Back
[00:04:46] <furrywolf> argh imagemagick can be frustrating.
[00:07:07] <zeeshan> still on it?
[00:07:08] <zeeshan> haha
[00:08:06] <furrywolf> I have it nicely recoloring the high resolution image, but I can't make it stick the scale label on it.
[00:08:45] <zeeshan> you should email the stuff to the makers of the microscope
[00:08:47] <zeeshan> :-)
[00:08:51] <furrywolf> I can add the label perfectly to simple images, but it won't add the label to the results of compositing the previous images, and I can't figure out why.
[00:15:00] <furrywolf> argh... I might be hitting a bug.
[00:22:27] <zeeshan> hm!!
[00:22:30] <zeeshan> gimp is powerful
[00:22:33] <zeeshan> but such a pain to use
[00:22:41] <zeeshan> mainly cause im a noob =D
[00:31:02] <zeeshan> finally learned how to animate
[00:31:15] <zeeshan> http://i.imgur.com/sarKomJ.gif
[00:31:29] <furrywolf> I'm failing at some part of understanding imagemagick's layering
[00:31:39] <zeeshan> i can see that i need to get some more data points when the shape changes a lot
[00:32:11] <zeeshan> you really like your command line software dont you!
[00:41:48] * zeeshan will need to think about this
[00:41:53] <zeeshan> its 1:30 am. brain slow :)
[00:41:58] <zeeshan> and i need food
[00:46:03] <furrywolf> for some reason I am failing to understand, if I do other operations later, my perfectly-overlayed image ends up with the contrast cranked up to 100% or something similar, despite my doing no such operations of any kind
[00:47:46] <zeeshan> all i know is you need to find a job
[00:47:52] <zeeshan> where you can utilize all these skills!
[00:48:12] <furrywolf> ... because I need to specify -respect-parenthesis. because -ignore-parenthesis-and-do-things-broken seems to be the default.
[00:49:57] <zeeshan> so i thought about the algorithm
[00:50:37] <zeeshan> find max scale value out of all the images, , multiple each images scale's value by its scale max. divide all the images's height data by the maximum scale value
[00:50:46] <zeeshan> this will normalize them all equally
[00:51:56] <furrywolf> convert -respect-parenthesis \( -size 1600x1200 -background transparent -fill white -strokewidth 2 -stroke black -pointsize 36 label:"$(echo "Z scale:"$(od -An -t f4 -j 0x59cd18 -N 4 025a.tif)"um")" \) \( 025a.tif.jpg \( 025a.tif.png -resize 1600x1200 \( xc:black xc:red xc:yellow xc:green1 xc:cyan xc:blue xc:black +append -filter Cubic -resize 600x30\! \) -clut -alpha set -channel a -evaluate set 90% +channel \) -compose overlay -composite \) -compose d
[00:52:04] <furrywolf> now that's an imagemagick command.
[00:52:10] <zeeshan> rofl
[00:52:19] <zeeshan> i dont know how you do it.
[00:52:27] <zeeshan> its crazy.
[00:52:39] * furrywolf waits for it to slowly upload
[00:54:15] <furrywolf> http://imgur.com/M7zsdxL
[00:55:35] <zeeshan> impressive
[00:55:43] <zeeshan> thats a powerful software
[00:55:50] <zeeshan> and its crazy you know how to work it :p
[00:56:16] <furrywolf> ... sorta
[00:58:19] <furrywolf> what it really needs is some bump mapping too...
[00:58:21] <zeeshan> i wonder if LeelooMinai's vfd has burned down
[00:59:02] <zeeshan> it 'bump maps' in the actual software
[01:00:00] <furrywolf> I need to find the slope at any point in the image and use that to tweak the color to exagerate the 3d effect.
[01:00:00] <furrywolf> but first, I need sleep.
[01:01:33] <zeeshan> youre pretty much reverse engineering their software
[01:01:33] <zeeshan> lol
[01:01:44] <zeeshan> (analysis part of it)
[01:03:51] <furrywolf> no, I'm just inventing how to make pretty pictures from depth data
[01:06:01] <furrywolf> funky. I did something wrong, but the result is cool. :P
[01:06:21] <zeeshan> pic?
