Back
[00:31:00] <Contract_Pilot> Another busy, busy, day! Did get a DM8010 drive in today to fix got it done and tested and packaged for return.
[00:40:43] <Contract_Pilot> Going to have to kill the ebay ad for repairs when i start my lcnc conversion and move away from mach.
[00:48:37] <Contract_Pilot> Then will have to set up a test axis.
[02:21:52] <Deejay> moin
[02:41:42] <Contract_Pilot> mon DJ
[02:41:56] <Contract_Pilot> well evning dj
[02:42:06] <Deejay> good morning (9 am here)
[02:42:08] <Contract_Pilot> been busy
[02:42:34] <Contract_Pilot> More terrorists that use ak-47 = more biz
[02:48:55] <Contract_Pilot> 500 magazines in the last few days
[02:52:23] <Praesmeodymium> yeah thats a lot of work
[03:23:32] <gonzo___> they are talking about banning 'certain' semi auto rifles across europe
[03:24:37] <gonzo___> though in the UK they went years ago. Only smallbore SA left. And they will probably take the opertunity to ban them too
[03:27:31] <archivist> shotguns still too
[03:30:05] <gonzo___> there are noises coming from the police about section 1 shotguns
[03:30:58] <gonzo___> adding little extra requiorements (that are not actually law) to just make it too inconvenient to have
[04:34:08] <Jymmm> (NSFW) Da Booty...
http://i.imgur.com/ZmdMlcd.jpg ( lair82, Tom_itx: your morning wakeup call as requested )
[04:58:06] <ReadError> so.. question
[04:58:25] <ReadError> trying to use some loctite retaining compound that requires a anaerobic cure
[04:58:57] <ReadError> is there any way to emulate this in a jar or something?
[04:59:19] <ReadError> I know it would happen naturally if I used it properly
[05:48:42] <anomynous_> https://www.youtube.com/watch?v=Y2xv3XwOVaQ
[05:49:56] <SpeedEvil> ReadError: remove the obics
[05:50:10] <SpeedEvil> flush with argon, for example.
[06:16:18] <hsm> Hello guys
[06:17:37] <hsm> what is the best way to control feed overdrive, rapid feet overdrive and spindle speed overdrive?
[06:18:19] <jthornton> I seldom use them
[06:19:07] <hsm> Is it a good plan to control them with 3 MPG handwheels? (TTL) over 7i77 encoder input?
[06:19:29] <jthornton> just depends on your needs
[06:19:31] <_methods> most real cnc's have 3 knobs to control those
[06:19:57] <hsm> yes but i don´t want to control it in steps
[06:20:07] <jthornton> my "real" cnc's only one had a feed override knob
[06:20:21] <jthornton> just ask the real question
[06:20:41] <_methods> any recent cnc has spindle, feed and rapid override
[06:20:56] <hsm> yes this is what i want
[06:20:59] <jthornton> lol that lets me out then
[06:21:15] <XXCoder> yeah 3 knobs is good for those
[06:21:31] <hsm> why knobs? knobs = MPG?
[06:21:39] <XXCoder> spindle, g0 % override, and feed rate %
[06:21:45] <_methods> use a pot
[06:21:46] <hsm> yes
[06:21:48] <jthornton> no they could be encoders
[06:21:51] <_methods> or whatever the hell you want
[06:21:54] <XXCoder> yeah can be anything really
[06:21:58] <_methods> sliders
[06:22:02] <_methods> buttons
[06:22:09] <_methods> a sting
[06:22:10] <_methods> string
[06:22:12] <_methods> lol
[06:22:16] <XXCoder> haas has 5 buttons, 0, 25, 50 75 and 100%
[06:22:24] <hsm> okay but i had read that potentiometer are bad for this
[06:22:29] <XXCoder> haas sucks
[06:22:32] <hsm> yes
[06:23:11] <XXCoder> best ones has 0% to 100% in 5% or less incerments
[06:23:19] <_methods> most people use incremental knobs for a reason
[06:23:34] <_methods> so they can calculate their new feed/spindle speed after adjustment
[06:23:43] <hsm> okay, I buy 3 MPGs
[06:23:48] <_methods> a nice defined 10,20,30.....%
[06:23:49] <XXCoder> honestly 5% is fine actually
[06:23:57] <XXCoder> do anyone need any finer resolution?
