Back
[00:06:32] -!- rob_h has quit [Ping timeout: 244 seconds]
[00:54:39] -!- pink_vampire [pink_vampire!~kvirc@ool-4a5a8c6d.dyn.optonline.net] has joined #linuxcnc-devel
[00:55:04] -!- tobias47n9e__ has quit [Ping timeout: 240 seconds]
[00:59:43] <pink_vampire> seb_kuzminsky: hi
[01:00:31] <pink_vampire> my BF is here and he want to ask you something about configuration of the probe.
[01:02:25] <pink_vampire> Hi, Installed 'Probe Screen' but getting this error when trying to test the probe:
[01:03:02] <pink_vampire> "probe trip during non-probe MDI command"
[01:06:53] <andypugh> Are you using a probe routine with non-probe moves in it?
[01:08:22] <andypugh> If it’s (for example) the Verser Python routines, those send the G-code moves one at a time through the MDI interface.
[01:12:21] <skunkworks> getting there
[01:12:24] <skunkworks> http://electronicsam.com/images/matsuura/20160506_175211.jpg
[01:13:57] <andypugh> I made little brackets with tubes for where my cable loops crossed the hinge. I am not sure if that was overkill.
[01:14:37] <skunkworks> andypugh, if you wanted to advance the carousel componant manually - how would you do it?
[01:15:26] <andypugh> Hmm, good question. Mux tool+1 into it?
[01:15:50] <skunkworks> ok - that was the nebulous thought..
[01:16:14] <andypugh> You think it needs jog inputs?
[01:16:29] <andypugh> I can see that being needed for tool loading
[01:17:01] <skunkworks> yes..
[01:19:21] <skunkworks> the matsuura has a Geneva mech that runs a tool chain
[01:19:48] <skunkworks> https://www.youtube.com/watch?v=kwnIIWTb7U8
[01:21:38] <andypugh> Putting tools in that does look like you need to be able to move it, yes.
[01:22:18] -!- KimK [KimK!~Kim__@ip68-102-66-31.ks.ok.cox.net] has joined #linuxcnc-devel
[01:22:28] <andypugh> Am am going to be very bored in a hotel in Detroit most evenings next week, I can add that to the list.
[01:23:23] <Tom_itx> don't venture far from home in that town
[01:23:46] <andypugh> Well, I am actually staying in Dearborn
[01:24:35] <andypugh> And I do wonder how much worse than Pitsea Detroit really is. Except for the guns thing.
[01:24:57] <Tom_itx> i've never been there and really don't plan to
[01:25:56] <Tom_itx> nice job on the lathe btw
[01:32:06] <skunkworks> andypugh, that would be awesome!
[01:33:15] <andypugh> skunkworks: Remind me next week :-)
[01:33:27] <skunkworks> :)
[01:35:04] <skunkworks> the advance has a brake that is turned on while the advance motor is on. So a little ladder is required
[01:38:38] <andypugh> You mean off while the motor is on?
[01:38:52] <skunkworks> ye
[01:38:55] <skunkworks> yes -
[01:39:30] <andypugh> My Holbrook lathe spindle has a brake. It’s strange to see the spindle stop so fast. I can’t decide whether to use it.
[01:40:16] <skunkworks> run the forward motor -> stop by turning on the 'brake' relay witch still requires the forward relay to be one. So I need a time delay that when the forward pin goes low - the brake is turned on for a time - then the motor and brake relay get shut off
[01:41:07] <andypugh> skunkworks: It’s an easy extra output to add. But really the brake doesn’t need ladder, it is just “NOT (motor-fwd OR motor-rev)”
[01:41:43] <andypugh> We are talking out of synch, clearly.
[01:42:50] <andypugh> You can do what you describe in HAL or in Ladder, I think it comes down to where you are comfortable.
[01:43:53] <skunkworks> right
[01:44:14] <skunkworks> no - if the forward relay isn't on - the brake doesn't work
[01:44:53] <andypugh> That’s silly.
[01:45:14] <andypugh> Probably made sense without HAL.
