to grasp the flow of #avr
[06:02:18] <Corwin> sometimes? :D
[06:02:23] <creep> a fungi spore sticks between the threads and you have 3um error ;/
[06:03:19] <RikusW> metal is harder than a spore...
[06:04:01] <creep> i always think about spores like some peanuts with a hard shell
[06:04:27] <RikusW> vsync_: Just interested in medicine
[06:04:52] <RikusW> OndraSter: lol, not quite house yet :-P
[06:07:01] <creep> would you examine her too ?
http://www.sacramentomenshealth.com/fda-appears-poised-to-get-off-its-fat-behind/
[06:08:40] <creep> http://www.musingcontinuum.com/category/life-imitates-horror/page/2/
[06:08:48] <creep> flesh eating bacteria ;/
[06:09:07] <vsync_> nice
[06:12:14] <RikusW> we don't feed our cattle ANY antibiotics..
[06:12:27] <RikusW> I consider that just plain insave
[06:12:29] <RikusW> *insane
[06:13:05] * RikusW don't like the feedlot idea either
[06:13:57] <OndraSter> doesn't*
[06:14:54] <RikusW> why does some languages have to differentiate between singular an plurar ? :-P
[06:15:20] <OndraSter> almost all do :P
[06:24:57] <OndraSter> I know what the world lacks!
[06:25:01] <OndraSter> GSM serial console!
[06:25:06] <OndraSter> as in serial console that you can connect to
[06:25:10] <OndraSter> that works over GSM somehow
[06:25:28] <OndraSter> because when your network disconnects you have to drive there :P
[06:27:57] <Amadiro> http://www.sparkfun.com/news/384
[06:28:02] <Amadiro> holy shit, those probes are amazing
[06:28:19] <Amadiro> 0.25 microns, can test a single npn junction inside an AVR
[06:31:33] <RikusW> OndraSter: telnet over GPRS ?
[06:31:41] <theBear> OndraSter, it's called a mobile, i first did it last century
[06:31:42] <OndraSter> yes
[06:31:45] <OndraSter> heh
[06:32:14] <RikusW> or better ssh
[06:33:42] <theBear> i used to just do serial... gsm module looks exactly like a at modem over serial... was easier in the days when phones had hw serial interface and not usb
[06:33:47] <creep> http://en.wikipedia.org/wiki/Necrotizing_fasciitis
[06:33:48] <creep> hah
[06:33:57] <theBear> what's that house ?
[06:33:59] <creep> 2009 R. W. Johnson, South African journalist and historian, contracted the disease in March 2009 after injuring his foot while swimming. His leg was amputated above the knee
[06:34:17] <creep> 2011 Jeff Hanneman, guitarist for the thrash-metal band Slayer, contracted the disease in early 2011, allegedly after being bitten by a spider
[06:34:30] <creep> spiderbite, swimming in africa ;/
[06:38:09] <specing> the second one was probably in australia, where everything wants to kill you
[06:39:00] <theBear> i don't want to kill you, just some other people
[06:40:47] <specing> Im still not going there, ever.
[06:41:05] <theBear> i promise i won't kill you, ever
[06:41:31] <Roklobsta> where?
[06:41:44] <specing> Roklobsta: < specing> the second one was probably in australia, where everything wants to kill you
[06:41:56] <theBear> he said it 3 minutes ago
[06:42:09] <Roklobsta> yah, i had an emu mencaing the car the other day
[06:42:12] <Roklobsta> menacing
[06:42:17] <specing> 0.o
[06:42:54] <specing> I heard emus chase you
[06:42:57] <specing> in packs
[06:43:03] <specing> like raptors
[06:43:21] <RikusW> ostriches might be even more dangerous
[06:44:13] <theBear> you just gotta stay still, then they can't see you
[06:44:13] <Roklobsta> no emus are a bit smaller but they do attack
[06:44:50] <Roklobsta> i have been out on family bbqs and emus just come over and eat the food out of your hand as you are putting it to your mouth.
[06:44:59] <Roklobsta> they eat the suasages right off the hotplate too
[06:45:25] <impulse> is that why you kick them?
[06:45:26] <theBear> you gotta look more like a 10' tall bear, that puts them off, and there's a lotta neck to grab and throttle if you have to
[06:45:37] <theBear> hehe, nah, that's quokka socca
[06:45:56] <impulse> quokka?
[06:46:05] <Roklobsta> leave the cute quokkas alone
[06:46:09] <theBear> look it up
[06:46:29] <theBear> i don't hurt animals, and animals don't kill people ... i kill people
[06:46:32] <impulse> looks like a rat
[06:46:44] * theBear is like guns
[06:47:40] <impulse> you guys got some interesting wildlife there
[06:47:47] <theBear> i AM interesting
[06:48:38] <impulse> we have bears here too
[06:48:46] <Roklobsta> dropbears
[06:48:50] <OndraSter> lol
[06:50:12] <Roklobsta> ondra how many megs left?
[06:50:28] <Roklobsta> on your hippie 3G?
[06:51:00] <theBear> heh, i never noticed till now... all hippies only have 3g net these days
[06:52:28] <Roklobsta> ondra is using some free love 3g
[06:54:06] * RikusW only have 30MB...
[06:54:22] * specing has 3T
[06:54:43] <RikusW> *had ?
[06:55:10] <specing> had?
[06:55:17] <theBear> haddock ?
