Back
[00:00:49] <Aero-Tec> I used that scope hard for a good number of years in my TV repair shop, time for a new one
[00:00:57] <archivist> I used an manky 30 year old scope to get mine running top right
http://www.archivist.info/cnc/stage8/IMG_0268.JPG
[00:03:29] <Aero-Tec> mine was running for several years, not it is going nuts
[00:04:26] <Aero-Tec> yes it is good enough to get running, but is it good enough to hunt down pesky problems in a digital work
[00:04:33] <Aero-Tec> not so much
[00:04:47] <Aero-Tec> my CNC was running
[00:05:23] <Aero-Tec> oops now not not
[00:05:52] <archivist> I had pesky noise problems when I added the B axis and the old scope found that too
[00:06:55] <archivist> the signal was on the edge of the right level and the vfd was effecting it, pulled it up, fixed
[00:07:30] <Aero-Tec> the only signal problem I can find when it is acting up is the encoders
[00:09:04] <Aero-Tec> but if the motors are doing wild things then I guess they will mess with the sig steam of the encoders
[00:10:48] <Aero-Tec> yes old scopes can find nose, and they can do tons of things, but they can also miss some things if they happen fast enough
[00:10:58] <Aero-Tec> noise
[00:12:28] <jymmm> pcw_home: ping
[00:12:34] <Aero-Tec> long and the short of it is my scope is just plain wore out
[00:13:13] <jymmm> pcw_home: I know you're just using what's available,but check this out...
http://www.ntop.org/products/pf_ring/libzero-for-dna/
[00:13:15] <Aero-Tec> it is still good for a few things, much better then nothing, but still need replacing
[00:20:29] <Aero-Tec> what do you guys think of this one?
[00:20:33] <Aero-Tec> http://www.ebay.com/itm/ATTEN-ADS1102CAL-100M-Hz-1G-Digital-Oscilloscope-7-LCD-/140469625191?pt=BI_Oscilloscopes&hash=item20b4a46167
[00:21:54] <alex4nder> Aero-Tec: if you're going to get a 'cheap' scope, get one that runs Linux
[00:22:21] <Aero-Tec> ?
[00:22:38] <Aero-Tec> what would you suggest?
[00:23:37] <alex4nder> if you want that style of scope, Hantek
[00:24:56] <alex4nder> /tekway/voltcraft
[00:24:58] <Aero-Tec> just looking for one that was stand alone, not PC based
[00:25:11] <alex4nder> these are standalone.. the scope runs Linux
[00:25:12] <Aero-Tec> and would not cost lots
[00:26:12] <Valen> people seem to like these
http://www.dealextreme.com/p/rigol-ds1052e-5-6-tft-lcd-50mhz-2-channel-digital-color-storage-oscilloscope-30573?item=4 or something similar
[00:36:21] <Aero-Tec> looks like I could up grade my scope real cheap
[00:36:42] <Aero-Tec> some very nice scopes for cheap
[00:36:55] <Aero-Tec> 100 to 500 mhz
[00:44:13] <Aero-Tec> off to bed
[00:44:15] <Aero-Tec> night
[00:57:17] <alex4nder-> ner.
[02:02:35] <DJ9DJ> moin
[02:56:02] <jymmm> archivist:
http://www.youtube.com/watch?v=nUnnu8u_NHs&NR=1
[03:03:48] <alex_joni> going underground?
[03:04:35] <jymmm> alex_joni: It was issued to people during the war. Someone restored one. Thought it was interesting.
[03:06:04] <jymmm> alex_joni: As far as underground, not yet. Just got me some of this today and took forever to mix. Doing a test run and waiting for it to dry now.
http://www.rutland.com/productinfo/black-furnace-cement.html
[03:16:23] <Valen> they were actually suprisingly effective
[03:16:41] <Valen> for anything short of a direct hit
[03:17:36] <Valen> 300 hours to restore it
[03:17:46] <Valen> be easier to just get some new gal sheet rolled lol
[03:18:26] <jymmm> or just put out ads to gather parts that ppl still have
[03:19:28] <Valen> they would all probably be in about the same condition
[03:19:51] <jymmm> Maybe, but might be able to piece some together
[03:19:56] <Valen> its just a couple of rolled gal sheets and some ends with holes cut
[03:20:13] <Valen> probably cost ~$200-400 to order from a sheet metal place
[03:20:56] <jymmm> Not sure about that. Just plastic/fiberglass corrigated 16ft long are like $40 EACH these days. I can only imagine what galv would cost
[03:21:08] <Valen> cheaper
[03:21:21] <jymmm> maybe =)
[03:21:42] <jymmm> Man, I hope this stove cement works, this would be cool
[03:22:19] <Valen> do keep in mind they arent full sheets
[03:22:27] <Valen> a sheet is 2.1 x 1.2 meters
[03:22:45] <Valen> and you can get it in whatever lengths you want these days
[03:22:51] <Valen> its all continious extrusion
[03:23:56] <Valen> http://www.getprice.com.au/Corrugated-042mm-Zincalume-Gpnc_550--31104394.htm
[03:24:05] <Valen> $14 s ,eter
[03:24:07] <jymmm> Hell you can't even get a lot of things thee days. I had to goto two Ace Hardware stores just for the stove cement becasue the big orange and blue borgs dont carry it!
[03:25:00] <Valen> eh order it from the interweb ;->
[03:25:29] <jymmm> fuck that! shipping is too expensive. It was only $3.99 for a 1lb tub of it
[03:26:48] <jymmm> I only mail order if it's reasonable to do so
[03:27:12] <jymmm> flat rate/free shipping, ok.
[03:34:30] <archivist> and how much did the fuel cost getting your arse to the stores
[03:38:14] <Loetmichel> mornin
[03:43:03] <jymmm> archivist: Not much actually, as I've been planning my running around all at once =)
[03:46:48] <jymmm> Which was is better for high heat teflon or fiberglass coated lead wire?
[03:47:06] <jymmm> -was
[03:47:08] <archivist> lead melts
[03:47:29] <jymmm> leeeed, not the soft metal
[03:47:41] <jymmm> wire leads
[03:48:06] <jymmm> I see fiberglass coated wire leads inside coffee makers
[03:49:08] <jymmm> PTFE is only rated to 200C
[03:49:16] <Valen> glass
[03:49:30] <Valen> teflon is just plastic
[03:50:03] <jymmm> ok! fiberglass is rated around 482c/900f
[08:17:56] <tjb1> Went to work today…they aren't working today...
[08:19:37] <jymmm> then why did you go?
[08:20:50] <ReadError> lol
[08:25:07] <tjb1> The boss told me to go today
[08:25:15] <tjb1> apparently the guy at the warehouse doesn't work today...
[08:25:29] <tjb1> Separation of boss and actual workers.
[08:26:50] <JT-Shop> is it a holiday today?
[08:27:03] <tjb1> No they are just slow still
[08:27:16] <tjb1> I work at warehouse and they haven't started getting the containers yet
[08:29:03] <JT-Shop> I'm bored... I think I'll finish the conversion of the BP knee mill to emc
[08:30:13] <tjb1> Have fun
[08:30:27] <JT-Shop> but first and more important is to take my girlfriend for a walk in the woods so she can chase things
[08:31:22] <tjb1> No living animals in the woods here
[08:32:08] <jymmm> that would be called a cemetery, not woods.
[08:32:42] <tjb1> Seems like every hunter in the surrounding states come to my county to hunt
[08:32:49] <jymmm> ah
[08:33:53] <jymmm> I really would have thought gold made a better thermal conductor than copper for some reason.
[08:48:46] <tjb1> They need to make a CAD program that pulls drawings from your brain
[08:49:14] <jymmm> lol
[08:49:53] <tjb1> then make a CAM program that does it by itself
[08:50:07] <tjb1> Here is the part, here is what I have now…make it
[08:50:25] <jymmm> tjb1: Ever see the movie "Weird Science"?
[08:50:33] <tjb1> Don't think so
[08:51:07] <jymmm> tjb1: Two kids make a woman from their imagination
[08:51:56] <tjb1> how did that work out
[08:52:12] <jymmm> pretty good actually =)
[08:59:17] <tjb1> A 1/2-10 5 start is 2 turns per inch right?
[09:36:04] <tjb1> Anyone in here running a plasma table?
[09:38:32] <micges> few for sure, ask away
[09:39:22] <tjb1> Building my initial height sensor
[09:39:55] <tjb1> http://pastebin.com/raw.php?i=7eNXFaYc
[09:41:13] <tjb1> The way it works now, if I use any Z settings in the controller it will mess it up
[09:53:43] <skunkworks> tjb1: why don't you do it within hal?
[09:54:00] <tjb1> No idea how
[09:54:15] <tjb1> and not 100% sure yet that I will be using Linux
[10:02:56] <JT-Shop> tjb1: yes
[10:03:21] <tjb1> JT-Shop: Do you have a THC setup?
[10:03:28] <JT-Shop> yes
[10:03:50] <JT-Shop> what are you trying to do with the code?
[10:04:48] <tjb1> I know you are gonna say about a hal program, but is there another way to disable THC when program speed lowers?
[10:04:53] <jymmm> TAKE OVER THE WORLD!
[10:04:56] <tjb1> Like a simple output on the parallel port
[10:07:17] <JT-Shop> what kind of THC do you have?