[06:24:00] <_methods> is easier to calculate than some random pot setting
[06:24:17] <lair82> Jymmm, where are you finding this shit?
[06:24:27] <XXCoder> google is your friend lol
[06:24:59] <hsm> if you have resonance vibrations you can fix it with fine steps under 5%
[06:25:03] <ReadError> SpeedEvil yea I was thinking a jar + gas to displace oxygen
[06:25:14] <lair82> True, but I looked at that imgur site, and don't find any nice shit like that!!!!!!!!!!!!!!!
[06:25:29] <ReadError> but dont have any good gas bottles here to use
[06:26:18] <hsm> I will buy 5 of these
http://www.ebay.de/itm/Hand-wheel-Pulse-Encoder-100PPR-CNC-Mill-Router-Manual-Control-For-CNC-System-20-/201087639110?hash=item2ed1c1de46:g:X34AAOxyi-ZTZo2m
[06:26:52] <anomynous_> 5?
[06:27:03] <XXCoder> hsm: dunno dont think that is best choice
[06:27:17] <XXCoder> can easily bump and turn it from 100% spindle to 0% while its running
[06:27:24] <hsm> jog z, jog x, feed override, rapid override, spindle override
[06:27:33] <anomynous_> +1 XXCoder
[06:27:56] <XXCoder> hsm: best is single one for both jogs with X or Z button or knob
[06:28:11] <XXCoder> unless you feel need to control both at once
[06:28:55] <hsm> what would you do?
[06:29:01] <XXCoder> hsm: look for "selector knob"
[06:29:08] <XXCoder> it has defined positons
[06:29:43] <XXCoder> you do need one pulse encoder for jogs but only one encoder needed
[06:30:27] <XXCoder> example
http://www.alibaba.com/product-detail/LW28-20A-25A-32A-63A-125A_60101251592.html?spm=a2700.7724857.29.190.cc5Adt
[06:30:40] <XXCoder> not best example but it shows what I mean
[06:30:51] <hsm> i want to build a cycle lathe with handwheels
[06:31:27] <anomynous_> pulse wheels dont have any feedback
[06:32:35] <anomynous_> you can just keep running it and its not any different based on feeding power used ;) Also it just is not... easy to use like handknobs in a manual lathe
[06:33:43] <XXCoder> I dont know whats best for lathe, being mill machinist but yeah encoders isnt good idea for everything, just jog really
[06:34:03] <XXCoder> man I wanna get a encoder and build something ro control my machine as keyboard really sucks lol
[06:34:07] <XXCoder> anyway night
[06:35:09] <anomynous_> its good for getting your zero maybe? =) Or driving it close to probe for measurement... or something like that
[06:35:37] <anomynous_> if you want hand wheels, use hand wheels. Or mdi. My opinion. Or just make a program.
[06:35:58] <hsm> okay i will use two encoders for jog and 3 selector knobs for override
[06:44:04] <hsm> is it possible to use only 2 button switches for override ? Every push on one button makes the override 5% higher or lower
[06:44:29] <hsm> because of the massive input wiring for selector knop with 12 positions
[06:49:24] <alex_joni> howdy all
[06:49:57] <Tom_itx> morning
[06:51:35] <hsm> hi
[06:53:34] <anomynous_> hsm, is it possible to use a potentiometer that has fixed positions and round the value?
[06:53:46] <anomynous_> i dont know if there are such potentiometers, but there probably are ;D
[06:54:20] <hsm> good idea
[06:55:24] <Tom_itx> 38�F A 4.7 earthquake woke up a lot of people in southern Kansas about 1:42 a.m
[07:00:27] <anomynous_> only a pot may be unreliable
[07:00:28] <anomynous_> ;D
[07:02:03] * anomynous_ is thinking of the same thing that old stereos do when adjusting volyme
[07:03:16] <hsm> or i use quadratur encoder with just 20 positions
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Spindleoverride
[07:06:52] <hsm> but than i will have problems with indey point
[07:06:55] <hsm> *index
[07:51:16] <anomynous_> index point?
[07:54:42] <anomynous_> or weld... i mean solder resistors to the override knob.... i mean... whatever a fancy name is
[07:54:44] <anomynous_> in series
[07:54:46] <anomynous_> ;D
[07:54:54] <anomynous_> and measure that
[07:56:04] <anomynous_> maybe i didnt think that through. Or maybe i did. dunno
[07:56:43] <archivist> dont need an index on a jog encoder
[07:59:07] <gonzo_> another stupidly simple question for you all.....
