Back
[02:01:32] <Loetmichel> mornin#
[02:04:41] <Loetmichel> hrhr, co-worker has got me a present yesterday...
http://www.cyrom.org/palbum/main.php?g2_itemId=13817 :-)
[02:44:24] <DJ9DJ> moin
[03:43:37] <r00t4rd3d> they gave you a cup of tea?
[04:11:07] <Loetmichel> the cup
[04:11:15] <Loetmichel> note the size
[04:11:41] <Loetmichel> and the writing... "i am [forbidden] good"
[07:57:58] <archivist> retrofit cnc drill/mill fleabay 140898536372
[08:00:50] <TekniQue> I like the toolchanger
[08:02:49] <jdh> turret mill?
[08:03:51] <archivist> meddings was mainly a drilling machine company
[08:04:10] <archivist> never seen that thing till today
[08:16:54] <skunkworks> that reminds me of the k&t eb - only in-side-out
[08:23:29] <archivist> emco did something too
http://www.hellopro.co.uk/EMCO_Groupe_Emco_Maier-1620-noprofil-1001212-6328-0-1-1-fr-societe.html
[08:24:58] <skunkworks> ok - maybe a bit early to make connections like that..
http://www.practicalmachinist.com/vb/antique-machinery-history/k-t-milwaukeematic-mill-113966/#post424670
[08:25:26] <skunkworks> but you kinda see it
[08:25:45] <skunkworks> be cool to have one of those in the corner of the garage..
[08:26:56] <archivist> but that is a separate changer/caddy for the tools :)
[08:31:46] <skunkworks> I know I know
[08:32:13] <jthornton> http://imagebin.org/239944
[08:32:17] <skunkworks> there where a couple for sale on ebay a few years ago - but they looked rough
[08:32:29] <skunkworks> Yay!
[08:32:35] <skunkworks> What was the issue?
[08:32:37] <jthornton> no errors
[08:33:43] <jthornton> I did not have the experience necessary to code it
[08:34:09] <jthornton> but I have it now :)
[08:34:12] <skunkworks> A little help from your friends>
[08:34:14] <skunkworks> ?
[08:34:38] <jthornton> a lot of help from a new friend
[08:34:58] <jthornton> I never would have figured it out by guessing
[08:36:11] <skunkworks> cool
[08:37:33] <jthornton> check the user photo
http://linuxcnc.org/index.php/english/forum/30-cnc-machines/25963-noob-starting-out
[08:43:13] <archivist> a lot of the mechanical work is done there, just add a couple of lead screws and some motors and drives
[09:08:24] <skunkworks> everyone needs a car lift in the shop... right?
[09:08:25] <skunkworks> http://electronicsam.com/images/KandT/shop3.JPG
[09:09:50] <skunkworks> a few stitched together..
[09:10:10] <skunkworks> http://electronicsam.com/images/KandT/shop.JPG
[09:10:16] <archivist> how did you clear that much space!
[09:10:31] <skunkworks> Dad has been working very hard at that...
[09:11:00] <skunkworks> actually - that was where the 2nd k&t was going to go - so it was just a dirt pit. now it is a nice flat floor :)
[09:11:01] <archivist> what is outside in the rain
[09:11:41] <skunkworks> suprisingly nothing ;)
[09:12:35] <skunkworks> 4ish yards of concrete poured a couple of weeks ago
[09:14:11] <skunkworks> he actually has a trough to put the cables and hydraulics below floor level - to get rid of the speed bump. But he has to get cables and re-locate some pullys
[09:16:35] <skunkworks> if you look strait back between the car lift uprights - you can see the y axis assembly for the second k&t. We are slowly disassembling it.
[09:16:44] <skunkworks> on the stitched photo
[09:17:33] <skunkworks> (that area back there is about 4ft lower than the main floor)
[09:32:24] <alex_joni> skunkworks: looking good
[09:34:22] <jdh> how high is the ceiling?
[09:40:22] <skunkworks> 12ft
[09:43:16] <jdh> I could have used that this w/e.
[10:39:14] <L84Supper> is it really that hard to label active low signals with a line over the text or a / in front of it ?
[10:39:37] <L84Supper> or maybe a - in front
[10:42:17] <L84Supper> I've recently come across stepper where they don't seem to understand the difference between active HIGH or LOW in the data sheets, connection diagrams, connectors or on the silk screens
[10:42:37] <L84Supper> stepper/stepper driver
[10:43:35] <archivist> or just too lazy
[10:43:40] <L84Supper> I'm now surprised that they get Power and GND right
[10:44:06] <pcw_home> whats your problem man? you give it a pulse and it steps...
[10:44:34] <L84Supper> sure, any pulse will do
[10:45:20] <archivist> specially those induced from the vfd on the spindle
[10:49:30] <L84Supper> those are just helper pulses
[10:50:30] <L84Supper> or a secondary psychic/phantom servo loop
[11:35:55] <Loetmichel> re @ home
[11:40:16] <r00t4rd3d> http://www.indiegogo.com/robotdragonfly
[11:55:40] <L84Supper> look another terrorist tool!
[12:01:18] <IchGuckLive> hi all B)
[12:03:01] <alex4nder> r00t4rd3d: that looks awesome
[13:17:52] <IchGuckLive> by for today B)
[13:23:20] <andypugh> Result! I just found about £200 of high-density modelboard in the skip at work.
[13:24:02] <archivist> anything useful from a skip is a result
[13:24:49] <andypugh> Well, I have been trying to work out where to buy some from, because I have a pattern to make. So it's a real result.
[15:26:14] <Nick001-Shop> ssi - you around?
[15:36:46] <tjb1> JT-Shop: !
[15:52:47] <_ilias> hi all
[15:53:38] <andypugh> Hi
[15:53:53] <_ilias> I have a XYZ milling machine, but on the z axis there are 4 spindles independently activated by electrovalves
[15:54:06] <andypugh> OK, sounds like fun
[15:54:26] <andypugh> Do they all cut at the same time?
