Back
[02:53:33] <anonnumberanon> So what do we do about a wireless gamepad for robots?
[02:54:59] * MrCurious looks away http://gameimps.com/ps3-controller-linux-usb-290
[02:56:20] <anonnumberanon> This entails the robot needing to run Linux :)
[02:57:31] <anonnumberanon> I don't know why but I have an itch to make a Linux for an ARM platform as an exercise of brain masturbating futility.
[02:59:47] <anonnumberanon> Is there an advantage to running Linux on embedded other than being able to use all that's already been written for it as far as libraries and packages go?
[03:01:14] <MrCurious> or you can pay %20 cpu overhead, and just be happy shit works, optimize that shit later
[03:01:20] <robopal> isnt that enough?
[03:03:45] <anonnumberanon> yeah but my computer runs linux and it isnt cool. but robots should be cool alwayd
[03:12:46] <anonnumberanon> Are there specs on the ssc32 servo controller as far as bits of precision and jitter?
[03:13:50] <MrCurious> margo price ladies and gentlemen
[03:30:27] <anonnumberanon> i dont understand what they mean here by 0.5 us accuracy of their pwm:
[03:30:31] <anonnumberanon> http://vi.raptor.ebaydesc.com/ws/eBayISAPI.dll?ViewItemDescV4&item=151267427004&category=71394&pm=1&ds=0&t=1460273175644
[03:31:00] <anonnumberanon> 0.5 us of jitter?
[03:31:25] <rue_house> it means your servos will have the shakes
[03:31:41] <anonnumberanon> cause that would be quite good if true
[03:31:50] <rue_house> pff
[03:31:55] <anonnumberanon> rue
[03:31:59] <anonnumberanon> wtf
[03:32:20] <anonnumberanon> 0.5 is good enough for business isnt it?
[03:32:33] <rue_house> you people and your low resolution jittery signals
[03:34:10] <rue_house> Complete new code. (stability increased 10 times)
[03:34:20] <rue_house> so, it'll crash 10x less than it did before?
[03:34:29] <anonnumberanon> that made me cry of laughter too
[03:34:53] <rue_house> oh hum
[03:35:05] <rue_house> I need to work out code for an adc based on the tiny26
[03:35:22] <rue_house> needs to be really good timewise cause it'll be part of a servo loop
[03:36:38] <rue_house> its usb, and its got auto baud recognition?
[03:36:47] <anonnumberanon> how many chips do you need for 14bit res no jitter, also 32 PWMs while we're t it
[03:37:16] <rue_house> is that like saying you have a can of black paint with builtin undercoat that automatically adjusts its tint to the colour of what its painted on?
[03:37:34] <rue_house> I'm not an idiot trying to do 32 channels on one chip, DUH
[03:37:43] <anonnumberanon> i th8nk when you dont use usb it is an arduino shield
[03:38:06] <rue_house> ok then
[03:38:28] <rue_house> been a long time since I'v seen someone ACTAULLY do autobaud
[03:39:09] <anonnumberanon> must have some arithmetic in it
[03:39:09] <rue_house> so tell me, when 32 servos all update, whats the inrush current?
[03:39:23] <rue_house> no, its just a bit watching technique
[03:39:58] <rue_house> usually use the uart like a scope to determine the minimum bit times, and adjust to that rate
[03:40:15] <anonnumberanon> how many bits to read to know baud, worse case scenario?
[03:40:37] <rue_house> it usually require seeing an enter, 0x0A
[03:41:55] <rue_house> supports wireless joystick... that leaves a lot of ambiguty
[03:42:08] <anonnumberanon> my board can output 12 Amps. no problems on that side
[03:42:34] <rue_house> trickey power traces
[03:42:43] <rue_house> plane?
[03:42:43] <anonnumberanon> that shield, id say 6amp then starts gets hot
[03:42:57] <rue_house> do they have 4017 on there?
[03:43:12] <Jak_o_Shadows> yeah, i've watched uart signals on a scope and found that my system clock is wrong. That was fun.
[03:43:18] <anonnumberanon> i have been wondering about that
[03:43:37] <rue_house> 8 bits ea, could be 4017 or a shift register
[03:43:55] <rue_house> meaning they just need to use 4 timers
[03:44:14] <rue_house> gonna be interrupts in there, wonder if they overlap
[03:45:01] <anonnumberanon> so they listened to your teachings
[03:45:09] <rue_house> ?
