#robotics Logs
Nov 30 2017
#robotics Calendar
12:00 AM AchiestDragon: 6 motors per stm32 althoug i could prob do all 18 with one , i want to have enoug h spare for other leg sensors like strain guagues for load balancing and other proimaty sensors on each leg
12:03 AM AchiestDragon: and i still may opt to scrap the idea and switch to making a humanoid style upright robot , one about 3foot tall like a baby terminator
12:03 AM anonnumberanon: ah yeah, here I'll show you some that are like what you're describing:
12:04 AM AchiestDragon: other than im not looking forward to geting it to balance
12:04 AM anonnumberanon: https://www.youtube.com/watch?v=t8Ni5cB9FCc
12:04 AM anonnumberanon: in RoboCup, soccer, now that is getting a LOT of traction
12:08 AM rue_mohr: I dont thin the usa will be interested untill its baseball or hockey
12:08 AM rue_mohr: they love to step in all over on our hockey
12:10 AM AchiestDragon: well at least there not using realdolls in miss world yet
12:11 AM mrdata: lol
12:12 AM mrdata: do they use amputees?
12:12 AM AchiestDragon: robo football in the usa , maybe when they have robo cheer leaders also
12:13 AM mrdata: those seem to exist already, https://www.youtube.com/watch?v=Ebgrr4yLlCk
12:16 AM AchiestDragon: they look sweet
12:17 AM AchiestDragon: although flat surface only
12:28 AM AchiestDragon: micromouse has been going for a long time , what would a a cool robot sport would be a autonamus robot obstical course type race ,
12:33 AM anonnumberanon: for obstocles i think a hexapod would fair better wouldn't it?
12:34 AM AchiestDragon: yea , but also some tracked designs , dependant on the obsticals
12:40 AM AchiestDragon: say like a set of stairs , a doorway (opening) a box , a rough surface like a pile of rocks or aradomly placed bricks , a sloping enbankment and some sticks / cones to avoid hitting so it autonimusly has to avoid some objects and calaculate how to get over others , points based on time and loose time for hitting marker objects or boundry markers
12:44 AM anonnumberanon: what size robots though?
12:44 AM AchiestDragon: maybe 3 catogorays like urban (sort of common city objects like curb stones steps , doorways , and padestrians) , to extream wilderness , big rocks , water like streams , trees and such ,
12:49 AM AchiestDragon: well thered be problems with large robots , ie car size , but human sised ( fit though a door way ) down to say cat size , even smal robots can find ways to go where the bigger ones cant get so it leaves that bit open ) and you could have a size catogory
12:50 AM AchiestDragon: im still looking at trying to get onboard 3d vision and that makes the cpu requirement quite high and not that small
12:51 AM z64555: not if the vision system is modular
12:51 AM z64555: and seperate from the decision making CPU
12:51 AM anonnumberanon: you should assume you can't communicate with other machines
12:51 AM z64555: it'll still be high, but instead of having one CPU do everything, you have a co-processor specifically for high-workload sensors
12:51 AM anonnumberanon: so may have to cram it on the robot itself
12:53 AM AchiestDragon: yea i did buy a hp dl740 8 chip xeon 10gb ram server to do the number crunching as a local cloud server dam thing weighs 68kg and pulls over 1kw of power , and sounds like a hoover with a trubocharger
12:54 AM Luminax-Work is now known as Luminax
12:56 AM AchiestDragon: it should be posible with something like a dual quadcore xeon machine , and i beleve thats what they need to use in the self drive cars
12:57 AM AchiestDragon: althoug hallowing for it to be wifi linked so you dont need to carry the cluster core onboard should be posible
12:58 AM AchiestDragon: i just had prob with getting the video to grab at the right rates over usb , that caused a bottleneck and limited it
01:00 AM anonnumberanon: the Hololens does it onboard
01:00 AM anonnumberanon: will map for example a room, pretty fast
01:00 AM AchiestDragon: small robots not needing the same accuracy or range shoudl reduce the ammount of cpu converting the vison to 3d mapping down to something that could be onboard
01:02 AM z64555: just a matter of working on a lower image resolution
01:02 AM z64555: that's what I would think, at least
01:02 AM z64555: sicne its 3D, using a LIDAR + camera, or stereo vision?