[15:54:29] <_ilias> therefor you can either use one or all four of them for many copies of the same pattern
[15:55:16] <p0st4L> http://www.aliexpress.com/store/product/3-axis-CNC-controller-kit-3pcs-NEMA23-425-oz-in-Dual-shaft-stepper-motor-256-microstep/704350_385804768.html
[15:55:18] <andypugh> OK, so it's for making several parts, not for using several tools?
[15:55:18] <p0st4L> any input?
[15:55:34] <_ilias> yes
[15:56:03] <andypugh> p0st4L: Those very big steppers can be a bit slow.
[15:56:24] <p0st4L> it's for a V90 Fireball CNC
[15:56:39] <p0st4L> https://probotix.com/FireBall_v90_cnc_router_kit/
[15:56:57] <andypugh> Depending on gearing they might not out-perform smaller ones. You need to compare the speed/torque curves a bit carefully.
[15:57:17] <_ilias> but when I want to mill a single pattern longer than the workspace of one spindle head, then this spindle head must be lifted and another must descend, or something more complex than that
[15:57:19] <tjb1> p0st4L: Noo!
[15:57:20] <tjb1> Bad.
[15:57:32] <p0st4L> ohai
[15:57:33] <_ilias> any ideas on how can i do this?
[15:57:45] <p0st4L> sgt sarcasm, right?
[15:57:49] <tjb1> Yes
[15:57:59] <p0st4L> well, help a nigga out.
[15:58:02] <p0st4L> wifey got me the cnc for xmas
[15:58:07] <p0st4L> i need a motor kit fori t
[15:58:12] <tjb1> Should of built your own :P
[15:58:16] <p0st4L> nah
[15:58:16] <_ilias> is this procedure called gang-milling?
[15:58:21] <p0st4L> i'm sure i will later in life
[15:58:28] <p0st4L> but i need a working intro frame to get used to everything
[15:58:39] <cradek> _ilias: you could use tool offsets for this purpose
[15:58:52] <andypugh> _ilias: In theory you could do it with a toolchange. The tool offsets are programmable in X and Y as well as Z. In theory you could machine with Tool1, then switch to Tool4 and it would pick up where it left off.
[15:58:53] <tjb1> p0st4L: You need 3 or 4 motors?
[15:58:56] <p0st4L> 3
[15:59:03] <p0st4L> nema23 sized
[15:59:05] <tjb1> I vote this
http://www.cncrouterparts.com/3-axis-electronics-kit-p-74.html
[15:59:33] <p0st4L> $520, you're killin me dawg
[16:00:09] <p0st4L> that is a nice lil board setup they have though
[16:00:10] <tjb1> You won't regret it
[16:00:13] <_ilias> sorry for that
[16:00:17] <tjb1> Thats a Gecko G540
[16:00:30] <andypugh> p0st4L: Same kit, easier:
http://www.ebay.com/itm/3Axis-4-lead-Nema-23-Stepper-Motor-76MM-270oz-3A-driver-4-2A-128Mric-CNC-KIT-CE-/261143782954?pt=LH_DefaultDomain_0&hash=item3ccd61e62a
[16:00:52] <andypugh> (Ah, well, smaller steppers now I look)
[16:01:13] <p0st4L> still nema23
[16:01:16] <_ilias> yes but when the code is produced by s/w like aspire or artcam, it could be difficult to edit this code for me
[16:01:19] <tjb1> p0st4L: The G540 is the cheapest of the gecko drives
[16:01:59] <andypugh> _ilias: You might have to cut the workpiece into 2 parts in the CAM program, then run that with different tools.
[16:02:09] <p0st4L> hmm
[16:02:12] <_ilias> ah yes
[16:02:23] <tjb1> Here are a bunch of different kits -
http://www.kelinginc.net/CNCNEMA23Package3x.html
[16:02:57] <AR_> so
[16:03:14] <p0st4L> since the bed is only 12" x 18", i don't need very powerful motors, do i? (torque wise)
[16:03:17] <AR_> 2x4 aluminum extrusions or steel tubing (filled with something?) for CNC frame
[16:03:18] <AR_> ?
[16:03:51] <tjb1> AR_: what size
[16:03:57] <andypugh> p0st4L: I would buy _exactly_ the motors that Fireball sugest.
[16:04:02] <AR_> tjb1, small
[16:04:16] <tjb1> AR_: What kind of CNC and what do you want to do on it?
[16:04:20] <AR_> probably 10x10 work area max
[16:04:27] <_ilias> andypugh: that's correct... thx.. so there is no out of the box solution, is it?
[16:04:29] <tjb1> 10 inch?
[16:04:31] <AR_> yes
[16:04:37] <tjb1> This a router?
[16:04:42] <_ilias> is there
[16:04:43] <AR_> maybe mill circuit boards
[16:04:43] <andypugh> Machine it from solid?
[16:04:43] <_ilias> ?
[16:04:46] <AR_> other little things
[16:05:03] <tjb1> You could do steel with no filling or 1530 inch extrusions
[16:05:04] <AR_> it would be nice to do light aluminum work but i dunno if that will happen
[16:05:11] <AR_> yeah
[16:05:13] <tjb1> even 1515 or 1010 would probably work
[16:05:19] <andypugh> I would look for a lump of aluminium tooling plate and bolt slides to that.
[16:05:27] <AR_> how stiff are the aluminum extrustions
[16:05:41] <andypugh> Not as stiff as steel.
[16:05:55] <tjb1> If you are willing to do math you can figure that out :)
[16:05:55] <AR_> but will they allow for decent accuracy in a small mill
[16:06:20] <tjb1> tslots.com gives the modulus of elasticity for all of their extrusions
[16:06:32] <tjb1> + and bunch of other information
[16:06:33] <AR_> is tslots cheaper than 8020
[16:06:56] <tjb1> In my case it was
[16:07:23] <tjb1> They gave me free shipping because I am a student but that wouldn't matter as much in your case.