[03:45:11] <rue_house> pff
[03:45:37] <rue_house> I'm trying to stay awake, but I'm useless so I dont know why
[03:45:52] <rue_house> what is the good of being awake if all your able to do it keep yourself awake
[03:46:55] <rue_house> and why the hell do USB connectors have to be reinvented every 2 yeras?
[03:47:17] <anonnumberanon> the more the merrier
[03:47:38] <rue_house> ugh, you would think its apple
[03:47:54] <Jak_o_Shadows> I always liked mini
[03:48:17] <rue_house> "nono, thats yesterdays, connecotor, it looks the same, but there is a key so you cant use it with the old stuff"
[03:48:25] <anonnumberanon> my robot wireless controller huge breadboard mess took a hit or something
[03:48:37] <rue_house> hah, 1 loose wire?
[03:48:41] <rue_house> I hate that
[03:48:44] <anonnumberanon> since i showd 8t for job interview last month
[03:48:59] <anonnumberanon> i checked all the wires
[03:49:09] <rue_house> or one wire pushed it a bit too far and the insulation pushes the connectors appart and it dosn't make
[03:49:29] <rue_house> did it work for the interview?
[03:49:32] <anonnumberanon> on screen 4 out of five values from the pots dont react to moving said pots
[03:49:43] <rue_house> ouch
[03:49:50] <rue_house> are they stuck high? low?
[03:50:08] <rue_house> I take it they did work?
[03:50:11] <anonnumberanon> kinda high i think or kinda low
[03:50:24] <rue_house> did you do something with all the unused adc inputs?
[03:50:39] <rue_house> if not, try tieing them all low via 1k or something
[03:50:55] <anonnumberanon> i think i was using 5 so that didnt leave many unused ones
[03:51:06] <anonnumberanon> i mean i know
[03:51:18] <rue_house> I'v had LOTS of fun with adc channels that weren't the channels they were supposed to be, and were picking up the signal via capacitance from adjacent pins
[03:51:42] <rue_house> you have a realtime value dumper?
[03:52:08] <anonnumberanon> i think at the interview they were fine. i got the job lol
[03:52:16] <anonnumberanon> then lost it
[03:52:25] <rue_house> nice that it worked
[03:52:30] <rue_house> how did you lose it?
[03:53:02] <anonnumberanon> low budget startup priority was on the beta of their app
[03:53:09] <rue_house> oh
[03:53:33] <anonnumberanon> they had to let me go for a UI consultant for the smartphone app
[03:53:48] <rue_house> musical chairs
[03:54:24] <anonnumberanon> but yeah real time value dumper
[03:54:32] <anonnumberanon> what?
[03:55:09] <anonnumberanon> its just the thingy i had on my robot video
[03:55:57] <anonnumberanon> it's real time
[03:59:52] <rue_house> tels you the adc values as you cange them?\
[04:00:46] <anonnumberanon> yeah
[04:00:47] <rue_house> I need to build in force feeback for the gripers on my arms
[04:01:10] <rue_house> maybe I could change the pot for a current sense
[04:01:17] <rue_house> hmmm
[04:01:31] <rue_house> I think most of the servos use a 1.1V ref across the pot
[04:02:37] <anonnumberanon> you have a claw built?
[04:02:41] <rue_bed> the new arm will be fun, custom motor/pot servos
[04:03:05] <rue_bed> arms 5 thru 9 have grippers
[04:03:05] <Jak_o_Shadows> rue, looking forward to it
[04:03:21] <rue_bed> I'm planning to reuse my PD controller
[04:03:30] <rue_bed> tho I might port it to a m328
[04:03:41] <Jak_o_Shadows> Why not PID ? Why just PD?
[04:03:49] <rue_bed> I wasn't working out
[04:04:00] <rue_bed> causing awefull instability
[04:04:18] <rue_bed> integrators are trickey, they wind up and you have to reset them at just the right times
[04:04:22] <Jak_o_Shadows> But won't you have some sort of steady state error with just PD?