01:03 AM anonnumberanon: in retrospect if the hololens was to detect stairs i don't think it would do it correctly
01:03 AM * z64555 gets all the terms wwrong
01:03 AM anonnumberanon: would just detect a ramp or something
01:03 AM AchiestDragon: and its not like it needs to be done at a frame rate of a car that may be doing 80mph needing to range with enohg acuaracy to cover its stopping distance at that speed is it
01:04 AM z64555: what does the hololens use?
01:04 AM anonnumberanon: dunno, try it and tell us :)
01:05 AM AchiestDragon: i was just going to use sterio vision , as the lazer sensors dont work that well outside
01:05 AM z64555: I'm broke :(
01:05 AM anonnumberanon: hololens has 4 cameras, i think it's close to what the Kinnect was using
01:08 AM anonnumberanon: i was telling AchiestDragon z64555
01:08 AM z64555: kinnect was lidar + camera, and maybe a sonar
01:08 AM z64555: oh, ok
01:08 AM anonnumberanon: even though it's kind of cool i wouldnt buy it myself
01:08 AM anonnumberanon: i could afford it but that would be a big part of my savings ;0
01:08 AM anonnumberanon: BUT I Get to use it at work
01:08 AM z64555: 4 cameras? wierd.
01:08 AM AchiestDragon: the other fact cpu requierments aside , is the 2d (flat plain lazer range 300degree fov laser sensors are still in the £90 each mark where a pair of webcams are like £5 each
01:08 AM anonnumberanon: actually sorry, the reason for that is you got cameras for the environment, and the cameras that detect your hand gestures
01:10 AM AchiestDragon: anyway one step at a time , need a to get the cad sorted and get a sutable hexapod frame that has enough "payload capacaty for experimental controls
01:11 AM z64555: frame can carry the weight fairly easily, its the servo's you'd want a close look at
01:11 AM AchiestDragon: and idealy 3d printed
01:12 AM * z64555 takes a back seat
01:12 AM * anonnumberanon same
01:13 AM anonnumberanon: although he's using good motors
01:13 AM anonnumberanon: now the gearing is the question, how
01:13 AM anonnumberanon: electronics should not be too hard and have been done many many time
01:13 AM z64555: spur gears are easiest, if a bit vulnerable
01:15 AM AchiestDragon: driving threaded m3 bar , as a linear actuator , from a 12v motor if the threaded bar is good and not rough , then it should provide enough force so that i cant stop it moving by hand , so thats about 5 to 7kg maybe more
01:17 AM AchiestDragon: the threaded bar would be close to bendding and the plastic breaking would be more of an issue i think
01:18 AM z64555: why would it bend?
01:19 AM AchiestDragon: i need to print one and test it to desruction to get some idea of how much force it actualy gives , the first hexipod used sat positioning jacks , there rated at 1.5 ton
01:20 AM anonnumberanon: brb dinner
01:20 AM AchiestDragon: it shouldent bend if the plastic holds
01:20 AM AchiestDragon: bed for me been up all night again and its 7;20 am
02:19 AM rue_bed: my advice to you is 'compliance', the ability for a robot to fall and the actuators to give enough that the whole thing dosn't snap
02:19 AM rue_bed: I did some 3d printed gearbox cases
02:19 AM rue_bed: made making a gearbox really easy
02:21 AM rue_bed: I wanted to use pots with long shafts for making servos, but they just cant be had for a reasonable price
02:21 AM rue_bed: it turns out that multiturn pots are only rated for a few hundred sweeps
02:22 AM rue_bed: as well as most trim pots, so their out
02:22 AM rue_bed: I'm working on ideas for etching thin sheetmetal
02:22 AM rue_bed: waiting for some materials to arrive, if it works, I can photoetch encoder wheels
02:23 AM rue_bed: I need to play with weak lye mixtures for etching aluminum
02:25 AM anonnumberanon: you would like to keep torque and precision while having compliance?