[01:06:40] <furrywolf> uploading. slow.
[01:10:44] <furrywolf> my connection might be sucking too badly.
[01:11:34] <zeeshan> lol
[01:11:37] <zeeshan> how big is the file
[01:12:01] <furrywolf> just a normal jpg
[01:12:18] <furrywolf> 280kb
[01:12:28] <zeeshan> haha
[01:12:39] <zeeshan> http://i.imgur.com/acG78Sq.gifv
[01:12:42] <zeeshan> i guess you dont wanna open that :)
[01:13:03] <zeeshan> http://i.imgur.com/acG78Sq.gif
[01:13:07] <zeeshan> this one might be less intensive
[01:14:02] <zeeshan> its a bit more stable now
[01:14:12] <zeeshan> the outside is remaining red
[01:14:17] <furrywolf> http://imgur.com/hXhYFVU
[01:14:33] <zeeshan> that looks kreepy
[01:14:33] <zeeshan> haha
[01:14:35] <zeeshan> creepy
[01:18:09] <furrywolf> that data still doesn't look properly scaled?
[01:18:17] <zeeshan> it is
[01:18:38] <furrywolf> why does it have all the way from red to blue in every frame, even the last one, which is almost flat?
[01:19:44] <zeeshan> i multiplied each image's scale value by the bitmap file so a value like .791 became 111.11 for example, then i found out what the maximum scale value was, and then i divided that value by all the matrices. so the 111.11 value became .521 for example
[01:20:00] <zeeshan> then i use that value to get the color scheme.
[01:20:10] <zeeshan> basically if .521 is red, it'll stay red in all the images
[01:20:50] <zeeshan> it should have red in the last one too
[01:20:57] <zeeshan> because only the diamond shape is changing
[01:21:00] <zeeshan> not the area around it
[01:21:10] <zeeshan> actually wait that doesnt make sense
[01:21:48] <zeeshan> it should all go blue
[01:21:51] <zeeshan> maybe thats an error.
[01:21:54] <zeeshan> in the actual data itself
[01:22:13] <zeeshan> the scope takes like 5000 images in 1 sec
[01:22:34] <zeeshan> changing the focus of each image. so if something is out of focus in the image , it will be an error point
[01:23:49] <zeeshan> i also noticed, my sample slightly moved during the test
[01:23:55] <zeeshan> =/
[01:24:21] <furrywolf> back to the lab! :P
[01:24:24] <zeeshan> haha
[01:24:30] <t12> whats this thing for?
[01:24:31] <zeeshan> the main purpose of this test was for proof of concept
[01:24:43] <zeeshan> im using a cold plate to freeze the action
[01:24:52] <zeeshan> the first 15 images are over a span of 10 minutes
[01:25:01] <zeeshan> you can see that there isn't _much_ change
[01:25:08] <zeeshan> t12 my thesis
[01:25:32] <t12> i mean wahts actually being videoed and why etc
[01:25:38] <t12> microscoped
[01:25:39] <t12> photoed
[01:26:10] <furrywolf> I tried adding more shading, but the lazy way I tried didn't work well, and I'm too tired to implement a better method.
[01:26:27] <zeeshan> sleep dude
[01:26:27] <zeeshan> lol
[01:26:44] <zeeshan> t12:
http://i.imgur.com/0UE3gYQ.png
[01:26:59] <zeeshan> im indenting a shape memory polymer - adhesive- metal laminate
[01:27:10] <zeeshan> and trying to study it's shape recovery behaviour among other things
[01:27:13] <zeeshan> as a function of temp
[01:27:24] <t12> oh so all the images are like
[01:27:26] <t12> depth maps
[01:27:29] <zeeshan> yes
[01:27:44] <zeeshan> the depth is changing as a function of time, and temperature
[01:28:00] <zeeshan> so id like to correlate those with the depth
[01:28:13] <zeeshan> for different types of smp's
[01:28:18] <zeeshan> and non smp
[01:28:20] <t12> how are you actually judging depth
[01:28:35] <zeeshan> the microscope uses a focusing technique to determine depth
[01:28:45] <t12> z stacks and sharpness?