[06:55:27] <specing> Rawr
[06:55:39] <theBear> that's flake, not haddock
[06:55:41] <RikusW> obviously you used a lot of it already :-P
[06:56:42] <OndraSter> about 57MB left
[06:59:04] <OndraSter> I like Vodafone: "You have just reached 90% of your FUP. For further informations look at
http://vodafone.cz/something" lol
[06:59:25] <Roklobsta> 3T on 3G?
[07:07:05] <bluebie> Hopefully a quickie - I've just started making use of the reset pin on tiny85 for IO - I've programmed rstdisbl fuse, and it seems to work, but it's seemingly remaining an input pin - when I write to it, it outputs current at about the same rate as other pins when they're using the internal pullup
[07:07:50] <bluebie> a few minutes googling hasn't turned up anything juicey. Is there something I should be looking at, to switch this pin to being an output, beyond the usual DDRB register?
[07:12:08] <RikusW> good question..
[07:12:21] <RikusW> bluebie: and how will you be turning off RSTDSBL ?
[07:14:43] <RikusW> bluebie: did you maybe program DWEN too ?
[07:14:54] <OndraSter> mmm playing with RSTDSBL is fun
[07:15:11] <bluebie> RikusW: Where's your sense of adventure!
[07:15:16] <bluebie> I bootloaded it first :)
[07:16:13] <RikusW> bluebie: do you have a HVSP programmer too ?
[07:16:21] <bluebie> I'm pretty sure I didn't program DWEN
[07:16:29] <bluebie> but to be honest, I can't be totally 100% certain
[07:16:35] <bluebie> um
[07:16:48] <RikusW> what fuse values did you set ?
[07:16:56] <RikusW> hex..
[07:17:03] <bluebie> yeah there's a dragon over on the table but I don't know how to use it and it doesn't look to have any chip sockets soldered in or anything
[07:17:18] <bluebie> I actually did it through a GUI so I don't have the fuse bytes at hand
[07:17:55] <RikusW> you'll have to solder a 2x20 or 2x 2x10 connectors into the dragon
[07:18:13] <RikusW> AVR Studio does have docs explaining HVPP/HVSP connections
[07:18:18] <bluebie> anyhow, is the implication that this is quite unusual and I'm not just missing some obvious extra step normally needed in addition to the rstdisbl fuse to use pb5 as an output?
[07:18:46] <RikusW> ADC maybe overrides the pin ?
[07:18:50] * RikusW goes checking
[07:19:55] <RikusW> seems ADC0 might be doing that...
[07:20:19] <bluebie> I haven't touched the ADC at all since reset o_o
[07:20:34] <bluebie> got a page number in a datasheet or something I can go read more about this?
[07:21:05] <RikusW> ports chapter
[07:36:53] <bluebie> maybe I've just shorted the pin some time and burnt out the chip :/
[07:56:47] <OndraSter> is there any arduino with 64 pinned chip?
[07:56:53] <OndraSter> so I know on which size I should base it
[08:04:01] <Amadiro> OndraSter, I don't think any of the standard arduinos (uno, leonardo, duemilanove, mega) have 64-pin chips
[08:04:12] <OndraSter> thanks
[08:04:53] <Amadiro> Some of the more exotic ones I haven't looked at might (arduino pro, lilypad, nano, mini...)
[08:05:31] <Amadiro> no, lilypad seems to have very few pins
[08:06:12] <bluebie> yeah lilypad are atmega 168 or 328, but with few pins broken out on the PCB
[08:28:13] <OndraSter> so, have mythbusters already tested the "no right angles" at different speeds and tracks?
[08:29:14] <theBear> not sure what yer mean, but well, they're not exactly scientists, or aware of you know, things like scientific method
[08:29:57] <OndraSter> :P
[08:30:00] <OndraSter> but they do all the myths!
[08:30:37] <theBear> all depends what you mean by 'do' :) they certainly mention a myth at the start of every segment, there's no denying that
[09:01:28] <megal0maniac> jadew: The OLS app is looking good. Neat, seemingly bug free (without actually having the OLS, but I did try to make it fall over) Protocol .js files are nicely commented
[09:28:41] <jadew> thanks for trying it out and for the input megal0maniac_afk
[09:34:34] <jadew> going afk as well o/
[11:34:39] <tld> Any experience with how much (protective) resistance you can usually add to 3.3V or 5V serial lines?
[11:39:45] <Casper> it depend on both end
[11:40:03] <Casper> usually, you want the lowest resistor possible
[11:40:59] <Casper> ex, if the pin is able to supply 20mA and 5V, figure out which resistor will cause a 20mA load when shorted to 5V and GND
[11:41:42] <tld> was hoping I could use something like that as a rule of thumb.
[11:42:18] <tld> (if it can take 20mA continuous, figure out resistor to limit to 19mA, use the next bigger value resistor I have… or some such)
[11:42:41] <Casper> you can go with max
[11:42:58] <tld> just curious if I'd typically run into troubles with something like that, or if there's a rule of thumb thing… basically limiting factor becomes RC-time dependent, so depends on the rest of the circuit?