[10:07:55] <jymmm> T H C !!!
[10:09:06] <tjb1> diy arduino
[10:10:21] <skunkworks> http://www.cnczone.com/forums/stepper_motors_drives/151504-diy_torch_height_control.html
[10:10:47] <tjb1> Thats my thread :)
[10:10:52] <tjb1> http://pastebin.com/raw.php?i=piCtWQuv
[10:11:01] <tjb1> Thats the arduino code for the height control
[10:12:30] <JT-Shop> so you made an up down THC?
[10:12:53] <tjb1> I have it working on an old stepper motor from a printer
[10:13:16] <tjb1> I still need a way to prevent it from diving on corners
[10:13:44] <tjb1> I can make it not drop after the voltage drops below a certain point
[10:14:07] <JT-Shop> I use a couple of hal pins to compare actual velocity to commaned velocity
[10:14:15] <JT-Shop> but you don't like hal I see
[10:15:00] <tjb1> Not that I don't like it, don't know how to use it :)
[10:15:22] <tjb1> I did read a bit into the manual but didn't get much of it
[10:16:01] <JT-Shop> how are you controlling the Z axis?
[10:16:34] <JT-Shop> JT-Shop> so you made an up down THC?
[10:16:47] <JT-Shop> from the ardunio thingy
[10:17:11] <tjb1> When I have the gecko, its going to intercept the z step and direction
[10:17:18] <tjb1> right now I have to use a motor controller
[10:17:45] <jymmm> JT-Shop: Got a 1LB tub for $3.99 yesterday =)
http://www.rutland.com/productinfo/black-furnace-cement.html
[10:18:01] <tjb1> Each time this is activated -> motor.step(1, BACKWARD, DOUBLE); it steps 1 time in the direction
[10:18:01] <JT-Shop> so your ardunio will handle all the software functions of the THC?
[10:18:29] <JT-Shop> good score jymmm
[10:18:35] <tjb1> Im run the 50:1 voltage through it
[10:18:45] <tjb1> and the range in the current program is 450-550
[10:18:52] <JT-Shop> your tip voltage?
[10:18:56] <tjb1> which is about 3.5v after the division
[10:19:07] <tjb1> Yes
[10:19:11] <JT-Shop> I see
[10:20:00] <JT-Shop> would it be simpler to just have the ardunio output two pins, Up and Down?
[10:20:12] <tjb1> As inputs into Linux?
[10:20:18] <JT-Shop> aye
[10:21:19] <JT-Shop> somehow you have to tell the ardunio what is center and how wide it is
[10:21:56] <tjb1> center?
[10:22:07] <JT-Shop> target voltage
[10:22:20] <tjb1> I have a potentiometer hooked up now
[10:22:42] <JT-Shop> hooked up to?
[10:23:00] <tjb1> 5v, ground, and an analog input
[10:23:16] <tjb1> Well its not set up yet but thats how I will adjust the range
[10:23:23] <JT-Shop> to the ardunio as the target voltage input?
[10:23:37] <tjb1> Yes, it will adjust from 0 - 1023
[10:24:00] <JT-Shop> sounds simple enough
[10:24:18] <tjb1> I have to run the tip voltage through a voltage divider before the arduino still
[10:24:19] <tjb1> It can't handle 7v
[10:24:40] <JT-Shop> are you pushing rope up hill because you don't know about the THC-Up-Down comp?
[10:25:29] <tjb1> Ha what?
[10:25:39] <skunkworks> JT-Shop: does that comp take into account current and commanded velocity to?
[10:25:49] <skunkworks> Too
[10:26:22] <JT-Shop> it's the same one I use but takes up/down inputs instead of volts converted to velocity
[10:27:11] <JT-Shop> so if you have a pot for target voltage and the ardounio has two pins up and down the rest could be handled by the comp
[10:27:30] <JT-Shop> tjb1:
http://gnipsel.com/shop/plasma/plasma.xhtml
[10:27:49] <tjb1> I would also need to compare voltage in the arduino?
[10:27:57] <JT-Shop> skunkworks: current is set at the plasma torch
[10:28:24] <JT-Shop> aye, that is how you determine if you want up, down or nothing
[10:28:41] <tjb1> Sounds simple enough
[10:28:47] <JT-Shop> skunkworks: yes it takes in commanded velocity
[10:28:52] <tjb1> Does that take into account corners?
[10:29:02] <tjb1> Guess you just answered that
[10:29:14] <JT-Shop> doesn't care about corners if your velocity is within range
[10:29:31] <tjb1> How does that touch off work?
[10:29:58] <JT-Shop> I have a floating head with a micro switch connected to the motion.probe-input
[10:30:37] <tjb1> You just set the distance between the torch and the micro switch and then add the pierce height?
[10:30:43] <JT-Shop> when I run that it sets my Z0 using G92
[10:31:31] <JT-Shop> the G1 Z0.070 (Raise up to Z0) line gets the torch to Z0 after the probe move
[10:32:01] <JT-Shop> then the G92 sets that point as Z0
[10:32:07] <tjb1> So you have .07 between the torch touching the material and setting the switch off?
[10:32:10] <JT-Shop> then it raises up to pierce height
[10:32:32] <JT-Shop> aye that is how far it moves to trip the micro switch
[10:32:50] <tjb1> and the computer will handle the corner lock by ignoring the inputs?
[10:33:16] <JT-Shop> yep
[10:33:26] <JT-Shop> this shows the floating head
http://i47.photobucket.com/albums/f163/johnplctech/Plasma%20Cutter/P1010030.jpg
[10:33:35] <JT-Shop> and the touch off switch is the lower one
[10:33:58] <skunkworks> you would think that it would be better to offset the voltage when the velocity slows down - so it it still does torch hight compinsation.. too hard?
[10:34:30] <tjb1> Thanks JT-Shop, guess I will move this direction
[10:34:37] <JT-Shop> not possible to control voltage directly only with distance and speed
[10:35:00] <tjb1> That back switch your home and limits?
[10:35:08] <JT-Shop> aye
[10:35:22] <tjb1> Thats some thick stock
[10:35:26] <JT-Shop> more photos in the album
[10:35:29] <tjb1> 1"?
[10:35:40] <JT-Shop> that is a slat :)
[10:35:51] <JT-Shop> might be 11ga material on top of it
[10:36:02] <skunkworks> right - I am saying - in the comp - as the velocity slows down around a corner - the comp knows that the voltage will change x amount to automatically compinsate.
[10:36:10] <skunkworks> (if that made sense)
[10:37:02] <skunkworks> so instead of locking the torch head - it keeps adjusting at the same hight.
[10:37:18] <skunkworks> maybe I am not understanding..
[10:37:19] <tjb1> It will only lock on tight radiuses right?
[10:37:22] <JT-Shop> the comp controls the voltage with the distance from the tip to material but when you get too slow you don't want the tip to dive into the material so you don't adjust
[10:37:35] <tjb1> Large circles it shouldn't slow down enough
[10:37:56] <JT-Shop> it doen't care what kind of move your doing it only compares the actual velocity to the commanded velocity
[10:38:40] <tjb1> Does that comp require the arc start and motion start inputs?
[10:38:50] <skunkworks> right - because when it slows down - the voltage increases.. So the comp would see the the velocity slow down and the voltage increase to still keep adjusting the torch hight automagically
[10:38:53] <JT-Shop> the acceleration of your XY axes control when it can't move very fast
[10:39:25] <JT-Shop> skunkworks: yea and you don't want that or the tip will dive into the material trying to fix the voltage
[10:40:00] <JT-Shop> tjb1: it uses the arc ok output from the plasma torch
[10:40:11] <skunkworks> does the voltage go off scale?
[10:40:21] <tjb1> so the torch_on bit is in the software
[10:40:56] <JT-Shop> the torch on is M3 with S>0
[10:41:04] <JT-Shop> torch off is M5
[10:41:15] <JT-Shop> skunkworks: what do you mean?
[10:41:21] <tjb1> So it just requires the torch up and down and then arc ok
[10:41:34] <JT-Shop> iirc yea
[10:41:41] <tjb1> Which will be going to the computer anyway...
[10:41:49] <skunkworks> heh - I must be thinking wrong. ;)
[10:42:00] <tjb1> Doesn't the DB25 only have 2 inputs?
[10:42:19] <JT-Shop> do you mean the parallel port?
[10:42:40] <tjb1> Yeah
[10:43:04] <JT-Shop> depends on how you configre it if the hardware allows
[10:43:16] <JT-Shop> http://linuxcnc.org/docview/html/hal/parallel_port.html
[10:43:28] <tjb1> I would need motion start, torch up and torch down which is 3
[10:44:16] <JT-Shop> so 2 spare if you configure the parallel port as output
[10:44:24] <JT-Shop> see diagram in link
[10:46:04] <tjb1> 10-13, 15?
[10:47:08] <JT-Shop> sounds about right
[10:48:00] <tjb1> So I'm assuming the G540 uses pins 2-9
[10:49:29] <tjb1> Yep, 2-9
[10:50:04] <tjb1> 10-13 are inputs, 1, 17 are outputs
[10:56:17] <tjb1> Thanks JT
[11:00:49] <JT-Shop> your welcome
[11:04:40] <jdhNC> anyone use ngcgui for pocketing, or just straight moves for squaring stock?