[01:28:46] <zeeshan> i was sayin above it takes like 5000 images per second
[01:28:56] <zeeshan> while changing the focus for each image
[01:29:06] <zeeshan> and computes a depth map from it
[01:29:17] <zeeshan> i was skeptical about it first, but i machined some plates to known depth
[01:29:26] <zeeshan> and it was within .01 mm
[01:29:29] <t12> whats the scope
[01:29:33] <zeeshan> vhx5000
[01:29:37] <t12> yeah it should be good to the limits of the objective
[01:29:51] <zeeshan> i thought it was 2000, but its actually the 5000
[01:29:54] <zeeshan> not sure what the diff is :p
[01:31:04] <t12> so you're trying to reverse engineer its storage format?
[01:31:16] <zeeshan> furrywolf and me have cracked most of it
[01:31:16] <zeeshan> haha
[01:31:39] <zeeshan> furrywolf found the data locations of the hidden info/images
[01:32:10] <zeeshan> we had to do all this, because they don't let you export a 3d point cloud out of the software.
[01:34:46] <t12> those look like neat microscopes
[01:35:51] <zeeshan> furry, i m actually animating the 3d images now
[01:35:59] <zeeshan> it really shows it a lot better on whats going on
[01:46:26] <zeeshan> http://i.imgur.com/hpvNhJS.gif
[01:46:31] <zeeshan> you can see the error in the scope better now :)
[01:48:08] <zeeshan> whoops error in that
[01:48:45] <zeeshan> http://i.imgur.com/SZjelxl.gif
[01:50:48] <furrywolf> you seem to be holding the bottom of the indentation still rather than the rest of the piece
[01:51:11] <zeeshan> it might be how the microscope works
[01:51:16] <zeeshan> it tries to keep the bottom constant
[01:51:20] <zeeshan> i dunno
[01:51:54] <furrywolf> so the microscope also normalizes it so the deepest part is z=0 then?
[01:52:00] <zeeshan> yes
[01:52:02] <zeeshan> thats what im thinking
[01:52:08] <zeeshan> cause if you notice the scale
[01:52:11] <zeeshan> the blue always is 0.
[01:53:26] <furrywolf> so add maxscale-scale to every point for each image? (i.e. if the max is 130, and that image's is 50, add 80 to every point)
[01:54:32] <zeeshan> yea thatll work
[01:55:57] <zeeshan> well the bad news is
[01:56:01] <zeeshan> it looks like my temperature plate isn't working
[01:56:02] <zeeshan> :[
[01:56:11] <zeeshan> this bloody thing starts recovering
[01:56:31] <zeeshan> so im likely losing data between the time i go from the indenter to the microscope.
[01:57:24] <zeeshan> i think i need to plot the value at the tip
[01:57:32] <zeeshan> for each interval and see what sup
[01:57:41] <zeeshan> fak im tired!
[01:58:11] <furrywolf> did you get it so your rim stays in one place instead of the bottom?
[01:58:29] <zeeshan> ?!
[01:58:38] <zeeshan> what rim
[02:00:22] <furrywolf> the outside of the image area
[02:00:52] <furrywolf> I need to get to bed.
[02:00:56] <zeeshan> no i didnt set it like that yet
[02:00:59] <zeeshan> me too!
[02:01:00] <zeeshan> :p
[02:15:56] <aventtini> hello
[02:16:17] <aventtini> my first time on the irc chanel
[02:16:21] <aventtini> channel
[02:17:29] <furrywolf> aventtini: heyas! unfortunately, it's after bedtime for most people here, and that's where I'm about to head too.
[02:17:47] <aventtini> here to :D
[02:17:48] <furrywolf> zeeshan: I got a first version of 3d shading done, but now I'm going to bed.
