Back
[01:11:52] <rue_house> yes
[01:12:03] <rue_house> i SHOULD HAVE GOT THE GENNY UP AND GOING WHILE IT WAS NICE OUT WITH POWER
[01:12:50] * synja gives you a look to which you know the exact meaning, raising one eyebrow with half a smirk
[01:13:27] * synja shifting his posture to one hip, while crossing his arms
[01:13:33] <synja> to give you a mental picture..
[02:14:25] <rue_house> hmm, I should be able to write this code almost without thinking
[02:14:30] <rue_house> maybe I'm getting old
[02:14:43] <rue_house> (void)rue_mohr
[02:15:46] <z64555> void (*rue_mohr)(void);
[02:16:26] <rue_house> hah, I'm a function
[02:17:14] <rue_house> why dosn't python use cammelCase?
[02:17:20] <rue_house> why do they use underscores?
[02:19:09] <synja> __danger
[02:19:11] <z64555> just a matter of style
[02:19:22] <rue_house> WHY NOT CAMMELCASE
[02:19:26] <rue_house> arg
[02:20:04] <z64555> If you want to be really nasty youUse_both()
[02:20:08] <rue_house> so, the object for a cnc machine will have a text entry,
[02:20:21] <rue_house> a stopwatch
[02:20:28] <rue_house> an ncfile object
[02:20:43] <rue_house> I suppose I need a history window
[02:21:03] <rue_house> or maybe the history window should be part of the serial object
[02:23:15] <rue_house> or maybe a serial history should be a derived from a history window
[02:24:28] <synja> so you're trying to automate the cnc by rendering designs into code that can be interpreted by the cnc?
[02:24:42] <rue_house> no
[02:24:51] <rue_house> I'm writing an nc spooler for grbl
[02:24:57] <rue_house> cause a good one dosn't exist
[02:25:31] <synja> no idea what that is
[02:26:27] <rue_house> cnc machine
[02:26:42] <synja> well i know what that is
[02:26:52] <z64555> the spooler.
[02:28:03] <z64555> https://en.wikipedia.org/wiki/Spooling
[02:28:09] <z64555> It's similar in concept to a stack
[02:28:12] <Jak_o_Shadows> rue_, I use camelCase in python. Ignore them, they are wrong
[02:28:19] <Jak_o_Shadows> In practice, libraries are sorta a mix
[02:30:38] <Snert> camelcase is great.
[02:30:57] <Snert> I wouldn't do it any other way most of the time.
[02:31:47] <rue_house> I like space compression, in that sense, cammelcase wins
[02:31:53] <rue_house> if (0) {
[02:31:56] <rue_house> }else {
[02:31:58] <rue_house> }
[02:32:59] <Snert> a leading _ tends to blend with the baseline. Depending on what font is in use, it tends to get hidden and blend with the baseline.
[02:33:09] <Jak_o_Shadows> I ONLY use underscores for subscripts. Which is a habit LaTeX taught me. Only really useful when you're working on something fairly mathematical
[02:33:35] <Jak_o_Shadows> ./in matlab, because seriously, who could be bothered making a named struct thing?
[02:33:48] <z64555> If I'm stuck with C, and am trying to do OOP, I like to use camcase + underscores
[02:34:03] <z64555> math_add, math_subtract, etc.
[02:34:26] <rue_house> you seen my C
[02:34:36] <z64555> shows it as a method/member as the camelcase
[02:34:42] <z64555> Yeah, I've seen it
[02:34:49] <rue_house> data objects
[02:37:01] <Jak_o_Shadows> I need to improve my state machines in C
[02:37:10] <z64555> I know enough about C and C++, that I can probably implement C++ in C
[02:37:31] <z64555> with the exception of the templates
[02:38:00] <z64555> Jak_o_Shadows: state machines mentioned
[02:38:04] <z64555> whats up?
[03:03:22] <rue_house> the power is gonna go out again soon
[03:03:34] <rue_house> its flashing on and off and its annoying
[03:04:08] <rue_house> usually the recloser dosn't cycle so many times
[03:09:20] <Jak_o_Shadows> aha
[03:09:37] <Jak_o_Shadows> Just a general purpose state machine. I have several steps in my processing that I need to do.
[03:11:03] <Jak_o_Shadows> Thing is, there's a little bit cof communication with the outside world. Only a tad though
[03:14:29] <z64555> When does the communication occur? which states can do communication?
[03:15:12] <Jak_o_Shadows> Hmm. comm occurs at outside requests - i don't know when. largely the comm just wants me to send the output of the entire processing.
[03:15:29] <Jak_o_Shadows> But it'd be DEAD handy to be able to set inputs to bits of the provcessing, and look at the outputs
[03:16:07] <Snert> visual feedback is best tool.
[03:16:29] <Snert> love my blinkys for that.
