Back
[06:14:16] <STS_Patrik> anyone knows a none-GPL modbus library? Master RTU RS-485 for 8-bit AVR
[10:44:58] <antto> i implemented support for flashing firmware to my atmega2561's stk500v2-based bootloader from my own app
[10:45:15] <antto> it works.. BUT the flashing process is uber slow.. for some unknown reason
[10:45:26] <antto> (compared to when i flash via avrdude)
[10:45:37] <antto> halp
[11:01:16] <antto> does avrdude send packets one after another without waiting for the bootloader to reply?
[11:11:48] <specing> ah look, atmel changed their logo since the last time I've looked
[11:12:08] <LeoNerd> Hrm.. it does look slightly different
[11:14:24] <specing> lol, neither mouser nor farnell nor digikey have realtek switch chips
[13:29:09] <guanche> if I have AC (~20V), can I make a transistor, or otherwise using discrete analog parts, switch at ~4Hz?
[13:30:18] <LeoNerd> .. I'm not sure I follow the question
[13:32:24] <guanche> I made a bike's phone charger with a brushless, now I'm experimenting with a permanent magnet dc motor to be used with leds (on another bike)
[13:33:12] <guanche> so I've divided it's voltage a cruising (bike) speed, 20.5V / 3.3 for the leds, giving me six that I use in series for it to lit
[13:34:09] <guanche> now, I'd like some of these switching at a lower frequency, and as I'm using just AC, just wondering what I could do to achieve this with discrete components
[13:36:13] <LeoNerd> How does 4Hz relate to any of this?
[13:36:30] <Jordan_U> I was highlighted in this channel again, but unfortunately the message appears to be beyond my scrollback buffer. Anyone know what the message to me was?
[13:36:42] <guanche> 4hz is my target frequency? so they would ligth up 4 times a second?
[13:37:00] <LeoNerd> Ah, so that's not related to the AC input frequency?
[13:37:46] <guanche> yes, I though first on a sort of counter, such that each, whatever the division cycles we get, it fires
[13:37:58] <guanche> I mean, the led or leds
[13:38:06] <LeoNerd> Surely then, just rectify the incoming AC, use that to power a 4Hz oscillator
[13:38:29] <guanche> and can it be done without dc?
[13:39:01] <LeoNerd> No? but that's what the rectifier/cap smoother are for
[13:39:08] <LeoNerd> This is basic electronics, nothing AVRish
[13:39:22] <guanche> yes, but a linear regulator does require power
[13:39:36] <LeoNerd> AC off the dynamo => bridge rec => cap smoother => standard oscillator of coice
[13:39:37] <LeoNerd> choice
[13:40:01] <LeoNerd> Possibly via a regulator if you want one - for efficiency you might consider one of the boost/buck universal things, rather thna a lossy linear
[13:40:05] <guanche> can you tell me a part number such that i can look for it?
[13:40:09] <LeoNerd> Not offhand no
[13:40:16] <LeoNerd> There's about a billion of them, any will suffice
[13:42:51] <guanche> I'more trying to get it out of the incoming ac if possible, so will do more research
[13:43:06] <guanche> and with discrete components, let's see what I can find
[13:44:02] <LeoNerd> You can't, out of the incoming AC. it's AC
[13:44:10] <LeoNerd> Seriously, just convert it to DC. It's not hard
[13:45:02] <guanche> well, transistors can switch on ac with a sreies diode it seems
[13:45:36] <LeoNerd> Yes but you'll lose state on each half-cycle so the thing will be completely screwed up
[13:45:47] <LeoNerd> I've already told you the right way to do it; why do you persist in trying to research the wrong way?
[13:46:23] <guanche> I'm not saying your ways isn't wrong, or isn't acceptable in any way
[13:46:47] <guanche> is just that I want to do something and you want me to do it any other way
[13:47:04] <LeoNerd> I'm just not sure what you hope to gain doing it this way
[13:47:17] <LeoNerd> If you put in a single diode, your output will be totally off for half a cycle.
[13:47:39] <LeoNerd> If you do the proper 4-diode B.R. you'll be on for both half-cycles but lose state inbetween, thus messing up the output frequency.
[13:47:52] <LeoNerd> If you then add a smoothing cap you're now /at the suggestion I originally made/
[13:47:53] <guanche> knowledge, the transistor switching part wouldn't be the only on the resulting thing, if possible to do it at all
[13:48:18] <LeoNerd> ?
[13:48:45] <guanche> nah, leave it, thanks so much for your time which I have said before
[13:48:55] <guanche> *which I should have said before
[16:36:55] <autrilla> How much memory do the stdlibs (if there is such a thing, newcomer here) take on runtime?
[16:38:55] <ColdKeyboard> Is this
http://www.elecrow.com/premium-pcb-service-p-1061.html Elecrow's PCB premium service of $1 for 1cm x 1cm or what?