[04:04:33] <rue_bed> yes
[04:04:58] <rue_bed> but its all software, I can change it when I'm motivated
[04:05:16] <rue_bed> make it 10x more stable than the last version :)
[04:05:20] <anonnumberanon> so what is the input into your PD?
[04:05:36] <anonnumberanon> lol
[04:05:39] <rue_bed> serial
[04:05:52] <Jak_o_Shadows> Is this just angular position of each joint? That's what you're controlling?
[04:06:04] <rue_bed> control over serial, feedback by adc
[04:06:12] <anonnumberanon> well played. but you arent Chinese only they can achieve factor of 10
[04:06:16] <rue_bed> yup, 200 degrees per pot
[04:06:24] <rue_bed> heh
[04:06:44] <Jak_o_Shadows> So the base arm joints perform worse? Because they're mrore non-linear?
[04:07:11] <rue_bed> I'm still assembling mechanics
[04:07:27] <rue_bed> my next issue is the wrist assembly
[04:07:46] <rue_bed> its going to be cable drive, and I'v not found any cables to use yet
[04:07:56] <anonnumberanon> youtube has some good videos on that
[04:08:11] <rue_bed> ;) I have a direction
[04:08:29] <rue_bed> have you seen it so far?
[04:08:53] <Jak_o_Shadows> nope
[04:08:59] <anonnumberanon> the newest bin of metal parts? no
[04:09:16] <rue_bed> http://ruemohr.org/~ircjunk/robots/arm6/p1070916.jpg
[04:09:26] <Jak_o_Shadows> Rue, is your PD controller online? I get a decent amount of the theory of PID controllers, but I'm at a bit of a loss of how they actually get coded
[04:09:33] <rue_bed> thats the wrist, see directory for mre
[04:09:45] <rue_bed> I dont know if the code is
[04:10:13] <rue_bed> Jak_o_Shadows, yes, real-world implementation is a challange
[04:10:33] <rue_bed> esp when you need to deal with saturation and windup
[04:11:03] <Jak_o_Shadows> Yeah. Timing as well - I've never seen an implementation that didn't assume a constant tick rate.
[04:11:50] <rue_bed> PI!?
[04:11:57] <Jak_o_Shadows> yeah.
[04:11:58] <rue_bed> maybe its PI and not PD!?
[04:12:08] <rue_bed> http://ruemohr.org/~ircjunk/robots/buddy_III/servo32/adc_pwm_PI_uart4/
[04:12:17] <rue_bed> been 3 years, ouch...
[04:12:21] <anonnumberanon> make a quadcopter youll learn it in a hurry
[04:13:05] <rue_bed> it'd be easier on a mips where the words are larger and the instructions go by faster
[04:13:30] <rue_bed> PI, huh, sorry
[04:14:13] <rue_bed> oh yea, each channel has a status led
[04:14:39] <Jak_o_Shadows> If I were doing it, It'd be on a little stm32f10blah controller
[04:16:17] <Jak_o_Shadows> Hmm. Isn't one way of avoding windup just turning off the I until the error is smaller?
[04:16:19] <rue_bed> the project was started a LONG time ago
[04:16:32] <rue_bed> there are lots of ways
[04:16:53] <rue_bed> error = cmdpos[0]-AdcValues[1];
[04:16:53] <rue_bed> if (ABS(error) < INTTHRESH) { In[0] = limit(In[0] + error, -INTTHRESH, INTTHRESH); }
[04:16:53] <rue_bed> else { In[0] = 0; }
[04:16:53] <rue_bed> error = (Kp*error)+(In[0]/Ki);
[04:16:53] <rue_bed> pwm0 = limit(ABS(error), 0, 255);
[04:16:54] <rue_bed> if (error > 0) SetBit(DIR0BIT, DIR0PORT);
[04:16:56] <rue_bed> else ClearBit(DIR0BIT, DIR0PORT);
[04:17:09] <Jak_o_Shadows> Yeah, you limit your error to an amount right?
[04:17:54] <rue_bed> yup, looks like I turn the integrator off if over too much error
[04:18:56] <Jak_o_Shadows> I think you might also be assuming unit timesteps.
[04:19:16] <Jak_o_Shadows> Which, as long as you don't have any physical meaning to values, is probably fine actually.