[08:00:05] * archivist hides
[08:00:50] <gonzo_> when the g code interpreter in axis gets the feedback position from the machine, will it use that to adjust the speed of movement on all axis and the rates that it runs througc commands
[08:01:10] <archivist> it does
[08:01:16] <gonzo_> or does the interpreter just spit out positional commands and it's the hal job to keep up?
[08:01:45] <archivist> they are all working together
[08:01:58] <archivist> except for stepper systems
[08:02:25] <anomynous_> http://1drv.ms/1SHtEUf i just drew this knob hack with my pro electronics design software.
[08:02:28] <gonzo_> ok. so if I have, for example, a slow X axis, and was cutting a diagonal, the feedback of the actual posn will be used to slow the Y axis too
[08:02:54] <archivist> it will correctly coordinate the move
[08:03:31] <archivist> anomynous_, use a better image hoster :)
[08:03:41] <gonzo_> ok, prob such a basig principal that it wasn;'t documented anywhere (not that I remem,ber)
[08:03:51] <gonzo_> additiona Q
[08:03:51] <anomynous_> archivist, it has a feature. If you go too fast or too slow, you have to change fuse.
[08:04:32] <gonzo_> foes the interpreter run in axis, or is that just the display prog that shows what is going on and has a few inputs
[08:04:36] <archivist> anomynous_, I get an crane image and a reminder to upgrade my browser....wtf
[08:04:42] <anomynous_> :/
[08:04:47] <anomynous_> its a png
[08:05:30] <archivist> gonzo_, that is a display mainly
[08:05:37] <anomynous_> http://1drv.ms/1I12e5W
[08:05:43] <anomynous_> that should be smaller. its jpg
[08:06:13] <archivist> that site is fsked, it is not your image
[08:06:18] <gonzo_> was wonderiong about the hal entry (example files) that sends the feedback from encoders to axis.something
[08:06:47] <gonzo_> does there also need to be a line that sends the feedback pos to the interpreter then?
[08:06:57] <archivist> badly named gui
[08:07:22] <archivist> an axis A has little relation to the gui called axis
[08:07:42] <gonzo_> ok
[08:08:01] <archivist> the motion controller is the part doing the work live
[08:08:43] <gonzo_> so the hal serts up paths/nets to the motion controller and that talks to the display prog?
[08:10:16] <archivist> hal is in between almost everything axis-hal-motion-hal-hardware
[08:17:14] <gonzo_> thought that was the case. but seeing things like:
[08:17:25] <gonzo_> net homeswitches => axis.0.home-sw-in
[08:17:37] <gonzo_> caused me to doubt myself
[08:45:13] <Jymmm> (NSFW) lair82 Here's a bonus for you...
http://i.imgur.com/t21XbfG.jpg
[08:49:41] <archivist> gonzo_, and then you realise you can connect things in different ways with hal and its components, enabling things like hobbing machines
[08:51:39] <Jymmm> hobbing machines ???
[08:54:11] <archivist> a method of cutting gears
[08:54:22] <Jymmm> Ah, ok
[08:57:15] <Tom_itx> you hobbit
[08:57:29] <Jymmm> you troll =)
[09:01:10] <Jymmm> (Semi-NSFW) YAY for sweaters!
http://i.imgur.com/6JT9quJ.jpg
[09:01:40] <_methods> won't you be my neighbor
[09:01:48] <Jymmm> lol
[09:09:41] <Tom_itx> _methods is there an easy way to rotate parts around in an assembly to when setting their constraints?
[09:10:41] <Tom_itx> oops.. gotta run. check in later.
[09:15:57] <lair82> Jymm, again, where in the hell are you finding those? I looked around on that site, and nothing as nice as those could be found.
