#robotics Logs

Nov 21 2017

12:00 AM z64555: coulda sworn I saw a few online awhile ago
12:03 AM z64555: uh, try searching for "robotic" servos
12:03 AM z64555: https://electronics.stackexchange.com/questions/213737/are-there-double-sided-servos
12:03 AM anonnumberanon: they all have an output shaft on one end and on the other end they have just a bearing
12:11 AM z64555: i dunno what to tell ya, then
12:18 AM anonnumberanon: heh
12:18 AM anonnumberanon: it is a weird fact
12:19 AM anonnumberanon: is a long steel shaft very expensive relatively speaking? who knows
12:25 AM z64555: hey, that's a whole 2 cents more per unit! Too expensive!
12:25 AM z64555: (sarcastically guessing)
12:31 AM anonnumberanon: https://www.aliexpress.com/item/F17664-JMT-10-Pcs-2-100mm-Metal-Model-Axle-Gear-Shaft-Diameter-2mm-DIY-Toy-Accessories/32665715155.html?ws_ab_test=searchweb0_0,searchweb201602_3_10152_10065_10151_10344_10068_10130_10345_10324_10342_10547_10325_10343_10546_10340_10341_10548_10545_10541_10562_10084_10083_10307_10178_10060_10155_10154_10539_10312_10059_10313_10314_10534_10533_100031_10103_10073_10594_10557_10558_10596_10595_10142_101
12:31 AM anonnumberanon: 07,searchweb201603_2,ppcSwitch_5&btsid=626a706a-69b2-4bff-97d5-051151986d3c&algo_expid=a4d805b2-326c-4b8a-8fb0-bd3b804845c3-31&algo_pvid=a4d805b2-326c-4b8a-8fb0-bd3b804845c3&rmStoreLevelAB=0
12:31 AM anonnumberanon: you're right
12:32 AM anonnumberanon: although they would have to use different potentiometers because as it stands, the output shaft's butt connects to the potentiometer but in the other scenario, it would go through the pot and all the way out to the other end of the servo
12:32 AM anonnumberanon: https://goo.gl/58hG6h
12:33 AM anonnumberanon: shortened url ^
12:34 AM zhanx: what happened to anniepoo
12:54 AM anonnumberanon: this little thing is sooooooo cute :)
12:54 AM anonnumberanon: https://goo.gl/7SCFKZ
01:06 AM Jak_o_Shadows: zhanx, no commits on github for 3 weeks. 'bout the same on instagram
01:07 AM zhanx: thanks Jak_o_Shadows
01:31 AM i-make-robots: hello
01:31 AM i-make-robots: what is the technical name for a coupler between a motor shaft and a flat surface like a wheel or plate?
01:32 AM i-make-robots: i'm looking for them online and finding nothing.
01:33 AM anonnumberanon: something like a servo horn?
01:33 AM anonnumberanon: https://www.google.com/search?client=firefox-b-1-ab&biw=1408&bih=671&tbm=isch&sa=1&ei=ldYTWvr3MIOzggegoKnwBQ&q=servo+horn&oq=servo+horn&gs_l=psy-ab.3..0i67k1j0l5j0i7i30k1l2j0i30k1j0i5i30k1.6043.7733.0.8123.5.5.0.0.0.0.77.313.5.5.0....0...1c.1.64.psy-ab..0.5.310...0i7i5i30k1j0i13k1.0.PgVNr3fZ6Fs#imgrc=6BzQt8j-ys07mM:
01:36 AM i-make-robots: a wheel with a cut for a key on the inside si pretty perfect. a set screw on the smaller models I need.
01:38 AM anonnumberanon: http://www.robotshop.com/media/catalog/product/cache/1/image/900x900/9df78eab33525d08d6e5fb8d27136e95/a/c/actobotics-set-screw-futaba-servo-shaft-coupler-1-4.jpg
01:38 AM Jak_o_Shadows: second ever PLA print
01:39 AM anonnumberanon: cool how does it look
01:40 AM anonnumberanon: http://www.ebay.co.uk/itm/5mm-x-6-35mm-Small-Rigid-Shaft-Coupler-Stepper-Servo-Motor-CNC-Coupling-Nema-17-/181965700273
01:43 AM Jak_o_Shadows: haven't got bed 100% level
01:43 AM Jak_o_Shadows: seems ok
01:54 AM rue_mohr: i havnt got to bed either
01:56 AM rue_mohr: Jak_o_Shadows, .... pla is the easier plastic, what you been printing with?