[11:44:01] <Casper> the problem is more on the "max232" side than on the avr side
[11:44:15] <Casper> the avr have strong driving capability
[11:44:23] <Casper> the other component might not
[11:44:38] <Casper> so might need an higher resistor
[11:44:56] <Casper> but atleast the avr have a very high input impedance, so a resistor isn't an issue
[11:45:31] <tld> no max232, just wiring up a 3.3V xbee to a 5V avr, tempting to go higher on resistance "in case something bad happens", and started thinking about rules of thumbs for doing this kind of thing.
[11:46:38] <Casper> then your issue is the 5-> 3.3... are you sure that the xbee is 5V tolerant? last time I read about it, 5V kill it
[11:47:21] <OndraSter> xmega? :P
[11:49:51] <tld> I'm sure the Xbee *isn't* 5V tolerant, so I'm scaling down to 3.3V.
[11:50:00] <tld> which is partly what got me thinking about this.
[11:50:27] <tld> using a 10k + 20k resistor-divider to drop the voltage, added a zener for smaller overcurrent-situations (too paranoid perhaps)
[11:50:43] <tld> so I started thinking about 3.3V -> 5V line, Xbee tx to aver rx.
[11:50:52] <tld> if I should do something similar, simply for protection
[11:50:54] <OndraSter> not needed
[11:51:08] <tld> (who knows, I might accidentally set the pin to output and high… )
[11:51:11] <Casper> time to go shopping, bbl
[11:51:13] <OndraSter> avr has high from 1v8?
[11:51:17] <OndraSter> lol
[11:51:39] <tld> this is 5V, so if I'm setting the avr pin to a 5V high, it'd be a bad day for the xbee.
[11:51:56] <OndraSter> :P
[11:54:29] <tld> "Dead xbees don't transmit." <-- movie title?
[11:55:10] <OndraSter> lol
[12:03:57] <Horologium> 3.3V xener diodes to GND
[12:04:48] <tld> I figured 3.6V zeners
[12:05:34] <Horologium> use similar on USB connections when doing vUSB work.
[12:06:02] <tld> good point.
[12:06:13] <tld> brb
[12:10:05] <tld> my basic thought went something like "People use 10k+20k divider for stepping down, and if so, 10k might be okay, and given that, I could survive higher overvoltages."
[12:11:12] <tld> 10k, rather than, say 500 Ohm, would limit the speeds I could use, but if I'm using a 10k as part of a voltage-divider in "the other direction", it might be zero-loss.
[12:11:20] <Horologium> voltage divider would work too.
[12:12:05] <Horologium> if you had a pair of 10K resistors then your output from the divider comes out at 2.5V if fed 5V on the high side, obviously. that should be in the HIGH range for 3.3V input.
[12:12:52] <Horologium> using 10K and 20K gives less wastage and higher output voltage.
[12:12:56] * Horologium is just rambling.
[12:13:53] <tld> The way things are now, my 5V -> 3.3V is a 10k+20k divider, with a 3.6V zener, for "overkill over-voltage protection".
[12:14:05] <Horologium> should work.
[12:14:25] <tld> Just wondering about doing the same on the 3.3V -> 5V side as well, instead of just 3.6V zener and 500 Ohm (or thereabout)
[12:15:03] <tld> (if it limits higher speeds, that'd already be done by the 5V -> 3.3V side, and as long as I don't drop below the 5V high… hmm.)
[12:15:09] <Horologium> shouldn't need to do that with 3.3V to 5V as the 5V side doesn't care about lower voltage, so long as it is high enough to hit the HIGH state.
[12:15:38] <Horologium> at most you might have an issue with the 3.3V side not providing high enough voltage to drive the 5V side.
[12:15:43] <Horologium> in which case, use a transistor.
[12:15:57] <Horologium> drive the transistor with the 3.3V side.
[12:16:21] <Horologium> 10K or 20K pullup on the 5V side with the transistor pulling to GND.
[12:16:34] <tld> hmm, should work nice. :)
[12:16:45] <Horologium> that way even if you turn the 5V side on as an output it won't fry your 3.3V side.
[12:17:02] <tld> or… if I worry this might about overvoltages, perhaps I should just re-do things to use optoisolators?
[12:17:50] <tld> I'm not too worried about normal operation of the thing, I'm more curious about handling it if someone introduces shit and fan, close to my circuit. ;)
[12:17:56] <tld> which means all of this is probably overkill.
[12:18:29] <Horologium> or you could put a 74HTC part between them driven at 3.3V and all is good.
[12:19:21] <tld> how are they with overvoltages? (ESD and the like?)
[12:22:47] <Horologium> http://rocketnumbernine.s3.amazonaws.com/uploads/2009/04/level-converter-mosfet.jpg bidirectional level shifter.
[12:23:42] <Horologium> http://www.maximintegrated.com/images/appnotes/3007/3007Fig02.gif
[12:23:58] <Horologium> this should work for shifting either direction really.
[12:24:23] <Horologium> although, you will need a pullup resistor on the output side.
[12:25:13] <tld> any guesses as to what speed-ranges such a solution would work well?
[12:27:34] <Horologium> well, with the transistor method, it is all dependent on the transistor...and the other components as well, but, primarily how fast can that transistor switch.
[12:28:20] <Horologium> how are you with small surface mount packages?
[12:28:40] <tld> getting there. ;)
[12:28:52] <tld> not good with them yet, but usually ends up okay with the stuff I try
[12:29:44] <Horologium> is this a one off or hope-to-production device?
[12:30:22] <Horologium> and what kind of comms bus are you using? SPI I think you mentioned or am I totally off?