[01:45:28] <skunkworks> yes - it could be the forward or reverse relay is always on - with the brake relay activated.. I don't know
[01:45:43] <andypugh> Like the 555 timer in my millng machine PSU made sense, until I realised HAL could do it for free.
[01:45:50] <skunkworks> heh
[01:46:00] <skunkworks> hal/classic ladder is pretty aweome
[01:46:03] <skunkworks> awesome
[01:46:16] <andypugh> I found a fun trick with LUT5.
[01:46:24] <skunkworks> the K&T gear shift was done in hal/comp. the tool changer was done in ladder
[01:46:46] <andypugh> If you feed the output to one of the inputs then LUT5 becomes “stateful”
[01:46:57] <pink_vampire> my BF working on wire some stuff in the panel.
[01:47:59] <andypugh> pink_vampire: You are slipping too easily into gender roles, tell him too cook dinner while you do the wiring.
[01:48:39] <pink_vampire> andypugh: I can't pull the wires.
[01:49:17] <andypugh> You might break your fingernails? I know _exactly_ what you mean :-)
[01:49:32] <pink_vampire> also that.
[01:49:50] <pink_vampire> but I can't close the screws tight.
[01:50:01] <andypugh> (I have always had longer fingernails than my girlfriends)
[01:51:02] <pink_vampire> she bite them?
[01:52:24] <andypugh> skunkworks:
http://codepad.org/zupMW54p
[01:53:42] <andypugh> pink_vampire: I started to keep them long as a kif to use them as srewdrivers for my favourite toy, Meccano.
[01:54:13] <pink_vampire> omg!!!!!!!!
[01:54:21] <pink_vampire> I love the gyro.
[01:54:31] <pink_vampire> the best thing ever!!
[01:54:49] <andypugh> (kid, not kif. I couldn’t be less Kif, they are obligate carnivores from the Chanur series by Cherryh)
[01:55:03] <andypugh> pink_vampire: Cickspring?
[01:55:20] <pink_vampire> noo
[01:55:35] <pink_vampire> one sec I will give you a link
[01:56:54] <pink_vampire> http://www.sears.com/4-v-max-gyro-8482-screwdriver/p-00938462000P?sid=IDx01192011x000001&gclid=CKikl5ruxswCFYMehgod1KkPHg&gclsrc=aw.ds
[01:58:35] <pink_vampire> andypugh: ^
[01:59:04] <pink_vampire> there is a video in the page that show how it work.
[01:59:45] <andypugh> Ah, right.
[01:59:53] -!- skunksleep has quit [Ping timeout: 265 seconds]
[02:01:16] <andypugh> pink_vampire: Look at the Bosch Ixo too, been out for about 10 years, while you were (I assume) asleep. :-)
[02:02:34] <pink_vampire> the coolest thing about the gyro is there is no right and left switch
[02:02:59] <andypugh> Ah,
[02:03:22] <pink_vampire> the way you tilt it is't how it rotate
[02:04:43] <andypugh> Right, clever. Pointless and unnecessary, but clever.
[02:05:36] <andypugh> It lets you twist the wrong way because you are in an unusual place relative to the driver just like the real thing. :-)
[02:15:14] <pink_vampire> Hi, Pink`s BF here.
[02:16:35] <pink_vampire> Just finished wiring some stuff, we tested the probe and for some odd reason the Z axis moves when you probe the X or Y axis
[02:17:10] <pink_vampire> Besides that, we are getting an error: "user probe error".
[02:17:52] <pink_vampire> The same setup works perfectly in Mach 3...
[02:18:03] -!- skunksleep [skunksleep!~AndChat14@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[02:19:02] <pink_vampire> andypugh: any thoughts?
[02:19:54] <andypugh> None at all. That’s just inexplicable.
[02:20:14] <andypugh> What G-code is being run?
[02:21:22] -!- teepee has quit [Ping timeout: 244 seconds]
[02:21:54] <pink_vampire> andypugh: I'm not running any G code, I simply click the buttons in the Probe Screen interface.