01:57 AM rue_mohr: ridgid shaft couplers are a bad idea
01:59 AM rue_mohr: a stock tiny26 must be set to 1Mhz
02:10 AM Jak_o_Shadows: Traditionally, ABS
02:11 AM Jak_o_Shadows: I'm printing that sprocket shaped not-sprocket.
02:11 AM Jak_o_Shadows: Only 90mm diameter, so it's a reasonably simple print
02:12 AM Jak_o_Shadows: If it doesn't break from the stress, it'll definitely be the right hcioce to print it
03:39 AM Jak_o_Shadows: BEAUTIFUL print
09:02 AM DenDen562: hey
09:03 AM rue_mohr: hay
09:04 AM rue_mohr: hmm, if I paid my credit card on nov 19, why did they give me interest on the 20th?
09:12 AM DenDen562: because they are bad
09:13 AM durrf2: some places have like a fukken 5 day processing period
09:13 AM durrf2: so they dont "know" you paid till 5 days later
09:48 AM Snert: system is geared around making $ on the float.
10:03 AM DenDen562: yep
01:37 PM veverak: seems like I've got solution for battery management
01:37 PM veverak: http://www.ti.com/product/bq25703a
01:56 PM veverak: now how to make a board
02:13 PM robotustra: veverak, what pitch?
02:13 PM veverak: pitch?
02:16 PM robotustra: what is the traces width and what is the distance between pins on yout chip
02:16 PM robotustra: 1.27 mm?
02:16 PM robotustra: 1mm?
02:17 PM robotustra: veverak, did you ever make PCBs?
02:18 PM veverak: yeah
02:18 PM veverak: QFN chip
02:20 PM veverak: robotustra: sec
02:21 PM robotustra: how do you make your pcbs?
02:21 PM veverak: CNC mill
02:24 PM robotustra: and you can do 1 mm resolution?
02:33 PM veverak: nope
02:33 PM veverak: solving that problem just now
02:33 PM veverak: wait, 1mm ? maybe, depends
02:36 PM robotustra: https://imgur.com/jOKAqFi it's like 1 mm pitch
02:36 PM robotustra: do it with laser printer
02:38 PM veverak: I am thinking more for a adapter
02:38 PM robotustra: adapter should be cheap I think
02:38 PM veverak: https://www.itead.cc/qfn-qfp-tqfp-lqfp-16-80-to-dip-adapter-breakout-board.html
02:40 PM i-make-robots: https://www.instagram.com/p/Bbvp4Ijgzs8/?taken-by=imakerobots
02:41 PM veverak: http://www.proto-advantage.com/store/product_info.php?products_id=3100132
02:41 PM veverak: this
02:41 PM veverak: exactly
02:42 PM robotustra: i-make-robots, show me your whole robot
02:44 PM i-make-robots: https://www.instagram.com/p/BbsaNaRgoab/?taken-by=imakerobots
02:44 PM i-make-robots: uh, ok...
02:44 PM i-make-robots: https://www.instagram.com/p/Ba68yt6AKv8/?taken-by=imakerobots here's an old pic
02:44 PM i-make-robots: before I installed the limit sensors
02:44 PM i-make-robots: or the gears.
02:44 PM i-make-robots: I supposed it's time for a new pic.
02:46 PM robotustra: i-make-robots, do you drive your joints directly from stepped motors?
02:46 PM robotustra: no reduction?
02:47 PM robotustra: you have bir drivers
02:47 PM robotustra: big*
02:49 PM AchiestDragon: you know if you use adaptors you may find its not much more to actualy get a custom pcb made
02:50 PM i-make-robots: robotustra - planetary gears, ~1 deg backlash
02:50 PM i-make-robots: currently the is one custom metal part, an adapter on the shaft to distribute the torque through the plastic. without it the shaft key eats the plastic.
02:50 PM robotustra: 1 degree is a lot
02:51 PM robotustra: should be no backlash
02:51 PM i-make-robots: in planetaries? 1 is acceptable.