[16:07:29] <andypugh> _ilias: Not id nothing is built into the CAM.
[16:07:37] <AR_> well, i am a student :P
[16:07:59] <tjb1> Well I was getting 10' long extrusions so shipping was almost $200 for me
[16:08:08] <andypugh> _ilias: There is certainly no G-code to automatically change tools as you get near the edge of the job.
[16:08:18] <AR_> lol oh
[16:08:54] <tjb1> Problem I had with 80/20 was them forcing me to deal with a distributor that kept no stock
[16:09:18] <tjb1> I wasn't ordering special extrusion either, I feel that 3030 and 1530 are probably the most common inch extrusions
[16:09:42] <AR_> ah
[16:09:48] <AR_> well, here is my issue
[16:09:57] <_ilias> ok thanks a lot
[16:10:04] <AR_> a few years back i was young and stupid and thought i could cheaply piece together a cnc mill
[16:10:16] <AR_> so i ordered a bunch of crap off ebay and put it together with wooden fram
[16:10:19] <AR_> frame*
[16:10:40] <_ilias> what cam s/w are you people using?
[16:10:41] <AR_> i have 2 small long linear rails with 1 carriage for each (pretty small)
[16:10:58] <tjb1> _ilias: I use sheetcam but I am plasma
[16:10:59] <AR_> i have a nice 16" hardened rod with 2 round bearings
[16:11:12] <AR_> 7/8 diameter
[16:11:17] <andypugh> I did much the same. Actually I bought an Indian knock-off of a cheap Chinese lathe, and CNC-ed that. You probably wasted less money than me.
[16:11:26] <AR_> and a small like 3/8 diameter rod with bearings
[16:11:39] <AR_> so i'm still hoping to piece together SOMETHING
[16:11:52] <AR_> but i really am tempted to just buy some nice things :P
[16:11:53] <andypugh> _ilias: I am too busy making machines to actually _use- them :-)
[16:12:04] <tjb1> _ilias: Many people on cnczone use vetric
[16:12:07] <_ilias> tjb1: happy with it?
[16:12:09] <_ilias> :)
[16:12:19] <tjb1> *Vectric
[16:12:28] <AR_> if i got a block of aluminum about 1.5" thick, with through holes for the bearings on the rods
[16:12:35] <AR_> that could make an X axis
[16:12:41] <andypugh> The only CAM I have is SheetCAM. It's OK for 2.5D stuff.
[16:12:41] <tjb1> It does its job, I haven't bought the full version yet
[16:12:51] <AR_> mount a Z on that (which i have a ballscrew and linear slide for)
[16:12:54] <_ilias> i use vectric two... but the drill bank in artcam is very tempting for my configuration
[16:13:15] <_ilias> *too
[16:14:01] <_ilias> although i haven't figured out out how to configure it yet
[16:14:26] <tjb1> AR_: Shouldn't be too hard to make one for routing PCBs
[16:14:38] <AR_> yeah
[16:14:38] <tjb1> But if you want to do aluminum it will need to be quite a bit more rigid
[16:14:44] <AR_> thats what i keep telling myself, tjb1
[16:14:51] <AR_> but i keep trying to build around what i have
[16:14:58] <AR_> and then i get deeper into trouble
[16:18:31] <andypugh> AR_: There are only so many corners you can cut. Some parts of an adequate CNC machine do cost some money.
[16:18:45] <AR_> yeah
[16:18:59] <AR_> for precision, you need precision parts
[16:19:19] <AR_> i often wonder how lucky they had to get when building the first precise machines
[16:20:50] <andypugh> Luckily a machine (especially a lathe) can male parts better than itself. What I don't understand is how you make machines _bigger_ than your machine.
[16:21:38] <AR_> haha
[16:25:54] <mrsun> andypugh,
http://i.ytimg.com/vi/2ivuCRavE8I/0.jpg heres one way to make bigger parts then the spindle of a lathe can handle =)
[16:26:00] <_ilias> my machine is similar to this
https://www.exfactory.com/Detail.aspx?recnum=RL-010501
[16:26:50] <andypugh> mrsun: I am not sure whether to be apalled or impressed.
[16:27:01] <mrsun> andypugh, =)
[16:27:03] <mrsun> spindle raiser =)
[16:27:38] <andypugh> _ilias: That's fairly hefty.
[16:28:15] <_ilias> yes it is
[16:28:22] <mrsun> and ofc with some tricks you can machine alot bigger stuff in a small lathe then it can handle =)
[16:28:35] <andypugh> Did the original control automatically choose the optimum spindle?
[16:28:43] <mrsun> say you want to make a spindle for a bigger lathe .. just lay it on the side and bore it in the lathe using the lathes spindle as boring head
[16:28:48] <_ilias> but changing the controller was cheaper than buying a new one
[16:29:06] <DJ9DJ> gn8
[16:30:17] <_ilias> the machine was configured for only one task, that was using all spindles
[16:30:54] <andypugh> I guess extending the Y slide is too difficult?
[16:31:14] <_ilias> after a while, the controller broke... changed it to linuxcnc and now want to make full use of it
[16:31:18] <andypugh> (considering that it is a problem that can be solved in software, after all)
[16:32:03] <_ilias> for movement on the Y axis the table is moving
[16:32:05] <andypugh> The spindles all move together, I assume, but you can optionally retract some?
[16:32:12] <_ilias> yes
[16:32:58] <andypugh> I think the only option is to persuade the CAM that it is two jobs, cut with two different tools.
[16:33:08] <_ilias> using electovalves the can be lowered and raised independently along the moving z-axis
[16:34:04] <toastyde1th> on fanuc that's done with either a g52 offset or defining as many work offsets as you have spindles
[16:34:15] <_ilias> I will try to solve my problem using your suggestion
[16:34:48] <toastyde1th> on the cam package you just say they're all at 0,0,0
[16:35:00] <toastyde1th> and that T1 works on offset 1, t2 on offset 2, etc
[16:35:04] <toastyde1th> that way the cam doesn't really care
[16:35:11] <toastyde1th> then on the machine you set your offsets up with a different y
[16:36:10] <toastyde1th> or just keep changing the local offset rather than work offset, but i don't know how emc does that
[16:36:17] <_ilias> so each job should have its own tool?