[16:39:44] <autrilla> ColdKeyboard, We Elecrow will quote it for you in 1~3 working days.
[16:39:58] <ColdKeyboard> :\
[16:40:39] <ColdKeyboard> Can someone recommend PCB fabrication? I need to fabricate 3x4cm board and need no more than 10-20 boards. Any cheap solutions?
[16:48:44] <Casper> what about seeeed?
[16:57:26] <Tom_itx> ColdKeyboard, hackvana can fix you up
[16:59:12] <Tom_itx> also laen's board service can do it
[16:59:18] <Tom_itx> i forget his webpage...
[17:00:05] <ColdKeyboard> OSHpark?
[17:00:19] <ColdKeyboard> I mailed Mitch from hackvana for quote :)
[17:00:33] <Tom_itx> oshpark is laen's site
[17:00:35] <Tom_itx> yes
[17:01:09] <ColdKeyboard> I don't like that OSHpark is too pricy for 3 boards. Quality looks really nice thou
[17:01:23] <aandrew> hackvana++
[17:04:25] <Tom_itx> you'll have to wait for HK post on mitch but they're good boards
[17:04:32] <Tom_itx> i've used him several times
[17:05:13] <Tom_itx> itead is ok i suppose
[17:05:21] <Tom_itx> it's down on the list though
[17:08:22] <autrilla> Would you use an AVR to control an RS485 network, gather data from devices and connect to an internet server over GSM to upload and download related data or would you use something else?
[17:09:21] <LoRez> If the AVR has sufficient storage space for the data, it can probably handle that.
[17:10:07] <LoRez> depends on how much work you want to do to implement the "connect to the internet server" part
[17:10:58] <autrilla> LoRez, the connect to the internet part is not that difficult, SIM900 modules handle that nicely
[17:11:19] <autrilla> I just need to give it a couple of commands over serial with the HTTP request I want to send, and it's done
[17:12:13] <autrilla> I'm worried about the data limits on AVRs, I'm going to be storing 256 characters * a max of 256 devices, ~64kB. Not sure if I want to use EEPROM or something similar for that, since it wears out.
[17:12:21] <LoRez> I'd probably only use an AVR for that if I was highly power constrained.
[17:13:38] <autrilla> What would you use then?
[17:13:50] <LoRez> probably a Pi and python
[17:13:54] <autrilla> It has to be something that can be sold and look professional
[17:14:07] <autrilla> Pi looks hackish
[17:14:39] <autrilla> Plus, it's about three times more expensive than the AVRs :P
[17:15:35] <Jordan_U> autrilla: You can make your own board with an ARM chip, you don't need to use a pi.
[17:16:35] <Jordan_U> (Though if you're looking at running linux, I don't know how hard it is to bootstrap linux on a custom ARM board).
[17:16:48] <autrilla> I'll try with an AVR first, see if it can handle it...
[17:17:51] <LoRez> $(hardware dev costs) + $(hardware production costs) + $(software dev costs) + $(software maintenance costs) + $(hardware replacement costs)
[17:19:05] <LoRez> $(field upgrade difficulty costs)
[17:19:13] <autrilla> Yeah, if I wrote to EEPROM at one write per 5 minutes, the memory will last 200 days. Not good.
[17:19:29] <autrilla> Maybe I don't even need to write it
[17:19:34] <autrilla> Just get it and upload it
[17:20:21] <LoRez> http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en539039
[17:21:31] <autrilla> LoRez, I guess low-level code will be needed for that
[17:22:03] <LoRez> everything is low level code on an AVR :)
[17:22:36] <LoRez> it's just SPI.
[17:22:46] <LoRez> most AVRs do most of that for you
[17:23:07] <autrilla> Well I'm not crazy enough to code this whole thing in ASM
[17:23:32] <autrilla> I even find C too low level
[17:24:32] <Jordan_U> autrilla: No ASM needed at all.
[17:25:24] <Jordan_U> autrilla: But if you really find C too low level, then you probably should be considering a pi (rather than even your own ARM based board).
[17:26:11] <autrilla> Jordan_U, I'm obviously more comfortable writing with as many resources as I could need, but cost is important too.
[17:26:38] <autrilla> But C is fair enough, definitely better than assembly
[17:33:11] <hackvana> ColdKeyboard: Hi!
[17:33:20] <ColdKeyboard> hi hackvana
[17:33:50] <Tachyon`> oh hello hackvana
[17:33:53] <Tachyon`> you get around -.o
[17:35:09] <hackvana> Tom_itx: I offer China DHL (2 days), HK DHL (3 days), EMS (5 days) and HK Post (1 week to UK, 2 weeks to USA/Australia). An option for everyone :-)
[17:35:37] <hackvana> Tachyon`: Hi there!
[17:36:18] <LeoNerd> So, hrm... how much should I trust that, if my new solder station claims "380C", the tip actually is...?