[02:18:01] <aventtini> its 8 a clock
[02:18:03] <aventtini> :D
[02:18:11] <zeeshan> furrywolf: you're addicted
[02:18:13] <zeeshan> lol
[02:18:20] <furrywolf> zeeshan:
http://imgur.com/kXSLjYm
[02:18:24] <aventtini> im learning manual
[02:18:26] <aventtini> :D
[02:18:32] <zeeshan> wow thats impressive
[02:21:17] <furrywolf> lol
[02:21:36] <furrywolf> http://imgur.com/QvMxmzC too dark, still needs work, but it shows the idea is good.
[02:55:51] <zeeshan> http://i.imgur.com/7Mflttm.png
[02:56:05] <zeeshan> looks like i need to repeat the tests a couple of times to reduce the scatter in the beginning.
[02:56:18] <zeeshan> theres no way the indent is shrinking and then growing lol
[03:05:51] <archivist> zeeshan, but what I want to know, can we start making our own scopes once we know what is going on
[03:05:59] <zeeshan> haha
[03:06:01] <zeeshan> i dont thinkso
[03:06:07] <zeeshan> well its possible
[03:06:13] <zeeshan> but you know it's all about the optics
[03:06:17] <zeeshan> and the fancy ccd cams
[03:07:00] <archivist> actually possibly mechanics and software
[03:08:13] <archivist> looking for peak frequency in the data at each image
[03:08:43] <archivist> fourier transform to do that
[03:09:08] <archivist> I think that is how autofocus works
[03:09:21] <zeeshan> it must
[03:09:30] <zeeshan> i just looked at the specs its 50frames/s
[03:09:37] <zeeshan> so its interpolating massively
[03:10:19] <archivist> I have seen one in action at a show with that little gear I made
[03:11:32] <zeeshan> yea its pretty cool
[03:11:42] <zeeshan> i hope i can get better data!
[03:13:16] <archivist> there are some measurements I thought it was picking the incorrect point though, iirc it was a bit out on the hole diameter
[03:14:19] <zeeshan> i personally think its accuracy is about 0.02mm
[03:14:52] <zeeshan> thats what i was gettin when i measured known depths in solid metal
[03:15:11] <zeeshan> ok i need to sleep!
[03:16:05] <archivist> but it is time to get up!
[03:16:08] <zeeshan> hehe
[03:25:24] <Deejay> moin
[04:19:14] <anonimasu> hello
[04:19:39] <anonimasu> can i somehow force classicladder to show me timers(ms) ones..
[04:37:59] <archivist> define show, I would connect them to pins if possible and then show in some way
[04:40:56] <archivist> http://linuxcnc.org/docs/html/ladder/classic_ladder.html#_the_variable_windows seems to show something
[05:46:36] <eventor> first use , test
[09:08:44] <JT-Shop> is possible to sharpen drill bits on a surface grinder?
[09:09:40] <SpeedEvil> Technically, sure.
[09:10:04] <SpeedEvil> Arrange a jig to hold in the appropriate orientation, do a pass, move, repeat
[09:10:15] <JT-Shop> what about practically?
[09:10:17] <SpeedEvil> It may take less time to rub it against some paper
[09:10:35] <SpeedEvil> A standard grinder will be much, much faster
[09:10:47] <JT-Shop> standard?
[09:11:10] <SpeedEvil> That is - assuming you don't just use the surface grinder freehand and get at the wheel while spinning
[09:11:15] <SpeedEvil> bench grinder
[09:12:30] <JT-Shop> I suck at sharpening by hand
[09:13:03] <archivist> get one of those cheap jigs
[09:13:05] <SpeedEvil> How much have you done?
[09:13:08] <JT-Shop> I've seen those jigs for sharpening end mills for a surface grinder
[09:13:34] <archivist> I mean drill bit jigs for bench grinders
[09:13:35] <JT-Shop> I've sharpened a few dozen over the years by hand
[09:13:44] <SpeedEvil> Not enough.