[04:19:20] <rue_bed> the calcs are called on a regular time interval
[04:19:21] <anonnumberanon> you been busy rue
[04:19:33] <veverak> yay
[04:19:38] <veverak> been long time till I used scad
[04:19:44] <veverak> and I think quality of code is bad
[04:19:45] <veverak> http://www.thingiverse.com/thing:1479754
[04:21:03] <Jak_o_Shadows> Are you actually firing an interrupt and running the calcs, or are you just waiting until enough time has passed.
[04:21:07] <Jak_o_Shadows> It looks more like the latter
[04:21:57] <rue_bed> the interrupts clear a flag that has the main loop redo the calcs
[04:22:14] <rue_bed> its timed in a disjointed way
[04:22:41] <Jak_o_Shadows> Yeah. So they have a minimum wait between calcs.
[04:22:45] <Jak_o_Shadows> Is there a maximum?
[04:23:25] <rue_bed> ah, its based on the adc conversions, after 8 cnversions the calc are redone
[04:23:51] <rue_bed> I think at the time I was taking it that the conversions take a fixed amount of time
[04:23:57] <rue_bed> not so sure of that now
[04:24:02] <anonnumberanon> they should all takes exactly the same time
[04:24:09] <rue_bed> successive approx?
[04:24:24] <Jak_o_Shadows> I would think those might be ok - that's above my head anyway.
[04:24:29] <Jak_o_Shadows> You have uart in there as well though
[04:24:40] <rue_bed> uart is handled by main
[04:24:44] <rue_bed> so its interruptable
[04:24:49] <Jak_o_Shadows> righto
[04:26:29] <rue_bed> I wonder why D was making the loop unstable...
[04:26:57] <Jak_o_Shadows> I suspect windupness?
[04:27:08] <rue_bed> I recall that I spend a long time trying to get PID and gave up
[04:27:13] <rue_bed> D cant
[04:27:24] <rue_bed> its just V-V(t-1)
[04:27:32] <Jak_o_Shadows> Actuator limits?
[04:27:40] <Jak_o_Shadows> Dunno, I should play around in simulink some
[04:28:17] <rue_bed> I think the same code did a fine job of linear position control of a standard solinoid
[04:28:36] <rue_bed> look for the solinoid video on youtube by ruenahcmohr
[04:29:09] <rue_bed> it flipped out when it accidently latched, but it was underdamped and so recovered
[04:29:52] <rue_bed> I was impressed by the ability to cntrol a solinoid like that
[04:30:20] <rue_bed> esp since it only took like an hour to pull it all togethor
[04:30:56] <Jak_o_Shadows> I oughta get some of those strip things I have working.
[04:35:11] <anonnumberanon> bed
[11:18:18] <veverak> http://veverak.org/~squirrel/laptop/scad/beta3.png
[11:18:34] <veverak> camera gimbal is too big
[11:18:42] <veverak> still
[11:19:10] <veverak> But I suppose that only thing that will help me now is smaller servos
[11:20:01] <veverak> *are
[11:59:35] <rue_house> ah, is that how that works
[11:59:53] <rue_house> how do you turn on the ruler?
[12:00:01] <rue_house> can it be a cubde grid?
[12:00:35] <rue_house> a green, blue and red one
[12:00:45] <rue_house> translucent cube grid
[12:00:49] <rue_house> yea
[12:00:53] <rue_house> yea....
[12:20:55] <rue_bed> nobody doing anything yet today?
[12:21:51] <rue_bed> I think I'd like to make a temptrary gripper for the mecha and get its arms going
[12:23:01] <rue_bed> haha I had a dream I found parts for its shoulder
[13:31:53] <veverak> rue_house: actual openscad?
[13:31:56] <veverak> :)
[13:45:09] <rue_shop3> yay! my power booster works
[13:45:23] <rue_shop3> 12V -> 24V
[13:46:41] <rue_shop3> without blowing up
[13:58:17] <rue_shop3> "its designed to achieve both maximum bigness and maximum smallicity"
[14:01:20] <veverak> http://veverak.org/~squirrel/laptop/scad/animated.gif moving!