[17:36:32] <Tachyon`> well, more so after a few seconds
[17:36:35] <LeoNerd> I don't feel like I'm getting anywhere near that; this doesn't feel like it's soldering well
[17:37:01] <Tachyon`> I dunno actually, mine has a sensor but some just assume the temperature is what you set
[17:37:07] <Tachyon`> does it change immediately or do you see it heat up
[17:37:20] <Tachyon`> when you set it
[17:37:25] <LeoNerd> I see it change... it claims "real"
[17:37:37] <LeoNerd> It's definitely reading /a/ real temperature, I just don't feel like it's solder-melting hot
[17:37:38] <Tachyon`> hrm, suppose it could be out of calibration
[17:37:46] <LeoNerd> It takes a -long- time to melt the stuff.. like, maybe 15 seconds?
[17:37:56] <LeoNerd> But I don't know if that's just because the tip is brand new
[17:37:59] <Tachyon`> you can get those infra red laser thermometers you fire at things ike that to read tempearture
[17:38:10] <LeoNerd> Ah; I don't have one of those
[17:38:11] <Tachyon`> that's what I use to test/calibrate mien
[17:38:12] <Tachyon`> ahh
[17:38:27] <Tachyon`> can't really think of any other easy way of checking
[17:38:43] <autrilla> There isn't any other
[17:39:45] <autrilla> But 15s is a veeery long time
[17:40:45] <Tachyon`> http://www.ebay.co.uk/itm/310839836935
[17:40:49] <Tachyon`> they're pretty cheap now it seems
[17:41:49] <Tachyon`> hrm, seems those ones really /are/ in london
[17:42:35] <autrilla> You could make an IR thermometer yourself too :P
[17:42:43] <LeoNerd> Yeah; I'm worried that I'll melt things if I have to hold it this long
[17:43:20] <autrilla> When I've had problems with my irons is because I left them on for too much
[17:43:24] <LeoNerd> I'm also managing sometimes to tear off pieces of solder off the board while it's still almost solid, and the iron can move them while still solid
[17:43:34] <LeoNerd> It like... doesn't melt
[17:43:49] <autrilla> LeoNerd, what solder are you using?
[17:44:36] <LeoNerd> 0.7mm 63/37
[17:44:45] <autrilla> That's just weird
[17:44:53] <LeoNerd> The other markings on the reel are all in Chinese so I can't read the rest ;)
[17:44:55] <autrilla> Melting point for those are around 180 celsius
[17:45:09] <LeoNerd> Yeah, I know... My station is currently claiming 417
[17:45:12] <LeoNerd> C
[17:45:28] <LeoNerd> Ah, OK.. now it's melting better
[17:45:38] <LeoNerd> Maybe the tip was just too new, covered in some manufacturing oil crap
[17:46:25] <autrilla> I wonder why they only provide the melting temperature
[17:46:35] <autrilla> At what temperature does it boil?
[17:47:25] <LeoNerd> ... I don't want Pb/Sn vapour in my kitchen, thank you ;)
[17:48:05] <autrilla> Turns out somewhere around 1000-2000, no worries
[17:48:34] <autrilla> But at high temperatured it does smoke quite a lot
[17:51:19] <LeoNerd> Oh huh... random question: is it possible this is solid solder without flux in it? How would I tell?
[17:56:34] <autrilla> LeoNerd, what?
[17:56:40] <autrilla> Solid solder without flux?
[17:57:04] <autrilla> Sn63Cu37 melts pretty well
[17:57:07] <autrilla> Anyway, off to sleepp
[18:02:29] <LeoNerd> Oooh.. surely I can calibrate this by turning it down until it literally does not melt the solder at all
[18:04:45] <LeoNerd> Well, I have it down at 200 and it's still melting. so it can't be far off
[18:06:15] <LeoNerd> Still juuust about melts it at indicated 165
[18:07:57] <LeoNerd> ... and also at 150 it -barely- still manages it. I suspect the readout isn't very accurate, though it is at least changing
[18:29:48] <RikusW> LeoNerd: the flux will leave a residue
[18:30:07] <RikusW> and cause the white smoke on a hot tip (350C +)
[18:30:23] <RikusW> Jordan_U: use zlog
[18:30:26] <RikusW> zlog
[18:31:08] * RikusW is off to sleep..
[18:31:40] <LeoNerd> Ah, .. oh he went
[18:56:13] <LeoNerd> https://twitter.com/cpan_pevans/status/531968586387320833 Managed it :)
[19:31:10] <rue_house> LeoNerd, remember, with great power comes... just because you can smt things dosn't mean everything should be smt, some things are better off as thruhole
[19:31:24] <rue_house> like if you need a resistor to jump .6" between its elads
[19:31:26] <rue_house> leads
[19:31:59] <rue_house> never mount a connector via solder...