Back
[09:18:43] <Duality> hi
[09:19:16] <Duality> i need working udev rules for avrisp2 (MKII) could someone share them with me :)?
[09:32:59] <DO9XE> Duality, do you have the olimex version?
[09:34:04] <DO9XE> i have and i hade to patch and compile avrdude, because there is a bug which causes the device to deny working, because the usb endpoints were faked in a strange way and avrdude uses fixed endpoints
[09:54:06] <Clietv> I'm looking up those attiny for the first time. 6 pins sot23. How does one program those chips?. I see a reset line, vcc, ground and a few i/o ports
[09:54:19] <Duality> isp?
[09:54:41] <Duality> the programing pins double as general i/o
[10:11:48] <Clietv> It uses some weird TPI interface but I have jtagice3 only. Any way to program those chips?
[10:33:22] <Ad0> which attiny ?
[10:47:30] <Clietv> attiny9
[11:40:39] <Duality> does anyone know a way to turn a font 90 degrees ? 5x7 for a ledmatrix.
[11:42:20] <specing> octave
[11:44:42] <Duality> GNU octave ?
[11:45:02] <Thrashbarg> Duality: 5x7 would be difficult to rotate and keep legibility. 8x8 would be simpler... I'd reckon the idea is to load the seven or eight bytes that make up the character then shift all of them sequentially, and the bit that comes out the end gets shifted into the output....
[11:45:54] <Thrashbarg> if that's too much for your controller to do (probably is) you could have a separate font bitmap that's pre-calculated with rotated letters
[11:46:14] <Duality> i see
[11:46:57] <Duality> i have a ledmatrix (that i didn't make) that is multiplexed in such a way that the normal fonts i find online are rotated 90 degrees clockwise
[11:47:22] <Thrashbarg> so scanning vertically and across, not the other way around
[11:48:24] <Thrashbarg> so you'd want to make a small program to recalculate your font bitmap
[11:49:21] <Duality> yea but now i have to figure out how :)
[11:50:38] <Thrashbarg> what you're trying to do is scan out the same bit across seven bytes, with those bits going to your destination byte
[11:51:57] <Thrashbarg> that is, bit 5 of the bytes, 4, 3, etc, become byte 5, 4, 3, etc
[13:22:16] <fractallife> hello my friends
[13:22:31] <fractallife> how's it going ?
[13:30:16] <Lambda_Aurigae> if you are on irc you have no friends.
[13:33:17] <fractallife> are you enemy?
[13:33:43] <fractallife> man eating alien ?
[13:34:59] <fractallife> i was looking into summoning a simple timer with an atmega, and studied this stuff's usage
http://www.ia.omron.com/data_pdf/cat/h3cr_l084-e1_2_1_csm1013792.pdf?id=193 i have found that it is overcomplicated and not too logical in operation :/
[13:36:01] <fractallife> the full operation can not even be understood from the diagrams and i think some have errors too
[13:36:49] <fractallife> and flicker mode is a $200 ne555 timer circuit with enable and optocoupler output
[13:38:43] <Lambda_Aurigae> building a timing device with an AVR is fairly simple.
[13:39:21] <Lambda_Aurigae> how you interface it will depend on the voltage and amperage of the load.
[13:39:41] <fractallife> but it needs a menu too to select things and store stuff in eeprom
[13:39:48] <fractallife> little details will take long time
[13:40:08] <Lambda_Aurigae> I have driven 250V 5A triacs with an AVR in the past.
[13:40:33] <fractallife> i'd use my zvs ssrs that need >5mA to turn on
[13:40:36] <Lambda_Aurigae> depending on the complexity, a simple menu with 4 buttons and a 16x2 LCD shouldn't take more than 2 hours to build, program, and test.
[13:41:01] <fractallife> so it is optoisolated
[13:41:12] <fractallife> oh i was thinking about 2 buttons
[13:41:17] <fractallife> sounds enough
[13:41:23] <Lambda_Aurigae> that SSRS is likely just an optoisolator and an SCR.
[13:41:32] <fractallife> yes
[13:41:38] <Lambda_Aurigae> depends on how complex you want your menu system to be.