[14:01:23] <veverak> :)
[14:02:30] <rue_shop3> yep, not done that yet
[14:03:24] <rue_shop3> now you just need to move the camera around at the same time
[14:05:25] <veverak> not really
[14:05:29] <veverak> just wanted to see min/max
[14:05:31] <veverak> :)
[14:05:36] <veverak> point is
[14:05:41] <veverak> I still don't like the way it looks
[14:05:48] <veverak> make it smaller
[14:05:50] <veverak> !
[14:12:45] <robopal> nice veverak, what program did you use to make it?
[14:13:03] <veverak> openscad
[14:13:17] <veverak> also
[14:13:19] <veverak> http://www.thingiverse.com/thing:1479754
[14:13:24] <veverak> will upload rest later
[14:13:27] <veverak> :)
[14:24:43] <Snert> might haveta trim servo arms down to fit the groove for them. Nice universal mount.
[14:44:57] <anonnumberanon> Nah ill prolly be coding some k&r C and looking up company of my next job interview
[15:20:20] <flyback> <rue_shop3> "its designed to achieve both maximum bigness and maximum smallicity"
[15:20:26] <flyback> and super canuckness
[15:34:43] <deshipu> veverak: someone already modded the tote he built at the workshop, and shared an improvement in the foot design!
[15:34:47] <deshipu> veverak:
http://imgur.com/a/lS6vK
[15:35:44] <veverak> nice
[15:35:47] <veverak> :)
[15:35:55] <veverak> that's good one
[15:36:10] <veverak> deshipu: I made progress on camera gimbal
[15:36:15] <veverak> http://veverak.org/~squirrel/laptop/scad/animated.gif
[15:37:10] <deshipu> compact
[15:37:49] <deshipu> have you considered attaching the camera to the second servo instead? so that the lens is closer to the center of rotation?
[15:38:03] <veverak> deshipu: well, which center? :)
[15:38:13] <veverak> this way it's closer to center of rotation of 'z axis' servo
[15:38:15] <veverak> :D
[15:38:16] <deshipu> the tilt
[15:38:20] <veverak> also
[15:38:31] <veverak> poin it, that this is more compact
[15:38:33] <deshipu> be careful about the dimensions of those servo horns
[15:38:37] <deshipu> they vary greatly
[15:39:35] <veverak> if I would do it other way
[15:39:47] <veverak> I would have to make much bigger space between camera and "closer" servo
[15:39:55] <veverak> so it can rotate 90 degress up
[15:39:57] <veverak> :)
[15:40:28] <veverak> this way, the space can be small because the arm is long -> big arc of movement -> it won't hit top of servo
[15:40:34] <deshipu> you could stack the two servos vertically
[15:40:38] <veverak> so, it's smaller practicaly
[15:40:42] <deshipu> which would give you a better view too
[15:40:45] <veverak> verticall?
[15:40:50] <veverak> *vertically
[15:40:58] <deshipu> I will make a sketch
[15:41:01] <veverak> good
[15:44:05] <deshipu> http://paste.sheep.art.pl/62ddeada-544c-46f5-9384-c6599f45a47a/%2Binline
[15:44:10] <deshipu> something like this
[15:44:33] <deshipu> then the lens is not farther away from both axes than the thickness of the servo
[15:45:31] <deshipu> of course, if you move the top servo to the side, and put the camera on its horn, you can even have the camera ideally on both axes
[15:45:54] <deshipu> at the cost of the whole thing being much wider
[15:46:20] <deshipu> I think the extra height might actually be good for such a small robot
[15:47:50] * veverak want's it small :/
[15:48:11] <deshipu> I guess you will have to test anyways
[15:48:13] <deshipu> and see
[15:48:22] <veverak> yeah
[15:54:23] <deshipu> I have a loose idea for a "brain slug" 3d-print, with an eye on a pan-and-tilt, and OpenMV in it, that you could wear on your head, and it would look at people's faces
[15:55:15] <veverak> lol :D
[15:57:04] <deshipu> https://www.youtube.com/watch?v=uVvVqtTN6QE
[15:57:06] <deshipu> one of those
[16:23:36] <rue_shop3> ok garden is taking shape
[16:25:12] <rue_shop3> next I'll work on putting this 24V converter into the mower...