02:51 PM i-make-robots: i can do a lot with an arm that has 1 degree.
02:52 PM i-make-robots: i'm working on getting strain wave gearboxes.
02:53 PM i-make-robots: also, personally, I'm really fucking tired of people saying 1 is a lot. Do better, then show me how. :)
02:54 PM i-make-robots: i'm not saying you're wrong. it is a lot if you want a really tight machine. but come on, plastic parts. printable robot.
02:54 PM robotustra: I have no lashbacks in my cnc mill
02:55 PM robotustra: and that's why I will redo my robot in metal to avoid lashback from the very beginning
02:56 PM robotustra: and I would use timing belts in joints
02:56 PM AchiestDragon: its a hell of a lot of effotr to make "radial " gearboxes with antibacklash gears , its not so bad on linear ones and usualy you can mod the fix in
02:56 PM robotustra: not planetary gearboxes
02:56 PM robotustra: yes
02:56 PM i-make-robots: and that's cool. I'm trying to make a robot anyone can build. not everyone has a CNC mill. timing belts are a thing everyone will complain about in your arm. trust me, I already got an earfull.
02:59 PM robotustra: as you wish
02:59 PM robotustra: I will use some wabble drive also
03:01 PM robotustra: like this one https://www.youtube.com/watch?v=POSKD3lSD8E
03:01 PM robotustra: it can be done without lashbacks
03:07 PM i-make-robots: harmonic drive? lol. show me it working with real load. every example i've seen has no load, because load jams the fucker. they're shit.
03:08 PM i-make-robots: nice idea, terrible in execution. like the wankel engine of gearboxes.
03:09 PM i-make-robots: i tried them. spent a lot of money on R&D. couldn't do it. but i sincerely wish you the best of luck. i'd love to be proven wrong.
03:09 PM i-make-robots: i wouldn't let my ego get in the way of a working solution.
03:11 PM robotustra: it's not a harmonic drive
03:11 PM robotustra: it's wabble drive
03:12 PM robotustra: it's cycloidal drive
03:13 PM robotustra: when I'll mill it in metal I'll do a real load test of cause
03:13 PM i-make-robots: i recognize it's a cycloidal drive. Sometimes peple call them harmonic, like I did. it's a mistake, of course. harmonidDrive is a trademark for strain wave drives in the USA.
03:14 PM i-make-robots: look, you do you, right? don't listen to me. what do I know. https://hackaday.com/2015/05/23/hackaday-fail-a-3000-prototype-that-doesnt-work/
03:15 PM i-make-robots: idk why they put a pic of a plantary. that's dumb.
03:15 PM robotustra: nothing is flexing in wabble drive
03:16 PM robotustra: I know what is harmonic drive of cause
03:16 PM i-make-robots: that's what you think. but you're wrong. even metal flexes, and when it does the wabble will bind and jam.
03:16 PM i-make-robots: when the gearbox has no load, the metal doesn't flex because (obv) there's no force pushing back. once there is a load, BAM.
03:17 PM robotustra: I don't want to argue with you about anything does not exit yet
03:17 PM i-make-robots: I BUILT ONE.
03:17 PM robotustra: exist*
03:17 PM i-make-robots: i built more than one.
03:17 PM i-make-robots: it already existed.
03:17 PM robotustra: harmonic?
03:17 PM i-make-robots: ugh! yes.
03:17 PM robotustra: I built 1 wabble drive and it works
03:17 PM i-make-robots: i've made every kind except strain wave, which i'm now trying to source from china.
03:17 PM i-make-robots: show me, please. with load on the output shaft.
03:18 PM robotustra: this guy on video I showed could break my hand
03:18 PM robotustra: for instance
03:18 PM i-make-robots: WITH LOAD OR NOTHING :)
03:18 PM robotustra: and it's PLA printed
03:19 PM i-make-robots: i'm not interested in a gearbox that has nothing attached. show me it being actually used.
03:19 PM i-make-robots: please.
03:19 PM robotustra: come in half of year
03:19 PM robotustra: I'l show you
03:19 PM robotustra: now I'm busy with cncn lathe
03:19 PM i-make-robots: hahahahahaha.... ok. and if it doesn't work, you write to me and tell me i was right.