02:25 AM anonnumberanon: i haven't seen it done, or just didn't realize it was there if I have
02:35 AM anonnumberanon: rue_bed, what software to draw CNC models and is it the same as for 3d printing? also I would like to find a gears simulator or something of the sort
02:36 AM anonnumberanon: geargenerator.com :)
02:44 AM Jak_o_Shadows: anonnumberanon, any decent CAD package will be good for CNC & 3d printing
02:44 AM Jak_o_Shadows: the next step up is probably sheet metal work
02:48 AM anonnumberanon: Jak_o_Shadows, I'd like to find a combination of gears and a motor that is easy to source, cheap, strong, and would give 50kg.cm of torque at 55.6RPM at the output, which is 0.18seconds for 60 degrees
02:49 AM anonnumberanon: i could make several different scenarios and compare prices vs torque vs strength (metal gear vs plastic for example)
02:50 AM anonnumberanon: the only known variables here are the minimum torque and speed i listed, maybe it's shooting high though but I'll discover that in the study i assume
02:53 AM anonnumberanon: Jak_o_Shadows, for sheetmetal work i was planning on cnc aluminum cases
03:26 AM anonnumberanon: 4bearing per servo :X
03:32 AM anonnumberanon: Okay how does this look? https://i.imgur.com/lHyrVcG.png
03:33 AM anonnumberanon: http://geargenerator.com/#130,130,65,360,0,7,425167.4,8,1,8,2,4,27,-90,0,0,16,4,4,27,-60,1,1,8,2,4,20,-60,2,0,40,10,4,20,0,3,1,8,2,4,20,0,4,0,40,10,4,20,0,5,1,8,2,4,20,0,6,0,32,8,4,20,0,0,0,2,-260
04:14 AM Jak_o_Shadows: for sheetmetal work with folds and stuff I meant
07:54 AM AchiestDragon: rue_bed: pots along the shaft or using pots in general after time they get noisy and have deadspots , especialy in devices that move "all" the time , optical encoders have issues also , like dirt buildup or lighting conditions if not covered
07:58 AM rue_bed: anonnumberanon, no, your teeth should get larger as you get to the output, and the train should be folded back on itself
07:59 AM AchiestDragon: the main cost factor are encoders that detect direction or absolute position , but with motor control you should only need a rotation count , as long as you ensure the motor is in a stoped positition and have a referance end /start switch /sensor for a "0" / home position then you just use an up /down counter to count the nos of pulses with up / down tied to eather left right rotation
07:59 AM rue_bed: anonnumberanon, I use librecad, not for gears tho
08:00 AM rue_bed: and then I deal with z in openscad
08:02 AM AchiestDragon: there is a gear lib for openscad , its a bit of a pain to use from what i remeber , saying that not as bad as the cyclic cam lib i did https://www.thingiverse.com/thing:111396
08:04 AM AchiestDragon: anyway plastic gears are going to be out for what i need there not going to be anything like strong enough and they would wair too quick in use , metal gears on the other hand are rather expencive ,
08:08 AM AchiestDragon: https://www.thingiverse.com/thing:5505 and https://www.thingiverse.com/thing:3575 should help for doing openscad gears
08:12 AM AchiestDragon: i need to find the older metric fastner lib i used that would show nuts and bolts with there threads , than the openscad one currently in mcad/dev on git
08:24 AM AchiestDragon: ok so could do with a better tooth profile but my attempt at an antibacklash gear
08:28 AM AchiestDragon: all those links have .scad files in the thing files that you can customize the parameters
08:29 AM AchiestDragon: https://www.thingiverse.com/thing:98393 ,,,
08:31 AM AchiestDragon: i was hoping to do a 3d printed clock at the time other than my printer packed up , https://www.thingiverse.com/thing:92629
08:34 AM AchiestDragon: some of those are freecad FCStd files ,, i seem to think the reason i stoped using freecad was because it was easyer to just use scad than switching between the two packages all the time
09:15 AM rue_mohr: I have a box of about 1500 gears, so...