[09:17:03] <Jymmm> lair82: pornz, it comes free with the internetz ;)
[09:23:18] <zeeshan> Tom_itx: middle mouse button
[09:25:12] <_methods> Tom_itx: not really, i usually just put a temporary constraint on parts to make them easier to move around
[09:25:23] <_methods> or get them facing the way i need them to make assembly easier
[09:36:43] <zeeshan> man work is getting awesome here
[09:36:52] <zeeshan> im doing some serious FEA! :D
[09:44:05] <duc> zeeshan: one area where I wish I used at all for my ME degree
[11:37:59] <Tom_itx> _methods, what i find is when 2 parts are facing each other it makes it hard to select the ajoining faces for a constraing
[11:38:00] <Tom_itx> t
[11:39:00] <Tom_itx> and i can't figure out how to get the rotate working when in the mate dialog
[11:39:32] <archivist> I select one, rotate/zoom/whatever select the other
[11:41:28] <archivist> iirc just use the outer area for the move/rotate
[11:42:08] <archivist> it just knows the context is rotate and not a face/edge
[11:43:42] <archivist> this is one of those things one is used to doing rather than explaining
[11:44:21] <Tom_itx> yeah
[11:45:11] <Tom_itx> when i have the mates dialog up though it won't let me rotate
[11:46:36] <archivist> and you can also use the top menu to do moves and rotates and the mate menu remembers where you are when you drop back out of the move/rotate
[11:46:59] <Tom_itx> i must not have enough toolbars showing then
[11:47:06] <archivist> not used it for a few months
[11:47:45] <archivist> yes adding your favourite tools is a must :)
[11:48:12] <Tom_itx> do you know what toolbar that's under? view?...
[11:49:11] <archivist> I would need to go and fire up the windows laptop
[11:49:31] <Tom_itx> view gets you the standard views but not dynamic rotate
[11:52:09] <Tom_itx> i guess i could have multiple viewports
[11:53:17] <JT-Shop> use the arrow keys
[11:53:58] <Tom_itx> viewports will get it
[11:54:15] <Tom_itx> i'd forgotten about that
[11:54:58] <Tom_itx> do you generally do sub assemblies then have a main assembly or do it all in one?
[11:57:14] <archivist> depends on the job
[11:58:00] <JT-Shop> the more complex the machine the more sub assemblies I make
[12:09:15] <renesis> tom_itx: i do it how i think it should be assembled
[12:09:47] <renesis> so if you were to build up a sub-assy before installing into something else, thats how i model it
[12:11:41] <Tom_itx> sounds reasonable
[12:11:55] <archivist> looks like I need to leave the laptop on charge to get it to fire up
[12:28:17] <CaptHindsight> anyone have a good source that makes custom sanding belts ?
[12:28:56] <Tom_itx> i had a place in NY that makes buffing wheels etc for me. they also deal in abrasives so they might
[12:29:29] <_methods> Tom_itx: what do you mean it makes it hard to select the faces?
[12:29:53] <CaptHindsight> got am old custom belt sander recently that needs belts in various grits
[12:29:55] <_methods> sorry customer here
[12:29:58] <_methods> i'll bbiab
[12:30:04] <Tom_itx> np
[12:30:43] <Tom_itx> CaptHindsight,
http://www.divinebrothers.com/
[12:30:56] <Tom_itx> maybe
[12:31:02] <CaptHindsight> _methods: you need to change the angle of the part to be able to select to faces that you want to mate if they are facing or opposite each other
[12:31:15] <CaptHindsight> to/the
[12:31:43] <Tom_itx> yeah, and i can't figure out how to use 'rotate view' when in the mates dialog
[12:33:08] <CaptHindsight> it depends on the CAD package, but I either have the button to freely rotate or some let you right click to get to a view menu
[12:33:23] <CaptHindsight> while in the mates dialog
[12:33:48] <CaptHindsight> Tom_itx: are you using SW?
[12:33:56] <Tom_itx> yeah
[12:34:25] <Tom_itx> i just opened 2 panes and got it done
[12:35:23] <CaptHindsight> I'm starting it up here
[12:35:59] <Tom_itx> ^^ they may not do belts
[12:36:13] <CaptHindsight> they do
[12:36:25] <CaptHindsight> Met-L-Brasive® Belts
[12:37:00] <Tom_itx> ahh there it is..
[12:38:32] <CaptHindsight> Tom_itx: while in the mate dialog I click on the free rotate button and it lets me rotate to choose a surface currently not in view
[12:39:14] <Tom_itx> what toolbar is that on?
