#linuxcnc | Logs for 2014-03-31

Back
[00:06:21] <tjtr33> Tom_itx: it works in master.
[00:06:44] <tjtr33> Tom_itx: I added 5170 to the var file, set at 0.0000, then ran my Master rip, and set it to 51.70, verified it with (debug,#5170), exited linuxcnc, then viewed the .var file, it was saved ok,
[00:07:18] <tjtr33> Tom_itx: then reopened linuxcnc, and did (debug, #5170) and got 51.70. It works without hacking any code in master ( i had tested in a release version, my bad )
[00:10:46] <tjtr33> Tom_itx, now everyone running master can use the variable. but the allocation is unique to your implementation. Some mem map is needed like in the .cc file, like a startup popup or letter to the group, dunno how to make the special stuff puplic/acepted.
[00:12:46] <tjtr33> !later Tom_itx ^^^
[00:12:46] <the_wench> will tell Tom_itx when he/she joins next
[00:48:36] <tjtr33> Tom_itx, you can add #5170 ( for example) and do #5170 = #5400 ( yournewvar = currenttool ) and that will save ok. but there is no POC info except in andypugh's SQL tool table system. i'm still reading about that.
[02:05:05] <Deejay> moin
[06:09:30] <cpresser> is there a source for binary packages for the simulator on debian (wheezy)
[06:10:09] <cpresser> all ubuntu packages i tried (via buildbot) have at least one unmet dependency
[08:34:25] <zeeshan|2> R2E4: im asking a friend for it, he will check tonight
[08:34:30] <zeeshan|2> and get back to me
[08:46:33] <R2E4> cool thanks
[08:49:11] <R2E4> I wonder if I can change the drive for a more mainstream drive that would work with the fanuc servo.
[08:51:52] <zeeshan|2> i dont thinkso
[08:52:26] <zeeshan|2> all those alpha,beta drives look like theyre the only ones that are compatible with their i series controllers
[08:52:35] <zeeshan|2> (at least according to fanuc)
[08:52:46] <zeeshan|2> and if you look at the motor codes supported by those drives
[08:52:51] <zeeshan|2> i didnt see your motor on there
[08:53:01] <zeeshan|2> i should really be saying servo codes
[08:53:38] <R2E4> I will check again, the servo amp is A06B-6044-HXXX
[08:54:31] <zeeshan|2> A06b 1003
[08:54:35] <zeeshan|2> B300
[08:54:37] <zeeshan|2> was the motor code right
[08:54:39] <zeeshan|2> servo code
[08:54:43] <R2E4> Thats the motor
[08:55:23] <zeeshan|2> http://www.hofstragroup.com/media/product_images/productimage-picture-ge-fanuc-6-3000-ac-servo-motor-a06b-0128-b575-0008-719.jpg
[08:55:36] <zeeshan|2> visually thats what i saw on the machining centers and turning centers and grinders
[08:55:40] <zeeshan|2> the 'red' cover
[08:56:13] <zeeshan|2> i wish i had one of those motor and drives
[08:56:20] <zeeshan|2> if you can get them working with linuxcnc let me know!
[08:56:23] <zeeshan|2> i might convert my lathe to use it
[08:56:33] <zeeshan|2> indexing the spindle with that servo is a joke ;p
[08:57:25] <R2E4> My motor and spinle together is about 36" long...
[08:57:29] <R2E4> ITs huge.
[08:57:37] <zeeshan|2> yea
[08:57:45] <zeeshan|2> prolly cause its 20hp
[08:57:49] <zeeshan|2> or something ridiculus like that?
[08:57:49] <zeeshan|2> ;p
[08:57:53] <R2E4> Might as well be 2" long. Not working....hehe
[08:58:36] <zeeshan|2> haver you heard of the emag vl7?
[08:58:56] <zeeshan|2> http://www.emmequattroaxles.it/immagini/progetto_automazione/vl7.jpg
[08:59:11] <zeeshan|2> that was the baby i worked on most of the time
[08:59:21] <R2E4> yeah, nice lathe
[08:59:48] <R2E4> Thats a nice machine
[08:59:50] <zeeshan|2> they had robots in between each center
[09:01:37] <R2E4> I was so close.
[09:02:10] <R2E4> Theres not enough documentation on this that i can get a handle on it. Thats what is frustrating.
[09:04:51] <R2E4> That cd of Fanuc docs is newer products.
[09:11:55] <skunkworks> R2E4: did you talk to fanuc yet
[09:12:16] <R2E4> No, there not open.
[09:12:19] <R2E4> yet
[09:12:36] <R2E4> MO is in pacific time?
[09:15:10] <R2E4> I dont remember setting the scale of my spindle...... 10v / 6000 rpm = .0016 scale factor
[09:54:09] <zeeshan|2> 1700/60
[09:54:33] <zeeshan|2> 28.3333*200hz = 5666RPM
[09:58:27] <R2E4> http://irmtl.com/vm40/SP-schematics.pdf spindle vm40 schematics
[10:33:41] <pcw_home> R2E4: DR1 closed? 10V at DA2 (and 7I77 GND at E)
[10:37:20] <R2E4> DR1 is closed, thats what enables it and I hear contactors. yup, 10v DA2, yes gnd from 7i77 at E.
[10:37:50] <pcw_home> S2 installed?
[10:40:01] <R2E4> I couldnt find where S2 is. I would have to remove the drive I think. however, it would have to be for the machine to work before cause I dont have those wires OVR1 and OVR2 to connect.
[10:45:05] <pcw_home> there was no front panel feed override potentiometer?
[10:47:13] <skunkworks> that would do it - wouldn't it...
[10:47:40] <R2E4> I didnt see one, didnt specifically check though.
[10:48:41] <R2E4> you mnean on control panel of machine?
[10:48:56] <pcw_home> Yes
[10:49:46] <R2E4> There was no pots on the controil panel. There was feedrate select pushbuttons though.
[10:49:51] <R2E4> But not a pot
[10:50:23] <pcw_home> Hmm I would check for S2 then
[10:51:19] <pcw_home> I would also get the scaling set so that you are not at 10V
[10:52:22] <R2E4> like this? http://www.linuxcnc.org/docs/2.4/html/examples_spindle.html#r1_3
[10:52:57] <pcw_home> not really
[10:53:30] <pcw_home> the idea is the same. the implementation is different
[10:57:35] <pcw_home> setp hm2_5i25.0.7i77.0.1.analogout5-maxlim MaxRPM
[10:57:36] <pcw_home> setp hm2_5i25.0.7i77.0.1.analogout5-minlim 0
[10:57:38] <pcw_home> setp hm2_5i25.0.7i77.0.1.analogout5-scalemax MaxRPM
[10:57:39] <pcw_home> plus you will need a abs comp to convert the signed RPM to positive only
[11:06:33] <R2E4> comparator clamp?
[11:06:55] <pcw_home> ?
[11:07:23] <R2E4> abs comparator so its just a positive signal.
[11:08:29] <R2E4> ah abs component....
[11:08:33] <R2E4> sorry
[11:08:35] <Jymmm> today is a holiday?
[11:10:11] <pcw_home> Its a hollowday
[11:12:08] <Jymmm> César Chávez Day State holiday
[11:12:50] <Tom_itx> what did he do?