[16:36:33] <toastyde1th> maybe i misunderstood
[16:36:56] <toastyde1th> you have a number of spindles on y, right?
[16:37:00] <toastyde1th> and they all move together?
[16:37:08] <_ilias> yes
[16:37:13] <_ilias> along the X
[16:37:14] <toastyde1th> and you want to have them with different tools?
[16:37:31] <toastyde1th> can they move independently on y as well?
[16:37:36] <tjb1> I love it when pipes start leaking..
[16:37:36] <toastyde1th> or are they bolted onto the rail
[16:37:47] <_ilias> they are bolted
[16:37:50] <toastyde1th> cool
[16:38:00] <_ilias> they only move independently along z
[16:38:03] <toastyde1th> so each tool gets its own work offset
[16:38:07] <_ilias> yes
[16:38:09] <toastyde1th> done
[16:38:39] <_ilias> I see what you mean but my problem is
[16:39:23] <_ilias> that when the job extends beyond the workspace of one spindle... how can I automate the process of using a neighbouring spindle?
[16:40:04] <toastyde1th> you will have to do that.
[16:40:14] <toastyde1th> i.e. you will have to know that spindle 1 machines the right half
[16:40:19] <toastyde1th> and spindle 4 does the left half
[16:40:23] <_ilias> ok
[16:40:35] <toastyde1th> so you do all your cam work on t1, offset 1
[16:40:40] <toastyde1th> right side finished
[16:40:42] <_ilias> and split the job in 2 halves?
[16:40:46] <toastyde1th> switch to t4, offset 4
[16:40:49] <toastyde1th> and machine the left side
[16:40:50] <toastyde1th> yep
[16:41:33] <_ilias> yes I see your point ... than you guys a lot
[16:41:37] <toastyde1th> depending on the cam system, it should be that hard, just a wee bit time consuming
[16:41:41] <toastyde1th> *shouldn't
[16:41:58] <_ilias> in vectric aspire there is a job splitting option
[16:42:08] <toastyde1th> awesome
[16:42:34] <_ilias> I don't know though if in artcam the drill bank would be of any help
[16:43:38] <_ilias> anyway... this is beyond our scope
[16:46:28] <_ilias> is this procedure called gang milling?
[16:46:41] <toastyde1th> negative
[16:46:51] <toastyde1th> it doesn't have an explicit name that i'm aware of
[16:47:04] <_ilias> what is gang milling?
[16:47:06] <toastyde1th> gang milling would be using a bunch of cutters at once; it's usually reserved for horizontal milling machines
[16:47:18] <toastyde1th> where you have a long arbor, and the milling cutters look like big circular saws
[16:47:31] <toastyde1th> some people refer to multiple spindles, like on your router, as gang milling
[16:47:42] <toastyde1th> where you space out the milling heads and drive all of them at the same time
[16:48:01] <toastyde1th> that way for each machine cycle, each spindle makes a part
[16:48:22] <_ilias> I get it... this is the way the machine was operating
[16:48:45] <_ilias> all parts were identical though
[16:48:46] <toastyde1th> it's almost always how multispindle machines operate
[16:48:57] <toastyde1th> since most cnc machines are used to make many parts
[16:49:04] <toastyde1th> it makes sense to have many spindles make the same part
[16:51:03] <_ilias> that's clear to me now
[16:52:05] <_ilias> so we are talking about g55 g56 etc.?
[16:52:25] <toastyde1th> yes
[16:52:28] <_ilias> ok
[16:52:36] <andypugh> It might even make sense to remove some spindles, if you don't need that capability. However being able to do toolchanges really quickly is one nice feature you do have.
[16:52:36] <toastyde1th> alternatively
[16:52:45] <toastyde1th> emc lets you define specific cycles
[16:52:48] <toastyde1th> o-codes
[16:53:16] <toastyde1th> you might want to read up on that and define either custom m-codes or an o-code to modify where 0 is
[16:53:20] <toastyde1th> and not use up work offsets
[16:53:25] <toastyde1th> if you NEVER move your spindles
[16:53:30] <andypugh> toastyde1th: I think in this case having _tool_ offsets for each spindle probably makes more sense than using different work offsets.
[16:53:32] <toastyde1th> if you do move them from time to time, the work offsets are a better way to go
[16:53:43] <_ilias> andypugh: I thought of that
[16:53:48] <skunkworks> andypugh: agreee
[16:53:51] <skunkworks> I agree
[16:54:01] <toastyde1th> andypugh, sorry, i'm not an emc guy, most controls don't have x-y modification as part of their tool offset
[16:54:12] <skunkworks> you can have xyzabcuvw offsets for each tool
[16:54:14] <toastyde1th> just length and radius
[16:54:18] <_ilias> i am not that confident on using tool -offsets
[16:54:49] <andypugh> You can have frontangle on your mill too. I am not sure what you would do with it :-)
[16:55:03] <toastyde1th> but if emc does have that, it makes a lot more sense and is more convenient than using work offsets
[16:55:18] <toastyde1th> easier to have multiple fixtures on the table
[16:55:54] <_ilias> i looking at tool offsets now
[16:55:59] <_ilias> i am
[16:56:06] <_ilias> just a minute
[16:59:33] <_ilias> so... if I correctly set the tool table it could be as easy as issuing a Tn M6 command?
[16:59:37] <marl_scot> hi guys, i curently have a reprap printer, and would like to build a basic cnc machine to run beside it, any pointers to designs/howtos for really basic and cheep to build units anyone would recomend, im not looking for anything too big, initally maybe 18" x 18", mostly would beusing it for cutting tempaltes/brackets from wood and plastic sheet
[17:00:25] <andypugh> _ilias: yes.