[02:22:33] <andypugh> That will be running some G-code “under the hood”
[02:22:49] <andypugh> Is this the Verser Probe-screen?
[02:23:37] <pink_vampire> andypugh: Yes, it came from his github page.
[02:23:47] <andypugh> Link?
[02:24:47] <andypugh> Got it
[02:24:54] <pink_vampire> http://i.imgur.com/V0fxqcM.png
[02:25:20] <pink_vampire> andypugh: See the error we are getting at the bottom of the screen.
[02:25:43] <pink_vampire> https://github.com/verser-git/probe_screen
[02:26:13] <pink_vampire> ^^ That`s where I have downloaded and installed for her.
[02:26:16] <andypugh> There are G-cdoe routines included here:
https://github.com/verser-git/probe_screen/tree/master/macros
[02:26:47] <pink_vampire> andypugh: These .NGC files are in there.
[02:27:04] -!- teepee [teepee!~teepee@unaffiliated/teepee] has joined #linuxcnc-devel
[02:27:30] <andypugh> I suspect if you put a (MESSAGE, I be running G-code, I be) in each one of those .ngc files you will see them being run.
[02:28:02] <pink_vampire> down.ngc xminus.ngc xplus.ngc yminus.ngc yplus.ngc
[02:28:58] <pink_vampire> Also how come there are only 5 .ngc files, but 22 buttons?
[02:30:21] <andypugh> I think the Python calls them in various different ways
[02:30:51] <pink_vampire> Look for an example at yplus.ngc, there is no Z in the file, but the Z axis moves down during the probing process, (and never goes back up).
[02:33:31] <pink_vampire> Is there any documentation for Probe Screen?
[02:33:50] <andypugh> Maybe it also calls G-code directly? Lots if G90 etc in
https://github.com/verser-git/probe_screen/blob/master/python/probe_screen.py (I have not found the definitions)
[02:34:24] <andypugh> Apart from:
https://github.com/verser-git/probe_screen/blob/master/README.md ?
[02:34:39] <andypugh> It appears not
[02:35:46] <pink_vampire> Read that, used it to do the install of Probe Screen.
[02:36:21] <pink_vampire> No one here uses that?
[02:37:14] <andypugh> Not yet, I keep meaning to install it
[02:37:39] <pink_vampire> It`s easy enough, I can probably do that for you.
[02:38:36] <pink_vampire> andypugh: ^
[02:41:19] <andypugh> It’s a matter of getting round to it, and needing it.
[02:41:36] <pink_vampire> andypugh: All I need you to do is to copy the 'Macro' and 'Python' folders under your 'mymill' and send me your .ini file, so I can edit it.
[02:41:45] -!- ktchk [ktchk!~eddie6929@n219079181044.netvigator.com] has joined #linuxcnc-devel
[02:41:59] <andypugh> So far I have got by by just typing G38.2 a lot in the MDI window
[02:42:40] <pink_vampire> andypugh: If this works properly, you will save lots of time when you do a set-up.
[02:43:15] <andypugh> pink_vampire: is this pink_vampire or the BF? I do know how to edit INI files, I am one of the devs :-)
[02:44:37] <skunksleep> I have been using cradeks probing routines. Been meaning to try that screen though
[02:44:43] <pink_vampire> BF here... sorry, didn`t mean to offend you or anything, I have no clue in this field, and I'm no programmer, just trying to help her out.
[02:46:06] <pink_vampire> This thing not working right, drives her bananas.
[02:47:57] <pink_vampire> andypugh: She is like a cute little ballistic missile ready to explode...
[02:50:06] <andypugh> It _sounds_ like noise on the probe signal is triggering warnings on non-probe moves. The routines have to do some non-probe moves to position the probe.
[02:51:30] <andypugh> You might want to look at whether vibrations trigger the probe, and possibly look at passing the probe input through a software debounce. 3mS delay isn’t a huge loss of spatial accuracy.,
[02:54:02] <pink_vampire> andypugh: If there was a noise or vibration problem, than Mach 3 shouldn't have worked either...
[02:55:08] <andypugh> Unless the Mach config had a debounce. Or perhaps Mach just doesn’t care if the probe input changes state when not probing.