[11:13:10] <Tom_itx> i want my own holiday!
[11:13:25] <Jymmm> Cesar Chavez was an American farm worker, labor leader and civil rights activist, who, with Dolores Huerta, co-founded the National Farm Workers Association.
[11:16:05] <Jymmm> Tom_itx: You can have Tomorrow
[11:17:30] <R2E4> 0 -10v, so using the forqwarfd rwverse input determines the spindle rotation direction... i think.
[11:18:42] <skunkworks> pcw_home, do you think a rolling average would work better for figuring acceleration?
[11:19:04] <skunkworks> of the last x samples
[11:21:28] <pcw_home> Yes
[11:21:57] <skunkworks> http://en.wikipedia.org/wiki/Moving_average
[11:22:17] <skunkworks> The problem can be overcome by iterating the process three times, with the window being shortened by a factor of 1.4303 at each step.[3] This removes the negation effects and provides a better behaved filter.
[11:22:37] <skunkworks> does that mean they take the average of the 3 averages?
[11:23:40] <pcw_home> sounds like it but a simple SMA would be a big improvement
[11:23:52] <pcw_home> (and its easy as well)
[11:24:00] <skunkworks> so average of 10 samples, 7 and 5.. then average those 3 togather?
[11:24:27] <skunkworks> I think I can write a comp to do that pretty easy..
[11:24:56] <pcw_home> SMA is just average of last N samples
[11:25:14] <skunkworks> right
[11:25:33] <skunkworks> but they talk about inversions if you don't itarate though 3 times...
[11:26:21] <pcw_home> (so you maintain a circular buffer if the N samples and at the pointer position you subtract the old and add_ overwrite the old with the new)
[11:26:34] <pcw_home> of the N
[11:26:38] <skunkworks> right - I can wrap my head around that.
[11:27:11] <pcw_home> mod N the pointer
[11:29:17] <pcw_home> should be much better (and more consistent) since the average will not
[11:29:18] <pcw_home> depend on the synchronization between accel data and sampling interval
[11:29:39] <pcw_home> (the problem the block average has)
[11:31:12] <pcw_home> (it will still be delayed 1/2 average period so the plot will show the accel late )
[11:32:45] <pcw_home> (I guess you could add a 1/2 average period delay function as a utility to the moving average comp)
[11:32:49] <skunkworks> that is ok - (already have that issue)
[11:33:49] <archivist> stock markets use moving averages too, use two over different (5 and 10 days) periods and look for the cross over to buy/sell
[11:34:48] <pcw_home> the delay function is almost the same code as the SMA (just 1/2 the length and no adds, just fetch/store)
[11:35:19] <pcw_home> so you could feed you velocity through and get you plots aligned
[11:36:11] <skunkworks> oh - delay the velocity.. duh.
[11:36:42] <skunkworks> I was wondering how I was going to look into the future to align the acc with the incoming velocity..
[11:37:14] <pcw_home> looking into the future is hard
[11:37:21] <skunkworks> I agree.. ;)
[11:38:27] <Gigs-> what is this thing used for? http://www.ebay.com/itm/YUASA-550-003-INDEXING-HEAD-SPIN-JIG-For-5C-Collets-NO-RESERVE/351032297734?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222003%26algo%3DSIC.FIT%26ao%3D1%26asc%3D21144%26meid%3D5876928570519962153%26pid%3D100005%26prg%3D9373%26rk%3D2%26rkt%3D6%26sd%3D111309685130&rt=nc
[11:38:31] <Gigs-> sorry for long link
[11:39:40] <pcw_home> the lowpass filter comp is a Exponential Moving Average (EMA)
[11:39:50] <archivist> any regular shape but on bar type material
[11:40:10] <XXCoder2> http://www.chaski.org/homemachinist/viewtopic.php?p=193948
[11:40:15] <skunkworks> pcw_home, yah - that didn't work well
[11:40:16] <XXCoder2> might be helpful
[11:40:58] <archivist> Gigs-, eg this was made on a dividing fixture http://www.collection.archivist.info/archive/DJCPD/PD/2014/2014_03_29_measuring_tools/IMG_1752.JPG
[11:41:00] <Gigs-> I'll take a look
[11:41:23] <pcw_home> Yeah you want equal weights not exponentially weighted
[11:41:23] <Gigs-> I was searching for indexing table type things and these things keep coming up
[11:41:33] <Gigs-> i.e. geneva drive type deal
[11:41:59] <XXCoder2> might help too http://www.kanabco.com/vmsbbs/index.php?topic=1249.0 dunno
[11:42:32] <pcw_home> so the pegholes along the top ate a vernier of sorts?
[11:42:37] <pcw_home> are
[11:42:56] <archivist> digital vernier :)
[11:42:57] <Tom_itx> yes
[11:43:36] <pcw_home> 0 to .5 a hole pitch?
[11:43:41] <archivist> must tale pictures of my 4 ft digital vernier calipers (every 1/4")
[11:43:48] <archivist> take
[11:46:18] <archivist> 36 outer holes every 10 degrees, digital vernier to 1 degree
[11:46:48] <pcw_home> Yeah I kind of guessed that
[11:47:34] <pcw_home> must be a pretty old idea
[11:50:32] <R2E4> pcw_home: The gnd/AOUT on 7i77, that is using the field power gnd?
[12:03:05] <pcw_home> No its 5V (PC) gnd
[12:03:51] <archivist> 4ft digital vernier calipers http://www.collection.archivist.info/archive/DJCPD/PD/2014/2014_03_31_tools/IMG_1776.JPG no makers name
[12:04:25] <IchGuckLive> hi all B)
[12:05:59] <XXCoder2> hey'
[12:08:57] <pcw_home> wow that really neat (but surprisingly large steps)
[12:09:45] <XXCoder2> its 4 ft long wow
[12:09:48] <archivist> got it at a model engineer exb many years ago, never used it yet :)
[12:10:15] <IchGuckLive> this tool is for timber man carpenter that make houses
[12:10:38] <archivist> too solid for a carpenter
[12:10:47] <XXCoder2> metal shop?
[12:11:22] <archivist> casting /patterns rough measurement maybe
[12:12:06] <archivist> must post the images to the oldtools mailing list
[12:15:46] <XXCoder2> I have few strange stuff from dad
[12:16:14] <XXCoder2> like screwdriver that has tiny screwdriver that rotates when handle is pressed in
[12:16:46] <IchGuckLive> archivist we had one for 3meter in the shop
[12:16:49] <XXCoder2> so basically its screwdriver that screws regular screws with mini screw in it or something
[12:17:16] <XXCoder2> or maybe its just to help lock or something. so weird
[12:17:52] <archivist> XXCoder2, tappet adjusting tool?
[12:18:00] <XXCoder2> no idea, got picture?
[12:19:05] <archivist> do a google images search for tappet adjusting tool, there are many variants
[12:19:11] <IchGuckLive> we used it for setting up sensors on liniar robots
[12:19:12] <XXCoder2> ok lemme look
[12:19:28] <skunkworks> pcw_home, what did you mean by 'mod N the pointer'
[12:19:40] <XXCoder2> interesting byut no
[12:20:01] <XXCoder2> it looks like regular screwdriver but has mini screwdriver in it too, that rotates
[12:21:01] <Einar> XXCoder2: Ratchet screwdriver?