[13:41:57] <Lambda_Aurigae> I find 4 buttons to be just about right for menu systems...back, forward/enter, up, down
[13:42:14] <fractallife> i have a dead ultrasonic cleaner crap here that had an 555 with an 5V supply in it
[13:42:22] <fractallife> i'd like to make that work first
[13:42:25] <Lambda_Aurigae> I work with such menus on a daily basis on printer control panels.
[13:42:46] <Lambda_Aurigae> 555 timers are simple and many dozens if not hundreds of examples are available online.
[13:42:47] <fractallife> i'm puttin in a 4 digit 7 segment display and an atmega
[13:43:21] <Lambda_Aurigae> replacing a 555 with a microcontroller shouldn't be difficult at all.
[13:43:22] <fractallife> after that i have to fix my induction cooker, that has a crappy temperature control
[13:43:44] <fractallife> i need to put a working temperature sensor in, and a temperature controller with timer
[13:43:50] <Lambda_Aurigae> 2 buttons to set a time on and time off with a 4 digit display should be simple.
[13:43:55] <fractallife> yea
[13:44:02] <Lambda_Aurigae> one button to enter/advance and the other to cycle 0-9 and back to 0.
[13:44:04] <fractallife> :)
[13:44:13] <fractallife> if i feel like it i might put a pwm in it too
[13:44:18] <Lambda_Aurigae> make the active digit flash at 1Hz or so.
[13:44:22] <fractallife> so like power control
[13:44:39] <Lambda_Aurigae> remember that not all power likes to be controlled with PWM
[13:44:41] <fractallife> but it does not really need it, it is fine at 100%, we want to clean that stuff
[13:45:02] <Lambda_Aurigae> some devices have fits if you pulse their power or pulse it wrong.
[13:45:47] <Lambda_Aurigae> and doing pwm through an SSR could be problematic unless it is built for such...their switching times are not always the best.
[13:46:26] <fractallife> hm yes, well my zvs ssr can do it without problems, though i need to set the pwm freq low to be somewhat accurate
[13:46:38] <fractallife> because it will only start at zero
[13:46:50] <Lambda_Aurigae> ok...I gootta get back to work.
[13:46:52] <Lambda_Aurigae> laters.
[13:46:55] <fractallife> bb
[13:46:58] <fractallife> what do you do?
[13:48:18] <Ad0> Clietv:
http://minisystem.blogspot.no/2012/01/programming-attiny10-with-avrisp-mkii.html
[13:48:30] <Ad0> if you read down you see the connections you need to make
[13:56:29] <fractallife> hey Ad0 what's up ?
[13:58:13] <fractallife> hehe this is interesting
http://2.bp.blogspot.com/-nKRPydvCu3k/TyA9Kn1kK2I/AAAAAAAADEY/RWoMZa655tQ/s1600/IMG_5289.JPG
[14:00:05] <fractallife> Ad0, you wrote that to me? i already have a programmer i made a dapa cable last year :)
[14:00:48] <fractallife> used an ide hdd connector and a standard 40/80 pin cable
[14:01:48] <fractallife> i have grounded every second line so there is no chance of any bit error due to cross-talk ;)
[14:04:26] <Ad0> hey fractallife
[14:04:46] <Ad0> lol
[14:04:59] <Ad0> do you need to go assembler to even use an attiny9
[14:05:31] <Ad0> I used a ds1b20 + uart library and I crossed 4k
[14:05:52] <fractallife> well
[14:06:05] <fractallife> c is not really good with 512 byte ram, you may try ...
[14:06:27] <fractallife> if you have weird errors then you ran out of memory :)
[14:06:46] <Ad0> for sending temp data to uart I used this much in C++
[14:06:50] <Ad0> Program Memory Usage :3562 bytes 10,9 % Full
[14:07:03] <Ad0> Data Memory Usage :105 bytes 5,1 % Full
[14:09:51] <fractallife> lol
[14:10:06] <fractallife> c++ on an mcu ?
[14:10:10] <Ad0> yeah man :D
[14:10:17] <Ad0> why not
[14:10:28] <Ad0> it all boils down to small code anyway
[14:10:28] <fractallife> it will waste all the resources of your tiny
[14:10:29] <Ad0> I tested
[14:10:42] <Ad0> same byte size on both C and C++
[14:10:47] <fractallife> i do C only
[14:10:57] <fractallife> hm okey...