[12:30:59] <tld> serial
[12:31:10] <tld> and somewhere in between I suppose.
[12:31:18] <tld> *this* is a one-off, but hoping to bring some of the ideas forward
[12:34:11] <Horologium> unclocked serial then?
[12:34:17] <tld> yeah
[12:34:18] <Horologium> one line each direction.
[12:34:38] <Horologium> then a MAX3000E is overkill.
[12:34:43] <Horologium> it's 8 lines bidirectional.
[12:35:38] <tld> thinking maybe opto would be worth considering…
[12:36:44] <Horologium> MAX3372E
[12:37:52] <Horologium> http://www.maximintegrated.com/datasheet/index.mvp/id/3253
[12:39:42] <tld> hmm
[12:39:45] <tld> that looks *very* nice
[12:40:04] <tld> and I should be okay with at least the sot23 and tssop I think
[12:40:05] <Horologium> it does EXACTLY what you want.
[12:40:06] <tld> thanks. :)
[12:40:15] <Horologium> and at the voltages you are running, should run 8MHz
[12:40:15] <tld> many many thanks. :)
[12:40:20] <Horologium> or more.
[12:40:29] <Horologium> maxim makes a lot of nifty interface devices.
[12:40:32] <Horologium> AND
[12:40:40] <Horologium> if you live in the right countries, you can get free samples.
[12:40:58] <Horologium> not sure if they send freebies to Norway though, but can't hurt to ask.
[12:41:12] <tld> They FedEx them here.
[12:41:23] <Horologium> I get my hobby max233 chips free there.
[12:41:29] <tld> (I've been sampling maxim parts occasionally for 10 years or so)
[12:41:44] <Horologium> have half a dozen or so I've sourced as samples over the past few years.
[12:41:50] <Horologium> then there ya go.
[12:42:34] <Horologium> I love the max233 chips because I don't need external capacitors for rs232 level shifting.
[12:42:52] <tld> yeah. :)
[12:42:54] <Horologium> simplifies the hell out of board design.
[12:42:58] <Horologium> but they are pricy.
[12:43:06] <Horologium> compared to a max232 and a few caps.
[12:43:21] <Horologium> about double the price, so it's a tradeoff.
[12:43:36] <Horologium> ok, methinks is bath time.
[12:44:15] <tld> thanks for input. :)
[12:59:03] <SuperMiguel> any one familiar with webbotlib ??
[13:35:24] <SuperMiguel> so how do you get multiple servos to move at the same time??? when i use more than 2, lets say 3, one of them is out of sync with the other 2
[13:35:57] <OndraSter> show somebody the code
[13:37:35] <SuperMiguel> OndraSter, eve something with arduino, something simple like this
http://pastebin.com/vrNa8M7X
[13:38:23] <OndraSter> not me, I am almost out of my monthyl FUP
[13:38:28] <OndraSter> I have got about 24MB left :D
[13:50:49] <Horologium> SuperMiguel, so, how about the underlying code?
[13:51:45] <Horologium> if it is arduino then I might suggest asking in the arduino channel.
[13:52:32] <SuperMiguel> Horologium, well how would you do it? on regular avr code?
[13:52:34] <Horologium> if I want servos to move at the same time I drive them from the same port and pulse them both together that way.
[13:52:40] <SuperMiguel> lets say you have 20 servos and u want to move them at the same time
[13:52:50] <SuperMiguel> ummm
[13:52:55] <SuperMiguel> how about using different ports?
[13:53:17] <Horologium> I didn't mean same pin, but same port.
[13:53:23] <Horologium> 8 pins to a port.
[13:53:32] <Horologium> it depends on how close you want them to move.
[13:53:46] <Horologium> and, remember, not all servos move at the same speed, even of the same kind.
[13:54:18] <SuperMiguel> Horologium, well if they are programable servos you can set speed no?
[13:54:28] <SuperMiguel> but ya fairly close to each other
[13:54:33] <Horologium> what kind of servo are you using even?
[13:55:08] <Horologium> I'm talking about hobby servos, the kind that's used on model airplanes and such.
[13:55:43] <Horologium> that take a 1ms to 2ms pulse to determine the position it is to move to.
[13:56:28] <Horologium> there are many other kinds of servos but one needs to have a common frame of reference before we can discuss what you are using.
[13:58:20] <SuperMiguel> Horologium, ya that kind
[13:58:21] <SuperMiguel> http://www.servocity.com/html/hs-5645mg_digital_torque.html
[14:02:15] <Horologium> not seeing how those are exactly "programmable"
[14:02:21] <Horologium> even though they claim to be.
[14:02:57] <Horologium> aahh..
[14:02:59] <Horologium> there it is.
[14:03:03] <Horologium> you need their programmer.
[14:04:32] <Horologium> so, yes, in theory, you could adjust them to run at the same speed.
[14:05:11] <Horologium> the problem with driving them from something like an arduino comes into how efficient the driver code is.
[14:06:01] <Horologium> so, you have to pulse all the lines together as closely as possible.
[14:06:09] <Horologium> and as accurately as possible.
[14:06:58] <Horologium> at 1.5mS the servo should be centered, at 1mS it should be full swing one way and at 2mS full swing the other side.
[14:07:07] <Horologium> send the pulses out at 20mS intervals.
[14:09:50] <Horologium> using a timer interrupt routine with the timer set to 20mS you can send the pulses out at the right interval.