[11:07:03] <Connor> I used it to do a db9 the other night.. I had to modify the db25 for it.. found that it doesn't take into account tool diameter though.
[11:07:30] <jdhNC> it doesn't?
[11:07:38] <Connor> well.. the db25 didn't.
[11:07:52] <Connor> Not sure if it's suppose to read the tool table or what..
[11:07:57] <jdhNC> how fast will your spindle turn?
[11:07:58] <Connor> I need to research a little and find out.
[11:08:12] <Connor> I can get it up to 3k RPM
[11:08:21] <Connor> well.. 3.1k
[11:08:27] <Connor> but, limited it to 3k.
[11:09:01] <Connor> I wouldn't run it over 2500 with the stock bearings for any length of time..
[11:09:24] <jdhNC> I did the same. Pretty nifty. Not sure I like the spindle estop there though. Guess it would be ok with spindle feedback.
[11:10:20] <Connor> I put my primary estop below the control box. Of course, mine isn't mounted to the side of the spindle anymore.
[11:10:43] <jdhNC> did you keep the chip guard? I semi-crushed mine last night on the vise
[11:11:00] <Connor> Heck no. I pulled that thing off Day 1.
[11:11:37] <tjb1> JT-Shop: Can I run the torch up, down, and arc ok through the g540 to the computer?
[11:11:40] <Connor> and bypassed it. The wire for which BTW, can be used with a really for a failsafe
[11:11:58] <Connor> it's on the AC side of the circuit.
[11:12:21] <jdhNC> my spindle almost starts up when you start linuxcnc until the relay kicks in
[11:12:53] <JT-Shop> tjb1: I guess if the g540 has inputs connected
[11:13:04] <tjb1> It has 4 inputs and 2 outputs
[11:13:05] <alex4nder-> hey
[11:13:07] <JT-Shop> btw ngcgui has a pocket routine
[11:13:25] <tjb1> Ill use 1 input for the homing switches, then the other 3 for the thc connections
[11:13:35] <jdhNC> JT: oh, I didn't see that one in there.
[11:14:19] <Connor> Yea. I have the same issue. or DID. I'm setting it up so my break out board is disabled, and I have a push-to-start button. When I push it.. enables the BOB, allowing the BOB to send the signal to charge pump to latch the relay.
[11:16:09] <jdhNC> the 'enable' relay on the spindle control?
[11:16:27] <Connor> No. The one on the charge pump.
[11:16:45] <jdhNC> maybe I should get one.
[11:16:52] <Connor> http://cnc4pc.com/Tech_Docs/1_C1GR4_1.pdf
[11:17:12] <JT-Shop> seems like DB25 does use the tool diameter from the tool table
[11:17:23] <jdhNC> JT: it does
[11:17:38] <Connor> Yea.. but, I had my tool set at .125 and it didn't take it into account ??
[11:18:55] <Connor> I'm adding DPDT relay to the mix though. 1 side will be for 12v to handle the failsafe spindle relay (and 12v to switch the reversing relay at some point) other side will be to handle the enable/disable of the stepper drivers.
[11:18:56] <jdhNC> it has g41/g42 in the iquad thing it uses
[11:19:19] <Connor> what is g41 / g42 ?
[11:19:30] <jdhNC> I can't come up with a good scenario for usign stepper enable/disable
[11:19:50] <Connor> jdhNC: I can. No heat when your machine is off..
[11:20:06] <Connor> I don't want the drivers being driven while I have the machine OFF
[11:20:16] <jdhNC> I guess. I was going to kill A/C
[11:20:56] <jdhNC> g41/42 are cutter radius compensation
[11:21:14] <jdhNC> mine get hot. These drives are not very bright.
[11:21:34] <Connor> I'll have to go back and look and see why I had to manually account for the diamater.
[11:21:49] <Connor> I just added a new variable for tool diamater.
[11:21:58] <Connor> I don't always use the tool table.
[11:31:54] <tjb1> Anyone know of a PCB service that will make 1 for like 20-30?
[11:36:27] <Loetmichel> tjb1: double sided?
[11:36:37] <tjb1> no
[11:36:50] <Loetmichel> make it yourself
[11:36:58] <Loetmichel> its not THAT complocated
[11:37:04] <Loetmichel> compicated
[11:37:48] <tjb1> You talking about the printing it on to something then ironing that on and then acid burning the copper off?
[11:38:21] <Loetmichel> s/ironing/expose to UV/develop
[11:38:25] <Loetmichel> but yes
[11:38:50] <tjb1> Well I found a company that will do boards under 30^2 for $17
[11:39:39] <tjb1> Well…$35 shipping fee, I don't think so
[11:39:41] <Loetmichel> i know companys that will do std double sided 100mm*160mm for less than 5 Eur.
[11:39:54] <Loetmichel> with print and solder resist
[11:40:08] <Loetmichel> but only if you order 50 identical
[11:40:18] <tjb1> I don't need 50 :)
[11:40:51] <Loetmichel> for a single piece its most of the time simpler to do it yourseld, especially if its single sided
[11:41:14] <Loetmichel> if you have a CNC mill its even simpler
[11:41:15] <tjb1> Ill probably end up using protoboard
[11:41:25] <Aero-Tec> one can use oxygen peroxide I heard
[11:41:40] <Loetmichel> Aero-Tec: not solely
[11:41:58] <tjb1> I might spring for a board around $30
[11:42:06] <tjb1> but I will just protoboard it
[11:43:30] <Loetmichel> there are two/three comm en used etching agents for copper PCB board: FerrousIIIcloride, sodium persulfate and for the impatioent: a mix of HCl and H2O2
[11:44:00] <jdhNC> and carbide
[11:44:18] <Loetmichel> carbide?
[11:44:35] <Aero-Tec> I used to make boards but used ferric chloride to etch it
[11:45:04] <jdhNC> carbide bit in the mill.
[11:45:12] <Loetmichel> i have used sodium persulfate
[11:45:21] <Loetmichel> ....before i got the mill ;-)
[11:45:26] <Loetmichel> jdhNC: ok ;-)
[11:45:47] <Aero-Tec> high speed dril bit will work but will not last, but they are real cheap
[11:46:16] <Aero-Tec> I would just make it myself
[11:46:24] <Aero-Tec> and no CNC needed
[11:46:57] <Aero-Tec> I made ton of boards with just a dremal drill press
[11:47:15] <Loetmichel> Aero-Tec: fot the isolation milling
[11:47:19] <Loetmichel> NOT the drilling
[11:47:56] <Aero-Tec> you etch the boards that will center punch the copper and tell you where to drill
[11:47:57] <Loetmichel> http://bambuser.com/v/2639261
[11:48:13] <Loetmichel> result:
http://www.cyrom.org/palbum/main.php?g2_itemId=13197
[11:48:25] <Aero-Tec> your right
[11:48:38] <Aero-Tec> but then you do not etch
[11:48:41] <Loetmichel> right
[11:48:58] <Loetmichel> like i said: i etched a lot: BEFORE i got the mill ;-)
[11:49:14] <Aero-Tec> if you etch, you can drill the holes using a dremal drill and press
[11:49:26] <Loetmichel> the thing is called a dremel
[11:49:39] <Loetmichel> or a minidrill to avoid any (C)
[11:49:49] <Aero-Tec> lol
[11:50:14] <Aero-Tec> is that the generic term
[11:50:28] <Loetmichel> etching isnt that complicated bur requires a set of tools anyway
[11:50:42] <Loetmichel> as far as i know: yes
[11:50:57] <Tom_itx> etching is rather easy
[11:51:16] <tjb1> I would love to have the stuff to etch
[11:51:26] <tjb1> But Its going to take all my money to buy the materials for this plasma cutter
[11:51:40] <Aero-Tec> well I used the real deal as the other one were not available back in the day
[11:51:57] <Tom_itx> tjb1,
http://tom-itx.dyndns.org:81/~webpage/etching/etch_index.php
[11:52:02] <tjb1> I love it when a company won't email you back about a quote...
[11:52:03] <Loetmichel> Tom_itx: i know
[11:52:04] <Loetmichel> http://www.cyrom.org/pce/aetz2.avi
[11:52:29] <Tom_itx> can't open that here
[11:52:40] <tjb1> Thats pretty cool Tom
[11:53:01] <Loetmichel> Tom_itx: no problem, its just me etching some amplifier boards in timelapse
[11:53:11] <Tom_itx> i visualized that :)
[11:53:19] <tjb1> Guess I will have to email a different person at that company to get a freakin quote.
[11:53:45] <Tom_itx> unless it's just a test or something i'm doing quick, i'll send em off and have em made
[11:54:21] <Loetmichel> Tom_itx: the etchtank looks familar. mine:
http://www.cyrom.org/palbum/main.php?g2_itemId=2856
[11:54:30] <Aero-Tec> Tom_itx: what etchent are you using?
[11:54:43] <Tom_itx> i use sodium or ammonium persulphate
[11:54:44] <Tom_itx> i have both
[11:54:51] <djdelorie> LoetMichel: I use curpic chloride etchant, not one of the three you mention
[11:55:00] <Tom_itx> iirc one works better with etch pencils
[11:55:24] <Loetmichel> djdelorie: isnt that whats generated when using HCl and H2O2?