03:20 PM robotustra: haha, you arabs are so naive ppl
03:20 PM i-make-robots: ?
03:20 PM robotustra: I think it's because of lots of sun
03:20 PM i-make-robots: wtf do arabs have to do with anything? that got racist fast.
04:53 PM zzzoid: How do I calculate G(q) in 3d space for gravity compensation?
04:54 PM zzzoid: I have calculated the distance to centre of mass for each link using FK.. and calculating the link length on the xz plane
04:55 PM zzzoid: But of course the gravity on each link depends on its rotation..
04:55 PM zzzoid: How can I implement this
05:00 PM z64555: zzzoid: what are you using to model acceleration due to gravity? As a constant accleration?
05:00 PM zzzoid: Yes 9.81
05:01 PM z64555: Ok.
05:01 PM z64555: Torque on a link due to gravity is from all child links center of mass
05:02 PM zzzoid: yes I have calculated that
05:02 PM zzzoid: Except each link has a different axis of rotation
05:03 PM z64555: are each links one degree of freedom?
05:03 PM zzzoid: So a link that rotates about y axis has force but no torque, unless another link changes its rotation
05:03 PM zzzoid: Yes 4 links total of 4DOF
05:03 PM z64555: ok, so each pivot between links is just one axis
05:04 PM zzzoid: Yes J1-J2-J3-J4-ee and joints rotate on y-z-z-y
05:08 PM z64555: Ok then. Torque due to gravity is cos(a) * v_com * g
05:08 PM z64555: where a is the angle between v_com and horizontal
05:08 PM z64555: v_com is the vector from the link's pivot to the center of mass of the link and its children
05:08 PM zzzoid: What is v_com?
05:10 PM z64555: remember, torque is defined as the perpendicular force at a radius away from a pivot
05:13 PM z64555: Should the center of mass be perfectly perpendicular to horizontal (which is vertical) then the torque about that link's pivot is effectively zero
05:14 PM z64555: *the torque due to gravity, that is
05:14 PM zzzoid: But what if all links are horizontal
05:14 PM zzzoid: But the pivot doesn't allow rotation in that configuration
05:15 PM zzzoid: How can I cancel out the force of gravity
05:15 PM z64555: the torque will still be there, but instead of being countered by a servo, its countered by the material strength of the pivots
05:16 PM z64555: Since you can't counter the torque, just skip down to the one that can handle it
05:16 PM zzzoid: but in this case the torque is just a value that goes to the motor, so if I send a torque then the link will rotate
05:17 PM zzzoid: But I only want to send torque values to compensate for gravity
05:17 PM zzzoid: And since the links can be in any configuration at a given time
05:17 PM zzzoid: could I use the Jacobian for this somehow?
05:18 PM z64555: So the links can be rotated so that their pivot is parallel with the gravity vector?
05:18 PM z64555: or is there a mechanical stop on each pivot to prevent them from going past some angle?
05:18 PM zzzoid: Well yes, the DOF allow any configuration.. of the end-effector..
05:18 PM z64555: that doesn't clarify anything
05:19 PM zzzoid: By rotating each pivot about it's axis, the next pivots axis changes
05:19 PM zzzoid: E.g. a link with DOF about the y axis is not affected by gravity, but should this link rotate
05:19 PM zzzoid: Then it will be
05:20 PM z64555: define the x, y, z axes with respect to gravity and the horizon
05:20 PM zzzoid: That is to say if the previous link rotates about Z
05:22 PM z64555: not all coordinate systems are the same. Some have Y as up, some have Z as up, and others have Z as down
05:30 PM zzzoid: Sure,
05:30 PM zzzoid: But I'm not sure how this solves the problem of calculating the compensation torques at each joint
05:35 PM z64555: no I mean. I'm asking what your axes are defined as
05:35 PM z64555: so I don't know if your Z is up, down, or running sideways
05:36 PM z64555: also since the pivot can be rotated such a way where the servo can't counteract torque due to gravity
05:38 PM z64555: Then there's another angle you need to keep track of, which is the angle between the vector that points along the pivot's axis, with the gravity vector
05:38 PM z64555: s/with/and
05:40 PM z64555: So you have two reference angles. One angle is between the link and the horizontal plane. The other is the angle between the link and the vertical vector
05:43 PM z64555: the two angles are in a spherical coordinate system situated around the link's pivot under evaulation
05:51 PM z64555: also the two angles are not co-planar, but instead are orthonormal
07:51 PM computer: hi
07:51 PM computer: is it possible to output say a 80 digit number from a digi comp I or II ?