[14:12:27] <SuperMiguel> Horologium, but how would i send a big package to all of them?
[14:12:58] <Horologium> turn on the pin, wait the required amount of time, turn off the pin...or pins.
[14:13:20] <Horologium> do that 50 times a second.
[14:14:21] <Horologium> on an avr you would need to write to 4 different ports.
[14:14:37] <SuperMiguel> but if i turn on pin PB1, and turn PB2, and turn PB3, it takes 20ms from 1 to 2 and 20ms more to PB3
[14:15:00] <Horologium> write to all 4 ports, then do the wait.
[14:15:06] <SuperMiguel> let me try
[14:15:08] <Horologium> then write all 4 ports again.
[14:15:26] <Horologium> you just cut your wait time down enough that the write times are accounted for in the delay.
[14:27:10] <SuperMiguel> Horologium, so i would have to build my own library to control the servos that way?
[14:28:46] <Horologium> http://winavr.scienceprog.com/comment/164
[14:28:50] <Horologium> yes.
[14:28:55] <Horologium> there is a quick look.
[14:29:07] <Horologium> err..quick code segment.
[14:29:27] <SuperMiguel> thxs
[14:33:54] <Horologium> http://www.robot-electronics.co.uk/htm/sd20tech.htm there is a chip to do it...a preprogrammed pic.
[14:34:41] <Horologium> google is full of such things.
[14:44:50] <OndraSter> I am such a poor person
[14:44:57] <OndraSter> no interwebz
[14:44:57] <OndraSter> :(
[14:46:50] <Horologium> how do you get on irc? carrier pigeon?
[14:50:24] <tld> valuable lesson of the day: don't forget strain-relief.
[14:50:38] <Horologium> don't want to strain yourself after all.
[14:50:43] <Horologium> or your cables.
[14:51:46] <tld> row of 16… I fix one, it's neighbor goes, and so on.
[14:51:53] <Horologium> hehe
[15:05:32] <Horologium> SuperMiguel, the other thing one could do is replace the chip on each servo, turning them into real programmable servos with SPI interface maybe.
[15:17:26] <OndraSter> who stole my arduino!
[15:17:30] <OndraSter> I am using its FTDI only :P
[15:50:02] <OndraSter> hmm
[15:50:11] <OndraSter> USB CDC from atmel is over 9kB when compiled
[15:50:18] <OndraSter> my target is 4kB
[15:50:18] <OndraSter> is it doable in C?
[15:51:14] <specing> yes
[15:51:23] <OndraSter> USB CDC AND bootloader
[15:51:25] <OndraSter> in 4kB
[15:51:28] <specing> *C with inline asm
[15:51:29] <OndraSter> hehe
[15:51:43] <OndraSter> void main() { asm(".... everything ..."); }
[15:51:46] <specing> yep
[15:52:00] <OndraSter> I can do it in asm directly that way :P
[15:52:48] <Horologium> OndraSter, using vUSB or hardware USB?
[15:52:57] <OndraSter> hardware of course
[15:53:11] <OndraSter> I need to fit into 4kB because xmega32a4u has 4kB bootloader :)
[15:53:21] <Horologium> look at the teensy code.
[15:53:29] <OndraSter> teensy does not use xmega
[15:53:31] <Horologium> they have usb bootloader on there as I recall.
[15:54:23] <OndraSter> mayyybe when my internet comes back up
[15:54:27] <OndraSter> I have got few megabytes left :(
[15:54:33] <OndraSter> and the admins are probably for the weekend gone
[15:54:33] <OndraSter> HOW AWESOME
[15:54:53] <OndraSter> phone says 15.9MB of FUP left
[15:55:00] <OndraSter> yeah... about 10MB I can be sure of
[15:55:12] <OndraSter> there was few 0.x% difference between phone and vodafone themselves
[15:57:26] <OndraSter> riku
[15:57:28] <OndraSter> he is gone.
[15:57:35] <OndraSter> megal0maniac_afk, how big is Rikus' bootloader?
[15:57:37] <OndraSter> compiled that is
[15:58:05] <OndraSter> I am at 1.8kB without ISP and without CDC itself :/
[15:58:08] <OndraSter> only USB enumeration
[16:01:52] <OndraSter> I am done with today's internet
[16:01:53] <OndraSter> gn
[16:01:56] <OndraSter> "gn"
[16:03:23] <lethjakman_lapto> hey, I was wondering if there's a good inexpensive source for buying AVR chips?
[16:03:41] <lethjakman_lapto> mouser is kinda expensive, and avenet's shipping is kind of annoying
[16:05:28] <Horologium> there is always digikey.
[16:10:36] <lethjakman_lapto> hm, I'll check out how much their shipping is. I've always felt their prices were higher than mouser
[16:11:07] <Horologium> futurlec pricing wasn't bad when I last ordered from them.
[16:11:25] <Horologium> but it took 3 to 4 weeks to get stuff here from their shipping location in taiwan or wherever it was.
[16:12:07] <Horologium> I think they had them shipped over by canoe.
[16:12:20] <lethjakman_lapto> lol, probably had to paddle it all the way to your door
[16:12:47] <lethjakman_lapto> I'll look through them, I thik I'll probably stick with avenet....