[12:21:11] <XXCoder2> mini part is.
[12:21:25] <pcw_home> for a running average you need a circular buffer so the pointer has be incremented each time but wrap at N
[12:21:26] <XXCoder2> press handle in and it rotates
[12:21:32] <archivist> we want pictures
[12:21:37] <skunkworks> oh - righ
[12:21:39] <skunkworks> right
[12:21:41] <XXCoder2> dunno where it is now
[12:22:25] <Einar> Like this: http://www.powertoolsdirect.com/faithfull-spiral-ratchet-screwdriver-bits-range
[12:23:09] <archivist> I spent an hour a couple of times on google images trying to find what this is for http://www.collection.archivist.info/archive/DJCPD/PD/2014/2014_03_31_tools/IMG_1774.JPG
[12:23:14] <XXCoder2> nah it dont look different than regular screwdriver. with part of blade actually mini screwdriver. iuts all kinds of weird
[12:24:46] <Einar> Archivist: Looks like it can be used to determine how much out of level a surface is?
[12:25:01] <IchGuckLive> Einar: did you got your magic lathe gcode to run
[12:25:33] <XXCoder2> wonder why one end has mirror?
[12:25:41] <XXCoder2> (arch's pic)
[12:25:49] <Einar> Magic? There may be demons in the lathe, but no magic so far.
[12:25:50] <archivist> sort of a clinometer but only +- 5 degrees and at 90 to eash other
[12:26:44] <archivist> XXCoder2, so you can view along the axis and dont need to be above, common on surveying levels
[12:26:57] <XXCoder2> ogh cool
[12:27:11] <Einar> Or on weapons?
[12:27:34] <archivist> not sure for weapons
[12:27:48] <Einar> Where you don't want to stick your head up higher than needed.
[12:28:50] <archivist> http://www.ebay.com/itm/Dumpy-Level-Top-Bubble-7-Sextant-For-Sale-Gift-/320704118948
[12:29:15] <archivist> never buy a polished one like that, ruined
[12:29:48] <archivist> methinks brand new made in india junk
[12:30:39] <Einar> Ich: I don't remember what you mean. But the lathe is pretty much behaving now. It does not reach the feed speeds the original brochure says. But according to the plots from my logic analyzer neither did the original controller.
[12:31:12] <XXCoder2> heh sometimes restoring destroys value. I recall one old stuff show where guy who removed real ugly black paint from one funture. apparently its rare one $6000 or more... if black paint was still on. without it, $200
[12:31:16] <Einar> I did some logging work on it before I ripped the old guts out.
[12:33:21] <Einar> Would there be a point offering the old Boxford controller on fleabay or somewhere else? It seems a pity to scrap it if that's what keeps another lathe from working.
[12:34:22] <archivist> some seem to fight over the bits
[12:35:04] <archivist> but they dont know hoe to retrofit :)
[12:35:18] <XXCoder2> oh freakin funny https://scontent-a-sea.xx.fbcdn.net/hphotos-frc1/t1.0-9/10006580_683879768343246_3230987_n.png
[12:35:57] <Einar> Or they do education and the material is built around how the controller works.
[12:37:38] <Einar> Archivist: Is that how your web pages work? To document that something (not so common item) exists and where it is?
[12:38:32] <archivist> it was just a collection database originally, but it sort of grew
[12:39:02] <archivist> now people find old service manuals on there
[12:40:30] <archivist> I also chuck all my digital pictures on it now to document and sometimes show
[12:42:02] <Einar> The strangest instrument I have is a planimeter. Beautifully built, but I don't have a clue using it. :-)
[12:42:50] <Einar> It's one of those things I can show and offer a prize for guessing what it is for. And noone won yet.
[12:43:07] <XXCoder2> https://en.wikipedia.org/wiki/Planimeter
[12:43:27] <XXCoder2> do I win? lol
[12:43:41] <Einar> Did I show it to you? ;-)
[12:43:54] <XXCoder2> nah lol
[12:44:56] <Einar> We have those kind of quizzes in our "Badger Club".
[12:45:52] <archivist> I have one too
[12:46:12] <Einar> Why am I not surprized....
[12:46:42] <archivist> if its in a box, I am a sucker for it
[12:47:13] <Einar> I think we both would enjoy looking around in the other's drawers.
[12:47:23] <XXCoder2> lol
[12:47:38] <XXCoder2> in some places that'd be quite dirty statement
[12:47:50] <Einar> ??
[12:48:02] <archivist> draws=underwear
[12:48:03] <IchGuckLive> i just got all stl uploaded to the 3d printer will see if they are build tomorrow when i hit the entry door
[12:48:04] <Einar> Not my native language you know.
[12:48:48] <XXCoder2> I forgot what region has drawers = underwear
[12:49:28] <CaptHindsight> the 50-60's :)
[12:50:34] <archivist> http://www.urbandictionary.com/define.php?term=drawers
[12:50:49] <IchGuckLive> today a big hail storm hit HK it is realy rear 36 since weather recording
[12:51:02] <XXCoder2> arch it just mak4s it worse LOL
[12:51:11] <archivist> :)
[12:52:29] <IchGuckLive> only for your info there is a new CREO 3D CAD online from PTC it is limited to 60 parts !!! Free
[12:52:44] <XXCoder2> 60 parts?
[12:53:08] <archivist> at once in an assembly?
[12:53:22] <Einar> Women's underwear!?! I was NOT implying that archivist or I use such items!
[12:53:25] <IchGuckLive> PTC has the Assemply structure and you can put 60 items togeter
[12:53:32] <CaptHindsight> http://www.3ders.org/articles/20140331-germany-startup-twobears-launches-biofila-silk-and-linen-3d-printer-filament.html at least it looks better, still takes forever
[12:53:44] <XXCoder2> Einar: really its old usage, not current, I was joking lol
[12:54:05] <XXCoder2> 60 unique parts
[12:54:33] <archivist> Einar, I know, that's the fun of using other people languages :)
[12:55:27] <IchGuckLive> CaptHindsight: around your factrory new CNC builder in town http://xrcnc.com/
[12:57:51] <Einar> I have to. If I were restricted to using mine I would be sitting in a corner mumbling.
[12:59:10] <XXCoder2> weird cncs in site
[12:59:24] <XXCoder2> like 8 router motors
[12:59:29] <XXCoder2> other one just 4
[12:59:42] <Einar> Usually it works out. On my first visit to USA I thought "making hay" referred to harvesting or something. It did not!
[13:00:18] <XXCoder2> normally drawers just mean drawers so youre fine this time :D
[13:00:32] <Einar> After that it could only get better.
[13:03:06] <Einar> The lathe have a central oiler. If I connect a wiper motor and a wire to the handle it could be electrically operated. Is there a way in LinuxCNC to give a pulse after ### metres of movement?
[13:03:55] <Einar> It will vary widely how much it moves between power on/off, so that will not be useful.
[13:05:02] <IchGuckLive> Einar: you can make a comp and tell it the number of movement and it makes a sum()
[13:05:09] <IchGuckLive> so then you can pulse out
[13:06:09] <IchGuckLive> like the THC it interacts with the motion fb
[13:06:58] <Einar> Can the sum counter be saved so it will continue counting upon next run?