[09:13:56] <SpeedEvil> You need to sharpen a lot of drillbits at once to get a reasonable skill
[09:14:02] <JT-Shop> I had a jig once for bench grinder but it was a cheap one and didn't work well
[09:14:09] <SpeedEvil> Well - at least enough to get proficient at it
[09:14:21] <SpeedEvil> doing one every two years craply will never help
[09:14:30] <JT-Shop> I don't want the skill, I just want to sharpen a >1/2" bit from time to time
[09:14:48] <SpeedEvil> Then get a jig or a little machine
[09:15:20] <SpeedEvil> Or make a leeeeetle CNC sharpener :)
[09:15:30] <SpeedEvil> There is totally a market for that on kickstarter
[09:15:48] <SpeedEvil> Arbitrary geometry say up to 40mm tool sharpener
[09:16:27] <SpeedEvil> takes in carbide rod, out pops drills.
[09:16:44] <JT-Shop> I have plans for a drill bit sharpener but don't have time to make one
[09:16:45] <CaptHindsight> http://its.foxvalleytech.com/MachShop1/Bench/sterlingdrl/Vtrough.jpg
[09:17:49] <JT-Shop> looks like a dedicated drill sharpening machine
[09:17:50] <CaptHindsight> http://www.micro-machine-shop.com/drill_sharpener_1.jpg
[09:18:39] <JT-Shop> I had one of those in the past
[09:19:00] <JT-Shop> don't think the one I had looked as nice
[09:21:12] <archivist> the nice thing about hand grinding drill bits is adjusting the hole size deliberately
[09:21:16] <JT-Shop> this any good?
http://www.tormek.com/en/jigs/dbs22/
[09:21:38] <JT-Shop> lol I can't even get the cutting edges the same height
[09:22:22] <CaptHindsight> by the time I'm done the bit has lost 1/2" in length :)
[09:22:46] <JT-Shop> by the time I'm don't I've bought a new one
[09:28:14] <CaptHindsight> JT-Shop: do you cast steel or lower melting point metals?
[09:28:58] <JT-Shop> I plan on going up to brass/bronze
[09:50:37] <SpeedEvil> https://www.youtube.com/watch?v=otCpCn0l4Wo - not strictly on topic
[09:50:45] <SpeedEvil> err
[09:50:46] <SpeedEvil> not that
[09:50:57] <SpeedEvil> https://www.youtube.com/watch?v=otCpCn0l4Wo
[09:51:01] <SpeedEvil> doh
[09:51:34] <SpeedEvil> https://www.youtube.com/watch?v=qvx0jXbKkV0 - that one.
[09:51:41] <SpeedEvil> Insanely complicated watch/orrrery
[10:03:42] <andypugh> SpeedEvil: Tourbillons are silly anyway, but that one is sillier than most
[10:04:15] <SpeedEvil> Indeed - quite awesome - but I don't know that I actually remotely want one
[10:04:29] <andypugh> (A tourbillon helps get a better number for the difference between the positions in the chronometer tests, but has no effect on real-world accuracy when the watch is worn)
[10:05:12] <SpeedEvil> http://www.ebay.com/itm/Metal-Lathe-mini-ManSon-restored-antique-micro-Monarch-10EE-jewelers-watch-/231414756807 - would be more my silly purchase
[10:08:51] <andypugh> http://www.lathes.co.uk/manson/index.html
[10:09:51] <SpeedEvil> neat
[10:13:45] <andypugh> There is another on eBay with no back cover. $764 no bids
[10:16:35] <SpeedEvil> Yeah - I'd probably buy one no question at the original price - $58
[10:17:00] <SpeedEvil> But I can't justify full-size lathe money on a toy :)
[10:17:27] <andypugh> I don’t think that they are _exactly_ a toy, but te ergonomics are a bit wrong for a machine of that size.
[10:19:06] <Cromaglious]> Morning
[10:19:34] <SpeedEvil> Yes - it's not a toy.
[10:19:40] <SpeedEvil> It's a usable micro-lathe.
[10:20:19] <Cromaglious]> SpeedEvil: URL?
[10:20:52] <SpeedEvil> http://www.lathes.co.uk/manson/index.html http://www.ebay.com/itm/Metal-Lathe-mini-ManSon-restored-antique-micro-Monarch-10EE-jewelers-watch-/231414756807
[10:22:27] <Cromaglious]_> nice looking lathe
[10:23:24] <andypugh> Smaller than it looks
[10:23:32] <Cromaglious]_> Off to the train museum this morning.. Gonna try to get some pictures of their lathes nd mills and any CNC they may have
[10:24:44] <archivist> Cromaglious, which museum ?