[12:39:32] <CaptHindsight> if the free rotate button is not already on your desktop then add it
[12:40:06] <CaptHindsight> http://help.solidworks.com/2013/English/SolidWorks/sldworks/t_rotate_view.htm they have info and a pic here
[12:42:49] <CaptHindsight> when I get anew version one of the first things I do is make sure that Pan, Rotate, Zoom etc is on my desktop
[12:43:07] <Tom_itx> seems like a must
[12:43:43] <CaptHindsight> you'd think it would come as default but you have to tweak a few things to make it easier to use
[12:44:24] <Roguish> get one of these. full 3D motions without the keyboard at all. works in ProeE, SW, MasterCam, Autocad......
http://www.3dconnexion.com/products/spacemouse/spacenavigator.html
[12:44:43] <archivist> a right click also brings up a context sensitive menu where it may be available too
[12:44:59] <Roguish> inexpensive too. only $100. worth every dollar.
[12:45:06] <Tom_itx> yeah my bud had one of those for acad or very similar to it
[12:47:18] <CaptHindsight> http://www.incredible-adventures.com/l-39/l-39-space03.jpg I have one of these I use as 3d mouse for full effect :)
[12:53:05] <Tom_itx> how do you put those on a custom toolbar?
[12:53:28] <Tom_itx> i got em up top now
[12:54:13] <Tom_itx> in the command manager..
[12:56:53] <Tom_itx> yeah that's alot better
[13:15:09] <lair82> Hey Guys, just wondering, I have a 7i49 driving my new servo drive from Automation direct, and when I start Linuxcnc up, before I do anything I looked at the pwmgen value, and it is showing me -6.9 MVDC. Is there any way to actually get that to zero? I tried the adjusting the BIAS, but I cannot get it to a true zero.
[13:16:58] <lair82> The drive is reacting to this, and moving the servo. If I put a 7mv offset in the drive, it almost comes to a halt, but still creeps, very, very slowly.
[13:17:58] <pcw_home> small motion is expected (Even if you zero the motion it will still drift as temperature changes)
[13:19:26] <pcw_home> The fix is not to enable the drives until LinuxCNC is running and the PID loop is in control
[13:20:06] <pcw_home> (this is also a safety issue)
[13:20:10] <lair82> Linuxcnc is running, and using the PID loop,
[13:20:47] <lair82> Halmeter shows "0" for the value, but the actual reading at the Aout is -6.9 MV
[13:21:30] <lair82> Right now I only have a 1 for the P and a 1 for the FF1
[13:21:46] <pcw_home> That should be fine (the drives should not creep when however)
[13:22:55] <pcw_home> should not creep when PID is enabled I should say, creeping means the loop is not closed
[13:23:05] <lair82> If I enter the opposite voltage into the drive offset parameter it almost completely stops
[13:23:39] <lair82> The DRO on the screen is moving up and down for the axis
[13:24:26] <lair82> I am not sure on the scale yet ( haven't made it that far) but is does corellate with the servo rotation
[13:24:32] <pcw_home> is motion about right
[13:25:06] <pcw_home> dont worry about MV offsets, they are not an issue
[13:25:45] <pcw_home> continuous creeping means the feedback loop is not closed
[13:25:48] <lair82> The servo drive/motor is operating properly per their directions, I guess maybe in general it is opposite of what linuxcnc is expecting maybe?
[13:28:09] <pcw_home> is this a PNCCONF created config?
[13:28:23] <lair82> Nope
[13:28:52] <lair82> I am using one of the 7i52 encoder inputs for the feedback
[13:30:25] <pcw_home> In any case the PID setup is not working for some reason
[13:30:26] <pcw_home> If it was, either there would be no creep (correct feedback) or it would runway (reversed feedback)
[13:34:08] <pcw_home> I would check
[13:34:09] <pcw_home> input scaling (does the DRO read correctly)
[13:34:11] <pcw_home> PID HAL wiring
[13:34:12] <pcw_home> does the PID output connect to the right place (a pwmgen)
[13:34:14] <pcw_home> and does the output make sense (With P=1 the static ouput will equal the error)
[13:34:15] <pcw_home> is the PWMGen enabled?
[13:37:22] <lair82_> At the machine now, here is my hal or the a axis
http://pastebin.com/ytTKB2ec, the pwmgen is enabled
[13:37:34] <lair82_> hal for the a axis
[13:43:22] <lair82_> pcw_home that is another question, the scaling right now the servo motor is clamped to the table, not mounted in the rotary table, but regardless the drive is sending the 7i52 a 1:1 encoder signal 2500 ppr quad, so 10,000 pulses
[13:45:09] <lair82_> When I put the servo in the rotary, I need to adjust the ENCODER_SCALE in my INI so a command of 360 degrees of the table actually equals one full revolution correct?