09:15 AM rue_mohr: as long as I dont want matching sets I'm good
12:01 PM HumanSheeple: Tom_L do you have any idea of any component that has a wormwheel worm gear combo on I can find for scrap?
12:03 PM theBear: HumanSheeple, what kinda size/strength you looking fro ?
12:04 PM HumanSheeple: I really don't care at this stage, I just want SOMETHING so I can demonstrate wormwheels in conjunction with laser lenses to show micrometer adjustments
12:04 PM HumanSheeple: if I can demonstrate a successful test I might get funding for some precision brass ones
12:05 PM HumanSheeple: Like I mean a CD drive but these are square cut gears
12:05 PM HumanSheeple: you know what a worm wheel is right?
12:05 PM HumanSheeple: I mean do car windscreen wiper motors use them?
12:06 PM HumanSheeple: Do guitar keys use them?
12:06 PM HumanSheeple: Anything, really
12:11 PM theBear: worm gear, yeah
12:11 PM HumanSheeple: brilliant, what scrap parts would have one?
12:12 PM theBear: umm, guitar tuning pegs (the bits you twist to tune) got real rough big ones in em
12:12 PM theBear: in a minute i should remember some really common thing that has em... i should
12:13 PM theBear: oh, and basic Techno Lego (tm+tm) has a worm about 1.5" long that fits any of the tiny,small-mid,big-mid and giant sized cogs/gears that it also got
12:14 PM HumanSheeple: Got anything that's a bigger gear, so the smaller rotation fraction of a fraction of a degree
12:14 PM theBear: might be good for a nice big visible one, plus you could choose coloured blocks to hold it all together (the holey techno ones, plus the little + shaped black axle/shafts that fit all the gears
12:14 PM HumanSheeple: Like 70mm 80mm something like that I'm thinking
12:14 PM theBear: the biggest gear in that stuff is maybe 60mm dia from distant memory
12:14 PM theBear: and the worm is about 8-10mm dia, so it's a BIG reduction
12:16 PM HumanSheeple: So you reckon lego techno is the way to go or guitar keys?
12:16 PM theBear: plus cheap bits and pieces, tho for them alone you probly have to order direct from lego one way or another (vs buying a big expensive car/whatever set with them in)
12:16 PM HumanSheeple: I'm thinking something I can pick up from the flea-market for one euro
12:16 PM HumanSheeple: something like a CD drive
12:16 PM theBear: HumanSheeple, depends what you think really... tho the guitar ones aren't much of a stepdown
12:16 PM HumanSheeple: or some scrap car part
12:16 PM HumanSheeple: yeah ok, I want really really really really fine tuning
12:16 PM theBear: mmmm, i still feel like i'm forgetting something "everyone has somewhere" with em inside
12:17 PM HumanSheeple: so one turn equates to 0.001 degee of the big wheel
12:17 PM HumanSheeple: I've heard old blenders have them
12:17 PM HumanSheeple: washing machines
12:17 PM theBear: does it gotta be visible, or just in vs out ratio for your purpose ?
12:17 PM HumanSheeple: I really don't care as long as I can turn something that makes this lens move a tiny tiny amount I'm happy
12:18 PM theBear: gotta be gears ? cd/optical whatsits already got a nice servo/electromag controlled lens in them that can do TINY moves
12:20 PM HumanSheeple: I just want some twisty knobs so I can tweak things
12:20 PM theBear: if it didn't have to be worm, you can find a HUGE reduction gearing in a "ac synchronous motor" as found in disco-ball setups, frost-free fridge timers, errr, electronics surplus
12:20 PM HumanSheeple: Eventually yes I'd like servo/stepper control
12:21 PM HumanSheeple: disco ball setup?