[10:25:05] <Cromaglious]_> Orange Empire Rail in perris, CA
[10:25:24] <archivist> not been there
[10:26:38] <Cromaglious]_> nwither have I, except for 30 minutes last night unloading car
[11:41:24] * anonimasu sighs
[11:42:05] <anonimasu> anyone got a clue how to get a better resolution timer then the clasicladder 100ms base ones?
[11:49:16] <archivist> anonimasu,
http://www.linuxcnc.org/docs/html/man/man9/oneshot.9.html
[11:50:09] <archivist> or if measuring then
http://www.linuxcnc.org/docs/html/man/man9/timedelta.9.html
[12:01:03] <furrywolf> so, it poured last night, but it's bright sun today... should I head out and look for yard sales, or decide they all got rain out?
[12:01:56] <Tom_itx> it's sunny... go.
[12:04:24] <furrywolf> well, the sun seems to be being replaced by clouds again... I guess I should wait a bit and see which way it goes.
[12:18:50] <archivist> if you wait too long the bargains will have gone
[12:46:24] <zeeshan> :O
[12:52:23] * archivist wonders what aventtini is doing with his connection
[12:53:20] <pcw_home> lousy wireless connection?
[12:53:32] <furrywolf> zeeshan:
http://imgur.com/gs2bb3K I tweaked the coloring a bit last night
[12:53:34] <furrywolf> bbl, garage sales
[12:53:38] <SpeedEvil> And by this of course I mean the point in the universe where veterinary medicine, soap bubble making, and butt sex cross paths. See people who bought J-lube also bought Dawn manual pot & pan. Why? Because these are the soap bubble enthusiasts. Other people bought J-lube and the Extra Large SensaFirm Balled Cock 11 Inch Mocha. That's a whole nother segment of the market that comes together in this tiny speck of the Venn diagram with the soap
[12:53:38] <SpeedEvil> bubblers AND the veterinarians who need to do obstetrical work and artificial inseminations in large animals. We are all just 6 purchases away from each other in the Amazon universe!
[12:53:39] <SpeedEvil> err
[12:53:43] <SpeedEvil> nvm
[12:53:53] <zeeshan> looks like glue :)
[12:54:16] <SpeedEvil> Stupid mouse.
[12:54:25] <archivist> pcw_home, he quits rather than times out
[12:54:47] <pcw_home> funny irc client?
[12:55:54] <SpeedEvil> Some iphones do that IIRC
[12:56:06] <SpeedEvil> if you switch away from the browser, it kills the web-IRC session
[12:56:19] <zeeshan> yep
[12:56:24] <zeeshan> mine does that.
[12:56:26] <pcw_home> thats nice
[12:57:01] <furrywolf> I'm still waiting for the other one to upload so I can leave. stupid connection.
[12:57:40] <furrywolf> http://imgur.com/gvWqxce yay, finally. bbl.
[12:58:30] <SpeedEvil> That is a very bad skin condition
[13:00:08] <zeeshan> lol
[13:01:40] <archivist> lance it quick
[13:19:34] <Tecan> http://www.thingiverse.com/thing:112718 << threadless ball screw
[13:21:01] <SpeedEvil> yeah - it doesn't really work unless you also have a positioning thingy
[13:21:19] <SpeedEvil> resolver, or ...
[13:21:49] <SpeedEvil> I have wondered about using one of them.
[13:22:22] <Jymmm> SpeedEvil: is one bearing at an angle?
[13:22:32] <SpeedEvil> Combined axes and 'ballscrew' - made from a scaffolding pole
[13:22:38] <SpeedEvil> All the bearings are, I think
[13:23:18] <pcw_home> yes all bearings are canted limited force though
[13:23:59] <SpeedEvil> For very, very limited force, it might be sort-of-repeatable-enough
[13:24:04] <SpeedEvil> As long as you zero often
[13:24:07] <pcw_home> I used those many years ago for a wafer boat loader
[13:24:24] <pcw_home> (with potentiometer feedback)
[13:25:18] <Jymmm> http://thingiverse-production.s3.amazonaws.com/renders/d6/f2/40/95/cf/ball_screw_8mm_10_deg_preview_featured.jpg
[13:26:20] <Jymmm> Interesting, I wonder if you had tapered bearings would it still work.