[11:55:38] <djdelorie> eventually, but it's a different etchant.
[11:55:53] <djdelorie> You can use HCl/H2O2 to *make* CuCl etchant, or you can make it other ways
[11:56:08] <tjb1> Ha, this company doesn't even have a 800 number
[11:56:10] <Loetmichel> thats what i meant
[11:56:20] <djdelorie> I made mine with HCl/H2O2 and about half a pound of copper
[11:56:22] <Aero-Tec> my tank was much like yours but much bigger
[11:56:38] <Aero-Tec> I could do over 20 boards at a time
[11:56:52] <djdelorie> HCl/H2O2 is itself an etchant. My tank no longer had H2O2 in it
[11:56:58] <Loetmichel> Aero-Tec: for a "quick etch in between" one can improvise:
[11:57:03] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=9125
[11:57:49] <Loetmichel> (brought my UV exposer to the company to do sole prototypes, got some empty boxe4s and got going ;-)
[11:58:43] <Loetmichel> hrhr, is just see, tere's a exam of my apprentice on the table ;.-)
[11:58:53] <tjb1> Even better, I call them for a quote and the woman that won't answer my emails is the one who did the phone quote
[11:59:23] <Aero-Tec> had a bubble floor, drilled a whack of hole in plex that would fit in the bottom of the tank, had stand off glued to the tank side and hold down rails that locked into place at the top
[12:00:07] <Aero-Tec> one tube went through the faults floor and it would bubble up through all the holes
[12:01:22] <Loetmichel> Aero-Tec: i just used a pvc pipe and a hot needle
[12:01:33] <Loetmichel> -> done for the bubble string ;-)
[12:02:02] <Loetmichel> s/pipe/tube
[12:02:02] <tjb1> Is there any way to get a shipping quote from McMaster online?
[12:02:46] <Aero-Tec> so were you using photo resist?
[12:04:14] <Aero-Tec> I used to spray my photo resist onto the boards with a airbrush, also did the spin thing for smaller boards
[12:06:05] <Aero-Tec> I had a small commercial operation going for a bit, back before the net and easy access to board makers
[12:06:13] <Aero-Tec> way back in the day
[12:14:49] <IchGuckLive> hi all
[12:27:43] <JT-Shop> tjb1: email them for a quote
[12:29:42] <jdhNC> JT: what is the ngcgui pocket routine called?
[12:29:57] <IchGuckLive> JT-Shop: how is the weather ? here its thundery al day
[12:30:28] <IchGuckLive> jdhNC: G12/13 G150
[12:32:16] <IchGuckLive> jdhNC:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?G12-13
[12:33:14] <jdhNC> nifty (and the g150 one)
[12:38:16] <IchGuckLive> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?G150
[12:38:49] <IchGuckLive> jdhNC: i added also a python gui as you see for this
[12:38:54] <Connor> I'm liking ngcgui
[12:39:03] <IchGuckLive> it writes into axis
[12:39:30] <JT-Shop> sunny here IchGuckLive
[12:39:38] <IchGuckLive> in the new master there wil be this remappt includet
[12:39:47] <IchGuckLive> JT-Shop: B)
[12:39:54] <Connor> JT-Shop: I think I saw someone say you had a collection of ngcgui scripts ?
[12:40:16] <JT-Shop> for a lathe
[12:40:29] <Connor> oh.
[12:40:35] <JT-Shop> all scripts are here
http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=tree;f=nc_files/ngcgui_lib;h=cb5bf8056dd34daf3c0af901a012d4b2d2bfb379;hb=refs/heads/v2.5_branch
[12:40:35] <Connor> any for threading ?
[12:40:37] <IchGuckLive> mhabler is on this
[12:41:14] <JT-Shop> threading on the lathe yes and rigiding tapping too
[12:41:56] <Connor> http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=blob_plain;f=nc_files/ngcgui_lib/db25.ngc;hb=refs/heads/v2.5_branch
[12:42:38] <Connor> okay, this one.. I used it the other day.. and it didn't take into account my tool diameter, even though I had it specified correctly in my tool table..
[12:43:24] <IchGuckLive> Connor: we are not perfect
[12:43:48] <Connor> IchGuckLive: Yea.. I know.. I was just trying to figure out how to fix the issue. :)
[12:44:20] <IchGuckLive> as most of us use also a cam this is on sidestack
[12:45:50] <IchGuckLive> i got castomers that are making a sketch for a simple sqare generate the code abd manuell transfer by USB to the NC as they dont input 4lines into the mashine by hand
[12:45:53] <Connor> I use CAM too, but, it sucks having to run into the office to do a cad then cam it out for a simple task..
[12:46:12] <IchGuckLive> B) O.O
[12:46:16] <IchGuckLive> as i speek
[12:46:26] <jdhNC> Connor: did you select the tool first?
[12:46:38] <IchGuckLive> therfore im teatching 8k students a year
[12:47:28] <Connor> No, You specify the tool in the gui..
[12:47:35] <Connor> It asks which tool number..
[12:47:43] <jdhNC> Connor db25 calls iquad that does radius compensation
[12:48:06] <jdhNC> I tried db25 briefly last night, but I had a .25" end mill in.
[12:48:15] <IchGuckLive> you need to save the ngc and reload it
[12:48:33] <jdhNC> also, it complained about no % or no program end.
[12:48:49] <Connor> jdhNC: You don't have it loaded into the gui correctly.
[12:49:08] <jdhNC> what part?
[12:49:22] <jdhNC> I added a postamble setting/file with just an M2 and it was happy after that.
[12:49:23] <Connor> That's the other thing.. the sample file to get it to load didn't work right.. I ended up having to copy the files into my nc_files directory.
[12:49:43] <jdhNC> I copied the dir structure in my nc_files also and modified the path
[12:50:17] <Connor> [RS274NGC] SUBROUTINE_PATH = ../../../nc_files/ngcgui_lib:../../../ngcgui_lib/utilitysubs USER_M_PATH = ../../../nc_files/ngcgui_lib/mfiles
[12:50:33] <jdhNC> I just hardcoded my path
[12:50:34] <Connor> that would always throw a error.. it didn't like the [RS274NGC] part..
[12:50:48] <jdhNC> should be on a separate line
[12:51:00] <Connor> It is. IRC didn't spit it out right.
[12:51:56] <Connor> I'll do some more testing later.
[12:52:14] <Connor> I need to finish the e-stop, push-to-start and limit switch wiring..
[12:52:20] <jdhNC> I cut/pasted it from the doc. Worked fine except the path
[12:52:41] <Connor> might be the location I have it.. I'll try again tonight.
[12:52:48] <jdhNC> I would like limit switches too... and I think I want a separate home for X anyway (in the middle)
[12:53:10] <jdhNC> I went ahead and mounted the 7i37TA so I have lots of extra inputs.
[12:53:12] <Connor> I made my limit switch for X and mounted in the middle.
[12:53:27] <andypugh> You need a very long flag for X in the middle, or it doesn't know which way to home..
[12:53:37] <jdhNC> if it is in the middle, it's not a limit switch?
[12:53:38] <Connor> jdhNC: You ever going to post up some photos or make a build thread on the zone ?
[12:53:48] <jdhNC> connor: doubtful!
[12:54:05] <jdhNC> I'm not that organized.
[12:54:16] <Connor> jdhNC: Well.. at least take some pictures.
[12:54:31] <Connor> andypugh: Mine is typical setup with switch in middle with a flag on either side to trip it..
[12:54:38] <jdhNC> yeah, I will do that. I'm hoping to make the ballscrew mounts this week for x/y
[12:54:39] <Connor> I was going to use it for both home and limit.
[12:55:23] <jdhNC> oh, the left/right flags are the limits... gotcha. I just wanted to avoid full travel for homing, but it's so fast I guess it doesn't matter.
[12:55:26] <Connor> If that works, then, I'll install one for the Y.. the Z will probably have two switches.. top one for top end limit, and a adjustable lower one..
[12:56:04] <jdhNC> I think I got X up to 150ipm last night with the vise mounted in the middle.
[12:56:06] <Connor> jdhNC: Yea, I'm using those little adjustable endstops as the flags.
[12:56:30] <Connor> jdhNC: with the stock screws ?
[12:56:37] <jdhNC> yeah
[12:57:17] <jdhNC> 120ish on Z with the ballscrew and 2:1 drive
[12:57:20] <Connor> jdhNC: Yea, before I tightened everything up to get all the backlash out.. I could do about that. Now, 70IPM can cause it to stall.
[12:57:45] <jdhNC> I have X tight. Y has no anti-backlash screws in the nut.
[12:57:53] <Connor> I may have the pre-load on the jam nuts too tight.
[12:58:26] <Connor> I also sometimes get a awful resonance squeal sound at high speeds sometimes..
[12:58:38] <Connor> everything is oiled.. so, I'm not sure what that is.
[12:59:04] <jdhNC> I'll try dropping to half-stepping and see if that helps torque
[12:59:52] <Connor> The biggest thing that helped mine was the acceleration..
[12:59:52] <IchGuckLive> jdhNC: 8wire steppers ?
[13:00:02] <jdhNC> just 4wire
[13:00:11] <Connor> 570oz-in.. same as what I have.