[13:07:24] <IchGuckLive> that woudt be a python interact
[13:08:43] <CaptHindsight> IchGuckLive: looks like they are up north
[13:09:12] <IchGuckLive> i dont know if there are maybe free VARiables in the var files that maybe you coudt use to hold this in the
[13:09:39] <pcw_home> use the integ comp (on motion.current-vel) trigger at some level to pump the oiler (an reset the integrator)
[13:09:46] <IchGuckLive> CaptHindsight: i thoudt it is your next street
[13:11:26] <Einar> I may try that. More as an educational move than me not being able to pull the lever. At least for some years. It would be easier to debug than the toolchanger comp.
[13:14:02] <Einar> Another use for a long time counter is to pump coolant for 15min. if it have been sitting for more than a week. I try to on my mill, and it keeps the emulsion from becoming a stink bomb.
[13:14:20] <Deejay> hrhr
[13:15:03] <Einar> A thin layer of oil on top makes anaerobic critters multiply.
[13:15:14] <Einar> Aerating it helps a lot.
[13:15:32] <alex-bv-ro> hi all, here is a question you probably had allready.
[13:16:22] <alex-bv-ro> it is more ubuntu related than linuxcnc related, did anybody got the intel NM10 express graphics chip to work correctly?
[13:18:50] <pcw_home> Didn't know there was a problem with the NM10. Thats the one on the Atom D525 MBs that are commonly used for LinuxCNC
[13:21:00] <alex-bv-ro> I have an intel D2550 with nm10 chipset
[13:21:14] <alex-bv-ro> everything works great except monitor resolution
[13:21:36] <skunkworks> pcw_home, http://imagebin.org/302901
[13:21:47] <skunkworks> rolling average over 10 samples
[13:22:02] <skunkworks> (not logged data yet - just seeing if it works as expected)
[13:22:07] <IchGuckLive> alex-bv-ro: what os are you using
[13:22:14] <IchGuckLive> the live cd
[13:22:46] <alex-bv-ro> ubuntu 10.04
[13:23:05] <alex-bv-ro> i even tried 12.04 (wich by the way works until the real time kernel is loaded)
[13:23:17] <IchGuckLive> lnuxcnc live cd or a coustem build linuxcnc
[13:23:33] <alex-bv-ro> i have tried both
[13:23:39] <alex-bv-ro> latest livecd
[13:23:45] <IchGuckLive> did you update to 2.5.3
[13:24:24] <alex-bv-ro> yes, but the problem is ubuntu related
[13:24:25] <alex-bv-ro> :(
[13:24:29] <IchGuckLive> alex-bv-ro: where are you in the world
[13:24:38] <alex-bv-ro> romania
[13:24:52] <alex-bv-ro> Brasov, Transilvania
[13:25:08] <IchGuckLive> alex-bv-ro: http://buildbot.linuxcnc.org/
[13:25:09] <skunkworks> alex-bv-ro, have you for grins updated the bios?
[13:25:22] <IchGuckLive> get the synaptic open its the packet manager
[13:25:30] <IchGuckLive> go to packet source
[13:25:43] <IchGuckLive> 3rd party
[13:25:48] <skunkworks> IchGuckLive, it isn
[13:25:51] <pcw_home> I thought the D2550 used the PowerVR graphics
[13:26:02] <skunkworks> IchGuckLive, it isn't a linuxcnc issue.
[13:26:19] <pcw_home> (Cedar trail or whatever)
[13:26:28] <IchGuckLive> the 2.5.0 is not the best Atom support
[13:26:39] <alex-bv-ro> motherboard is http://www.asrock.com/mb/Intel/AD2550B-ITX/
[13:26:54] <alex-bv-ro> I tried everythink I could think of before writing here
[13:27:09] <IchGuckLive> there are graphic bugs nown in the issue list so it is best to update
[13:27:47] <IchGuckLive> alex first start the 122 rtai kernel and then update
[13:27:55] <alex-bv-ro> yes, I installed the latest updates
[13:28:18] <IchGuckLive> i think you messsed up with the 10.04 updates
[13:28:27] <pcw_home> I think the cedar trail MBs have issues with Linux graphics (D2550,2700,2800 etc)
[13:28:48] <Connor> skunkworks: You have a tool changer... I have a question... What do you do when you reboot linuxcnc / mill and still have a tool in changer.. do you have to deal with putting the tool back ?
[13:28:53] <alex-bv-ro> it's linux related, linuxcnc works great but in low resolution
[13:29:18] <Einar> Do you have another monitor to try? On one of my PC's it seems to work fine with one monitor, not with another. Obviously a bug, but if using another monitor works you might have a workaround.
[13:29:25] <Connor> in working on PeteFromTN toolchanger script.. I've identify a issue.. what to do if he reboots and has tool in spindle..
[13:30:05] <alex-bv-ro> Yes, other monitors work just fine, the only problem is, I just bought 3 identical monitors for my new operator panel for my linuxcnc machines
[13:30:15] <IchGuckLive> Connor: i got a pyvcp tool clamp open
[13:30:18] <Connor> when he comes back up, the carousel will home.. but not know what pocket the tool goes into.. and will cause a crash when he goes to change tools
[13:30:23] <IchGuckLive> and i replace it mauell
[13:30:24] <skunkworks> Connor, we normally put a dummy tool in the spindle.
[13:30:42] <skunkworks> I thought you could tell linuxcnc what tool is in the spindle
[13:31:02] <Connor> Well.. they're random and non-random..
[13:31:07] <Connor> his is non-random..
[13:31:14] <skunkworks> right
[13:31:20] <Connor> and.. talking about across reboots..
[13:31:26] <IchGuckLive> connor mdi tell
[13:31:34] <skunkworks> I think you would have to ask on the user list...
[13:31:40] <Connor> Trying to remove the "human" factor
[13:31:59] <Connor> I'm thinking... give him a button that just puts the tool.. and doesn't get a new one...
[13:32:07] <Einar> alex: What I ended up doing was to beg an old video card from the IT guys. That fixed it for me (I bought 2 touchscreens).
[13:32:23] <IchGuckLive> im off bye
[13:32:29] <Connor> and 2ndly.. storing the current pocket in non-volatile variable..
[13:33:00] <Connor> when he homes... it then checks if tool in spindle, and if so, checks that variable, and moves the carousel to that location.
[13:33:21] <Einar> Also because I read somewhere that onboard video is not so good for realtime performance. Then it seemed like a good solution for me.
[13:33:59] <Einar> Video cards gets thrown out early from the CAD guys (or PC gamers).
[13:34:04] <Connor> skunkworks: Thoughts on that ?
[13:34:54] <pcw_home> Onboard video is usually fine nowadays
[13:35:39] <skunkworks> Connor, sounds good - but you should ask on the mailing list. A lot more people on there running real machines
[13:36:19] <Connor> Kinda get lost on the list....
[13:36:35] <Connor> But, I'll see about posting up a question..
[13:36:43] <alex-bv-ro> i had problems in the past with this chipset but now the display is very bad
[13:36:44] <Connor> I plan on releasing this script into the wild after I'm done.
[13:37:07] <Connor> I just hope no one pokes fun at the way I did things..
[13:38:00] <pcw_home> can you sense if a tool is in the spindle (or in a toolholder?)
[13:38:08] <Einar> If I were to swap stepper to servo, would I need to go up in torque to not loose accel or speed? The torque/RPM curves are very different.