[16:25:59] <deshipu> there is this guy here trying to make a "line following" mower, over a wire buried into the lawn
[16:27:47] <veverak> :)
[16:44:16] <rue_house> I have a good setup for that
[16:44:22] <rue_house> picks up a wire really well
[16:44:35] <rue_house> you run about 10Khz or better thru the wire to make the pickup easier
[16:45:15] <rue_house> I have a good diagram around for a digital freq filter that will let you use different wires for differnt things
[16:45:23] <rue_house> one for follow, another for fence etc
[18:21:03] <anonnumberanon> deshipu lol
[18:29:54] <anonnumberanon> I just tested the voltages coming out of the pots. Moving the pots does not change the voltages, except for the one that still works.
[18:30:50] <anonnumberanon> So I know my ADC is not fried or something. sigh..
[18:30:56] <anonnumberanon> Kinda nice to know.
[18:31:19] <anonnumberanon> Problem is: there is absolutely no reason my "pots board would just stop working.
[18:39:22] <Snert> can you swap pots?
[18:40:42] <anonnumberanon> Oh when I remove power to the pots I think I still get voltage. :)
[18:40:52] <anonnumberanon> Just not from the working one
[18:43:34] <Snert> can you swap pots software wise?
[18:44:38] <anonnumberanon> Yes of course but I don't have my code upload equipment ready it's all in bags cause I moved.
[18:45:02] <anonnumberanon> I just tested the ground and it shows 4 milli volts on the pots board.
[18:46:03] <anonnumberanon> should be 0 ofc
[18:46:41] <Snert> I'd just kill the juice and ohm twixt the 2 grounds.
[18:47:18] <Snert> maybe add a 2nd ground wire for grins
[18:47:23] <anonnumberanon> There must be some "reverse voltage" or something.
[18:52:41] <anonnumberanon> Man I'm reading 9 milli volts coming out one of the ADC pins. Must be the ADC.
[18:55:23] <anonnumberanon> nvm
[18:55:32] <anonnumberanon> lol
[18:58:05] <anonnumberanon> Forget the ADC it's all a pots board problem
[18:58:17] <anonnumberanon> -says he, only 60% sure of his claim
[19:25:35] <rue_shop3> put 1/2 fsv on all the channels and see if they all read 512
[19:36:20] <anonnumberanon> I put the voltmeter to them and no matter pot position voltages don't change. Pot failure is highly unlikely but they just dont work :(
[19:37:02] <anonnumberanon> what is fsv?
[19:37:21] <anonnumberanon> forward sex voltage?
[19:38:46] <orlock> How about the resistance?
[19:39:28] <anonnumberanon> ah yeah ill test that
[19:58:30] <rue_shop3> full scale voltage
[20:01:20] <Tom_itx> "The server just went down on me" has a different meaning in an IT office then on a tennis court
[20:03:34] <Jak_o_Shadows> true
[20:06:49] <rue_shop3> ike
[20:07:31] <orlock> rue: best bang for buck batteries for a robot with a weight limit?
[20:14:53] <rue_shop3> see if you can find somene clearing out NiCd, or cell phnoe batteries
[20:14:54] <mrdata> berta voltaics?
[20:14:54] <mrdata> *beta
[20:15:03] <rue_shop3> people get rid of cell phones before the batteries are shot
[20:15:22] <rue_shop3> esp apple ones
[20:15:25] <ace4016> you mean normies do
[20:15:49] <ace4016> my battery is terribad in my phone; but i'm pretty sure i kept the phone for longer than expected
[20:15:57] <orlock> rue_shop3: I'd be buying new, i mean any specific chemistry?
[20:16:01] <orlock> NiCad? LiIon?
[20:23:40] <rue_shop3> most apple users get a new phone every time a new model comes out
[20:23:55] <rue_shop3> apple riggs it so that after 3 software updates, its useless
[20:25:09] <Jak_o_Shadows> true
[20:25:21] <Jak_o_Shadows> The hardware in new smartphones scares me
[20:25:36] <Jak_o_Shadows> 2 GB ram! Quad core! what!
[20:25:53] <anonnumberanon> Wow did you guys get this 18 minutes ago?
[20:25:55] <anonnumberanon> <anonnumberanon> In my mind I'm always on the tennis court or in the elevator with Monica.