[03:16:40] <Jak_o_Shadows> aha. this data is a tad complex for blinky
[03:16:45] <Jak_o_Shadows> through, I do have some
[03:19:22] <z64555> sounds like a coprocess, in which case it wouldn't be in the state machine
[03:19:43] <Jak_o_Shadows> how does it communicate with the state machine is my sticking point
[03:21:43] <z64555> It would be like a node that branchs from/to all major state nodes
[03:22:38] <Jak_o_Shadows> z64555, do you mind if I draw a diagram in a few days of what I'm thinking?
[03:22:40] <z64555> If this is implemented in C, it would be outside of the switch statement inside of the main loop
[03:22:49] <Jak_o_Shadows> Tonights work is to resolve a discussion I was having with deship.u
[03:22:57] <z64555> Sure, go ahead, but you may have to remind me
[03:25:20] <Jak_o_Shadows> oh yeah, would
[05:37:10] <goppo> with micro steppers like this
https://goo.gl/PvHVgy does anyone know how the white glider is meant to be used?
[05:37:24] <goppo> it looks like just a piece of plastic block
[05:38:28] <goppo> likewise
https://goo.gl/WYnSP0
[06:46:09] <mumptai> hi
[06:51:28] <veverak> starts to love clang-format
[06:51:33] <veverak> sorted my #include <3
[06:51:34] <mumptai> one hyppthesis, electronics limits the advanement of amateur robotics? what is contradicting this?
[06:54:05] <veverak> that's too wague question
[06:54:09] <veverak> really depends on context
[06:54:35] <mumptai> yes. let me explain
[06:54:39] <veverak> only thing that limits anything is lack of people with skills to advance it
[06:55:47] <mumptai> 15 years ago battery life, cpu power and sensor were damn expensive. now these things are cheap and available (and sometimes easy to use)
[06:57:07] <mumptai> and i still don't see the cool stuff going on in research arriving in amateur robotics
[06:58:53] <mumptai> it looks like the only new thing are webcams on robots driven by tablets and phones over wifi
[07:02:09] <mumptai> so, is it too hard to use or is just nobody interested?
[07:03:10] <mumptai> like, rpi + opencv is kinda smooth sailing
[07:03:38] <mumptai> arudino might handle low level interfaces well enough
[07:03:46] <veverak> hard to use may be proper question
[07:04:24] <Jak_o_Shadows> goppo, is the white bit not what actually moves?
[07:04:32] <goppo> Jak_o_Shadows, yes
[07:04:40] <Jak_o_Shadows> Like, it's a stepper. So the rod rotates. But the plastic is probably threaded, so it gives you linear motion
[07:04:42] <goppo> Jak_o_Shadows, the white bit moves
[07:04:57] <goppo> but how do you attach to it
[07:05:08] <Jak_o_Shadows> Slot over it somehow?
[07:05:11] <goppo> it's quite oddly shaped
[07:05:20] <veverak> mumptai: to make proper advance of computers in robotics still require high skill with computers in that manner
[07:07:09] <mumptai> so, cost and availability are of the table. so skill, time and will are the candidates
[07:07:22] <veverak> yep
[07:07:30] <veverak> or, I think so
[07:09:16] <mumptai> software skills, yes. electronics might just be good enough at the arduino level. the mechanical and systems engineering aspects remain
[07:09:50] <veverak> depends on what you do
[07:09:52] <mumptai> and maybe mechanical skill at lego level would also suffice at the beginning
[07:09:56] <veverak> :)
[07:10:56] <mumptai> you got to tie it all together, otherwise it won't work
[07:11:23] <mumptai> but that is the most basic thing, that is required
[07:17:13] <mumptai> so to advance, in this case, means to help people gain the required skills and build the will to apply it to robotics
[08:20:45] <malcom2073> mumptai: It's a numbers game. In order for things to appear and flourish in the amateur domain it has to become popular, otherwise it remains fairly niche, which greatly limits the breadth of feature and functionality since the number of people doing it are few and far between.
[13:26:10] <rue_house> wow, UPS battery died, that takes a lot
[13:31:05] * rue_house tries to restore the network
[13:32:44] <rue_house> well, the tent infront of the workshop collapsed this morning, I heard the members break as it went
[14:02:59] <mumptai> malcom2073, yes i agree, but it still seems to lack behind
[14:15:51] <rue_house> wow, sun is out in full force
[14:17:27] <anniepoo_> Hey Rue, rained this morning here, now it's warm and sunny, which is gonna make a mess, since there's snow
[14:18:21] <mumptai> +5°, overcast and now dark, propper IRC weather
[14:19:22] <anniepoo_> well, today's my day to really nail all the CAD work
[14:19:31] <anniepoo_> on Pomatia
[14:19:39] <mumptai> Pomatia?
[14:19:50] <anniepoo_> my robotic snail 8cD
[14:20:58] <mumptai> nice
[14:21:08] <mumptai> freecad? openscad?