08:12 PM z64555: digi comp?
08:17 PM anonnumberanon: sup
08:18 PM anonnumberanon: are you a baby?
08:18 PM anonnumberanon: computer, ?
08:22 PM computer: nice
08:22 PM anonnumberanon: cool display but 10 bucks
08:22 PM anonnumberanon: https://www.youtube.com/watch?v=D-zgtylBKUc
08:25 PM computer: z64555: it's a computer
08:25 PM anonnumberanon: in other news
08:25 PM anonnumberanon: i looked up stuff about pwm making and i forgot that I was making progress in one thing: looking at the assembly code!
08:31 PM z64555: "its a computer" doesn't really clarify things. "Its a specific computer model" might've been better
08:32 PM * z64555 waits for his slow internet to load info about the digi comp 1
08:33 PM rue_mohr: anonnumberanon, software pwm?
08:33 PM anonnumberanon: yeah rue
08:33 PM rue_mohr: on avr?
08:34 PM rue_mohr: !!!!!!!!! SEVERLY OPTIMIZED CODE !!!!!!!!!
08:34 PM rue_mohr: REFERNENCE ASSEMBLER OUTPUT WHEN MODIFYING
08:34 PM rue_mohr: THIS PWM FUNCTION
08:34 PM rue_mohr: gcc -O1
08:34 PM rue_mohr: **********************************************************/
08:34 PM rue_mohr: SIGNAL (SIG_OVERFLOW0) {
08:34 PM rue_mohr: if (++pwmcount == 0){
08:34 PM rue_mohr: SetBit(PWM1BIT, PWM1PORT);
08:34 PM rue_mohr: SetBit(PWM2BIT, PWM2PORT);
08:34 PM rue_mohr: SetBit(PWM3BIT, PWM3PORT);
08:34 PM rue_mohr: SetBit(PWM4BIT, PWM4PORT);
08:34 PM rue_mohr: };
08:34 PM rue_mohr: if ((pwm1 ^ pwmcount) == 0) { ClearBit(PWM1BIT, PWM1PORT); }
08:34 PM rue_mohr: if ((pwm2 ^ pwmcount) == 0) { ClearBit(PWM2BIT, PWM2PORT); }
08:35 PM rue_mohr: if ((pwm3 ^ pwmcount) == 0) { ClearBit(PWM3BIT, PWM3PORT); }
08:35 PM rue_mohr: if ((pwm4 ^ pwmcount) == 0) { ClearBit(PWM4BIT, PWM4PORT); }
08:35 PM rue_mohr:
08:35 PM rue_mohr: }
08:35 PM rue_mohr: oh osry, there is an upgrade
08:35 PM rue_mohr: !!!!!!!!! SEVERLY OPTIMIZED CODE !!!!!!!!!