12:21 PM theBear: ooh, little 20-turn trimpots got one inside 'em, but you'd need to hack them up to make it move something other than the internal tiny-pot
12:21 PM theBear: "disco ball motor" is the laymans term
12:21 PM theBear: they "all" look like http://img.frbiz.com/pic/z257d2c4-0x0-1/ac_motor_small_ac_synchronous_motor_110v_ac_0_8_1rpm_torque_10kg.jpg
12:22 PM theBear: it's a layered motor, very back (from where the shaft comes out) is a tiny flat motor that spins in sync with mains, then a super-big ratio gearbox layer (those little pins sticking out hold the gearbox cogs) and the final top cover
12:23 PM theBear: http://img.frbiz.com/pic/z257d2c0-0x0-1/ac_motor_110_127v_ac_2_5_3rpm_synchronous_motor_geared_motor_cw_ccw.jpg notice the rating/sticker 3rpm for 60hz mains
12:23 PM theBear: and they're dirt cheap to buy
12:23 PM HumanSheeple: well I had this ingenious idea I could have a big brass wormgear go around the outside of the laser so that the beam shoots through the middle of the bore
12:24 PM HumanSheeple: is that like a series of precision gears that repeatedly steps down?
12:24 PM theBear: hmm, big, if you can find a destroyed lathe or something you could use the whatsit-screw
12:25 PM theBear: also i suspect SOME domestic/cheap computer printers use the same kinda thing, long threaded shaft with motor off to one side (end of the shaft) and that moves the head which has some kinda thread inside it
12:25 PM HumanSheeple: It's like a shit CNC trapezoid thread
12:25 PM theBear: in the snchronous motors like i linked ? dunno about precision, but yeah, a few rounds of very small gear driving a very big one
12:25 PM theBear: if you moving a lens, that kinda makes sense
12:26 PM theBear: oh, and ballscrew is the term for those big cnc style ones
12:27 PM theBear: mmmm, surely small (laptop if not all) optical/cd mechs have something like that to position the head-unit/assy (along the radius of the disc plane) in them
12:27 PM theBear: whatever is in a given cd drive, it is gonna be able to move tiny amounts, cos that thing gotta move roughly with the resolution of the track on a given disc
12:31 PM HumanSheeple: Yes I've just looked inside a smashed CD drive there's a step down gearbox inside
12:31 PM HumanSheeple: but the gears are tiny
12:31 PM HumanSheeple: and they chatter a little bit
12:31 PM theBear: how big is the lens you gotta move ?
12:32 PM HumanSheeple: 11mmx2mmx1mm however the lens is intended to be housed inside a block of brass I will machine
12:32 PM HumanSheeple: this measures about 4mm x 12mm x 22mm
12:32 PM theBear: also often in these cases, you wanna have a nice solid setup that the lens/whatever can move up and down freely, taking all the weight/strain, then just "push/pull" that assy with your drive setup
12:33 PM theBear: you can machine a block of brass ? why not jsut set the lathe to auto-sideways and make yer own worm to match some cog you already got handy ?
12:35 PM HumanSheeple: That's the plan eventually, but for now I jsut want a "proof of concept"
12:36 PM theBear: well surely any ballscrew (cnc/auto-thinger on a manual lathe) or similar (like cd mech) is proof of concept enough ?
12:37 PM HumanSheeple: OK can you teach me a really easy way how to hoop up a 12mm thread to a bearing so when I twist the thread, the bearing rotates slightly?
12:37 PM theBear: if you got access to big expensive disco-lights, anything since maybe late 90s, they got a whole assembly if not several inside like you want, albeit with much bigger lenses
12:38 PM theBear: umm err, you mean the bearing firmly connected to the threaded rod, or "slightly" means a kinda loose but close fit ?
12:38 PM HumanSheeple: https://imgur.com/a/Nc1VR this is what I've got so far
12:42 PM HumanSheeple: or a pair of bearings enabling an "arc" of motion of the bearing
12:43 PM theBear: it gotta be a bearing ? otherwise you could just do some kinda cam-shaft thing
12:43 PM theBear: also, i understand the question re "rotates slightly" now
12:44 PM HumanSheeple: anything cheap, easily obtainable that can achieve one turn, one one hundredth of a degree of a 70mm disc I'm happy for proof of concept
12:45 PM theBear: howbout some kinda round=-thing, whatever, on the first shaft, then fix a bearing to that, maybe something as simple as a "screw" with a head big enough to catch the bearing inner tube, to a point on the round-thing
12:50 PM HumanSheeple: can u draw for me what u mean?