[14:21:53] <anniepoo_> solidworks
[14:22:02] <anniepoo_> and blender for the organic stuff
[14:23:21] <mumptai> i'm still looking for a m-cad with decent assembly function, as open source
[14:24:55] <malcom2073> mumptai: If you find one, let us all know please?
[14:25:31] <mumptai> the time-limited student versions, or some multi-1k price tags just don't work for open projects
[14:27:04] <mumptai> what actually turned out to work better than expected, using freecad to design single part attachment for the openbeam bars
[14:28:36] <mumptai> but that is avoiding assembly in the virtual domain completly
[14:28:50] <anniepoo_> Kevin's using Fusion360
[14:29:01] <malcom2073> I've heard pretty good things about fusion 360
[14:29:05] <anniepoo_> and liking it a lot
[14:29:14] <malcom2073> Coming from SW, I couldn't get into it though
[14:29:19] <anniepoo_> I don't want to pay the learning curve again
[14:29:42] <mumptai> but isn't its future pretty unsure?
[14:30:09] * z64555 checks his email to find the name of that one cad software
[14:30:13] <anniepoo_> I've spent a lot of time recently looking at snail shells
[14:30:24] <malcom2073> SW offers a veteran discount, it's only like $100
[14:30:27] <malcom2073> If that applies
[14:30:59] <mumptai> most robotics stuff i did took a 2-5 years of weekends and holidays. so i need some stability, and i also like to share the stuff i can't sell (which is a 100% for robotics)
[14:31:04] <anniepoo_> and it's obvious I have to choose between what works best both for HRI plans and for fitting the mechanism and anatomic fidelity
[14:31:17] <anniepoo_> it's a painful decision
[14:33:02] <mumptai> but the ones that been there before will recognize your effort (if that helps ..)
[15:12:40] <anniepoo_> yup
[16:46:42] <rue_house> hmm, sure is open with that tent collapsed
[20:31:01] <i-make-robots> hey, i have an a4988 driver breadboarded between an UNO and a 400 step NEMA17 motor. I can make it move one direction just fine. The other direction it doesn't move at all.
[20:31:06] <i-make-robots> why?
[20:31:27] <i-make-robots> the only difference is the voltage on the direction pin - when I make it HIGH I get no movement.
[20:33:45] <Jak_o_Shadows> oddd
[20:34:04] <Jak_o_Shadows> If you're sure you have it connected right, try swapping the motor leads around.
[20:34:10] <Jak_o_Shadows> To test if the motor wants to go back
[20:34:19] <synja> ^
[20:34:20] <i-make-robots> i tried a different UNO pin, a different spot on the same line of the breadboard, touching the wire directly to the pin on the a4988 driver...
[20:35:25] <i-make-robots> when i physically flip the wires for teh stepper, it drives the other way just fine.
[20:35:49] <synja> then it's not the motor
[20:35:51] <Jak_o_Shadows> Ah.
[20:36:01] <Jak_o_Shadows> So you think it's the stepper driver pretty much?
[20:36:04] <synja> itself
[20:36:13] <i-make-robots> maybe. i'll try another driver.
[20:36:14] <synja> ^
[20:36:31] <i-make-robots> fuckin thing was brand new...
[20:36:52] <Jak_o_Shadows> Do you have a logic analyzer or scope or anything?
[20:37:37] <i-make-robots> yep.... just a dead driver.
[20:38:18] <synja> mwhahaha. time for a refund
[20:42:35] <i-make-robots> maybe when i order another 200 from the OEM...
[20:42:56] <i-make-robots> i tested all these damn things when they came in from the manufacturer, so it's on me somehow. static? Dunno.
[20:43:30] <synja> i've never had static damage my equipment ever
[20:43:38] <synja> that's just an old engineer's tale
[20:43:41] <synja> ;)
[20:43:46] <i-make-robots> me neither, afaik.
[20:43:49] <i-make-robots> maybe thsi si the first.
[21:01:54] <i-make-robots> well...now it works.
https://www.instagram.com/p/BQHPCZ7jfS4/
[21:02:25] <i-make-robots> next step: do it without the sensor slipping on the motor shaft. some kind of "sense you're being pushed, don't fight the human."
[21:07:18] <malcom2073> i-make-robots: breadboarding stepper drivers is dangerous. If one of your wires pops loose, or doesn't have a good connection, it'll fry the driver
[21:07:59] <malcom2073> Just something to be aware of, be careful with it :)
[21:59:42] <rue_house> i-make-robots, hows the arm!?
[22:12:34] <rue_house> i-make-robots, you cant just leave pins you dont want to use floating, they will take on voltages that will stop the driver from working right
[22:14:07] <rue_house> hmm, ok, I think I'm understanding how I want put these classes togethor into an amp