[16:13:04] <lethjakman_lapto> has anyone here used a bus blaster/recommend a programmer? I'm curious what's out there
[16:13:16] <Horologium> I have a buspirate
[16:13:23] <lethjakman_lapto> do you like it?
[16:13:34] <Horologium> as for a programmer, I use either a homemade usb programmer or a dapa programmer.
[16:13:45] <lethjakman_lapto> you wouldn't happen to know if it can program the msp430 line do you?
[16:13:47] <Horologium> I like it for what it is but don't use it for an avr programmer.
[16:13:49] * lethjakman_lapto please don't hit me
[16:14:07] <Horologium> no clue on that one.
[16:14:22] <Horologium> I use the programmer built into the msp430 boards I have here for that.
[16:31:27] <tld> lethjakman_lapto: are you familiar with tayda?
[16:31:35] <tld> lethjakman_lapto: also, which AVRs do you want?
[16:31:36] <lethjakman_lapto> tld: no I am not
[16:31:49] <lethjakman_lapto> that's kinda hard, I'm not really sure because I have a lot of uses in mind.
[16:31:52] <tld> lethjakman_lapto:
http://www.taydaelectronics.com/
[16:32:01] <tld> lethjakman_lapto: They have atmega328p for $3.50
[16:32:07] <tld> and cheap shipping to most of the world.
[16:32:46] <tld> (they also have a couple of other chips, atmega8-16au (tqfp) for $1.89, and some attinies)
[16:32:55] <Horologium> rather limited selection though.
[16:33:01] <lethjakman_lapto> how much is cheap?
[16:33:14] <tld> couple of dollars usually?
[16:33:34] <lethjakman_lapto> hm, I think avnet still might be the cheapest...have you checked them out?
[16:33:36] <tld> a bit limited selection, but nice when you need a bunch of generic parts.
[16:33:43] <lethjakman_lapto> only thing is their shipping is around 8$ I think....maybe 6?
[16:33:50] <tld> 0.1uF decoupling caps? 1000 for $7
[16:33:51] <lethjakman_lapto> it is nice, it seems a lot cleaner
[16:33:52] <tld> that kind of thing.
[16:34:03] <lethjakman_lapto> bookmarked thank you :)
[16:34:31] <lethjakman_lapto> I can definitely use it for a few components
[16:34:50] <tld> I usually use them for the parts I try to keep in stock. Shipping takes a little while, but noticeably fast than shipping from china and the like.
[16:35:26] <lethjakman_lapto> ugh I just ordered a bunch of parts from china like 3 weeks ago and they're still not here
[16:35:29] <lethjakman_lapto> their shipping is terrible
[16:36:19] <tld> I'm wondering if there's been some kind of crap going on over there.
[16:36:43] <tld> I'm waiting for several things, and it just seems like stuff have been backlogged. There was a chinese holiday a while back, might be it.
[16:36:55] <tld> Oh, and if you have a choice between China Post and Hong Kong Post, choose the latter.
[16:37:24] <Horologium> www.arrow.com is another of the larger ones like digikey that I've dealt with in the past and been very happy with them.
[16:37:58] <lethjakman_lapto> Horologium: I just signed up with an account with arrow, I found it odd that they required a business name
[16:38:24] <lethjakman_lapto> tld: lol yeah I used china post....terrible.
[16:38:33] <lethjakman_lapto> when I was in china I saw tanks and crap...but never saw it on the news
[16:38:38] <lethjakman_lapto> they had soldiers parading through the streets
[16:38:50] <lethjakman_lapto> this was about 3 years ago though
[16:38:58] <tld> Horologium: Their selection of parts is limited. I mean, they only have 1 209 805 different types of caps! What's with that?!?
[16:39:04] <tld> Horologium: (actual number… impressive. :) )
[16:39:31] <tld> I've never been to China, and I plan on keeping it that way.
[16:39:34] <Horologium> 2.28 USD for the atmega328p at arrow.
[16:39:39] <Tom_itx> the 2560 is still a 5v chip right?
[16:39:59] <Horologium> Tom_itx, I believe so.
[16:40:17] <Tom_itx> i think i'm gonna get a cheap arduino 2560 to play with
[16:40:39] <Horologium> I still like the atmega1284p chips.
[16:40:47] <lethjakman_lapto> • 0 - 2MHz @ 1.8V - 5.5V, 0 - 8MHz @ 2.7V - 5.5V
[16:40:57] <lethjakman_lapto> http://www.atmel.com/Images/doc2549.pdf
[16:40:59] <Tom_itx> aren't they in the same family?
[16:41:07] <Tom_itx> less flash
[16:41:17] <Horologium> possibly, but the atmega1284p comes in a dip package.
[16:41:20] <lethjakman_lapto> the whole reason I never found pic's appealing....atmel's and TI's data sheets are way better.
[16:41:22] <Tom_itx> i'm not sure on those 2
[16:41:28] <tld> Tom_itx: 1.8 - 5.5 to 4.5 - 5.5V, depending on which chip, but all of them go to 5.5V.
[16:41:33] <Tom_itx> i kinda thought the 1280 and 2560 were
[16:41:54] <Horologium> 1280 and 2560 are the same chip with different amounts of memory I do believe.
[16:42:15] <Horologium> 644 and 1284 are similar, same chip with different memory.
[16:42:38] <Horologium> would be nice if there were a 2564, but, can't have everything.
[16:44:39] <tld> anyone looked at atmels zigbit stuff?