[14:11:27] <fractallife> aren't you concerned abour running out of ram? :)
[14:13:10] <Ad0> I suspect the C++ compiler is a bit different
[14:13:16] <Ad0> for the atmel toolchain
[14:13:31] <Ad0> but yeah, I dunno
[14:13:39] <Ad0> might not be worth screwing around in C++
[14:14:16] <Ad0> for instance, there is no new and delete operator
[14:14:51] <Ad0> I was hoping for nicer encapsulation but the same usage
[14:15:07] <Ad0> http://www.lugs.ch/lib/vortraege/small-c++.pdf
[14:15:07] <fractallife> if you get a large device with more ram, and flash then it may not be a problem
[14:15:28] <Ad0> I am working on an atmega 328P now to avoid hitting the wall
[14:15:33] <Ad0> but I have attiny in mind
[14:15:35] <Ad0> 88
[14:17:12] <Ad0> useless PDF
[14:18:04] <fractallife> your attiny will scream when it sees your c++ code ^^
[14:18:57] <fractallife> i already used atmega168pa with C, and had weird things gong on when i just stuffed things into ram, i didn't even notice i already ran out of ram
[14:19:32] <fractallife> i just carelessly used strings the same way ad on pc, and C eaten all ram :/
[14:19:48] <fractallife> so i needed to relocate my strings into flash and use them from there
[14:21:20] <Ad0> haha
[14:21:25] <Ad0> lots of crap to think abouit
[14:27:25] <fractallife> i didn't need asm yet, that 16k or flash seems pretty much for that smal lthing
[14:33:59] <fractallife> Ad0, hows your adc stuff going?
[15:15:04] <fractallife> anybody know which typenumber avr is the higher pin count version of the atmega168 ?
[15:15:23] <fractallife> (with same functionality)
[15:19:35] <fractallife> atmega169 looks similar, but only 16MHz
[15:21:06] <fractallife> oh oh answer to my q? :) Up to 20 MIPS throughput at 20MHz (Atmel ATmega329A/329PA/3290A/3290PA/6490A/6490P)
[15:21:43] <LeoNerd> Bah. Optoisolator is slow :)
[15:21:45] <LeoNerd> :(
[15:22:45] <LeoNerd> 62µs rise time. That doesn't feel right to me
[15:27:12] <fractallife> hehe
[15:27:20] <fractallife> 6N136 6N137 ;)
[15:27:57] <LeoNerd> Currently using a 4N28
[15:28:19] <LeoNerd> I notice my DMX isolator uses a 6N137 and that is happy at 250kBaud, so ... evidentally that is possible
[15:30:00] <fractallife> yes
[15:30:20] <fractallife> the 6n136 does not require external power and can work upto 12v i remember
[15:30:29] <fractallife> the 6n137 is for 5V
[15:30:35] <fractallife> but 10x faster
[15:30:44] <fractallife> 1/10Mbps i think
[15:37:43] <aandrew> de
[15:40:56] <fractallife> Ad0, in theory you could speed up a crappy 4n25 too
[15:41:51] <fractallife> a little bit by using 2 of them maybe and biasing them to always conduct, and then changing their led currents
[15:42:37] <fractallife> it may be tricky to bias it to work without adjusting
[15:43:01] <fractallife> t i mean LeoNerd
[15:49:13] <Duality> got my ledmatrix working :) although my avr might be not fast enough. or i just don't know how to get that speed and efficiency to reach nice displaying of text and leds :)
[15:49:38] <Duality> marquee function was easier then i though.
[15:49:45] <Duality> thought
[15:52:45] <fractallife> hi
[15:52:46] <fractallife> haha
[15:52:52] <fractallife> how many mips you need ?