[17:00:48] <_ilias> ok... now I totally get it
[17:01:11] <skunkworks> marl_scot: make something out of mdf?
[17:01:35] <marl_scot> skunkworks, thats whay i was thinking of
[17:02:11] <marl_scot> with screw drive instead of belt
[17:02:53] <marl_scot> but for my first one, would be more confident if i had some kind of design that has allready been built to start from
[17:04:05] <andypugh> I don't know of any that fit your description. Which isn't to say they don't exist.
[17:04:55] <_ilias> if the tools are already mounted and i want to continue the job with the next spindle, will i need hal_manualtoolchange?
[17:05:13] <skunkworks> there was a instructables iirc
[17:05:27] <marl_scot> what parts of the description would you say im failing on? i dont mind differant size/approch that was just my inital thoughts
[17:05:40] <_ilias> I guess i'll figure this out
[17:05:47] <skunkworks> marl_scot: like
http://www.instructables.com/id/How-to-Make-a-Three-Axis-CNC-Machine-Cheaply-and-/
[17:06:00] <_ilias> anyway, thank you people for your help
[17:06:49] <marl_scot> thanks skunkworks will study that one and see how i get on
[17:06:50] <_ilias> bye
[17:07:10] <andypugh> _ilias: Not manual toolchange, no. You probably need to sey up an auto toolchange to send down the right spindle.
[17:07:53] <marl_scot> could i ask peoples advice on torque for steppers? would 40Ncm be enough for plastic, would it maybe even do wood?
[17:08:17] <marl_scot> (just i can get hold of a few cheep 40Ncm steppers at the moment!)
[17:08:39] <andypugh> It depends on leadscrew pitch too.
[17:08:59] <andypugh> 40Ncm = 0.4Nm?
[17:09:55] <marl_scot> yup
[17:10:17] <andypugh> Sounds a bit low, but then you are talking about plastic.
[17:10:53] <andypugh> Were you going to drive direct, or through belts?
[17:11:56] <marl_scot> direct i think, as it requires less components
[17:12:32] <Jymmm> **************************************
[17:12:32] <Jymmm> * 48 HOURS TILL THE END OF THE WORLD *
[17:12:32] <Jymmm> * Thank you and have a nice day *
[17:12:33] <Jymmm> **************************************
[17:27:50] <Valen> sup peeps
[17:29:02] <Loetmichel> Jymmm: pessimist!
[17:29:35] <Loetmichel> just because the mayans handt more room on this calendar you think the world is coming to an end?
[17:29:37] <Jymmm> Loetmichel: Opposite actualy, thus the "have a nice day"
[17:29:43] <Loetmichel> hrhr
[17:32:14] <Valen> sucks for us
[17:32:25] <Valen> we are going to have the end of the world like 14 hours before you guys
[17:33:17] <p0st4L> lol
[17:33:26] <p0st4L> people sure are starting to act like it's the end of the world
[17:33:29] <p0st4L> http://www.ktvu.com/news/news/crime-law/highway-85-shut-down-after-gunfire-hits-government/nTbN5/
[17:33:35] <p0st4L> this is near me, in the bay area, ca
[17:33:41] <p0st4L> sniper shooting at a car on the fwy
[17:35:42] <Valen> i like the way that story is written, like the car has a mind of its own
[17:35:50] <andypugh> Valen: I think I figured out lube :-)
[17:35:56] <Valen> orly
[17:36:22] <andypugh> Let me find you a picture of my device.
[17:46:55] <skunkworks> I like rube goldburg devices...
[17:47:24] <Valen> marl_scot: torque also depends on how much reduction you run
[17:47:40] <Valen> more reduction = more torque, but less acceleration and top speed
[17:50:50] <marl_scot> yea, but speed isnt that big an issue for me, thinking is, if i'm not wanting a really fast machine, then i can use a samller drill motor and therefore build a lighter Z axis etc.
[17:51:31] <andypugh> Valen:
http://imagebin.org/240011
[17:51:56] <andypugh> It relies on there being a gap in the inner ball races.
[17:52:24] <andypugh> (The bearing has seals not shown, I will remove the rightmost one)
[17:53:54] <tjb1> andypugh: what is that
[17:54:31] <andypugh> It's a rotating ballnut mount
[17:59:13] <andypugh> tjb1: Later development of
https://picasaweb.google.com/lh/photo/EqgIdXH-kCH8I90oYKfm2dMTjNZETYmyPJy0liipFm0?feat=directlink
[18:00:09] <Valen> heh neat
[18:00:16] <toastyde1th> are you making a static screw machine
[18:00:21] <andypugh> Yes
[18:00:24] <toastyde1th> nice
[18:01:13] <andypugh> Unfortunately the bearings I need don't exist.
[18:01:38] <andypugh> So I am using that one, that does exist, but which I can't fit the ballnut into.
[18:01:41] <toastyde1th> AIR BEARINGS!
[18:02:03] <toastyde1th> :D
[18:02:42] <andypugh> It's an idea. But perhaps not a good one.
[18:03:55] <toastyde1th> hahah
[18:04:18] <Valen> http://scitechdaily.com/liquid-metal-used-in-ultra-stretchable-conducting-wires/
[18:04:36] <Valen> i figure it should also knock wire fatigue on the head
[18:05:30] <toastydeath> the only reason i suggested air bearings is that they're relatively easy to build
[18:05:37] <toastydeath> if you really get shit out of luck on finding something to fit
[18:05:58] <Valen> air for the rotating nut bearings?
[18:06:43] <toastydeath> no no, i meant if you can't find bearings with the contact angle or ID/OD you want
[18:09:10] <Valen> I've been thinking about air for the linear rail axies
[18:09:13] <Valen> those are expensive
[18:09:37] <Valen> but making them would probably be hard
[18:09:47] <Valen> Also air for the spindle would be nice
[18:10:23] <toastydeath> not as hard as you'd think
[18:10:27] <toastydeath> making the rail straight is the hardest part
[18:10:44] <Valen> that'd be the issue
[18:10:49] <andypugh> I have designed parts with integral air bearings. As you say, not difficult in principle.