[02:55:08] <pink_vampire> andypugh: How do you do a software debounce?
[02:55:51] <andypugh> http://linuxcnc.org/docs/2.7/html/man/man9/debounce.9.html
[02:58:23] <andypugh> Is that enough, or do you need more?
[02:58:51] -!- pcw_home [pcw_home!~chatzilla@c-50-143-148-115.hsd1.ca.comcast.net] has joined #linuxcnc-devel
[02:59:03] <andypugh> That tells you the function name and the loadrt syntax.
[02:59:17] <pink_vampire> andypugh: I see that it uses the loadrt command, so I assume it goes in the .hal file, right?
[02:59:21] -!- the_wench [the_wench!~the_wench@host81-149-189-98.in-addr.btopenworld.com] has joined #linuxcnc-devel
[02:59:37] -!- AndChat|144384 [AndChat|144384!~AndChat14@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[03:00:11] <andypugh> Yes, then you need to addf the function to a thread. (possibly base-thread for probing)
[03:00:48] -!- skunksleep has quit [Ping timeout: 265 seconds]
[03:01:09] AndChat|144384 is now known as skunksleep
[03:01:21] <andypugh> example:
https://forum.linuxcnc.org/forum/38-general-linuxcnc-questions/12062-joint-0-limit-error-setting-up-debounce?start=6#12083
[03:02:11] <andypugh> You would be using the probe pin, and probably don’t need the delsig, I gave that example as something that would go in a custom HAL
[03:03:43] <pink_vampire> So I need: loadrt debounce cfg=1 for only one channel, the probe, right?
[03:03:52] <andypugh> Yes
[03:05:01] <pink_vampire> And how will the code figure out that my channel or pin number is 11?
[03:05:14] -!- KimK has quit [Ping timeout: 276 seconds]
[03:05:16] <andypugh> You need to tell it.
[03:06:19] <andypugh> something like net probe-raw parport.0.pin-11-in => debounce.0.0.in
[03:06:33] <pink_vampire> Ok, let`s see if I can figure it out.
[03:06:36] -!- BeachBumPete [BeachBumPete!~IceChat9@2601:585:8201:5d87:5cf9:c1b6:3f97:5186] has joined #linuxcnc-devel
[03:06:57] <andypugh> and net probe-filtered debounce.0.0.out => motion.probe
[03:07:45] <andypugh> Don’t forget to set the debonce length and do the addf to make it actually run in a thread.
[03:08:26] <andypugh> I guessed the name of the motion pin for probe input.
[03:14:10] <pcw_home> motion.probe-input
[03:14:20] -!- BeachBumPete has quit [Ping timeout: 276 seconds]
[03:14:23] <andypugh> Thanks
[03:14:47] <andypugh> It’s 4am. too late to think. In fact i think it’s time to log off.
[03:14:54] -!- BeachBumPete [BeachBumPete!~IceChat9@2601:585:8201:5d87:5093:33ec:98b4:d3eb] has joined #linuxcnc-devel
[03:15:12] <pcw_home> 'nite
[03:15:40] <pink_vampire> andypugh: Give me a minute, almost done with the modification
[03:16:48] <andypugh> If I type a lot of random characters it means I have fallen asleep with my forehead on the keyboard.
[03:17:45] <pink_vampire> andypugh:
http://pastebin.com/746yzuQj
[03:19:32] -!- BeachBumPete has quit [Ping timeout: 276 seconds]
[03:20:13] <andypugh> Yo need to net the debounce.0.0.out to the same signal/net as motion.probe-input
[03:20:44] <andypugh> And I don’t see the debonce in the addf list
[03:22:46] <pink_vampire> Hmm...
[03:23:21] <pink_vampire> Is that base-thread ?
[03:24:33] <andypugh> I think that the base-thread is more appropriate. But given that I think that motion only looks at the probe-input every 1mS you should probaby set the debounce to be about half a millisecond.
[03:24:55] <pink_vampire> andypugh: This way: addf debounce.0.0 base-thread
[03:24:58] <pink_vampire> ?