[13:48:17] <pcw_home> AFAIK rotaries are scaled in degrees so the input scale would be encoder counts per degree
[13:50:03] <lair82> So 27.777 repeating, should be my scale
[13:50:22] <lair82> So I am way off, I have it at 1000
[13:52:34] <pcw_home> that seems pretty low, what is your rotary table gear ratio?
[13:55:49] <pcw_home> 27.777 would be the scale for motor shaft (10000/360)
[13:55:51] <pcw_home> but the 27.777 needs to be multiplied by the rotary tables gear ratio
[13:59:02] <lair82> I'm not sure what the reduction is, it is either 1/360 or 1/180, apparently they put multiple servos in this unit, with either 2000rpm or 1500rpm ratings
[13:59:29] <lair82> And there was no data on the servo I took out of this thing when we bought it
[14:00:27] <pcw_home> OK so thats either 5000 or 10000 for scale
[14:00:38] <lair82> Was just typing that
[14:00:50] <lair82> Ok, so that clears that up,
[14:01:25] <lair82> And sure enough, I gotta roll, have to go to another shop and look at a machine.
[14:02:15] <lair82> I will hit this again in the morning, I'm sure it is something simple, just working on a rotary axis vs a linear is new to me
[14:53:02] <Sync> zeeshan: who in the fuck thought it was a good idea to change the brake calipers between the hatchback and the wagon versions?!
[14:53:10] <Sync> just ragequit changing the handbrake cables
[15:12:21] <CaptHindsight> Tom_itx: load up the toolbar with your most used features
[15:20:53] <XXCoder> heys
[16:28:12] <Deejay> gn8
[16:41:17] <SpeedEvil> http://imgur.com/gallery/BfLdbMC on the topic of polymer-composite lathes
[16:43:59] * JT-shop takes five with Dave Brubeck
[16:46:07] <skunkworks> that gets stuck in my head
[16:46:40] <JT-shop> lol
[16:47:22] <skunkworks> I am in a cake mood atm
[16:48:18] <skunkworks> sheeps go to heaven goats go to hell
[18:04:25] * JT-shop tosses in the towel for the day
[18:44:19] <CaptHindsight> I wish to thank whomever is responsible for returning this channel to IRC vs a twitter feed
[18:46:10] <malcom2073> You're welcome
[19:41:13] <t12> dang when said and done
[19:41:18] <t12> i think my moglicing was a success
[19:50:53] <AndChat|407721> Anyone run into problems using zinc plated key stock versus SS 18-8 for servo motors
[20:01:27] <Topy44> are there any solidworks users here? or does anyone have a recommendation for an irc channel that might have sw users on it?
[20:01:31] <Topy44> kinda unsure where to look
[20:01:53] <Topy44> (yes its a metaquestion, but its a bit complex an issue to explain it "just in case" :)
[20:02:02] <roycroft> i would start by checking out #solidworks
[20:04:34] <naja452> Has anyone had experience with rack and pinion drives as apposed to belts in a cnc router?
[20:05:17] <CaptHindsight> naja452: they work fine in a router, lathe or gantry
[20:06:24] <naja452> Do yiu know if the manufacturing process lends its self to some drift across a long stretch of rack?
[20:06:47] <CaptHindsight> naja452: thats what encoders are for
[20:07:04] <naja452> :) say you have steppers.
[20:07:11] <naja452> Open loop
[20:07:43] <CaptHindsight> if you use open loop then you are at the mercy of the manufacturing tolerances and your abilities to assemble a machine properly
[20:08:20] <naja452> Also, how would encoders help with inconsistencies in a drive component?
[20:08:27] <Topy44> i know someone that uses a rack and pinion setup for the Z axis of an open loop cnc mill. its a headache.
[20:08:41] <naja452> If your ballscrew is messed up, your encoder wont know.
[20:08:54] <Topy44> naja452: you'd want positional feedback, not motor position feedback.
[20:09:19] <Topy44> like the glass rulers in large milling machines
[20:09:47] <Topy44> the advantage is that they are imune to backlash
[20:09:58] <Topy44> *immune
[20:10:00] <naja452> Sounds expensive.
[20:10:21] <CaptHindsight> expensive for whom?
[20:10:25] <Topy44> yes. :) though you could go with cheap chinese capacitive calipers
[20:10:40] <Topy44> they are surprisingly accurate. depends on how accurate you need to be.