[13:00:12] <jdhNC> dropped accel?
[13:00:24] <Connor> IchGuckLive: We're doing the same conversion. :)
[13:00:30] <jdhNC> did you see keling has 3.5amp 570's?
[13:00:49] <andypugh> I haven't seen a machine like this before:
http://www.youtube.com/watch?v=oKIntRHaBmo
[13:01:12] <Connor> jdhNC: Yea. lowered it.. on the Z it's really low, but, with it that low, I can get it up to 80IPM on stock screws direct drive.
[13:01:17] <Connor> jdhNC: No.
[13:01:48] <jdhNC> NEMA23 570 oz/in 3.5A 1/4. Dual Shaft Stepper Motor
[13:02:02] <IchGuckLive> andypugh: whow hevy
[13:02:27] <Connor> jdhNC: Wow, same price too.. wonder how they did that ?
[13:02:28] <IchGuckLive> andypugh: costom build for this part only
[13:02:39] <jdhNC> dunno, they have a 5amp 1/4" version also
[13:03:09] <Connor> andypugh: Would hate to get my hand stuck between that thing.
[13:03:40] <FinboySlick> Wow, I connect to the most interesting statements at times.
[13:03:51] <FinboySlick> Then again, I have a dirty mind.
[13:03:54] <jdhNC> that's what she said.
[13:05:43] <IchGuckLive> nema 23 570 oz 3.5A seems not to be parallel serial as they all same size got more Amp
[13:06:31] <Connor> jdhNC: Torque curve for our 570's torque nose dives around the 400 RPM point.
[13:06:39] <jdhNC> they both have 4 wires. Not many choices on how to wire them :)
[13:07:08] <Connor> No torque curve sheet for that other one yet..
[13:07:26] <Connor> I don't think I would use it for the conversion.. shaft too small.
[13:07:38] <jdhNC> 3v/5amp or 4.2v/3.amp
[13:09:03] <jdhNC> I drew up the new C41 from the dims in the manual. Plotted that (and the drives, mesa, PS, etc), taped it to the panel, punched, drilled, tapped.
[13:09:28] <jdhNC> and the dimensions for the C41 were wrong.
[13:09:36] <Connor> Cool. mine was all done by hand.
[13:09:39] <Connor> Oh. not cool.
[13:10:13] <IchGuckLive> im off by
[13:12:30] <jdhNC> they decided black/green was A/A- on one of them and B/B- on the other.
[13:13:14] <Connor> jdhNC: ?
[13:13:42] <jdhNC> those two motors. they swapped color codes for A & B
[13:13:47] <Connor> Oh.
[13:18:49] <Aero-Tec> is there a good gcode editor?
[13:19:08] <Aero-Tec> open source would be great
[13:19:19] <alex4nder-> g-code editor
[13:19:20] <FinboySlick> vim!
[13:19:20] <alex4nder-> ?
[13:19:36] <Aero-Tec> one with color coding
[13:19:40] <Aero-Tec> vim?
[13:19:52] <frallzor> isnt vime a coding software?
[13:19:55] <frallzor> *vim
[13:19:58] <frallzor> as in program
[13:20:10] <alex4nder-> I use emacs.
[13:20:11] <FinboySlick> Yeah, but it'll handle gcode just fine.
[13:21:29] <andypugh> gedit is hat I tend to use
[13:21:33] <Aero-Tec> does it have color coding options?
[13:21:37] <alex4nder-> yes
[13:21:43] <andypugh> I think it might
[13:21:57] <Connor> gedit does.. not sure if it has any for g-code though.
[13:22:14] <Tom_itx> you need dot to dot too?
[13:22:14] <jdhNC> it seems to
[13:22:20] <Aero-Tec> nice to have rems and some other thing marked with colors
[13:22:22] <alex4nder-> Aero-Tec:
http://pixpopuli.blogspot.com/2011/01/syntax-highlighting-for-cnc-g-code.html <- color coding g-code for emacs
[13:22:26] <andypugh> I think I have seeen mention of colour-coding for gedit somewhere
[13:23:08] <Aero-Tec> thanks
[13:24:04] <Connor> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Highlighting_In_Gedit
[13:26:38] <andypugh> I have that, though it doesn't make a great job when faced with string-style parameters
[13:29:42] <andypugh> Looking at the language-spec, it ought to work.
[13:30:09] <andypugh> It may be that axis-words are getting a higher-priority than the #<_xxx> pattern
[13:33:49] <tjb1> May have to get a water cooled torch for the tig…damn it gets hot
[13:35:08] <tjb1> or just wear a heavy glove on the tig hand
[13:55:54] <MDesade> OK, i got a response from CNC4PC. basically, i HAVE TO have an E-stop button wired in, press the button once, and then reset, and she will work....
[13:56:27] <andypugh> Interesting. Not necessarily a bad thing.
[13:56:56] <andypugh> But one might hope that it would have been documented.
[13:58:47] <Connor> MDesade: Which board is that ?
[13:59:32] <andypugh> I wonder if it is time to re-vist the G-code syntax highlighting? There seems to be a whole new XML format now. I think we have some missing G-codes too, as the current one is based on EMC 2.2
[14:06:10] <jdhNC> The board has a pin that allows you to enable/disable all the outputs at once. The board requires +5vdc in the EN pin. If it is not present, it will send all the outputs to ground.
[14:06:29] <Connor> Oh. the BOB. Yea.
[14:06:35] <Connor> C11 ?
[14:06:59] <Connor> You sure it sends all the outputs to grund ?
[14:07:30] <jdhNC> no clue, that's just what it says.
[14:07:36] <Connor> which board is this ?
[14:08:14] <jdhNC> c23
[14:09:03] <Connor> Yea, It's a bit smarter than the C10
[14:09:03] <jdhNC> nifty looking board other than the price.
[14:10:46] <jdhNC> got a pic of your X switch?
[14:11:29] <Connor> Not yet. I'm not even sure it's going to work. It was a micro-switch with roller from Radio Shack. Mostly as a test.. I'll redo with a heftier switch..
[14:12:19] <jdhNC> gotta be better than my dealextreme roller switches
[14:12:58] <Connor> Probably not.. it's SMALL and dinky..
[14:15:04] <Connor> I really wished I had a parport card that had all the inputs with internal connector..
[14:15:25] <Connor> Going to have to make a external cable go out and come back in the case to get more inputs..
[14:18:44] <jymmm> /join ##chemistry
[14:21:42] <jdhNC> connor:
http://www.dealextreme.com/p/diy-0-5a-125v-250v-micro-switch-black-10-piece-pack-120309?item=6
[14:22:03] <jdhNC> small, dinky, and not even a real roller
[14:22:09] <Connor> Yea, those are tiny too.. about the same. :)
[14:22:19] <jdhNC> I got some cnc4pc switches, not a whole lot better.
[14:22:41] <tjb1> jdhNC: how do those bent tabs work compared to rollers?
[14:22:56] <jdhNC> tjb1: no clue, haven't used them yet.
[14:23:11] <jdhNC> doubt it makes much difference though.
[14:23:14] <tjb1> Don't know if I will order from dx again...
[14:23:23] <jdhNC> tjb1: why not, it's fun
[14:23:36] <tjb1> Took me 2 months to get something one time
[14:23:38] <Connor> the A7 switches from CNC4PC look good.
[14:23:44] <jdhNC> I get cheap chinese crap I don't need every few months.
[14:23:50] <jdhNC> it's practically free
[14:24:21] <tjb1> When is the chinese new year?
[14:24:25] <jdhNC> when they finally show up, it's usually a surprise.
[14:24:45] <jdhNC> connor: I got the a7's
[14:24:50] <andypugh> Aero-Tec: I found a better language definition file. I suspect that they get used system-wide, not just by gedit.
http://cnc-club.ru/forum/viewtopic.php?f=33&t=80
[14:25:10] <tjb1> Do not order from DX near February 10th.
[14:25:21] <Tom_itx> chinese new year?
[14:25:23] <tjb1> Their mail system seems to shut down for the chinese new year...
[14:25:30] <Tom_itx> every year
[14:25:36] <jdhNC> http://www.dealextreme.com/p/t10-39-led-white-light-car-angel-eye-120mm-diameter-25522?item=10
[14:25:40] <andypugh> I found some micron-repeatability limit-switches on RS
[14:25:51] <jdhNC> I ordered one of those too. Thought it might work around the spindle
[14:26:22] <Connor> jdhNC: What's wrong with the A7's ?
[14:26:44] <jdhNC> connor: nothing afaik. They are just generic roller microswitches
[14:26:47] <Tom_itx> jdhNC devil's ring of fire!
[14:27:11] <andypugh> Here you are. Ruby plunger, repeatable to 1 micron, 10 million ops.
http://uk.rs-online.com/web/p/precision-position-switches/0341430/
[14:27:30] <tjb1> That angel eye really using a 194 bulb plug in?
[14:27:46] <jdhNC> tjb1: I'll let you know when it shows up
[14:28:08] <jdhNC> andy: kind of pricey for my use!
[14:28:11] <Tom_itx> it's probably like the ring lights for my binocular scope
[14:28:36] <Connor> okay, so, is their a way to but limits or exclusion zones in? Example: My vise hangs off the back of my table.. I have more travel on the Y on either side of it.. and if I use that travel, I can't got go all the way on the X without having to go around it..