[13:38:48] <Connor> pcw_home: Can sense if tool is in spindle.
[13:39:03] <Einar> Connor: If they do, they better come up with a very godd solution!
[13:39:16] <Einar> good
[13:39:34] <pcw_home> seems like a prox on the tool changer could log empty holders by doing a spin at startup
[13:40:09] <Connor> yea. but, would rather tool goto original slot.
[13:40:22] <Connor> otherwise tool table gets all messed up.
[13:40:48] <pcw_home> Sure but for safety it might be nice to verify
[13:43:21] <Connor> true.
[13:45:08] <pcw_home> not sure if it can be sensed easily or not
[13:45:35] <Connor> Yea.. Not sure either..
[13:45:49] <Connor> but.. I think my current idea is better than nothing..
[13:46:47] <pcw_home> skunkworks: looks good (a lot better than the boxcar average)
[13:48:03] <skunkworks> yes - but I am logging data and getting odd results. the calculated accelleration is extreme - and averaging that is crap. Would rolling avarage of the velocity -> then ddt that data?
[13:50:15] <pcw_home> Maybe thats better, worth a try anyway
[13:51:54] <andypugh> The best way I have found to get accelleration from noisy samples is a polynomial least-squares fit which you then explicitly differentiate.
[13:52:34] <pcw_home> a bit more difficult to do in real time
[13:53:30] <pcw_home> anyway a SMA should be a lot more consistent than the boxcar average
[13:54:57] <andypugh> So, if the polynomial fit is 12 + 13x - 14x^2 + 15x^3 the acceleration is -28x + 45x^2 which you then solve for the x at the point af interest. (Actually it is better to shift the data to site symmetrically around x = 0 and solve for x = 0)
[14:08:15] <andypugh> I think it probably could be done in real-time, but you would only have past data to fit to.
[14:15:02] <skunkworks> http://imagebin.org/302913
[14:15:16] <skunkworks> so - is the peaks the inversions they talk about?
[14:15:46] <skunkworks> 10 and 20 sample averages look pretty close to the same.
[14:16:29] <archivist> is that on a real machine or a sim
[14:16:35] <skunkworks> that is a real machine
[14:17:03] <archivist> do you have a real stiction/backlash problem
[14:17:10] <skunkworks> (linuxcnc running step/dir -> mesa encoder counter -> halscope
[14:17:33] <skunkworks> archivist: pure stepper openloop machine..
[14:18:52] <archivist> so any real hardware problem could show up
[14:20:47] <skunkworks> http://pastebin.ca/2689643
[14:20:52] <skunkworks> (don't laugh..)
[14:25:42] <pcw_home> You can do it without a loop
[14:26:16] <skunkworks> I am sure it could be done much better.
[14:26:48] <skunkworks> pcw_home, did you see the odd spikes at revearsal? Do you thing reiterating averaging help that?
[14:27:13] <cpresser> skunkworks: any you should check the value of PTA. you will get a segfault (in kernel space!) if PTA is >20
[14:27:34] <skunkworks> cpresser, sure - and it can't be 0...
[14:28:12] <pcw_home> skunkworks: I dont think so (though it would be easier to see whats going on if just X velocity was plotted)
[14:28:28] <skunkworks> oh - I can do that.. :)
[14:32:36] <skunkworks> http://imagebin.org/302917
[14:32:38] <pcw_home> to create the average without the loop, you just add 'in' to the average
[14:32:40] <pcw_home> subtract array ( ptr ) from the average and store 'in' at ptr
[14:32:42] <skunkworks> (not time corrected..)
[14:35:19] <pcw_home> but at 500 ms/div the time different sis not large on the plot relative to 10/20 ms
[14:36:40] <pcw_home> hard to tell if thats a stepgen artifact or encoder velocity artifact
[14:37:24] <pcw_home> (you should set the stepgen set and hold times very small relative to the step times)
[14:37:32] <pcw_home> setup
[14:37:34] <skunkworks> http://imagebin.org/302918
[14:37:53] <skunkworks> does that help?
[14:38:51] <pcw_home> are the stepgen direction setup and hold times set to minimums?
[14:39:01] <skunkworks> I would have to look
[14:39:12] <pcw_home> thats one possibility
[14:39:42] <pcw_home> (since it happens at reversals)
[14:40:59] <pcw_home> another possibility is the encoder velocity estimation setting velocity to 0 at reversals
[14:45:04] <skunkworks> setting dirhold and dirsetup didn't seem to effect anything
[14:45:20] <skunkworks> (to 0)
[14:46:09] <skunkworks> from the graph - it doesn't seem to be..
[14:46:17] <skunkworks> (setting velocity to 0)
[14:46:59] <pcw_home> what is the X steprate/scaling?
[14:47:49] <skunkworks> 1500steps/inch
[14:47:53] <skunkworks> software stepgen
[14:55:12] <pcw_home> if it encoder related you might try setting vel-timeout longer but it looks more like a stepgen artifact
[14:56:53] <pcw_home> (vel-timeout sets how long the encoder logic will wait for a count change before just giving up and reporting 0 velocity)
[14:57:58] <pcw_home> but that should make a simple deadzone around 0 velocity
[14:58:52] <skunkworks> it is set to .5 - I set it to 2 and didn't seem to change anything
[15:03:47] <skunkworks> have to play with it some more... but is close I feel...
[15:04:23] <skunkworks> jeeze - that is in seconds. .5 should have been long enough anyways
[15:04:58] <skunkworks> so - maybe a direction change issue with stepgen?
[15:05:49] <skunkworks> I could change linuxcnc to quadature...
[15:06:24] <skunkworks> see if it effects anything
[15:11:30] <pcw_home> or try the a hardware stepgen with PID
[15:12:35] <skunkworks> quadature has the same peaks
[15:14:21] <skunkworks> I guess I could feed a 7i80 stepgen back into a 7i80 encoder counter to see if I get the same peaks
[15:15:12] <pcw_home> yes or run the software stepgen in PID mode
[15:17:44] <skunkworks> oh
[15:17:49] <skunkworks> I could try that..
[15:17:53] <skunkworks> tomorrow though..
[15:17:55] <skunkworks> bbl
[15:18:13] <skunkworks> (increasing the scale didn't seem to effect it either)
[15:18:24] <skunkworks> if anything it was worse.. but hard to tell
[16:06:27] <XXCoder2> fucking wasps
[16:07:46] <andypugh> I suggest not trying that
[16:07:55] <XXCoder2> lol
[16:08:17] <XXCoder2> they are annoying me though. I cant mow back yard due to wasps
[16:11:47] <andypugh> Welding helmet, towel and duct tape?
[16:12:06] <XXCoder2> lol
[16:12:41] <XXCoder2> Star Wars laser system that lock on hz of wings and fire on wasps
[16:12:45] <XXCoder2> waps cant beat light speed
[16:15:02] <XXCoder2> if yioure curious http://www.nytimes.com/1992/07/21/us/star-wars-x-ray-laser-weapon-dies-as-its-final-test-is-canceled.html
[16:16:04] <XXCoder2> it closed down at 1992, dawn of internet. around time I started using internet
[16:16:15] <R2E4_> pcw_home: : changed it to port 5, set the scale and vcmd is varying so it is getting the correct voltage.