[15:53:23] <fractallife> you can always use shift registers :P
[16:11:36] <fractallife> hi whatsacompiler
[16:20:24] <fractallife> https://www.flickr.com/photos/maan2k
[16:20:26] <fractallife> awesome :)
[16:31:45] <Duality> fractallife: it is already multiplexed with shift register that is part of it being slow i believe, you have to select a row clock in 80 bits and that 7 times
[16:32:51] <fractallife> oh
[16:32:52] <Duality> maybe if i clocked it in with spi it would be faster
[16:33:01] <Duality> but i can't use spi on this pcb
[16:33:13] <fractallife> i clocked my test display at 256Hz with shift registers
[16:34:05] <fractallife> but it as not a limit
[16:34:10] <fractallife> it was just my timer
[16:34:54] <LeoNerd> http://home.leonerd.org.uk/local/screenie/optoiso-i2c.png
[16:34:58] <LeoNerd> That doesn't look too bad does it?
[16:36:42] <LeoNerd> Oh.. except that actually it does an immediate START-then-STOP :(
[16:36:49] <LeoNerd> Because of timing overlap
[16:38:58] <Duality> LeoNerd: what is it the timing of ?
[16:39:29] <LeoNerd> Top two traces are the output from the MCU, the bottom two are the output from the optoisolator for it
[16:39:32] <Duality> nvm didn't read well enough :)
[16:39:46] <LeoNerd> It's a protocol that looks like I2C
[16:40:06] <LeoNerd> (physically it's not I2C because I have two unidirectional pairs of data+clock, instead of shared buses)
[16:44:10] <LeoNerd> Since they're not bidirectional, the ACK bit is actually generated by the sender, and signifies if that was the final byte
[16:44:26] <LeoNerd> But it looks enough like I2C to a logic analyser, which was the entire point :)
[19:01:31] <ferdna> i am looking for something like this:
[19:01:37] <ferdna> https://www.tindie.com/products/FemtoCow/pogo-pin-iscp-spi-programmer-adapter/
[19:01:39] <ferdna> but of higher quality...
[19:02:07] <Lambda_Aurigae> that's nic
[19:02:09] <Lambda_Aurigae> nice
[19:03:47] <ferdna> Lambda_Aurigae, yeah
[19:21:31] <fractallife> hey
[19:22:13] <fractallife> Duality, anyway, i used fast chips and an udma cable ;)
[19:22:30] <fractallife> i was surprised how low delays i could get away with
[19:22:58] <fractallife> the less time required to update, the more time left for program
[19:31:58] <Ad0> http://support.connectblue.com/download/attachments/10551321/cB-OLx425+Bottom+side+J6.jpg?version=12&modificationDate=1416306847000
[19:32:07] <Ad0> WHY do people have tehse kinda soldering connections?
[19:33:09] <Ad0> to solder it directly on pads I guess
[19:33:59] <Lambda_Aurigae> smaller board.
[19:45:39] <fractallife> Ad0, i don1t get the question, btw what is that ?
[19:45:57] <fractallife> those edge vias are interesting
[19:50:23] <Ad0> yeah
[19:50:38] <Ad0> I wish it had holes but claerly it is for soldering on top of other PCBs
[19:58:07] <fractallife> i still don't see your problem, it has "holes" and solder pads
[19:59:01] <Lambda_Aurigae> the edge half-holes aren't quite as sturdy as full holes though.
[20:00:33] <fractallife> holes holes...
[20:00:58] <fractallife> stable solution is when you have a hole and loop the wire back in it.
[20:01:10] <fractallife> the wire will always break at the solder point.
[20:20:20] <fractallife> Ad0, Lambda_Aurigae here is an example of a secured power cable attachment
https://www.wayneandlayne.com/projects/battery-booster-shield/
[20:21:10] <fractallife> you pull the wire through at least 1 other hole to avoid stressing the solder joints
[20:24:05] <Lambda_Aurigae> hate to tell you this, but, that has been used since before either of us was born.
[20:24:30] <fractallife> probably
[20:24:46] <Lambda_Aurigae> heck, I do it all the time for pins on boards that are to be connected and disconnected a lot.
[20:25:40] <Lambda_Aurigae> I was just talking about the half-holes on the picture that was posted earlier...and the fact that they aren't as strong as full holes.
[20:26:35] <fractallife> it is as strong as the bond
[20:27:01] <fractallife> i'd dip the thing in industrial epoxy to make it secure.
[20:27:52] <fractallife> i did this to my SSR-s too.
[20:28:24] <fractallife> in case prototyping and you touch it with the tip of your tongue
[20:29:31] <fractallife> (accidentally)