[18:10:54] <Valen> it'd need to be very consistent in thickness
[18:11:03] <toastydeath> flat bearings can be lapped, then expoxied into place
[18:11:03] <andypugh> Just a gap, with an air supply, in fact.
[18:11:13] * toastydeath used to work at new way, which makes nothing but air bearings
[18:11:49] <toastydeath> the trick to aligning bearings for linear ways is vacuum
[18:12:12] <toastydeath> instead of hooking em up to a compressor
[18:12:25] <toastydeath> shims and tape, etc
[18:12:53] <toastydeath> or, if you know how much your epoxy material shrinks
[18:13:01] <toastydeath> per inch
[18:13:22] <JT-Shop> tjb1: ?
[18:13:27] <toastydeath> all you do is vacuum it straight to the rail and when the epoxy cures, you release the vacuum and it'll be at that height
[18:13:30] <tjb1> JT-Shop: Where are those pictures
[18:13:45] <JT-Shop> been making parts...
[18:13:53] <toastydeath> a third method is just to not care about gap at all, because the pressure will create the right gap in most circumstances
[18:16:27] <JT-Shop> well actually for the last 3 hrs I've been drinking and shooting with the other children
[18:17:12] <Valen> ooh did anybody say "hold my beer and watch this?"
[18:17:13] <JT-Shop> http://imagebin.org/240012
[18:17:46] <Valen> looks industrial
[18:19:16] <JT-Shop> we were quite reserved today it was Charlie's birthday... may he rest in peace
[18:19:57] <JT-Shop> removed the top delrin slide and replaced it with a roller... works nice and smooth
[18:20:05] <JT-Shop> well two rollers
[18:22:16] <Valen> toastydeath: you guys used graphite to get the air in, not really needed you think?
[18:23:03] <toastydeath> graphite was the surface
[18:23:13] <toastydeath> and we still made a few oriface and groove bearings, etc
[18:23:21] <toastydeath> brb though, food
[18:26:12] <tjb1> Very nice JT-Shop
[18:26:28] <JT-Shop> thanks]
[18:28:15] * JT-Shop is the pizza chef so time to cook
[18:29:18] <JT-Shop> http://www.youtube.com/watch?v=Myabbp4G5H0
[18:29:38] <kwallace> From scratch?
[18:29:45] <JT-Shop> yea
[18:30:05] <JT-Shop> the dough is ready now...
[18:31:05] <andypugh> Valen: I see a huge problem with those stretchy wires.
[18:31:29] <andypugh> We are about to run out of Indium. even without a new use being found.
[18:49:04] <kwallace> andypugh: Thanks a lot. I did a search on your basque reference from a couple of days ago but left my speakers on. :-0
[18:52:47] <andypugh> Squeaky?
[18:55:09] <kwallace> The link had music, Boom Bada Bomp Bada Bomp. Not the best in the office. I haven't heard Techno for a long while.
[19:02:17] <toastydeath> Valen, re: graphite, graphite is porous so using that allowed us to not drill holes for oriface bearings
[19:02:38] <toastydeath> building graphite bearings takes a lot of r&d and a very reliable source of graphite
[19:02:52] <toastydeath> otherwise every bearing is way different
[19:03:16] <toastydeath> whereas oriface bearings are harder to make en masse and perform slightly less well, but are way easy to make consistent and with minimal tooling
[19:05:58] <Valen> got any links for where to start?
[19:06:04] <toastydeath> hm.
[19:06:50] <toastydeath> I don't know of any websites, but if you can find a torrent version of this book
[19:06:59] <toastydeath> there is a chapter on fluid bearings, both air and hydraulic
[19:07:00] <toastydeath> http://www.amazon.com/Precision-Machine-Design-Alexander/dp/0872634922
[19:07:36] <toastydeath> and googling around will more than give you enough information to start
[19:10:44] <toastydeath> and if you do something that has preload to it, the design of the bearing is ultra-tolerant because the gap is determined by air pressure , not really the inlet restriction design
[19:11:02] <toastydeath> you can do essentially anything as long as it's preloaded and it will turn into an air bearing
[19:11:27] <Valen> doing it with 3 axies of stability could be an interesting challenge
[19:11:39] <Valen> What would you suggest using as a rail for a linear axis?
[19:12:01] <toastydeath> for an air bearing?
[19:12:06] <Valen> yeah
[19:12:14] <toastydeath> anything smooth
[19:12:20] <toastydeath> that you are comfortable machining
[19:12:27] <toastydeath> you can flycut aluminum
[19:12:41] <Valen> It'd be good for high speed lol
[19:13:06] <toastydeath> air bearings are nice in that there's little concern about surface contact
[19:13:11] <toastydeath> so it doesn't matter TOO much what you make it out of
[19:13:28] <toastydeath> iron can be scraped, obviously
[19:13:32] <toastydeath> and hard steel can be lapped
[19:14:00] <toastydeath> and a hobbyist can buy it hard, load a good insert into a flycutter
[19:14:02] <toastydeath> and get it close
[19:14:13] <toastydeath> then hand lap it using the same process as scraping
[19:14:42] <toastydeath> some people like to use granite for air bearings because it laps REALLY nicely and is very stable
[19:15:11] <Valen> what sort of gap between the bearing and the slide would we be looking at?
[19:15:39] <toastydeath> if you use a design that's preloading (has opposed bearings in every direction) you don't need to worry about the gap
[19:15:50] <toastydeath> the bearings should contact the surface
[19:16:13] <toastydeath> the air will deform the carriage enough to work just fine
[19:16:32] <toastydeath> if you want to set the gap yourself, .0002-.0005"
[19:16:45] <toastydeath> the bigger the gap, the higher the flow the bearing needs to be
[19:16:54] <toastydeath> and the bigger the gap, the less accurate the axis will be
[19:17:00] <toastydeath> and less rigid
[19:17:19] <Valen> so preload across the bearing not within the bearing?