[14:29:53] * djdelorie wonders if a HAL plugin could do that...
[14:29:57] <jdhNC> Connor: that would be nice, but not likely workable.
[14:30:24] <jdhNC> you coudl avoid the vise/column, but how would it know what the spindle was going to hit?
[14:30:59] <Connor> I'm just talking about limits.. just in case I do something stupid..
[14:32:22] <Connor> Can you change the soft limits dynamically ?
[14:32:49] <Tom_itx> limits are just that... limits
[14:32:54] <Tom_itx> they don't generally move
[14:33:00] <tjb1> Anyone know what the plastic hole plugs are that have a star cut in them to run wires through?
[14:33:01] <Connor> If so, I might just have a toggle button that limits the rear travel to the vise limit in one mode, then toggle to no vise
[14:33:05] <djdelorie> Connor: I suspect you can have a HAL plugin that reads X,Y,Z and asserts e-stop or one of the limit switch signals
[14:33:05] <jdhNC> where do you have your vise mounted?
[14:33:19] <Connor> jdhNC: Center slot.
[14:33:51] <jdhNC> me too, I didn't think it restricted travel though?
[14:34:08] <Connor> If you remove the way covers it does. :)
[14:34:20] <jdhNC> heh.
[14:34:30] <Connor> I'm going to redo those too.
[14:34:36] <jdhNC> I don't get the claimed Y travel out of mine, but I didn't really look at why
[14:34:48] <Connor> how short ?
[14:35:00] <Connor> the gib screws can stop the travel..
[14:35:14] <Connor> if they're adjust too far out.
[14:35:14] <jdhNC> 5"?
[14:35:20] <Connor> really ??
[14:36:22] <jdhNC> supposed to be 7?
[14:36:22] <jdhNC> 5.6 maybe
[14:36:23] <Connor> Oh wait.. yea.. the Y..
[14:36:23] <Connor> that's the one I'm talking about.. yea.. remove the way cover angle bracket..
[14:36:23] <Connor> and the rubber cover..
[14:36:23] <Connor> you can get the table to touch the back of the column.
[14:36:23] <Connor> to get 6 7/8" or so.
[14:45:45] <andypugh> I have thought about intelligent limits a bit. It seems like you could define a machine envelope with an STL file, and calculate limits that way. The hard part is moving the LinuxCNC soft limits out of joint-space into world-space.
[14:48:18] <jdhNC> you can do something like that when specifying paths for robot arms, but perhaps the CAMish path generator just uses those for generation.
[14:48:32] <andypugh> (I suggest STL as it is hand-creatable and easy for a machine to parse)
[14:56:15] <Aero-Tec> andypugh: thanks for the link
[15:00:37] <frallzor> STL? as in stereoli..... yada yada
[15:01:31] <frallzor> or something else? =)
[15:01:43] <djdelorie> Slower Than Light
[15:02:11] <alex4nder-> standard template library... andypugh is saying we should code our machine descriptions using C++
[15:02:15] <alex4nder-> by hand
[15:03:37] <andypugh> frallzor: Yes, the stereolith format. It's just points in space and the triangles that join them.
[15:03:43] <Aero-Tec> has anyone used
[15:03:46] <Aero-Tec> Avago ATS667LSGTN-T sensors
[15:04:03] <Aero-Tec> for limit switches?
[15:04:52] <Aero-Tec> they seem like a good option to switches
[15:05:27] <Aero-Tec> never ware out and will not have ware movement over time
[15:28:12] <jdhNC> looks good to me.
[15:28:21] <andypugh> I think JOn Elson used something similar to use his mill drive gear as an encoder
[15:29:05] <andypugh> I am not sure how repeatable it would be as a limit switch. And it might work all too well as a swarf detector.
[15:30:08] * djdelorie just pictured a whole array of swarf detectors, set up like the high energy particle detectors at CERN, around a mill...
[15:31:17] <djdelorie> "Look! We've discovered a new type of swarf!"
[15:31:54] <andypugh> :-)
[15:32:26] <andypugh> My machine is using
http://search.digikey.com/us/en/products/OPB742/365-1014-ND/374792
[15:32:55] <jdhNC> for encoder?
[15:33:08] <djdelorie> I imagine any optical detector would need to be protected against swarf and other debris...
[15:35:18] <andypugh> I have them for limit switches. And yes, they need to be somewhat shielded.
[15:36:49] <andypugh> actually, I think I used a slightly different design.
[15:37:19] <andypugh> They need to be the "focussed" style to work as a limit/home switch.
[15:38:01] <jdhNC> http://www.dealextreme.com/p/arduino-line-hunting-sensor-module-118057
[15:38:24] <jdhNC> kind of bulky packaging. I ordered a few of those to check also.
[15:39:02] <andypugh> An advantage of the focussed style is that the target can be close to a surface, like this:
https://picasaweb.google.com/lh/photo/d0KbRYcV8lTJh639Og5DfNMTjNZETYmyPJy0liipFm0?feat=directlink
[15:39:13] <jdhNC> https://picasaweb.google.com/lh/photo/d0KbRYcV8lTJh639Og5DfNMTjNZETY
[15:39:13] <jdhNC> myPJy0liipFm0?feat=directlink
[15:39:18] <jdhNC> oops.
[15:39:19] <andypugh> Which is detected by this
https://picasaweb.google.com/lh/photo/wLO_CcgbBmOMZ34L_s5pQtMTjNZETYmyPJy0liipFm0?feat=directlink
[15:40:08] <andypugh> (I encased it in epoxy putty, with an adjustable comparator circuit)
[15:41:16] <andypugh> And looking at another photo, this is the one I used:
http://www.optekinc.com/datasheets/OPB702.PDF
[15:41:46] <andypugh> I suspect that the "logic output" type would be a fair bit more convenient.
[15:45:34] <jdhNC> any chance grease or something on the target could cause it to miss?
[15:46:50] <andypugh> Yes. But so far they have been fine. Even the one on the cross-slide.
[16:24:38] <DJ9DJ> gn8
[16:32:13] <JT-Shop> silly sob didn't bring index into the electrical cabinet even though it is landed at the servo plug...
[16:38:53] <tjb1> If I were to intercept and apply my own stepper signals, LinuxCNC would no longer know the position of the stepper I modified correct?
[16:41:26] <andypugh> tjb1: Yes. Though LinuxCNC only has a vague impression of where a stepper is anyway
[16:42:00] <tjb1> Well if the code says to move .5, wouldn't it know and show that it has moved .5?
[16:42:21] <tjb1> I mean it won't know for sure that the stepper accomplished that, but it will show it "should" be there?
[16:44:32] <JT-Shop> you might want to study the stepgen component
[16:45:37] <tjb1> http://linuxcnc.org/docs/html/man/man9/stepgen.9.html ?
[16:45:47] <JT-Shop> aye
[16:45:53] <tjb1> Will do :)
[16:47:04] <JT-Shop> also
http://linuxcnc.org/docview/html/hal/rtcomps.html#_stepgen_a_id_sec_stepgen_a
[16:48:06] <tjb1> alright
[16:48:28] <JT-Shop> crap more eps's to fix
[17:06:00] <motioncontrol> good
[17:12:09] <jymmm> is Cosmoline still in common use?
[17:29:20] <Tom_itx> why wouldn't it be?
[17:40:35] <Aero-Tec> andypugh: your right Jon Elson did use them, that is where I found out about them
[17:40:57] <Aero-Tec> has anyone else used them for encoders and switches?
[17:41:25] <Aero-Tec> seemed like a good idea
[17:41:51] <Aero-Tec> was thinking on getting some and trying them out
[17:42:20] <Aero-Tec> was hoping to use them as spindle encoders for lathe and mill
[17:46:15] <Aero-Tec> yo
[17:46:19] <Aero-Tec> anyone here?
[17:46:46] * frallzor glares at JT-Shop
[17:47:01] <Aero-Tec> need a sample of conditional branching
[17:47:15] <Aero-Tec> what sample program should I look at
[17:48:30] <Aero-Tec> any simple sample of Gcode programing with if then whiles and stuff?
[17:56:43] <archivist> er the docs and samples that it comes with
[17:58:03] <Aero-Tec> going through them
[17:58:18] <Aero-Tec> was hoping for a short cut
[17:58:27] <Aero-Tec> simple samples
[17:58:57] <archivist> the docs are simple, its all I needed to write
http://www.archivist.info/cnc/standard_clock_n_teeth.ngc
[18:01:28] <JT-Shop> Aero-Tec: yes there are examples in the O Code chapter as well as example g code files
[18:03:55] <Aero-Tec> was looking for conditional branch example
[18:04:09] <Aero-Tec> there are lots of loops
[18:04:33] <JT-Shop> http://linuxcnc.org/docview/html/gcode/o-code.html#_conditional_a_id_sec_conditional_a
[18:04:40] <Aero-Tec> but a good if then branch example would be nice
[18:04:43] <andypugh> Do we have conditional branching?
[18:04:50] <JT-Shop> aye
[18:04:56] <andypugh> Ah, if / then /else
[18:05:03] <JT-Shop> just pasted you one
[18:05:08] <andypugh> Thought you meant "goto" _0
[18:05:17] <JT-Shop> if all esle fails...