12:52 PM theBear: no drawing software handy, but think of a car wheel, with one of the bolts going thru the middle of (and securing/fixing) the center of the bearing... make sense ?
05:21 PM i-make-robots: two things!
05:21 PM i-make-robots: thing 1: Jigsolve is now live at vancouver science world museum.
05:22 PM i-make-robots: it's twitch plays a jigsaw puzzle. http://twitch.tv/jigsolve
05:22 PM i-make-robots: thing 2: https://www.reddit.com/r/robotics/comments/7gpbsy/can_you_make_this_robot_arm_stronger_need/
05:22 PM i-make-robots: can you help me make this 3d printed arm stronger? I'm getting part failure in one spot and i'm not smart enough to fix it myself.
05:29 PM AchiestDragon: that does not look a good idea ,, the key is only short os it would almost act like a drill on the plastic , i would drop the key and drill the shaft to take a pin , and have the pin run so it is about the same dia as the plastic part , print the plastic part with an hole to accept the pin , the pin should be a tight fit though the shaft so it does not drop out and be at least 1.5mm dia
05:30 PM Tom_L: where is teh point of failure?
05:33 PM Tom_L: the side opposite the motor cracks out?
05:38 PM i-make-robots: as AchiestDragon said, the key strips out the plastic and makes a small hole bigger.
05:38 PM i-make-robots: I can do a version with a long set screw.
05:38 PM Tom_L: use 2 keys
05:38 PM i-make-robots: sure
05:38 PM Tom_L: and stronger material
05:38 PM i-make-robots: maybe a screw or two that run axially so the layers don't split.
05:39 PM i-make-robots: stronger material isn't really an option, but greater infill can be done.
05:39 PM i-make-robots: 70% infill is a 12h print :)
05:39 PM Tom_L: 100% at the stresss points
05:39 PM Tom_L: can you taper the infill?
05:39 PM i-make-robots: oddly, prusa's slicer doesn't like 100% infill. it's not an option.
05:40 PM i-make-robots: not in this version. maybe later they'll have varying infill.
05:40 PM Tom_L: taper it off away from the hole
05:40 PM Tom_L: but use alot near it
05:40 PM i-make-robots: agreed
05:40 PM i-make-robots: i'd love love love to not have any custom metal parts.
05:40 PM i-make-robots: i have some real work to do now but this weekend i'll update the fusion design and give it a try.
05:42 PM AchiestDragon: what is the dia of the shaft and is the slotted shaft section just something that you can replace with something diferent
05:46 PM Tom_L: how big is the shaft?