[19:17:25] <toastydeath> ya
[19:17:40] <toastydeath> air bearings are preloaded much in the same way a C clamp works
[19:17:49] <Valen> how do you think it'd compare for a bunch of amatures vs off the shelf linear rails?
[19:17:50] <toastydeath> one is fixed, the other one has a screw pushing it into the surface
[19:18:12] <toastydeath> depends; i know for a fact amatures can scrape and recondition machines if they try
[19:18:31] <toastydeath> it's going to be a LOT of work compared to linear rails.
[19:18:43] <toastydeath> you'll have to learn to scrape and lap
[19:18:55] <Valen> see that takes the edge off lol
[19:19:07] <toastydeath> and enough machine millwrighting knowledge, which is also available
[19:19:25] <toastydeath> it's a trade off; a rail is a one time cost, and is easy
[19:19:51] <toastydeath> you have to buy the equipment to make air bearing rails, but once you do it once, you can easily do it again for fairly cheap
[19:20:09] <toastydeath> equipment as in the stuff to measure and align a raw piece of stock into something worthy of guiding a machine
[19:20:48] <toastydeath> i can't speak to THAT personally, because i've never had to do it; but on the practicalmachinist there's a huge number of folks who have at least learned scraping
[19:20:57] <toastydeath> *practicalmachinist forum
[19:22:16] <Tom_itx> it's a painfully tedious job
[19:22:17] <andypugh> It seems to me that supported linear rail with woods-metal bearings cast round them, inflated with air, might just work a treat.
[19:22:50] <toastydeath> andypugh, if woods metal is porous enough, it should absolutely work
[19:22:54] <Tom_itx> our cmm had air bearings on it
[19:23:02] <toastydeath> and you can use stupid shit like nail polish to seal off the sides
[19:23:30] <andypugh> It's not porous at all. But it melts at a lower temperature than open-cell foam.
[19:23:53] <toastydeath> oh
[19:24:30] <andypugh> Not sure if that helps. And I have enough projects :-)
[19:24:33] <toastydeath> one of the things i am antsy to get out of college and move on with life for is to build air rails myself
[19:24:41] <toastydeath> to document it and show other people how
[19:24:58] <toastydeath> i know a lot about them from the professional side but not enough from the hand-work side
[19:25:10] <andypugh> I have scraped a few machine tools. My dad's lathe, the shaper, and my lathe.
[19:26:07] <andypugh> The guys who built my lathe scraped the front and back slides of the carriage quite well. Each was a pretty good fit to its rail. It is just a pity that they didn't both fit at the same time.....
[19:26:25] <toastydeath> aahahaha
[19:27:06] <toastydeath> there's another project i really want to take on aside from home built air rails
[19:27:08] <toastydeath> and that's capstan drive
[19:27:20] <toastydeath> for a homemade cnc jig grinder/boring machine
[19:27:34] <toastydeath> should be more accurate than any ballscrew
[19:28:13] <Valen> if you want accuracy just put a glass scale on the linear axis ;->
[19:28:36] <toastydeath> capstans are usually more accurate than the resolution of common glass scales
[19:29:03] <Valen> .001mm is off the shelf
[19:29:04] <toastydeath> very slow and can't handle cutting forces outside finish boring/turning/grinding
[19:29:44] <andypugh> One place I have thought that an air bearing would be useful and easy is a rotary table. Just hold it together with a spring washer, then feed air into the housing to lift the table off its seat. Turm off air to clamp.
[19:29:54] <toastydeath> the other issue is that the actual motion system and the scale aren't exactly coupled; you still get errors in following because the drive isn't accurate
[19:30:17] <toastydeath> and a well-made capstan is more accurate than .001mm
[19:30:26] <Valen> the "easy" way around that is to use 2 scales
[19:30:30] <toastydeath> in a temp controlled environment
[19:30:43] <Valen> one on the linear axis, the other on the motor for feedback
[19:30:50] <andypugh> The only capstan I have seen was on HMS Victory, and that hardly seemed accurate at all.
[19:30:59] <andypugh> (Maybe to the centi-fathom)
[19:31:04] <toastydeath> that doesn't help the problem of the drive not being accurate
[19:31:09] <toastydeath> you just see the errors after they happen
[19:31:13] <toastydeath> very very clearly
[19:31:15] <toastydeath> :D
[19:31:33] <Valen> I'm having difficulty finding any information on a capstain drive as it pertains to a CnC anything
[19:31:36] <Valen> its all tape decks
[19:31:59] <Valen> you use the linear scale for position feedback and the rotary scale for velocity, they can both feed into the one PID
[19:32:02] <toastydeath> it's a friction drive; it uses a lapped pin pinched against a lapped rail
[19:32:33] <toastydeath> Valen, none of that fixes the problem because the error has already happened by the time you are able to measure it
[19:33:07] <Valen> so how does your capstain drive magically have no error?
[19:33:09] <toastydeath> it's one of those things that doesn't matter for 98% of machines, but if you want to build a jig grinder it's an issue
[19:33:24] <toastydeath> it has error, but it's very constant
[19:33:55] <toastydeath> whereas a screw has a periodic error per turn in addition to a pitch error
[19:34:12] <Valen> with a linear scale you could calibrate those out
[19:34:36] <toastydeath> in theory, you can calibrate any error anywhere out
[19:34:37] <Valen> assuming your rotary axis has higher resolution than the linear scale interpolate between linear counts
[19:35:06] <toastydeath> in practice, that doesn't work and is why high accuracy machines still rely on primary accuracy rather than corrected accuracy
[19:35:40] <toastydeath> the more accurate your machine is without any sort of correction, the better any correction you apply after the fact will be
[19:35:50] <Valen> thats a given ;->
[19:36:29] <toastydeath> i've seen a few high accuracy hobby projects over the years, so i am hoping that it'll work
[19:36:36] <toastydeath> if not, i'll have a really impressive paperweight
[19:38:34] <Valen> hrh
[19:38:36] <Valen> heh
[19:39:06] <Valen> well I'd better get going
[19:39:11] <toastydeath> bai2u
[19:59:51] <tjb1> Ello.