[20:10:54] <CaptHindsight> that like asking how hard something is to do, or how boring the accounting conference was
[20:10:55] <naja452> Not on a 2 meter stretch though
[20:11:13] <Topy44> true. you can get them up to about 60cm or so afair
[20:12:39] <Topy44> why would you want to go rack and pinion? i'd rather go with a ballscrew
[20:12:48] <CaptHindsight> http://www.romiusa.com/heavy-duty-lathes/ they use rack and pinion for these lathes that are several meters long
[20:12:49] <Topy44> or do you need really high speed?
[20:13:10] <Topy44> hm, when i think about it, a 2 meter ballscrew is not trivial.
[20:13:28] <naja452> It's way cheaper, especially at these lengths
[20:13:39] <naja452> R&p that is
[20:13:59] <CaptHindsight> and they hold <0.001" over that travel using dual servos, one provides the preload for the rack
[20:14:41] <naja452> Preload? How do you mean?
[20:15:02] <Topy44> there is this odd way of using belts, not sure what its called. essentially you have 2 lengths of belt, one fixed to a rail, one on top of it, going around an idler, the pulley, and another idler on the carriage. the belt doesn't move. its essentially using a belt as a rack and pinion system. i could imagine that to be pretty accurate and very cheap. i don't have data to back that up though. :)
[20:15:26] <CaptHindsight> take up any lash in the pinion to rack, even when changing directions
[20:15:30] <Topy44> i guess you have 2 gears, slightly offset?
[20:15:31] <Topy44> right
[20:15:47] <Topy44> so when direction changes the "leading" one changes
[20:16:02] <Topy44> and the back one takes up slack
[20:16:15] <CaptHindsight> http://www.atlantadrives.com/systems1.htm Dual-Pinion Electrical Preload
[20:16:30] <Topy44> yeah that would be exactly what i just tried to describe :)
[20:16:36] <naja452> Wouldn't making sure that the pressure angle is shallow and tensioning the pinion into the rack take out the backlash as well?
[20:16:42] <CaptHindsight> you get a cookie
[20:19:02] <CaptHindsight> http://www.milwaukeegear.com/custom-gears/double-helical-gears.html as another option
[20:20:37] <Topy44> ah, i meant this setup btw:
https://www.youtube.com/watch?v=m_v1LCUSOwY
[20:20:55] <Topy44> (and this too could be doubled up for preloading if using servos instead of steppers)
[20:21:26] <Topy44> its a bit of a hack ofc. not saying its good engineering. :)
[20:21:38] <naja452> Yes, I watched that a few days ago. Interesting setup
[20:22:10] <CaptHindsight> if the belt stretch is not a problem for the design
[20:22:29] <Topy44> CaptHindsight: yes, however having that second length of belt at the bottom helps with that a bit
[20:22:30] <duc> naja452: r+p works fine on my plasma table.
[20:23:21] <Topy44> you could even machine a "rack" matching the belt profile for even better stability. or just screw down the belt every few cm.
[20:23:24] <duc> Purchased for cncrouterparts.com
[20:23:36] <duc> Been using for a few years with no issues
[20:23:49] <Topy44> so you'd only have to deal with a few cm of belt so stretching is less of an issue
[20:24:44] <CaptHindsight> bbl, out of cookies
[20:25:30] <Topy44> i want cookies :(
[20:25:52] <duc> Time to poke wife to make cookies
[20:25:55] <duc> Bastards
[20:39:35] <PetefromTn_> jeez man helping my kids with thier algebra homework is such a bitch LOL
[20:40:05] <renesis> i wish i could go back to algebra
[20:40:24] <PetefromTn_> heh I can help you with that ;)
[20:40:39] <renesis> what are they doing right now?
[20:41:23] <PetefromTn_> solve by elimination using multiplication
[20:41:40] <PetefromTn_> and slope intercept problems
[20:42:16] <renesis> im not sure what the first one is, but slope intercept problems are basically life, heh
[20:42:23] <renesis> gain and offset, yo!