08:35 PM rue_mohr: REFERNENCE ASSEMBLER OUTPUT WHEN MODIFYING
08:35 PM rue_mohr: THIS PWM FUNCTION
08:35 PM rue_mohr: gcc -O1
08:35 PM rue_mohr: **********************************************************/
08:36 PM rue_mohr: SIGNAL (SIG_OVERFLOW0) {
08:36 PM rue_mohr: if (++pwmcount == 0){
08:36 PM rue_mohr: p0 = pwm0; p1 = pwm1; p2 = pwm2; p3 = pwm3;
08:36 PM rue_mohr: if (power[0] != 0) SetBit(PWM0BIT, PWM0PORT);
08:36 PM rue_mohr: if (power[1] != 0) SetBit(PWM1BIT, PWM1PORT);
08:36 PM rue_mohr: if (power[2] != 0) SetBit(PWM2BIT, PWM2PORT);
08:36 PM rue_mohr: if (power[3] != 0) SetBit(PWM3BIT, PWM3PORT);
08:36 PM rue_mohr: };
08:36 PM rue_mohr: if ((p0 ^ pwmcount) == 0) { ClearBit(PWM0BIT, PWM0PORT); }
08:36 PM rue_mohr: if ((p1 ^ pwmcount) == 0) { ClearBit(PWM1BIT, PWM1PORT); }
08:36 PM rue_mohr: if ((p2 ^ pwmcount) == 0) { ClearBit(PWM2BIT, PWM2PORT); }
08:36 PM rue_mohr: if ((p3 ^ pwmcount) == 0) { ClearBit(PWM3BIT, PWM3PORT); }
08:36 PM rue_mohr: }
08:36 PM rue_mohr: glitch free
08:36 PM anonnumberanon: well on avr 1 microsecond only has 16 machine instructions to do things so it is hard to get a 1 microsecond delay, maybe on a typical arm dev board at 72 mghz it gives about 70 instructions per microsecond
08:37 PM rue_mohr: uh, software pwm is not fast pwm
08:37 PM anonnumberanon: so i may move to arm but i think but just coding in assembly i can push a 16 Mhz avr to do something close to a 1 microsecond pwm
08:37 PM rue_mohr: if you want fast pwm you use hardware pwm
08:37 PM z64555: rue: y u no pastebin
08:37 PM rue_mohr: z64555, :P
08:37 PM anonnumberanon: z64555, pastebin are for old people
08:38 PM rue_mohr: I'm not interrupting an important conversation
08:38 PM anonnumberanon: did you say interrupts?
08:38 PM z64555: computer: you might be able to extend it to print out 80 characters if you could move the display
08:38 PM rue_mohr: that fn should be called by a timer interrupt, yea
08:38 PM z64555: anonnumberanon: oh, what's the hip new code sharing thing, then?
08:39 PM anonnumberanon: rue_mohr, but there is a lot of overhead in that function, and the compiler would make it worse
08:39 PM anonnumberanon: I'm sure it doesn't create nice results.
08:39 PM rue_mohr: its pretty damn good at -O1
08:39 PM anonnumberanon: I don't know.
08:39 PM rue_mohr: want to see?
08:39 PM rue_mohr: ok
08:40 PM anonnumberanon: yes can you show oscilloscope?
08:40 PM anonnumberanon: there are a few criteria for niceness for this, 1> lack of signal jitter
08:41 PM rue_mohr: k, I'm not gonna paste that much, I think I miseed the -o1 when I did it
08:41 PM anonnumberanon: 2> if 5 outputs are told to turn off after 1 microsecond, they should do so
08:41 PM anonnumberanon: heh
08:41 PM rue_mohr: anonnumberanon, its glitch free, and I think I run it at about 300Hz
08:42 PM anonnumberanon: _|-|_|-|_
08:42 PM anonnumberanon: lol
08:43 PM anonnumberanon: rue do you use anything else than an mcu like the 4017s or other stuff?
08:43 PM rue_mohr: damnit, power went out
08:43 PM anonnumberanon: lol
08:43 PM rue_mohr: rlly?
08:43 PM rue_mohr: SERIOUSLY?
08:43 PM rue_mohr: its only like 6:44!?!?
08:43 PM anonnumberanon: what happn?
08:43 PM rue_mohr: damnit
08:43 PM rue_mohr: the power went out
08:44 PM rue_mohr: I have to go !@#$@#%$ with my generator, biab
08:45 PM anonnumberanon: z64555, do you see how rue_mohr calls "power outage" when you ask him a hard question?
08:45 PM rue_mohr: I dont know how long the ups running the modem lasts
08:46 PM anonnumberanon: ;)
08:46 PM z64555: maybe his brain is hooked up to the power grid, pulled in too much current and blew the substation
08:46 PM anonnumberanon: heh that reminds me of that movie with Depp
08:47 PM orlock: you mean Keanu?
08:47 PM anonnumberanon: no, Transcendence
08:47 PM orlock: Ahh
08:48 PM orlock: thanks for reminding me
08:48 PM orlock: NOT
08:49 PM anonnumberanon: i think this robot is doing pretty okay https://www.youtube.com/watch?v=eZGz7gXjvZk
08:51 PM anonnumberanon: <z64555> anonnumberanon: oh, what's the hip new code sharing thing, then?