[20:06:20] <toastydeath> oi
[20:44:00] <skunkworks> interesting...
https://www.youtube.com/watch?v=kh64OMb7DZg
[22:00:19] <tjb1> Wake up!
[22:00:27] <tjb1> Only a little over a day left to live
[22:18:59] <tjb1> Already at 16gb on my data plan :/
[22:21:34] <jdh> got unlimited?
[22:21:44] <tjb1> Yep
[22:21:51] <tjb1> Only 11 days into the cycle
[22:21:52] <jdh> do they throttle you?
[22:22:04] <tjb1> Downloading at 1.5 MBps right now
[22:25:31] <toxx> Hi folks, can anybody recommend me a reliable stepper driver for a machine like this:
https://www.wabeco-remscheid.de/index.php?cat=c12020_F1210.html
[22:25:41] <tjb1> Gecko.
[22:26:20] <jdh> depends on where you are and what you want
[22:26:27] <toxx> isn't gecko for servo drives?
[22:26:30] <tjb1> No
[22:26:39] <tjb1> They offer both
[22:26:54] <jdh> gecko 203 would be nice, but somewhat pricey.
[22:26:55] <archivist> I use this sort of chinese
http://www.arceurotrade.co.uk/Catalogue/Stepper-Motors
[22:27:25] <jdh> me too
[22:27:32] <tjb1> toxx: The G540 ends up being the cheapest out of Geckos and offers 4axis control
[22:27:44] <jdh> 540 isn't a good choice IMO
[22:27:56] <tjb1> Gets the job done :)
[22:28:26] <jdh> only 3.5amp
[22:28:28] <toxx> i ask for fablab munich / germany, we want good quality for this machine
[22:28:46] <jdh> then I'd go for a servo :)
[22:29:27] <toxx> well, the machine comes with stepper motors
[22:29:41] <jdh> but no drivers?
[22:30:05] <jdh> that's a BF20 or clone mill?
[22:30:51] <jdh> with a big Made in Germany sticker?
[22:31:06] <toxx> if you order it with drivers, you stick to there microcontroller/software solution, which is quiet pricey
[22:31:33] <jdh> looks like better version of a bf20
[22:31:36] <toxx> not realy a clone
[22:31:46] <tjb1> Looks like a G0704
[22:31:49] <tjb1> :P
[22:32:00] <jdh> size/design, but not junky chinese
[22:32:17] <jdh> a g0704 is a bf20-l or so.
[22:32:31] <tjb1> toxx: What are the stepper specs?
[22:33:14] <jdh> looks like 6amp
[22:33:15] <toxx> I don't know, maybe 3A
[22:34:02] <tjb1> jdh: looks?
[22:34:11] <jdh> on the web site.
[22:34:12] <tjb1> I didnt see any steppers in that link
[22:34:21] <jdh> under cnc upgrade kits anyway
[22:34:23] <toxx> archivist: what are your experiences?
[22:34:47] <archivist> they just work
[22:35:35] <tjb1> They are bipolar 6amps in the upgrade kits
[22:36:32] <tjb1> G201X, G203V, G120X and G213V will work from Gecko
[22:36:40] <tjb1> *G210X
[22:37:33] <jdh> or a chinese/leadshine
[22:37:33] <toxx> so geckodrive is state of the art?
[22:38:13] <jdh> I'd talk to <whatshisname> at rammstein
[22:38:54] <tjb1> They are probably mid level
[22:39:05] <tjb1> I don't know of what is higher, but there is a lot below them
[22:39:36] <archivist> there are limits to the micro stepping on some drivers limiting your speed
[22:39:47] <jdh> gecko is fixed 10x
[22:40:05] <archivist> yup why!
[22:40:09] <jdh> "because"
[22:40:18] <toxx> what do you mean with chinese/leadshine
[22:40:34] <archivist> those I pointed at
[22:40:43] <jdh> there are lots of chinese drivers. They are either made by leadshine, or clones of them.
[22:40:54] <toxx> i see
[22:41:03] <jdh> are you in .de?
[22:41:26] <toxx> yes
[22:42:06] <jdh> I'd talk to IchGuckLive when he is around. Might be more useful in finding things easily available there.
[22:42:45] <toxx> i noticed him
[22:43:33] <tjb1> usually pops in about 12 hours from now
[22:44:16] <toxx> ok, thanks for your infos
[22:44:18] <archivist> there are two other .de users who will join in the morning
[22:44:45] <jdh> it's pretty much morning there now isn't it? :)
[22:45:26] <archivist> 4am in uk 5 .de I assume
[22:45:34] <toxx> well, not much time left on this earth, i guess :)
[22:45:58] <jdh> perhaps I shouldn't bother taking vacation on friday.
[22:46:19] <archivist> ask Loetmichel and DJ9DJ
[22:46:40] <jdh> I might have been thinking about Loetmichel actually.
[22:46:42] <toxx> ok
[22:51:10] <jdh> http://itgetsworseforsysadmins.tumblr.com/image/27570541204
[23:01:34] <toxx> btw, does this work with the current linuxcnc version, or is it in developing state:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?ProbeKins
[23:04:17] <jdh> that entry hasn't been updated since Jan.
[23:05:25] <toxx> seems to be in the emc2-dev branch
[23:50:18] <tjb1> I'll pay someone to clean and sort the files on my computer and 3 hard drives
[23:50:20] <tjb1> Any takers?
[23:50:54] <toastydeath> hahah
[23:52:21] <tjb1> Guess you are a taker
[23:53:45] <toastydeath> NEGATIVE BLUE LEADER
[23:54:06] <tjb1> I am Red leader