[20:42:43] <PetefromTn_> I was not sure either
[20:43:16] <renesis> oh
[20:43:33] <renesis> its just scaling one equation so terms cancel
[20:44:02] <renesis> then you can solve for the remaining single variable, plug back into one of the original equations, solve for the other
[20:44:25] <renesis> heh, i had to do something like that on my statics test couple days ago
[20:45:02] <renesis> petefromtn_: when you have no other options, this shit all comes back real quick, eheheh
[20:45:29] <PetefromTn_> well I actually had to revert to youtube videos to find out what I was doing with it LOL
[20:45:54] <renesis> yeah but how long since you been in school or done theoretical work?
[20:46:09] <PetefromTn_> too long
[20:46:49] <renesis> going back to math classes after years off was kind of hellish
[20:47:15] <renesis> everything is fucked when your algebra and trig is weak
[20:47:29] <PetefromTn_> like right now we have a number that does not divide evenly and that is a problem
[20:48:00] <renesis> so multiply the other thing?
[20:48:10] <renesis> just guessing dont really know what youre doing
[20:48:35] <PetefromTn_> its 2X-3y=21 and 5X-2y=25
[20:49:02] <PetefromTn_> we multiplied the first one by 2 and the second one by 3 to eliminate the Y
[20:49:03] <renesis> multiple right by -2 and left by 5
[20:49:21] <PetefromTn_> but that gives us 19X=117
[20:49:36] <renesis> x=117/19 is the answer
[20:49:48] <PetefromTn_> it is?
[20:49:54] <renesis> ya
[20:50:03] <PetefromTn_> I thought that was wrong LOL
[20:50:22] <renesis> or 6 remainder 3
[20:50:51] <PetefromTn_> now you gotta plug it back in
[20:50:55] <PetefromTn_> and solvefor Y
[20:51:10] <renesis> heh, yeah i would just use 117/19
[20:51:23] <PetefromTn_> you gotta solve for both variables
[20:51:30] <renesis> but hard to say what a basic algebra class expects
[20:52:06] <renesis> 234/19 - 3y = 21
[20:52:48] <renesis> y = (21 - 234/19) / -3
[20:53:41] <PetefromTn_> yeah that is kinda what I got but it seems to not jibe with the rest of what the problems are
[20:53:51] <renesis> y = ((399/19 - 234/19) / -3, y = 155/19 / -3
[20:54:44] <renesis> y = -465/19
[20:54:57] <renesis> is this one of the last ones?
[20:55:10] <PetefromTn_> it is the last one
[20:55:21] <PetefromTn_> but none of the other problems were fractional answers
[20:55:29] <renesis> so theyre maybe just throwing some messy shit in to challenge kids
[20:55:40] <PetefromTn_> and their parents
[20:55:43] <PetefromTn_> bastards
[20:55:45] <PetefromTn_> ;)
[20:55:54] <renesis> i hate math classes
[20:56:38] <renesis> i like applied math, but math for the sake of math is really frustrating for me
[20:57:06] <renesis> i should be in the library right now studying math shit actually
[20:57:18] * renesis procrastinating
[20:57:26] <PetefromTn_> I like math but this stuff is maddening sometimes
[20:58:09] <renesis> http://thewaythetruthandthelife.net/index/2_background/2-1_cosmological/math/mat6/calcul11_files/s196.jpg
[20:58:15] <renesis> this is the kind of shit im doing right now
[20:58:16] <renesis> =(
[20:58:59] <renesis> but sometimes, there are sections that deal with real shit, like sprung damped mass, or methods that are used in electronics, or exponential growth
[20:59:24] <renesis> i do better at that stuff, probably just because i like it more
[20:59:43] <PetefromTn_> you have my sympathies mate
[20:59:50] <renesis> its almost over
[21:00:09] <renesis> if i want to double major in EE, only one more semester of math after this
[21:00:20] <renesis> if i do grad school, i think maybe two more after that
[21:00:26] <renesis> but for now its over
[21:01:49] <renesis> instructor doesnt get why i dont do well on more of the quizes because when i talk to him about stuff he gets the impression i know whats up
[21:02:01] <PetefromTn_> and here I am having trouble with my kids algebra homework sheesh
[21:02:30] <renesis> are they into technical stuff?
[21:03:19] <renesis> if you think they might end up in an engineering or science based field, you should encourage and help them with the algebra as much as you can
[21:04:09] <renesis> heh, my professor is always talking about how its too late for us, but we should focus on our kids
[21:04:12] <PetefromTn_> that is precisely what I am doing
[21:04:26] <PetefromTn_> but unfortunatly I SUCK at it LOL
[21:04:26] <renesis> cool, big up