[18:05:24] <archivist> but a while... is probably more sensible in a loop :)
[18:05:53] * JT-Shop seems to remember goto from a previous lifetime atari maybe
[18:06:22] <archivist> I still meet goto !
[18:06:25] <Aero-Tec> does EMC have goto?
[18:06:37] <jdhNC> goto is quite useful, in moderation.
[18:06:55] <JT-Shop> Aero-Tec: no
[18:10:03] <archivist> gcode is like assembler but worse and has no goto
[18:11:34] <JT-Shop> LOL
[18:11:54] <FinboySlick> I like assembler.
[18:12:28] <archivist> assembler has better variable syntax !
[18:13:17] <FinboySlick> Heh, if it were like C, could you malloc() more material on the part? That would be useful sometimes.
[18:13:55] <archivist> and goto the other side to get more work done
[18:41:23] <tjb1> Anyone here fluent in data sheets and optocouplers?
[18:43:37] <JT-Shop> tjb1: tip of the day... just ask your question
[18:43:41] <alex4nder-> yes
[18:43:44] <tjb1> http://www.datasheetcatalog.org/datasheet/motorola/4N26.pdf
[18:43:46] <alex4nder-> questions are good to ask
[18:43:54] <alex4nder-> what about it?
[18:44:15] <tjb1> What voltage can I run through the collector? non-led side
[18:45:07] <djdelorie> VCEO is 30 volts
[18:45:08] <alex4nder-> Vceo is 30 volts
[18:45:12] <alex4nder-> jinx
[18:45:28] <djdelorie> but it's the 150 mA limit you're more likely to be interested in
[18:45:52] <tjb1> So what does the E-C and C-B voltages mean?
[18:45:55] <djdelorie> VCEO is the max voltage when the opto is *off*, Ic is the max amps when it's *on*
[18:46:31] <djdelorie> Most optos, the base isn't connected so doesn't matter
[18:46:42] <tjb1> Sorry, Veco and Vcbo
[18:46:48] <djdelorie> VECO is the max *reverse* voltage (i.e. collector has -7 volts on it)
[18:47:19] <tjb1> and for an optocoupler to work each side needs its own power supply?
[18:47:42] <tjb1> Well for it to properly isolate
[18:47:50] <djdelorie> yes
[18:48:22] <djdelorie> the fundamental transfer is: for some current A through the led, you get some (higher) current B through the transistor (Ice)
[18:48:47] <tjb1> and what voltage can I use for the LED?
[18:49:14] <alex4nder-> tjb1: whatever it is you're trying to do, I'd run the design by someone before you even decide this is the right combinations of parts to do what you want.
[18:49:24] <tjb1> Max forward voltage of 1.5 and max 60 mA current
[18:49:26] <CareBear\> tjb1 : see "forward voltage"
[18:49:37] <CareBear\> tjb1 : so 1.5 V
[18:49:40] <JT-Shop> how can you continue a line in javascript?
[18:50:00] <tjb1> Im using this as an ohmic sensor on the tip of plasma to isolate it from the g540
[18:50:00] <djdelorie> again, wrong question. It's "what CURRENT do I put through the LED"
[18:50:05] <CareBear\> both
[18:50:06] <alex4nder-> tjb1: you just need to current-limit the LED
[18:51:18] <tjb1> So using a 12v supply I would Need a 200 ohm resistor
[18:52:15] <alex4nder-> compensate for tolerances
[18:52:19] <djdelorie> Given Given a preferred If of 20 mA, and 1.15 Vf, you need a resistor that drops 10.85V at 0.020 A = 542 ohms
[18:52:38] <djdelorie> er, 10 mA is preferred = 1085 ohms
[18:52:46] <alex4nder-> you don't just set everything to max and cross fingers.
[18:53:05] <djdelorie> yeah, max is "just before it explodes". You want the next page
[18:53:18] <alex4nder-> also optos are notoriously shitty at being linear, and they age.
[18:56:29] <tjb1> So you subtract the voltage forward from the input voltage
[18:57:31] <tjb1> Its showing .1085 watts so a 1/4 watt resistor?
[18:58:24] <djdelorie> sure
[18:58:46] <tjb1> Thank you for helping the electrically handicap :)
[19:04:17] <JT-Shop> lol
[19:06:50] <tjb1> I may have to have you all sign my table for all the help you have given :P
[19:07:52] <roycroft> with a cold chisel?
[19:08:41] <tjb1> sharpie
[19:09:03] <tjb1> don't need any dents or scrapes on it
[19:09:23] <tjb1> :)
[19:09:46] * JT-Shop has a drawer full of sharpies Dad gave me...
[19:09:59] <roycroft> sharpie will wear off
[19:10:11] <JT-Shop> yea
[19:10:36] <tjb1> clearcoat
[19:10:47] <tjb1> Ill powder coat the whole table
[19:11:32] <tjb1> Now I have to build a car sized oven...
[19:11:59] <djdelorie> just bring it to Texas and leave it outside :-)
[19:13:00] <tjb1> Not a big fan of warm weather
[19:14:48] <djdelorie> Aluminum?
[19:14:53] <djdelorie> Steel?
[19:14:56] <tjb1> Both
[19:15:16] <djdelorie> Aluminum can be chemically etched with plain HCl
[19:15:21] <JT-Shop> tjb1: you can use an infrared heater to do that if you don't bump it...
[19:15:28] <tjb1> Im not doing anything to the aluminum
[19:15:44] <djdelorie> how else are you going to get small signatures in there? ;-)
[19:15:47] <tjb1> JT-Shop: That would be sketchy with powdercoat
[19:15:59] <JT-Shop> it works I know
[19:16:21] * djdelorie wonders if a heat gun would work
[19:16:29] * JT-Shop has a powder coat setup
[19:16:48] <JT-Shop> never tried that...
[19:17:09] <tjb1> How do you get correct cure times on it all?
[19:17:50] <JT-Shop> you gotta kinda swag it and watch and use a non contact thermometer thingy
[19:18:03] <tjb1> You would have to over cure certain area where sections overlap
[19:18:07] <JT-Shop> it will wait till you get done if you don't bump it
[19:18:23] <Tom_itx> sounds like a big pita
[19:18:29] <tjb1> I have tried mixing colors and I can tell you that it doesn't work
[19:18:30] <JT-Shop> that it correct
[19:18:45] <JT-Shop> chow time guys
[19:19:03] <tjb1> peace
[19:19:13] <tjb1> I mixed yellow and red powders
[19:19:19] <tjb1> and I got red with a bunch of yellow dots
[19:19:34] <djdelorie> yeah, it's a powder coat, not a liquid coat
[19:19:47] <tjb1> It turns to liquid
[19:19:52] <tjb1> sort of
[19:20:09] <djdelorie> but you can't *mix* it while it's liquid
[19:20:21] <Tom_itx> sorta like laserjet toner
[19:20:47] <Tom_itx> charge holds it on the piece then you melt it to cure it
[19:20:49] <tjb1> Shooting it while the part is hot is a good way to put too much on it
[19:22:36] <tjb1> This is probably my favorite powder that I have
[19:22:36] <tjb1> https://www.google.com/search?sugexp=chrome,mod%3D19&q=black+stardust&um=1&ie=UTF-8&hl=en&tbm=isch&source=og&sa=N&tab=wi&ei=zNm6T7_WK4nTgQf5_M3jCg&biw=1680&bih=870&sei=ztm6T_SgINHggge9oqjDCg
[20:32:56] <Aero-Tec> Writing my first Gcode for EMC, I might just be falling in love
[20:33:15] <Aero-Tec> sure is nice
[20:33:33] <ReadError> writing it by hand?
[20:33:37] <joe9> any heekscnc users here? are you Z = 0, at the top? How do you get Heekscnc to generate gcode for that?
[20:34:03] <joe9> Aero-Tec: check out heekscnc and freecad. it is easier to write python than gcode.
[20:34:29] <Aero-Tec> yes, I hand code most every thing
[20:35:22] <Aero-Tec> I do have some cam software, but I just use it to gen some 3D code, sometimes, it is getting less and less
[20:37:15] <Aero-Tec> never did write in python
[20:37:54] <Aero-Tec> when I did web programing it was php and perl
[20:38:21] <Aero-Tec> there was others but I did not get into them
[20:38:49] <Aero-Tec> will have to check python out
[20:39:03] <Aero-Tec> hello sk
[20:39:11] <Aero-Tec> oops sw
[20:41:29] <roycroft> it seems to me that having z = 0 at the bottom of the part would be easier to deal with, unless your part blanks are milled very precisely to height before mounting for cnc operations
[20:45:16] <djdelorie> joe9: just give it negative Z numbers when it needs them
[20:53:54] <jdhNC> any suggestions for a reasonable feed rate & DOC for a .25 HSS 4 flute endmill and 6061?
[20:57:02] <roycroft> feeds and speeds are pretty easy to calculate
[20:57:08] <roycroft> doc depends entirely on your mill
[20:57:17] <roycroft> you'll have to experiment to determine that
[20:57:47] <jdhNC> yeah, I was just looking for a conservative starting point.
[20:58:59] <roycroft> if it's a small mill i'd start with 0.020" doc for 6061 as a starting point
[20:59:19] <roycroft> if it's bridgeport-size 0.060" easily
[20:59:32] <djdelorie> in general, which is better - light cuts and higher IPS, or heavy cuts at low IPS ?