[16:16:44] <andypugh> XXCoder2: http://en.wikipedia.org/wiki/Mosquito_laser
[16:16:48] <R2E4_> spoke with Fanuc, and first line support couldnt help aso he wanted to send me up which would cost 180/hr.
[16:17:04] <XXCoder2> dang
[16:17:41] <andypugh> It seems to even involve the same people :-)
[16:17:43] <R2E4_> ovr1 and two are correct.s They dont even go out on the connectors.
[16:17:43] <XXCoder2> interesing!
[16:18:52] <XXCoder2> andy I referred to star wars system because one of books used it to protect bees from africianized bees
[16:37:51] <skunkworks> pcw_home: I was thinking of setting up a software stepgen -> software encoder and calculate the accelleration from the velocity estimation. (but I can do that in simulator...)
[16:38:00] <skunkworks> to see if I get the same spikes.
[16:38:25] <skunkworks> so - I will try the pid tomorrow
[16:38:49] <skunkworks> *can't do that...
[17:02:33] <Deejay> gn8
[18:26:38] <JT-Shop> Tom_itx, I just sold my first air filter kit
[18:28:15] <XXCoder2> what aur filter?
[18:31:04] <XXCoder2> *air filter kit
[18:47:55] <JT-Shop> for the Spyder
[18:48:31] <XXCoder2> car?
[18:50:27] <JT-Shop> reverse trike
[18:50:56] <XXCoder2> ahh cool :)
[18:53:40] <PetefromTn_> Evening folks.
[18:53:48] <XXCoder2> hey
[18:54:00] <PetefromTn_> hey
[19:07:17] <JT-Shop> hey
[19:08:33] <PetefromTn_> JT-Shop Hey man howsitgoin?
[19:09:02] <JT-Shop> busy busy busy
[19:09:16] <PetefromTn_> sweet,
[19:09:20] <JT-Shop> you?
[19:09:55] <PetefromTn_> Well I am working on a quote right now for what would be a relatively large number of parts for my shop but only one job.
[19:10:06] <PetefromTn_> who knows if I will actually get it or not.
[19:10:21] <JT-Shop> we get about 10% of the quotes we do
[19:10:24] <PetefromTn_> Had a pretty damn stressful day at the other shop unfortunately.
[19:11:14] <PetefromTn_> man that sucks...
[19:11:46] <JT-Shop> I hate that
[19:11:53] <PetefromTn_> This quote is pretty ideally suited to my shops capabilities but there are a LOT of parts to make and not tons of time to do it in.
[19:11:56] <JT-Shop> I had a good time yesterday with my splitter
[19:12:04] <PetefromTn_> nice.
[19:13:11] <PetefromTn_> We had to setup some rather large weldments in the shop today. Getting everything square and plumb was difficult at best.
[19:13:24] <JT-Shop> http://www.gnipsel.com/images/splitter/splitter-28.jpg
[19:13:26] <zeeshan|2> big ass squares!
[19:13:29] <JT-Shop> http://www.gnipsel.com/images/splitter/splitter-29.jpg
[19:14:15] <zeeshan|2> how much does each one of those logs weigh
[19:14:15] <PetefromTn_> We used some rather large angle plates and a bunch of strap clamps.
[19:14:18] <zeeshan|2> in picture 28
[19:14:24] <JT-Shop> I used to have some 8' squares to set the mitter belts in the exhaust of the destroyers I worked on in the shipyard
[19:14:51] <JT-Shop> zeeshan|2, the 18" rounds?
[19:14:55] <zeeshan|2> yes
[19:15:16] <JT-Shop> pretty heavy, I split the bottom one in half before I picked it up
[19:15:26] <zeeshan|2> 100lb+?
[19:15:28] <JT-Shop> 75-100 lbs I guess
[19:15:32] <zeeshan|2> okay
[19:15:40] <zeeshan|2> btw log splitter looks nice.
[19:15:43] <zeeshan|2> i like the finishing touches
[19:15:54] <JT-Shop> until I get my lift built I do the squat lift on the rounds
[19:15:54] <zeeshan|2> the fact your anvil or whatever you call that part
[19:15:56] <JT-Shop> thanks
[19:16:00] <zeeshan|2> is red and looks shiny :D
[19:16:06] <JT-Shop> the wedge
[19:16:14] <zeeshan|2> yes the wedge
[19:16:18] <JT-Shop> built it from scratch
[19:16:27] <zeeshan|2> looks like a lot of work went into that
[19:16:43] <JT-Shop> yea you can say that, and I'm not quite done yet
[19:17:02] <JT-Shop> here is all the photos http://www.gnipsel.com/images/splitter/
[19:17:21] <zeeshan|2> jt-shop
[19:17:25] <zeeshan|2> quick q for you
[19:17:28] <JT-Shop> I'm pretty happy with it
[19:17:28] <zeeshan|2> http://www.gnipsel.com/images/splitter/valve01.jpg
[19:17:43] <zeeshan|2> you see those reinforcement plates inthe column?
[19:17:46] <zeeshan|2> they look like 1/2" plate
[19:17:49] <PetefromTn_> http://imagebin.org/302963
[19:17:50] <zeeshan|2> mayube 3/4"
[19:18:08] <zeeshan|2> when you fillet weld that, what amps is your mig welder set at?
[19:18:17] <zeeshan|2> is it a single pass weld or multipass
[19:18:31] <Connor> PetefromTn_: What the heck is that ?
[19:18:38] <zeeshan|2> also did you start with the edge that is being welded as < or [ ?
[19:18:53] <PetefromTn_> It is a weldment that we built for a local company
[19:18:54] <zeeshan|2> PetefromTn_:
[19:19:00] <JT-Shop> it's a mix of 1/2" 3/4" and 1" for the main plate
[19:19:00] <zeeshan|2> you need a serious welder to weld aluminum plate that thick
[19:19:04] <zeeshan|2> even with preheating
[19:19:15] <zeeshan|2> like a 20,000$ tig welder
[19:19:17] <PetefromTn_> It is kind of a yoke that holds a large cast gear atop a shaft.
[19:19:19] <Connor> That on your machine ?
[19:19:26] <JT-Shop> I welded it in stitches so it didn't warp
[19:19:33] <zeeshan|2> JT-Shop: yes
[19:19:40] <zeeshan|2> but whats the max amperage your welder can weld?
[19:19:47] <PetefromTn_> actually my 350 amp Tig unit can make welds like that.
[19:19:55] <zeeshan|2> b/s
[19:19:57] <PetefromTn_> No it is on the Haas VF8 at work.
[19:19:59] <JT-Shop> 212
[19:20:10] <zeeshan|2> JT-Shop: so then you had to do multiple passes
[19:20:13] <zeeshan|2> to get full penetration
[19:20:18] <PetefromTn_> why BS?
[19:20:19] <zeeshan|2> with a v groove edge?
[19:20:20] <JT-Shop> I have a 180 syncro weld tig
[19:20:29] <zeeshan|2> PetefromTn_: that looks like 1/2" or 3/4" plate
[19:20:33] <PetefromTn_> it is...