[13:28:19] <SpeedEvil> yes
[13:28:26] <SpeedEvil> The key is that they're canted to the shaft
[13:28:32] <SpeedEvil> the profile doesn't matter
[13:28:36] <Jymmm> furrywolf: I haven't been following the conversation, what are these drawings you've been creating for?
[13:28:56] <Jymmm> SpeedEvil: Well, I meant the edge of the bearing full enganged witht he shaft
[13:30:11] <Jymmm> trapezoidal shaped bearings
[13:30:42] <SpeedEvil> http://www.aliexpress.com/item/2pc-taper-roller-bearing-32912-Auto-Wheel-Tapered-China-Bearing-60x85x17mm/1917169055.html
[13:31:58] <Jymmm> yeah, like that =)
[13:32:37] <SpeedEvil> Probably smaller in most cases
[13:34:32] <Jymmm> I wouldn't use that for machining purposes. but thought it be cool for a tree house with a pole that you can slide down. just turn the pole to raise a basket of supplies. And by supploes I mena all the junk fod a kid could want =)
[13:35:02] <Jymmm> without ropes and such.
[13:35:40] <SpeedEvil> http://www.ebay.co.uk/itm/Skating-Shoes-608ZZ-Bearing-Inline-Skate-Wheel-Clear-Gray-WL-/281623656009?pt=LH_DefaultDomain_3&hash=item419213f649 might work well
[13:36:07] <SpeedEvil> You'd need a _lot_ of turns though
[13:36:33] <Jymmm> It's a kid, LOTS of energy, keeps em busy and out of trouble =)
[13:36:34] <SpeedEvil> Implying a fast rotation, and potential hazard (hair, ...)
[13:37:43] <Jymmm> Eh, hair is overrated.
[13:49:43] <JT-Shop> I don't know why I can't use my square 5c holder and a fixture to sharpen bits on the surface grinder
[13:50:15] <JT-Shop> I've sharpened thousands of point cutters on a surface grinder using fixtures to hold them
[13:51:29] <SpeedEvil> You can
[13:51:56] <SpeedEvil> It's just you'll need several-many positions if you want to make a normal shaped drill
[13:53:38] <JT-Shop> well the square 5c holder takes care of two sides so you only need two angles on each side to do all 4 angles and 2 sides
[13:53:49] <Jymmm> I love the "metalness" of this machine...
https://www.youtube.com/watch?v=cX-vIgvB1-o
[13:54:01] <SpeedEvil> Drill surfaces are not flat
[13:54:38] <JT-Shop> http://www.tormek.com/en/jigs/dbs22/
[13:54:57] <JT-Shop> it would be the same as they are doing with this machine I think
[13:55:10] <SpeedEvil> Well,do that then
[13:55:28] <SpeedEvil> In principle there isn't anything wrong - if you can get it setup accurately and registered
[13:56:14] <JT-Shop> I don't see the problem with setting it up and getting the drill square to the square holder
[13:57:03] <SpeedEvil> Go for it then.
[13:57:23] <JT-Shop> I need to cook a sack of crawfish first
[17:26:09] <Deejay> gn8
[17:38:42] <LeelooMinai> Infally I can access GS2 from my Windows PC:
http://i.imgur.com/Qc9qAP9.png
[17:38:51] <LeelooMinai> Finally even:)
[17:39:09] <LeelooMinai> All I had to do was: GS2 VFD -> RS485 -> SERIAL -> FTDI IC -> USB -> linuxcnc PC -> FTDI driver -> /dev/ttyUSB0 -> ser2net -> TCP/IP -> my Windows PC -> HW VSP3 -> virtual COM4 -> qModMaster
[17:39:12] <LeelooMinai> :p
[18:46:34] <JT-Shop> mmm, the mud bugs were good
[18:46:45] <jdh> steamed?