[20:25:58] <rue_shop3> not comprable to PC specs
[20:26:30] <rue_shop3> I wonder if I'm done welding for the day
[20:26:33] <anonnumberanon> Jak_o_Shadows, the joke is you look on your android at the available ram and you're left with 25% of that 2GB
[20:27:58] <rue_shop3> hmm
[20:28:09] <rue_shop3> - install 24V converter into lawn mowing robot
[20:28:16] <rue_shop3> - set up more potato planters
[20:28:29] <rue_shop3> - work on arm6s wrist
[20:28:37] <rue_shop3> - install led lighting in my room
[20:28:40] <anonnumberanon> <anonnumberanon> Results from tests: the pot that works shows resistance from 1k to 3k depending on pos. All others show infinite resistance.
[20:28:40] <anonnumberanon> <anonnumberanon> This really sucks it was supposed to be built like a tank.
[20:28:40] <anonnumberanon> * anonnumberanon should never be hired to build tanks
[20:28:40] <anonnumberanon> * Disconnected ()
[20:28:50] <rue_shop3> - install air line to sand blaster
[20:29:04] <rue_shop3> hmmm
[20:29:19] <rue_shop3> eating also might be a fine activity
[20:29:45] <Jak_o_Shadows> LED lighting is always nice
[20:30:47] <rue_house> always put series resistors from the wipers of your pots so that wiring errors cant damage them
[20:31:03] <rue_house> I'm betting you have a wiper and one end mixed up on the 3 that are burned out
[20:36:00] <anonnumberanon> I don't thinbk that's possible.
[20:36:26] <anonnumberanon> I remember making this thing very well. Spent a lot of time. Checked everything. It was working for a long time.
[20:36:42] <anonnumberanon> Until I took it in my backpack for a demo.
[20:38:15] <anonnumberanon> http://cdn.gurl.com/wp-content/uploads/2015/11/giphy-14.gif
[20:41:03] <anonnumberanon> Can pots actually burn though?
[20:56:32] <Snert> if they did, there would be accompanying symptoms.
[20:56:47] <Snert> I assume the pots are in dip cases?
[21:03:30] <anonnumberanon> I think I'll make the button just forward the working pot value to whatever signal I want it to control, for prototyping, until I make a new pot board.
[21:03:57] <anonnumberanon> Snert, I'll show you what model it is exactly.
[21:06:05] <anonnumberanon> It's almost exactly this one:
https://www.digikey.com/product-detail/en/bourns-inc/PTL20-15R1-502B2/PTL20-15R1-502B2-ND/3781294
[21:06:51] <Snert> oic not a digital pot.
[21:07:53] <Snert> still, voltages should be the same twixt working and non-working. I mean +V and grounds. This is something common to all 3 bad pots.
[21:08:45] <Snert> maybe a loose breadboard jumper.
[21:10:40] <Snert> since these are analog pots....
[21:11:29] <Snert> you could very easily mark the known good pot. then measure restistances and compare twixt the known good and a bad one.
[21:11:50] <Snert> it's just a pot....restatance mearurements are no different than any other pot.
[21:12:06] <anonnumberanon> http://imgur.com/Nwf3p81
[21:12:16] <anonnumberanon> These are it.
[21:12:20] <Snert> that would confirm whether or not a pot has failed.
[21:12:30] <anonnumberanon> compare twixt
[21:12:35] <anonnumberanon> What do you mean by that?
[21:13:04] <Snert> a pot in any form is just a variable resiatance.
[21:13:08] <Jak_o_Shadows> slide pots are annoyingly expensive
[21:13:27] <Snert> so you can measure the resistance and compare.
[21:13:42] <Snert> is it a 5k pot?
[21:13:51] <Snert> then you shuld be able to measure 5k.
[21:14:03] <Snert> on the good pot and the suspect bad pot.'
[21:14:27] <anonnumberanon> Yes you must have missed a few of my tests earlier.
[21:14:58] <anonnumberanon> The good one shows resistance from 1k to 3k per as the specs I built it for. The rest is infinite resistance everywhere.
[21:15:08] <Snert> you asked is it possible for a pot to burn.
[21:15:15] <anonnumberanon> Cannot see differences on the board between the wworking one and the nonworking one.
[21:15:23] <Snert> and that would be the way to confirm that a pot has not burned.