[19:20:38] <zeeshan|2> a 350amp welder can only heat up .350"
[19:20:40] <JT-Shop> I didn't go for full penetration
[19:20:45] <zeeshan|2> and with that much metal, itll absorb it
[19:20:48] <zeeshan|2> before aluminum will even melt
[19:20:52] <zeeshan|2> give me your address
[19:20:59] <zeeshan|2> ill send you some 1/2" aluminum plate
[19:21:06] <zeeshan|2> but weld 2 pieces for me and ill give your $100
[19:21:14] * JT-Shop heads inside to relax
[19:21:22] <JT-Shop> say goodnight Gracie
[19:21:25] <zeeshan|2> cu!
[19:21:30] <PetefromTn_> Goodnight JT.
[19:21:44] <zeeshan|2> 100$ on the table!
[19:22:26] <PetefromTn_> Connor Hey man I was paying a lot of attention to the Haas Toolchange cycle today to see how things work there.
[19:22:31] <zeeshan|2> PetefromTn_: what 350amp tig welder do you have?
[19:22:57] <Connor> PetefromTn_: Cool
[19:23:20] <PetefromTn_> I saw yout post to the mailing list about that guy also doing the toolchange.
[19:23:24] <PetefromTn_> changer.
[19:23:48] <PetefromTn_> Seemed like they kinda ignored ya unfortunately. I think you have some interesting things to contribute in that regard.
[19:24:28] <PetefromTn_> He was trying to do the whole thing in python apparently?
[19:24:59] <Connor> PetefromTn_: Yes.. as was I.
[19:25:23] <Connor> I started off with o-word, then was going to try python.. but.. it was too damn complicated, so went back to o-word
[19:25:27] <PetefromTn_> But you are not now right you are using the O command.s
[19:26:31] <PetefromTn_> I still need to get the encoder mount machined.
[19:27:06] <PetefromTn_> I like working in this shop but it makes getting anything done in my shop difficult and time consuming.
[19:27:56] <zeeshan|2> PetefromTn_: why talk shit!
[19:28:01] <zeeshan|2> and not back it up
[19:28:18] <Connor> PetefromTn_: I know that feeling.. I work from home.. feel like I can never get anything..
[19:28:20] <Connor> done
[19:29:23] <PetefromTn_> It sure is nice to work from home isn't it. I hope I can get this job I am quoting now here it would be a GOOD start to this thing for me.
[19:36:38] <humble_sea_bass> capital good is always good
[19:49:51] <PetefromTn_> Well Goodnight folks.
[20:18:43] <Connor_iPad> zeeshan|2: On ipad now.
[20:38:29] <R2E4_> uh oh. I canturnthe spindle by hand. It wasnt doing that before.
[21:32:02] <zeeshan|2> Connor: sorry gf called me
[21:32:02] <zeeshan|2> hehe
[21:51:55] <zeeshan|2> slowly getting organized
[21:51:56] <zeeshan|2> :D
[21:51:56] <zeeshan|2> http://imgur.com/a/ZShfB
[21:57:09] <XXCoder2> nice
[22:01:33] <zeeshan|2> as i was going through the tools, i found 1"-14 tap
[22:01:38] <zeeshan|2> not sure how the hell ill ever tap that by hand
[22:01:56] <XXCoder2> oh thats easy. use your hand to get tool lol
[22:02:33] <zeeshan|2> its hard
[22:02:38] <zeeshan|2> even 3/8 npt tap is kind of hard
[22:02:44] <zeeshan|2> 1/2npt i always had to power tap on lathe
[22:02:53] <zeeshan|2> but i guess taper taps require more force
[22:03:15] <XXCoder2> wouldnt know, I'm still a noobie lol
[22:05:53] <toastyde2th> big ass tap wrench and a center in a drill press
[22:06:06] <toastyde2th> generally I don't want to drive taps that big by hand
[22:06:35] <zeeshan|2> my biggest tap wrench is 24" in length
[22:06:46] <zeeshan|2> but its a pain in the but to use on a lathe or mill
[22:06:49] <zeeshan|2> cause it'll always hit the machine
[22:07:13] <zeeshan|2> hopefully soon i can power tap
[22:07:51] <zeeshan|2> i need help identifying two things:
[22:08:11] <zeeshan|2> http://i.imgur.com/iH4EkQb.jpg?1
[22:08:15] <zeeshan|2> in that drawer
[22:08:22] <zeeshan|2> what is that left most tool
[22:08:26] <zeeshan|2> with the weird teeth
[22:09:12] <XXCoder2> that looks familiar with special drill that extracts broken off screws but that big
[22:09:15] <toastyde2th> in the front of the drawer? next to the tap wrenches?
[22:09:29] <toastyde2th> or the back of the drawer with the spiral flutes
[22:09:29] <zeeshan|2> yes
[22:09:35] <toastyde2th> that's a chain breaker
[22:09:37] <zeeshan|2> front of drawer
[22:10:06] <XXCoder2> yeah looks same as small one for bicycle. pushes pin out of chain
[22:10:10] <zeeshan|2> LOL
[22:10:11] <zeeshan|2> are you serious
[22:10:17] <zeeshan|2> i didnt even know a tool for that existed
[22:10:27] <zeeshan|2> i've always driven the pins out with a punch
[22:10:44] <XXCoder2> its very effective. you dont need to completely remove pin.
[22:10:44] <zeeshan|2> so i guess thes works for a specific chain pitch
[22:11:21] <XXCoder2> whats the weird bits at back of drawer? lol
[22:11:33] <zeeshan|2> in that same one XXCoder2?
[22:11:36] <toastyde2th> XXCoder2, they're tap extractors
[22:11:38] <zeeshan|2> those are screw extractors
[22:11:44] <toastyde2th> spefically, ez outs
[22:11:45] <XXCoder2> guessed I was right lol
[22:11:51] <toastyde2th> they're pretty bad at what the do
[22:11:53] <XXCoder2> but damn that big
[22:11:54] <toastyde2th> *they
[22:11:59] <zeeshan|2> they work pretty good!
[22:12:10] <zeeshan|2> the square ones work better
[22:12:12] <toastyde2th> i always preferred a tap burner
[22:12:12] <zeeshan|2> thats whats in the black box
[22:12:30] <zeeshan|2> whats a tap burner
[22:13:13] <toastyde2th> it's an EDM for anything stuck in a hole
[22:13:16] <XXCoder2> comparing your toe with secrew extractor its in least 3/4 in thick
[22:13:18] <zeeshan|2> haha
[22:13:20] <toastyde2th> long, thin titanium electrodes
[22:13:28] <zeeshan|2> http://www.matcotools.com/ProductImages/SE25.jpg
[22:13:31] <zeeshan|2> those are whats in the black box
[22:14:09] <zeeshan|2> http://imgur.com/bXtnBaW
[22:14:10] <zeeshan|2> whats this
[22:14:14] <zeeshan|2> it says 5.00mm on it
[22:15:34] <toastyde2th> measure the thickness
[22:15:42] <toastyde2th> might be a 5mm block
[22:15:53] <zeeshan|2> thats weird to be a gauge block
[22:16:00] <zeeshan|2> cause the faces arent polished
[22:16:00] <toastyde2th> not a gage block
[22:16:10] <zeeshan|2> oh you mean a shim for a 123 block?
[22:16:10] <toastyde2th> just a regular ass 1-2-3 or parallel-style thing
[22:16:12] <Groguard> Hello, I had a couple question on end mills. What would be the best option for cutting wood quickly and cleanly? Looking for 1/8"h shank 1/8"cut.