[16:44:50] <Horologium> not I.
[16:45:19] <Horologium> nordic is the only wireless stuff I have played with.
[16:46:12] <Tom_itx> i wonder if the smaller shields fit the 2560 board
[16:46:29] <Horologium> if they follow the ardweeny spec they should.
[16:46:30] <lethjakman_lapto> https://avnetexpress.avnet.com/store/em/EMController/Microcontroller/_/N-100185/Ne-100000?action=products&cat=1&catalogId=500201&cutTape=&inStock=&langId=-1&myCatalog=&npi=&proto=®ionalStock=&rohs=&storeId=500201&term=atmega%2B2560&topSellers=&categoryLink=true
[16:46:39] <lethjakman_lapto> can anyone explain why the top two are more expensive than the third one?
[16:48:05] <Tom_itx> lethjakman_lapto the preface
[16:48:15] <Tom_itx> 8AU 16AUR
[16:48:21] <Tom_itx> opposed to 16AU
[16:48:27] <Tom_itx> packaging etc
[16:48:32] <Tom_itx> i suppose
[16:48:39] <lethjakman_lapto> preface?
[16:48:51] <Tom_itx> after the dash
[16:49:05] <lethjakman_lapto> it looks like the 3rd one is a better chip
[16:49:07] <lethjakman_lapto> to me at least
[16:49:07] <Tom_itx> the 2nd is 8Mhz
[16:49:18] <Tom_itx> the other 2 it's a matter of packaging
[16:50:04] <lethjakman_lapto> ah alright
[16:50:14] <lethjakman_lapto> that makes sense there
[16:50:47] <lethjakman_lapto> thank you
[16:57:48] <tld> Horologium: did you ever do any range-testing?
[17:00:27] <Horologium> with the chip antennas I get about 10 meters, line of sight, at 250Kb/s
[17:01:31] <tld> hmm
[17:02:16] <tld> they're so tempting, thanks to price, but 10 meter line of sight really isn't a lot.
[17:02:48] <Horologium> that's the little chip antennas.
[17:02:56] <Horologium> with real antennas they should get 100meters or so.
[18:51:33] <jadew> Horologium, what chip are you talking about?
[18:51:52] <jadew> I was just looking into RF AVRs just now
[18:56:56] <Horologium> jadew, nordic 2.4ghz chips.
[18:57:05] <Horologium> sparkfun sells them soldered to a breakout board.
[18:57:14] <jadew> thanks for the info
[18:57:27] <Horologium> they have different antenna options
[18:57:30] <Horologium> a chip antenna
[18:57:35] <Horologium> a trace antenna
[18:57:39] <Horologium> and an external antenna.
[19:00:44] <jadew> they don't seem cheap
[19:14:11] <Horologium> 20 to 30 dollars for a module.
[19:14:35] <jadew> yeah, that's a bit expensive for what I want to do
[19:15:25] <jadew> any idea how come wifi clients don't collide? I wonder if I can have multiple transmitters on the same frequency
[19:18:29] <Horologium> they do collide.
[19:18:36] <Horologium> but the protocol works around that.
[19:19:27] <Horologium> the nordic chips have multiple channels.
[19:19:37] <Horologium> the newer ones have dual receiver modules.
[19:19:55] <jadew> how can it work around that?
[19:20:07] <Horologium> not sure.
[19:20:14] <Horologium> I've never studied the spec.
[19:20:27] <jadew> it's an interresting topic
[19:21:16] <Horologium> and a rather complex topic.
[19:21:28] <jadew> I'm sure
[19:29:46] <Horologium> don't forget,,,vote early and often!
[19:29:53] <Horologium> the more times you vote the more your vote counts!
[21:42:49] <Richard_Cavell> Does AVR C provide malloc()?
[21:42:53] <Richard_Cavell> Or do you have to do that yourself?
[21:43:24] <Tom_itx> what for?
[21:43:34] <Richard_Cavell> for allocating memory
[21:43:37] <Tom_itx> manage your vars better
[21:44:06] <Richard_Cavell> Is that a "no, you don't get malloc()" ?
[21:44:08] <Casper> as Tom_itx said :D
[21:44:14] <Casper> there is one, but don't use it
[21:44:25] <Casper> malloc on avr is usually a sign of bad programmer
[21:45:10] <Tom_itx> Allocates 'size' bytes of memory and returns a pointer to it. Implemented, but not tested.
[21:46:06] <Richard_Cavell> well you might need it for code that's been ported from another platform
[21:46:52] <Tom_itx> not if it's rewritten right
[21:47:59] <Richard_Cavell> So do you just sort of assume the existence of RAM based on your chip choice and just address it?
[21:50:10] <Tom_itx> you can use progmem too
[21:52:23] <Tom_itx> is dean's site down?
[21:55:19] <Tom_itx> !seen abcminiuser
[21:55:20] <tobbor> abcminiuser was last seen in #avr on Oct 06 12:33 2012
[23:14:14] <Richard_Cavell> 'afternoon all
[23:14:22] <jadew> hey guys, any ideas on what those spikes might be:
http://dumb.ro/files/fg_bug.png
[23:14:25] <Xark> Richard_Cavell: Better is to allocate "worst case" size as a global.