[20:59:37] <alex4nder-> it depends
[20:59:43] <alex4nder-> on the mill, and your removal strategy
[20:59:58] <alex4nder-> I remove 0.20 DoC on my taig.. but I'm removing a .004-wide chip when doing it
[21:00:06] <alex4nder-> because I do trochoidal milling
[21:00:43] <roycroft> it also depends on the type of cutter you're using
[21:00:45] <joe9> djdelorie: what do you mean by "just give it negative Z numbers when it needs them"? I am generating the gcode from heeks.
[21:00:45] <alex4nder-> if you just ram the endmill into 6061 with like .. > 90 radial engagement, you're going to put a lot of strain on your endmill
[21:00:55] <alex4nder-> and you're going to get deflection
[21:01:01] <alex4nder-> so you'll get squealing and pain
[21:01:04] <joe9> djdelorie: do you mean just give the z values as -ve in Heeks.
[21:01:09] <alex4nder-> (and stalling, and broken endmills, etc.)
[21:01:21] <djdelorie> right, in heeks when it asks for Z values, you give it -ve values for "height"
[21:01:36] <joe9> djdelorie: ok, thanks.
[21:02:29] <alex4nder-> you also don't want to produce chips that are too small.. because you'll destroy your endmills
[21:02:35] <alex4nder-> (overheating, etc)
[21:02:35] <joe9> djdelorie: have you used heeks? do you know if the "set coordinate system" can be used to put Z=0 at the top of the piece?
[21:03:00] <djdelorie> I don't think heeks has a Z per se; it's a 2.5-D system
[21:03:32] <djdelorie> so for each milling operation, you tell it "start at 0, work towards -0.45" or whatever
[21:11:00] <joe9> ok, thanks. just to confirm, in such a scenario, the tool paths will be visible below the workpiece in the backplot, correct/
[21:11:06] <joe9> djdelorie: ^^
[21:11:48] <djdelorie> correct. Always test first :-)
[21:12:16] <joe9> djdelorie: ok, thanks.
[21:12:58] <skunkworks__> it was hard for me to get used to top of part being 0 as the old control only used positive positions.
[21:12:58] <joe9> i love FreeCAD. it is so much cooler to be able to use python to generate stuff.
[21:13:03] <djdelorie> http://www.delorie.com/cnc/bldc-frame.heeks is a sample
[21:15:00] <joe9> djdelorie: thanks, that helped.
[21:18:08] <Aero-Tec> tried to run my code
[21:18:44] <Aero-Tec> it says one can not use G95 mode when spindle is not going
[21:18:51] <Aero-Tec> but it was
[21:19:53] <djdelorie> the physical spindle was going, or the virtual spindle linuxcnc thought it was controlling was going?
[21:20:53] <Aero-Tec> it was running and the index is working
[21:21:06] <Aero-Tec> I had the lathe running
[21:21:44] <Aero-Tec> how to do you if it is in virtual mode?
[21:22:38] <djdelorie> well, for my cnc, the spindle is *not* controlled by linuxcnc, so just because it's running doesn't mean linuxcnc thinks it's running
[21:22:43] <skunkworks__> Aero-Tec: do yo have motion.spindle-speed-in IN FLOAT
[21:22:45] <skunkworks__> Actual spindle speed feedback in revolutions per second; used for G96 feed-per-revolution and constant surface speed modes
[21:22:47] <skunkworks__> hooked up?
[21:23:22] <joe9> djdelorie: do you notice this with heeks? it does not generate some toolpaths for "Operations". I have "Pocket" and "Profile" operation with the same tool. and it just does Profile now.
[21:23:33] <Aero-Tec> the encoder is hooked up and working
[21:23:34] <joe9> djdelorie: do you find heeks to be a bit unpredictable?
[21:23:36] <Aero-Tec> I checked it
[21:23:44] <joe9> djdelorie: or, is there a better way of using heeks?
[21:23:51] <djdelorie> you have to specify a start point
[21:23:57] <djdelorie> it won't pick one for you
[21:23:58] <joe9> from command prompt, directly or something like that.
[21:24:16] <djdelorie> right click on the operation in the left side pane, I think
[21:25:07] <Aero-Tec> skunkworks not sure what you mean by motion.spindle-speed-in IN FLOAT
[21:26:14] <skunkworks__> that was coppied and pasted from the hal manual. motion.spindle-speed-in is the pin (it is an 'in float')
[21:26:34] <joe9> djdelorie: start depth = 0, final depth = -0.1, step down = -0.03 -- sounds good to you?
[21:26:40] <joe9> djdelorie: for the pocketing operation.
[21:26:57] <joe9> rapid safety space = 2, clearance height = 2
[21:27:20] <djdelorie> those are also Z so do you really need to be two inches above your workpiece?
[21:27:34] <Aero-Tec> SO i SHOULD SET THAT?
[21:27:37] <Aero-Tec> OOPS
[21:27:46] <Aero-Tec> was doing gcode
[21:28:01] <joe9> djdelorie: just to be sure at this point. still playing so do not want to run into clamps and stuff.
[21:28:08] <skunkworks__> you need to hook your spindle rev/sec signal to it.
[21:28:19] <Aero-Tec> do I need to set g96 and g97?
[21:28:30] <djdelorie> joe9: I start my "start depth" a little above the part (0.02) and final -0.1, with step 0.04 but mine is wood so deeper cuts
[21:28:54] <Aero-Tec> that would be explained in the manual I would guess
[21:28:56] <joe9> is step = 0.04 or -0.04 ?
[21:29:02] <djdelorie> I *think* only clearance needs to be above clamps, rapid safety might allow for rapid Z
[21:29:02] <joe9> djdelorie: mine is wood too.
[21:29:10] <djdelorie> step is 0.04 because it's step DOWN
[21:30:49] <Aero-Tec> skunkworks do you have a page link you got that from?
[21:31:49] <skunkworks__> http://linuxcnc.org/docs/2.5/html/man/man9/motion.9.html
[21:32:07] <joe9> djdelorie: for pocket operation, I get a new window asking for the details. For the Profile operation, I just get the properties subwindow. Is there some config setting for that?
[21:32:13] <Aero-Tec> skunkworks I was able to use the scope to see the signal was working
[21:32:22] <djdelorie> unlikely
[21:32:33] <djdelorie> I've learned to just fill in the properties directly
[21:33:07] <joe9> how much do you normally set rapid safety to?
[21:33:21] <djdelorie> I sent you my one heeks file :-)
[21:33:30] <joe9> oh, thanks. sorry about that.
[21:33:43] <djdelorie> I'm not an expert. I just know a little bit more than you do ;-)
[21:34:22] <Aero-Tec> motion.spindle-index-enable I/O BIT
[21:34:24] <Aero-Tec> For correct operation of spindle synchronized moves, this signal must be hooked to the index-enable pin of the spindle encoder.
[21:34:41] <Aero-Tec> not sure how
[21:35:11] <Aero-Tec> it did not have a link to where it explained how to hook the signal
[21:37:01] <skunkworks__> threading and feed per rev are 2 different things. threading uses the index as a position while feed per rev uses the spindle rotations per second signal
[21:37:28] <skunkworks__> Aero-Tec: are you able to thread?
[21:38:19] <Aero-Tec> not sure
[21:38:24] <Aero-Tec> have not tried it
[21:38:27] <joe9> djdelorie: you know more than me in almost everything.
[21:38:34] <Aero-Tec> just testing thing
[21:38:40] <joe9> /ws 34
[21:39:12] <Aero-Tec> thought I would start with a quick and easy program using G95 for feed rate
[21:39:58] <Aero-Tec> go you want me to try the threading sample of a 1/4 thread?
[21:51:10] <skunkworks__> Aero-Tec:
http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=blob_plain;f=configs/nist-lathe/nist-lathe.hal;hb=HEAD
[21:52:11] <skunkworks__> this hal exaample shows the encoder setup for threading (example0
[21:56:49] <Tom_itx> is that part of the install example files?
[22:02:19] <Aero-Tec> I tried to run the threading example
[22:03:00] <Aero-Tec> after loading the tool table and a few other things I could load the code
[22:03:27] <Aero-Tec> skunkworks, it gets stuck at line 14
[22:03:51] <Aero-Tec> facing
[22:04:01] <Aero-Tec> whats up with that?
[22:30:18] <Aero-Tec> ok
[22:30:36] <Aero-Tec> was in G95 mode still and not G94
[22:31:14] <Aero-Tec> switch to G94 and it would work, but stopped again at threading
[22:31:37] <Aero-Tec> so it will not thread of do G95 mode moves
[22:33:00] <Aero-Tec> skunkworks: will you be around much longer?
[22:33:26] <Aero-Tec> will try and figure this out
[22:33:40] <Aero-Tec> but may need some help along the way
[22:39:54] <Aero-Tec> does EMC have a spindle seed readout at all?
[22:40:23] <Aero-Tec> how can one find out the RPM of the spindle?
[23:06:02] <jdhNC> add a pyvcp window that has spindle speed
[23:23:14] <Aero-Tec> new to all this
[23:23:34] <Aero-Tec> how does one add the window?