05:46 PM Tom_L: increase the shaft diameter, that will put the key further from the center which will decrease the effective load
05:49 PM Tom_L: that joint supports the load of all the others
05:49 PM Tom_L: and any forces applied to them
05:50 PM AchiestDragon: well lets see if i can find a pic ... you need something like this idealy https://a.pololu-files.com/picture/0J1107.1200.jpg?64e2b723aa828e9d3d96a6b98d47dc6b
05:57 PM AchiestDragon: although a quick fix would be to have a peace of fiberglass pcb with a the hole and slotfor the key cut into it and some holes to mount it onto the plastic , just ensure that the key is keying onto it rather than the plastic ,
05:58 PM AchiestDragon: making it that way from metal would be better
06:01 PM AchiestDragon: http://2.bp.blogspot.com/-z_0l1kwDMJA/TrgPJ-UkJOI/AAAAAAAAAeE/An1HIJio5Js/s640/02-UABMM2-New-Mount.jpg like that , but you could get away with it beeing only a about 1 to 2mm thick as long as the key is in contact with it
06:28 PM AchiestDragon: if you got some big metal washers that will fit the shaft file a groove for the key into them and drill 2 or 4 mounting holes to attach to the plastic like https://drive.google.com/file/d/1hARtNO8_kpsBof0YC7yUqTaoVEhuoJpo/view?usp=sharing
06:31 PM AchiestDragon: you may be able to retro fit that to the dammaged part if you can move the keyed shaft so that it mates with it rather than the plastic , else you will have to mod the plastic part so that it can fit
06:44 PM HumanSheeple: i-make-robots: might I suggest either 3d printing carbon fiber infused nylon, design hexagon honeycomb which will make it light and strong
06:48 PM AchiestDragon: youd have more luck sticking a small gear with big teeth on the shaft , heating it fuseing it into the plastic rather than a slotted key design of the original , it would be stronger but how stong do you need it to be
06:52 PM AchiestDragon: the other fix would be use a longer key , and a bigger slot for it in the plastic
08:16 PM rue_mohr: 8-|
08:16 PM * rue_mohr wakes up
08:29 PM anonnumberanon: rue_bed> anonnumberanon, no, your teeth should get larger as you get to the output, and the train should be folded back on itself
08:30 PM anonnumberanon: yes that web app didn't let me place the gears how I wanted i wanted to fold them though, it's also not in 3d, but for the gear sizing, I don't really know what to do, as per your recommendation
08:30 PM anonnumberanon: the way i have them setup here, the output ratio is perfect though
08:30 PM anonnumberanon: to turn 12000rpm input into a 60 rpm output
08:31 PM anonnumberanon: with torque being 200 times more than input, at the output
08:32 PM anonnumberanon: (neglecting friction)
08:50 PM rue_bed: I suggest keeping the ratio backdrivable
08:53 PM Tom_L: new belts for the drill press are too long :(
08:53 PM Tom_L: talked to my bud today and he's gonna help me mill things flat for assembly
08:55 PM rue_bed: cool
08:55 PM Tom_L: oh and turn down the ballscrew end for me
08:58 PM Tom_L: things are moving at a snail's pace
08:58 PM Tom_L: but they are moving
09:18 PM rue_bed: it seems like a good pace to me
09:18 PM rue_bed: ok, am I done with sleep?
09:19 PM Tom_L: you certainly should be
09:19 PM orlock: SLEEP... IS FOR THE WEAK!
09:19 PM rue_bed: yea, I know
09:19 PM Tom_L: took the drill press apart and cleaned and lubed it tonight
09:19 PM Tom_L: gonna find a shorter belt tomorrow
09:19 PM rue_bed: ok
09:19 PM Tom_L: i hope
09:20 PM rue_bed: I dont think I have a spare for mine..
09:20 PM rue_bed: I should set that up before it breaks
09:21 PM Tom_L: kinda hard to measure when it's broke
09:21 PM Tom_L: mine is getting close
09:21 PM rue_bed: yesno, an auto shop can usually deduce it
09:21 PM rue_bed: they know stretchout and stuff
09:21 PM Tom_L: industrial supply here got them a bit too long
09:22 PM Tom_L: i'll get one shorter at least
09:22 PM Tom_L: even if it's too short
09:22 PM Tom_L: cause there's an idler pulley between the two
09:22 PM Tom_L: i'll make sure they fit then save them for when one acutally breaks
09:29 PM anonnumberanon: rue_bed, what do you mean?
09:31 PM rue_shop3: belts stretch as they age
09:32 PM rue_shop3: the automotive stores are used to being brought broken belts and having to work out what size they were new
09:32 PM anonnumberanon: about the gears
09:32 PM rue_shop3: the courser gears take more toque better
09:35 PM Tom_L: i wish there was a way to fix the locator in the column in the drill press but i don't see any way to do it
09:35 PM Tom_L: the thing that keeps it from turning side to side
10:04 PM rue_mohr: SWEEET
10:04 PM rue_mohr: the line amp works
10:51 PM anonnumberanon: https://www.youtube.com/watch?time_continue=10&v=KPhkVPNKtVA
11:24 PM rue_shop3: huh
11:24 PM rue_shop3: pretty good