[22:16:12] <toastyde2th> yeah
[22:16:22] <zeeshan|2> that makes sense
[22:16:26] <zeeshan|2> lemme check it w/ a 123
[22:16:48] <toastyde2th> zeeshan|2, seeing as how 5mm is metric and 123s are imperial
[22:16:54] <toastyde2th> i doubt you'll find a good match, but good luck
[22:17:08] <toastyde2th> Groguard, try to find a router bit
[22:17:15] <toastyde2th> single or double flute
[22:17:18] <toastyde2th> single preferred
[22:17:29] <toastyde2th> not an endmill
[22:17:37] <Groguard> That would be good for fast milling?
[22:18:01] <toastyde2th> any milling in wood
[22:18:16] <toastyde2th> slow, fast
[22:18:19] <Groguard> Im looking more for a little less chipping, Im currently using a single flut up cut 1/8
[22:18:42] <toastyde2th> you could try a tool with a little less rake but chipping is sort of going to occur
[22:18:56] <toastyde2th> could also try polishing the blade on the bit
[22:19:39] <Groguard> Im trying to cut some fine wood parts, 3.175mm channels with only a few mm's on each side and its destroying those side parts
[22:19:52] <toastyde2th> what kind of wood
[22:19:57] <Groguard> red oak
[22:20:10] <toastyde2th> are you climb milling
[22:20:21] <Groguard> excuse me, its pine
[22:20:32] <toastyde2th> okay, you're probably fucked even with this suggestion given that it's pine
[22:20:36] <toastyde2th> BUT
[22:20:41] <toastyde2th> get a smaller endmill
[22:21:01] <Groguard> Yea, I figured that would be my only option, just wanted another opinion
[22:21:06] <toastyde2th> rough out all your material - when you go to do your finish pass, climb mill it
[22:21:30] <zeeshan|2> toastyde2th: it fits with a 123 block
[22:21:35] <zeeshan|2> the holes line up
[22:21:44] <zeeshan|2> and you were right its 5mm spacer
[22:21:50] <toastyde2th> well there you go, a compleltely useless distance to stack on top of a 123 block
[22:21:55] <Groguard> Alright, thanks. I will give it shot!
[22:21:56] <toastyde2th> but there it is
[22:21:57] <zeeshan|2> haha exactly
[22:23:33] <zeeshan|2> googling 123 block shim/spacer
[22:23:36] <zeeshan|2> doesnt really come up with much
[22:23:49] <zeeshan|2> it'd be nice to have a small kit that has various size shims premchined
[22:25:49] <toastyde2th> in practice even 123's don't get used a lot
[22:26:02] <zeeshan|2> i use them quite a bite
[22:26:02] <toastyde2th> at every shop i've worked at we had a bunch of blocks we cut and ground to size ourselves
[22:26:06] <zeeshan|2> because thats all i have
[22:26:25] <zeeshan|2> i have a 8 pair 123 block set for welding
[22:26:28] <zeeshan|2> works really well
[22:26:47] <zeeshan|2> http://www.desicodesign.com/meerkat/images/11-11-01_123_blocks.jpg
[22:27:32] <zeeshan|2> sometimes i clamp em onto the edge of say a 1x1" square tubing
[22:27:37] <zeeshan|2> so i can line up the sheet properly
[22:27:52] <zeeshan|2> that sits on top of a 1x1 square tubing frame
[22:29:14] <toastyde2th> nice
[22:29:32] <toastyde2th> we cut a bar of mild steel up into chunks
[22:29:39] <toastyde2th> 1", 2", 3", 5" etc
[22:29:44] <toastyde2th> and drilled a hole through them
[22:29:48] <toastyde2th> then ground em all flat and parallel
[22:29:51] <zeeshan|2> that sounds more useful
[22:30:02] <toastyde2th> so we can stack them up and stick a hold down bar through them
[22:30:10] <zeeshan|2> one day i will own a surface grinder!
[22:30:12] <toastyde2th> the holes are countersunk on both sides
[22:30:22] <toastyde2th> so the nut sits completely out of the way of the surface
[22:31:42] <zeeshan|2> toastyde2th: is a surface grinder a common thing at most machine shops?
[22:31:48] <toastyde2th> super common
[22:31:56] <toastyde2th> most places have at least one
[22:32:03] <zeeshan|2> is that the main grinder they use for surfacing
[22:32:09] <zeeshan|2> or are there other types of grinders?
[22:32:14] <toastyde2th> what do you mean by "main grinder"
[22:32:21] <toastyde2th> there are many types of grinders
[22:32:34] <zeeshan|2> well at my old work, we had lathe type grinders for cyndrical stuff
[22:32:50] <zeeshan|2> but i'd think that is more speciality
[22:32:55] <zeeshan|2> unless you were grinding dowels or something
[22:33:15] <zeeshan|2> i guess what im asking is what are the most used grinders
[22:33:17] <toastyde2th> for flat surfaces you generally have two types of grinder, horizontal spindle grinders (surface grinders) and vertical spindle grinders (blanchard grinders)
[22:33:22] <zeeshan|2> like mill and lathes are common in a machine shop
[22:33:25] <zeeshan|2> what are common grinders
[22:33:42] <toastyde2th> the reciprocating surface grinder is the super-common one
[22:34:09] <toastyde2th> if the shop does any volume of grinding at all, they'll also have at least one blanchard grinder
[22:34:17] <zeeshan|2> that blanchard grinder looks like the none that polished up the hubble telescope mirrors
[22:34:22] <zeeshan|2> none = one
[22:34:35] <zeeshan|2> basically a rotating table
[22:34:39] <toastyde2th> they're extremely different, despite the apparent similarity
[22:35:40] <tjtr33> blanchard grinders are common too
[22:36:07] <zeeshan|2> looks like blanchard grinders have a huge footprint compared to a regular surface grinder
[22:36:13] <toastyde2th> correct
[22:37:10] <toastyde2th> also are quite dangerous
[22:37:25] <zeeshan|2> why
[22:38:08] <toastyde2th> they're high-horsepower grinders with wheels spinning vertically
[22:38:17] <toastyde2th> so when they grab things, they really send them flying
[22:38:29] <toastyde2th> that's why you see the really thick case around the work area
[22:38:47] <zeeshan|2> ah
[22:38:49] <toastyde2th> and not some shitty thin glass thing that you'd see on a machining center or lathe
[22:38:58] <toastyde2th> you can buy 300, 400 hp blanchard grinders
[22:39:18] <toastyde2th> and their downfeed is very hard
[22:39:41] <zeeshan|2> http://www.practicalmachinist.com/vb/attachments/f44/93873d1387025845-blanchard-grinding-blanchard-grinder-96.jpg
[22:39:48] <zeeshan|2> leaves a nice fancy grinding pattern
[22:41:11] <XXCoder2> wow nice
[22:41:39] <toastyde2th> not a really good general purpose grinder, though
[22:41:54] <zeeshan|2> would be nice to have to resurfacing flywheels
[22:41:54] <zeeshan|2> :D
[22:43:19] <zeeshan|2> thanks for the tool identification
[22:43:20] <zeeshan|2> bed time!
[22:43:56] <toastyde2th> np, goodnight