08:52 PM anonnumberanon: sshing to a common known server and opening a sockets program that does chat
08:53 PM z64555: kinda like piratepad, huh
08:53 PM z64555: except more envolved
08:54 PM anonnumberanon: this is cool
08:54 PM orlock: multiplayer notepad
08:54 PM orlock: duh
08:55 PM z64555: MS word and OneNote also allows that, they call it "collaborative editing"
09:01 PM anonnumberanon: Funny how nobody uses it.
09:01 PM anonnumberanon: At least I've never seen anyone use this at work.
09:01 PM anonnumberanon: Iv used stuff like this to test a candidate for coding interview.
09:04 PM anonnumberanon: codepad
09:10 PM rue_mohr: hello? internet, are you still therE?
09:10 PM anonnumberanon: yes
09:12 PM rue_mohr: k, there are two ups's that have generator problems
09:12 PM rue_mohr: one isn't connected, the other just hates the generator power
09:20 PM anonnumberanon: :) maximum showoff: https://www.youtube.com/watch?v=Qqz28b6SLzQ
09:20 PM anonnumberanon: he's cool af too
09:22 PM rue_mohr: well, damnit, I didn't make tea yet
09:24 PM rue_mohr: nice interface
09:28 PM anonnumberanon: the code you posted SIGNAL(SIG_OVERFLOW0) is an ARM interrupt? AVR doesn't have this I think
10:13 PM anonnumberanon: http://www.avrfreaks.net/forum/disucussion-when-o0-appropriate
10:37 PM * anonnumberanon fires up avr-gcc
10:42 PM rue_: REAL POWER!
10:45 PM rue_ is now known as rue_mohr
10:45 PM rue_mohr: there is a breeze that leaves a trouble in the air, I dont think that'll be the only outage this week
10:45 PM rue_mohr: and I have to rework some backup power
10:45 PM rue_mohr: ugh, now I smell like gas
10:45 PM rue_mohr: the carb blocked up somehow, didn't want to start
10:45 PM rue_mohr: hut I need to clear it out again, right as the power came back, it started choking and coughing
10:45 PM rue_mohr: otherwise, a good generator run
10:45 PM rue_mohr: oh, I need to go deal with shop servers
10:45 PM rue_mohr: k, that stirred up the evening
10:51 PM rue_mohr: ok, there is a cable from the ups to the shop fileserver now
10:51 PM rue_mohr: that'll help[
10:52 PM orlock: a pull-cable attached to the fileservers cd-rom tray?
10:52 PM orlock: so it can pull-start the genny in case of an outage?
10:52 PM orlock: because that would be kinda cool
10:55 PM rue_mohr: hah
10:55 PM rue_mohr: yea, I dont know what happened there, the genny wouldn't start, its usually good about it
10:56 PM rue_mohr: had to dump the carb and ... well usually it starts and keeping it going smoothly is the problem
11:06 PM anonnumberanon: you internet does not cut off when the power cuts off?
11:07 PM rue_mohr: hell no
11:07 PM rue_mohr: I have ups' on the main internet equip chain
11:07 PM rue_mohr: but the generator only does the house
11:07 PM anonnumberanon: how come , what kind is it? the cable coming to your house from the ISP, that does not cut off?
11:07 PM rue_mohr: and the ups on the router and main switch dosn't like the genny power, so it keeps disconnecting till its battery is dead
11:08 PM rue_mohr: fiber to the home man!
11:08 PM anonnumberanon: that's reasonable
11:08 PM rue_mohr: passive FDM hubs on the street
11:14 PM rue_mohr: I need to software simulate the effect of a random network of interconnected timers, ideas?
11:14 PM rue_mohr: "random"
11:14 PM rue_mohr: each timer sends an impulse
11:15 PM rue_mohr: to n other timers
11:16 PM rue_mohr: so WTF is wrong with the shop em lighting pack TODAY
11:32 PM anonnumberanon: rue_mohr, just do it
11:55 PM anonnumberanon: What is the maximum frequency I can run at with this without modifications? It's an arm dev board. It doesn't have a crystal oscillator on the board. The model is stm32f030 and its datasheet says frequency "up to 48 MHz". https://www.buildyourcnc.com/images/IMG_9991-800.JPG
11:56 PM Tom_itx: pll?
11:57 PM Tom_itx is now known as Tom_L