[21:17:57] <Snert> and since more than 1 is not working, it seems unlikely that more than 1 pot has failed.
[21:18:18] <anonnumberanon> Yeah it has to be a forced failure on all 4 of them.
[21:21:14] <Snert> I bet you can take any suspect pot or "channel" and swap it into the good pot/channel.
[21:22:01] <Snert> that's soldered down already so forget that.
[21:25:32] <Snert> I like those lites - if I get some slide pots it'll have to be those.
[21:25:51] <anonnumberanon> I should probably stop being depressed about this. Order new ones and start work on something else.
[21:26:23] <anonnumberanon> In the picture you can see how I damaged the LED of the first one while making my first hook up heh.
[21:27:07] <Snert> looks like it's not lit at all
[21:27:32] <anonnumberanon> Yeah it's burnt.
[21:27:48] <Snert> yet, that is the working slider, right?
[21:28:08] <anonnumberanon> I unregulated current in it at 5V so it didn't last long...
[21:28:32] <anonnumberanon> Hehe interestingly enough yeah. But that has nothing to do with it. (OR DOES IT!!!)
[21:28:36] <anonnumberanon> I don't think so .
[21:28:56] <anonnumberanon> Are you saying you want me to burn the other LEDs to fix the pots lol.
[21:29:01] <Snert> I wouldn't think so. But swapping good and bad would tell for sure.
[21:29:22] <Snert> except those sliders aren't in sockets.
[21:29:41] * anonnumberanon calculates the amount of desoldering and resoldering involved
[21:29:58] <anonnumberanon> and having to relocate to garage for this...cause at parents' house heh
[21:36:21] <anonnumberanon> Anyway I'm back. And this robot will learn to walk again!
[22:10:36] <Jak_o_Shadows> Hey, rue_shop3, did you ever test how fast your air muscles could inflate/deflate?
[22:45:26] <rue_house> not speed no
[22:45:42] <rue_house> partly cause the construction is with small pipes right now
[22:46:05] <rue_house> and I didn't tune the servo system for the valves
[22:46:48] <rue_house> the valves are extremely nonlinear, tuning them is a choar
[22:47:14] <rue_house> I used a 4 part lookup table for ballparks
[22:47:49] <Jak_o_Shadows> ah, fair enough.
[22:48:10] <anonnumberanon> chore
[22:48:15] <anonnumberanon> :)
[22:54:17] <rue_shop3> I have one hour and then I cease to exist.
[22:54:23] <rue_shop3> for 5 days...
[22:54:41] <Jak_o_Shadows> damn, why?
[22:54:41] <orlock> ?
[22:54:45] <orlock> how does that work
[22:54:46] <rue_shop3> work
[22:54:50] <orlock> ahh
[22:54:59] <Jak_o_Shadows> oh, yeah, you're in a different timezone
[22:55:00] <orlock> working out in the great icy wastes?
[22:55:07] <rue_shop3> I cant get out of work mindset for 24 hours after its done
[23:12:06] <anonnumberanon> rue_shop3, what kind of job?
[23:13:27] <rue_shop3> electrician
[23:13:37] <anonnumberanon> perfect
[23:13:47] <rue_shop3> yea
[23:13:59] <rue_shop3> make work a cakewalk compared to the hobbies, right?
[23:14:22] <rue_shop3> its not tho, lots of politics and trying to work out how to get wire from here to there
[23:14:39] <anonnumberanon> customer facing too?
[23:15:01] <rue_shop3> politics, like customers asking for impossable things and trying to nicely explain why they are an idiot
[23:16:27] <anonnumberanon> yeah it's almost like it's easier to lie to an idiot than to explain to him how things are done
[23:16:39] <anonnumberanon> because they WILL want to know how it works, right?
[23:16:59] <anonnumberanon> then with their stupid non technical brain they'll think they understand
[23:17:28] <rue_shop3> its a good company and they pay good, so its good
[23:19:40] <anonnumberanon> yeayyy
[23:19:51] <anonnumberanon> you can buy things now!
[23:20:03] * anonnumberanon ducks
[23:21:00] <rue_shop3> heh
[23:21:08] <rue_shop3> I have a house, it eats all the money
[23:21:22] <rue_shop3> I wonder how the mortguage is doing