[19:21:52] <furrywolf> not too much at yard sales today... just got a few random tools and a toaster.
[22:20:54] <LeelooMinai> Hmm, looked at source code of that gs2_vfd linuxcnc hal driver and there's that: if ((haldata->motor_RPM < 600) || (haldata->motor_RPM > 5000)) haldata->motor_RPM = 1800;
[22:21:07] <LeelooMinai> Kind of a pita...
[22:22:37] <LeelooMinai> If not that everything would work fine. Right now I can still make it work, but I have to pretend in g-code that s1200, for example, means s12000
[22:23:01] <LeelooMinai> Not sure why they have to clip it on 5000:/
[22:24:45] <LeelooMinai> Unless there's some way of scaling the value coming from g-codes (?)
[22:27:50] <LeelooMinai> Hmm, maybe I can divide motion.spindle-speed-out
[22:29:29] <LeelooMinai> Also, do people use some, say, polycarbonate "shields" around the spindles? I thought of making something that moves with the table and protected the area around the workpiece and spindle end. Because if some bit breakes at those speeds, I am afraid I will end up with a shrapnel in my body (?)
[22:29:55] <SpeedEvil> not a bad plan
[22:30:17] <LeelooMinai> Well, I never saw anything like this though - I wonder why
[22:30:19] <SpeedEvil> transparent fexible PVC is another option
[22:30:54] <SpeedEvil> Or just box the whole thing
[22:31:00] <LeelooMinai> I mean I saw big CNC that are enclosed with doors, etc., but never saw this on small hobby machines.
[22:31:33] <SpeedEvil> The risk is generally smaller
[22:31:53] <SpeedEvil> you're probably not going to lose an arm if a quarter inch bit breaks
[22:31:56] <LeelooMinai> Yes, long time ago I thought about big box, but now I think I would rather have smaller protection that goes over the table (hinged etc.) which would also prevent all the chips and stuff polluting the machine.
[22:32:19] <LeelooMinai> SpeedEvil: I don't know... at 24k rpm it would be like a sharp bullet...
[22:32:25] <SpeedEvil> Sure.
[22:32:33] <SpeedEvil> Now compute the actual tip-speed
[22:32:38] <SpeedEvil> it's not _that_ high
[22:32:55] <SpeedEvil> If it gets you in an eye - yes - you're screwed
[22:33:41] <LeelooMinai> Hmm, well, lets see
[22:33:57] <SpeedEvil> Eyewear is probably a good plan if you don't have
[22:34:45] <LeelooMinai> I can only fit max, say, 7mm in there, so that's 22mm diameter
[22:35:09] <SpeedEvil> circumference
[22:35:25] <LeelooMinai> Right
[22:35:46] <LeelooMinai> So 484m / minute
[22:36:00] <LeelooMinai> 8 m/s
[22:36:04] <LeelooMinai> Well, no, thx:)
[22:36:17] <SpeedEvil> Sure - it's quite fast for a sharp thing thrown at you.
[22:36:29] <SpeedEvil> But it is really slow compared to the slowest bullet.
[22:37:17] <LeelooMinai> 28 km/h - right, maybe would not kill me:)
[22:38:04] <SpeedEvil> Noting of course that faiing cutters can go lots faster thna the tip speed - if wound up and springing
[22:38:31] <LeelooMinai> In any case, I want to have some protection on there
[22:38:43] <LeelooMinai> I don't want the chips falling on rails, etc.
[22:41:09] <LeelooMinai> Hmm, I don't know if to patch this 5000 out for myself or what
[22:44:45] <LeelooMinai> E, no
[23:00:58] <LeelooMinai> ok, finally everything is agreed on the display (including gs2):
http://i.imgur.com/oP3jkvk.png
[23:02:40] <LeelooMinai> Though when I set it to 24000 it's like starting a helicopter - pretty scary
[23:24:58] <MacGalempsy> :]
[23:57:57] <furrywolf> I broke a 1/8 carbide endmill once that I never found. heard the snap from the mill and the thud against the opposite wall of the room simultaneously. why I always wear eye protection even with tiny bits...