[03:25:07] <andypugh> So 20 for a 25uS base-thread
[03:25:32] -!- BeachBumPete [BeachBumPete!~IceChat9@2601:585:8201:5d87:9895:58a3:89fd:ca3] has joined #linuxcnc-devel
[03:25:43] -!- BeachBumPete [BeachBumPete!~IceChat9@2601:585:8201:5d87:9895:58a3:89fd:ca3] has parted #linuxcnc-devel
[03:25:57] <pink_vampire> andypugh: Not sure how this work
[03:26:11] <andypugh> debounce is confusing. It’s addf debounce.0 for every channel in debounce.0 (all one of them)
[03:26:24] <pink_vampire> andypugh: 20 debounces per 25 MicroSeconds?
[03:26:48] <andypugh> 20 samples at base-thread frequency
[03:27:32] <pink_vampire> Ok, so simply: addf debounce.0 base-thread ?
[03:27:39] <andypugh> Yes
[03:28:43] <andypugh> That would run debounce.0.0, debounce.0.1, debounce.0.N at base-thread frequency all with the same debounce count
[03:29:17] <andypugh> I think that debounce is the only component that works like that
[03:29:56] <andypugh> I remember being baffled by it several years ago when I firs started.
[03:30:59] <pink_vampire> andypugh:
http://pastebin.com/B7ugEyHA <-That`s all together now, am I missing anything?
[03:32:24] <andypugh> I still don’t see debounce.0.0.out going anywhere
[03:33:30] <pink_vampire> andypugh: So in my case: net probe-raw parport.0.pin-11-in => debounce.0.in
[03:33:35] <pink_vampire> and
[03:34:04] -!- kingarmadillo has quit [Ping timeout: 260 seconds]
[03:34:54] <andypugh> net pinkys-special-debounced-probe-signal deboince.0.0.out motion.probe-input
[03:35:07] <pink_vampire> :-)
[03:35:17] <andypugh> (just to stress that signal names have _no_ special significance
[03:37:13] <pink_vampire> So: net probe1 deboince.0.0.out motion.probe-input ??
[03:38:15] <andypugh> Well, you would want to spell “debounce” correctly, like I failed to.
[03:39:07] <pink_vampire> net probe1 debounce.0.0.out motion.probe-input
[03:39:54] <pink_vampire> So I don't need: net probe-raw parport.0.pin-11-in => debounce.0.in ?
[03:40:11] <pink_vampire> Or do I need both?
[03:40:13] <andypugh> yes, you do
[03:40:20] <pink_vampire> Ok
[03:41:02] <andypugh> The signal passes from the parport to the debounce,in then from the deboince.out to the motion probe input
[03:43:25] <andypugh> Right, I really need to sleep now.
[03:43:32] -!- andypugh has quit [Quit: andypugh]
[03:54:56] -!- KimK [KimK!~Kim__@ip68-102-66-31.ks.ok.cox.net] has joined #linuxcnc-devel
[04:04:42] -!- ktchk has quit [Quit: ktchk]
[04:56:53] -!- teepee has quit [Ping timeout: 244 seconds]
[04:58:16] -!- teepee [teepee!~teepee@unaffiliated/teepee] has joined #linuxcnc-devel
[05:56:24] -!- skunksleep has quit [Ping timeout: 260 seconds]
[06:01:30] -!- skunksleep [skunksleep!~AndChat14@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[06:13:07] -!- kwallace [kwallace!~kwallace@162.222.30.12] has parted #linuxcnc-devel
[06:58:18] -!- rob_h [rob_h!~robh@94.0.120.220] has joined #linuxcnc-devel
[07:34:59] -!- skunksleep has quit [Ping timeout: 276 seconds]
[07:46:48] -!- Mathnerd314 has quit [Ping timeout: 268 seconds]