[23:14:36] <jadew> at the begining of the second and third byte
[23:14:43] <Richard_Cavell> Xark: malloc() is hardly at the bleeding edge of C
[23:15:11] <Xark> Richard_Cavell: No, and it works on AVR, however it is not a great idea for reliability because it is difficult to make sure your stack doesn't collide with your heap.
[23:15:35] <Xark> Richard_Cavell: Depending on your chip you have 2K... :)
[23:15:53] <Xark> Richard_Cavell: Worst case memory booking is warrented.
[23:17:22] <jadew> regarding my issue, I can see them on the oscilloscope as well
[23:17:33] <Richard_Cavell> jadew: LOL
[23:17:34] <jadew> they're clear, full height
[23:17:46] <Richard_Cavell> "Anyway, back to me"
[23:18:18] <jadew> just wanted to add a bit of info on my problem and make it clear it's about that :)
[23:18:43] <Xark> jadew: Perhaps bus noise? What do your power rails look like on oscilloscope (clean)?
[23:19:33] <jadew> Xark, it can't be bus noise, because that wouldn't make it full size
[23:19:39] <jadew> at least I don't think it would
[23:19:49] <jadew> the power rails are pretty clean
[23:19:51] <Xark> jadew: It can (and certainly on OLS).
[23:20:20] <jadew> oh, you think it's an OLS issue?
[23:20:32] <jadew> let me disconnect the probes and see if it still shows on the scope
[23:20:38] <Xark> jadew: Well, I would suspect that, except you say you see it on oscope too.
[23:21:23] <Xark> jadew: I love my OLS, but I do tend to ground every other probe line to help with noise (when dealing with high frequencies). Also, don't leave any active probe lines unconnected (ground unused lines).
[23:22:04] <jadew> why? I would think the unconnected ones would just collect noise
[23:22:14] <Xark> jadew: I could see it being somewhat difficult to isolate that specific spike on an oscope also...
[23:22:30] <Xark> jadew: I have had them affect adjacent pins. YMMV, but it is something you can try.
[23:22:51] <jadew> well, they could affect the pins if they were in input, no pull up state
[23:22:56] <jadew> but my pins are output
[23:23:40] <jadew> I disconnected the probes tho, I lost the measurement on the osc tho :) give me a sec, must have moved the osc probe by mistake
[23:24:32] <jadew> yep, the spike is still there
[23:25:06] <Xark> jadew: How long does the spike last for?
[23:25:36] <jadew> 0.7 us on the first byte
[23:25:45] <Xark> jadew: What clock speed AVR?
[23:26:22] <jadew> 8mhz
[23:26:33] <Xark> Hmm, so .7us is the "shortest" duration spike you see?
[23:26:42] <jadew> yep
[23:27:00] <jadew> I also have a 25Mhz clock in there, but futher away
[23:27:06] <Xark> jadew: I also find it somewhat interesting that all three bytes shown in this PNG start HIGH (or with a spike).
[23:27:08] <jadew> driving the chip I'm communicating with
[23:27:20] <jadew> Xark, yeah, me too
[23:27:38] <Xark> jadew: I don't suppose you have a bus pirate or another chip with SPI to give a "2nd opinion".
[23:28:03] <jadew> actually I do, I built one just for this, good idea
[23:50:22] <jadew> well, now the signal doesn't go all the way down, since my dongle is pulling the lines low, while the chip I'm debugging is pulling it high
[23:50:31] <jadew> however, I don't see a spike there
[23:52:11] <Xark> jadew: Hmm, doesn't sound like an ideal test with the chips fighting. Is it possible you just need a puilldown/pullup or something?
[23:53:21] <jadew> the pins are set for output so I don't think that's needed
[23:53:24] <jadew> let me check something else
[23:54:57] <edboogie2011> hey guys what is the cheapest place to get barebones or regular spec PCBs, small quantitiy of a board the size of an index card, 4 pieces. no hurry needed, 2 to 3 weeks, okay. in USA?
[23:55:29] <edboogie2011> the place doesn't have to be in USA, but easy order and deliver to USA
[23:55:39] <Tom_itx> http://oshpark.com/
[23:55:43] <Tom_itx> probably
[23:55:47] <Tom_itx> $5 sq in
[23:55:57] <Tom_itx> or from capnkernel
[23:56:12] <Tom_itx> or itead
[23:56:15] <Tom_itx> or seeed
[23:56:40] <edboogie2011> thanks
[23:56:57] <Tom_itx> capn's boards are good
[23:57:04] <Tom_itx> maybe 4 weeks?
[23:57:11] <Tom_itx> about average for china boards
[23:57:49] <CapnKernel> edboogie2011: hi
[23:58:04] <Tom_itx> i want a comission
[23:58:06] <Tom_itx> :)
[23:58:21] <edboogie2011> CapnKernel: I'm not gonna use you for this small of an order
[23:58:23] <CapnKernel> Sorry Tom, Ed's already a customer of mine.
[23:58:28] <CapnKernel> Huh, why not?
[23:58:30] <edboogie2011> i am ordering test boards to build testing setups
[23:58:40] <CapnKernel> Small orders are my specialty :-)
[23:58:46] <edboogie2011> cuz its only 4 boards. what is the point of buying overseas for such a small order
[23:59:03] <CapnKernel> Lots of point! Let's talk about it in PM :-)
[23:59:07] <edboogie2011> lets save it for a 100+ board order for next time
[23:59:56] <jadew> Xark, I think I realized what it is