[09:06:38] -!- skunksleep [skunksleep!~AndChat14@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[10:41:12] -!- jthornton [jthornton!~john@198.45.191.246] has joined #linuxcnc-devel
[10:41:15] -!- JT-Shop [JT-Shop!~john@198.45.191.246] has joined #linuxcnc-devel
[10:41:49] -!- JT-JA13 [JT-JA13!~john@198.45.191.246] has joined #linuxcnc-devel
[10:50:48] -!- teepee has quit [Ping timeout: 244 seconds]
[10:51:12] -!- teepee [teepee!~teepee@unaffiliated/teepee] has joined #linuxcnc-devel
[11:02:15] -!- sel [sel!~sel@net77-43-27-64.mclink.it] has joined #linuxcnc-devel
[11:03:29] -!- sel has quit [Client Quit]
[11:21:49] -!- kingarmadillo has quit [Ping timeout: 244 seconds]
[12:10:46] -!- pink_vampire has quit [Quit: KVIrc 4.9.1 Aria http://www.kvirc.net/]
[12:49:33] -!- Komzpa has quit [Ping timeout: 240 seconds]
[12:58:59] -!- ktchk [ktchk!~eddie6929@n219079181156.netvigator.com] has joined #linuxcnc-devel
[13:43:51] -!- BeachBumPete [BeachBumPete!~IceChat9@2601:585:8201:5d87:9895:58a3:89fd:ca3] has joined #linuxcnc-devel
[13:51:47] -!- BeachBumPete [BeachBumPete!~IceChat9@2601:585:8201:5d87:9895:58a3:89fd:ca3] has parted #linuxcnc-devel
[14:04:45] -!- Roguish [Roguish!~chatzilla@c-50-143-183-159.hsd1.ca.comcast.net] has joined #linuxcnc-devel
[14:23:55] -!- ktchk has quit [Quit: ktchk]
[14:51:08] -!- Komzpa has quit [Ping timeout: 276 seconds]
[14:58:17] -!- Mathnerd314 [Mathnerd314!~quassel@supertux/Mathnerd314] has joined #linuxcnc-devel
[15:06:27] -!- kwallace [kwallace!~kwallace@162.222.30.12] has joined #linuxcnc-devel
[16:38:48] -!- liedman has quit [*.net *.split]
[16:38:48] -!- olli- has quit [*.net *.split]
[17:07:05] -!- kalxas has quit [Changing host]
[17:52:25] -!- Komzzpa has quit [Ping timeout: 244 seconds]
[19:15:36] -!- md-2 has quit [Ping timeout: 244 seconds]
[19:16:06] -!- teepee has quit [Ping timeout: 244 seconds]
[19:16:21] -!- kalxas has quit [Read error: Connection reset by peer]
[19:17:22] -!- teepee [teepee!~teepee@unaffiliated/teepee] has joined #linuxcnc-devel
[19:45:21] -!- anth0ny_ has quit [Client Quit]
[20:43:56] -!- Roguish_ [Roguish_!~chatzilla@c-50-143-183-159.hsd1.ca.comcast.net] has joined #linuxcnc-devel
[21:10:29] -!- Mathnerd314 has quit [Ping timeout: 260 seconds]
[21:18:24] -!- Komzpa has quit [Ping timeout: 246 seconds]
[21:47:56] -!- anth0ny_ has quit [Quit: anth0ny_]
[21:59:38] -!- andypugh [andypugh!~andypugh@cpc14-basl11-2-0-cust1010.20-1.cable.virginm.net] has joined #linuxcnc-devel
[22:30:55] -!- anth0ny_ has quit [Quit: anth0ny_]
[22:40:08] <andypugh> Which version of Python does the Lucif builbot use?
[22:40:18] <andypugh> (Lucid. Buildbot)
[23:29:02] <jepler> andypugh:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?MinimumSoftwareVersions
[23:29:09] <jepler> 2.7 I guess
[23:30:14] <andypugh> There is a discussion on the other channel about compiling 2.7 or master on 10.04
[23:30:27] <andypugh> I am not sure that I am giving good advice.
[23:34:42] -!- maurris has quit []
[23:44:58] -!- Mathnerd314 [Mathnerd314!~quassel@supertux/Mathnerd314] has joined #linuxcnc-devel
