#linuxcnc | Logs for 2012-04-13

Back
[00:00:34] <JT-Shop> usually X is positive and Y is negative and Z is negative numbers
[00:00:50] <joe9> your home is 0.00 and the limits are -50 to 51
[00:01:11] <JT-Shop> yea, ignore my not correct limits
[00:01:27] <JT-Shop> should really be -0.250 51
[00:01:37] <joe9> yes, that is what I was getting at.
[00:01:52] <JT-Shop> your going to make me fix that aren't you
[00:02:03] <joe9> and home_offset == min_limit, in that case?
[00:02:05] <JT-Shop> but I really don't care with the plasma
[00:02:26] <joe9> shouldn't home_offset = min_limit + certain offset
[00:02:26] <JT-Shop> pretty much as that would be on the switch
[00:02:28] <joe9> in your case.
[00:02:57] <joe9> to avoid the limit error.
[00:03:07] <JT-Shop> yea, to be sure you don't hit the limit switch you are correct
[00:03:34] <JT-Shop> no you only have to not sitting on the switch when the homing move is complete to avoid the error
[00:03:46] <djdelorie> but if your limit switch *is* the home switch, there can't be an offset
[00:04:04] <JT-Shop> there has to be an offset
[00:04:17] <JT-Shop> or you wil get an error every time you try and home
[00:04:17] * djdelorie still thinks it's either debounce or wrong homing velocity
[00:04:22] <JT-Shop> nope
[00:04:41] <djdelorie> there has to be an offset between the switch location and the final home location, but there can't be an offset between the limit switch and the home switch
[00:05:05] <JT-Shop> that makes no sense to me at all
[00:05:06] <djdelorie> his homing operation doesn't finish, so something is going wrong before it tries to move to the home location
[00:05:22] <JT-Shop> he has NO move to a home location
[00:05:30] <JT-Shop> it is set at 0
[00:05:47] <djdelorie> http://codepad.org/fNeNefP1 - HOME_OFFSET is 0 but HOME is 10
[00:06:10] <JT-Shop> yep
[00:06:12] <djdelorie> just like my setup, and mine works
[00:06:22] <JT-Shop> there is no move to home position
[00:07:22] <djdelorie> isn't the HOME=10 the move to the home position?
[00:07:47] <JT-Shop> yep, I'm full of shit, just refreshed my memory
[00:08:04] <JT-Shop> unfortunaly that is The position that the joint will go to upon completion of the homing sequence.
[00:08:23] <JT-Shop> if the limit switch is still tripped it will never happen
[00:09:27] <djdelorie> if the limit switch *is* the home switch, by definition that can't be the case
[00:09:42] <djdelorie> unless it has a HUGE hysteresis
[00:10:08] <djdelorie> 10mm is a big swing for a switch
[00:13:04] <JT-Shop> or final velocity is too high and your way over traveled the switch
[00:13:46] <JT-Shop> all I really know is what works on my plasma and what I read in the docs
[00:15:08] <JT-Shop> yep just reading back and I missed the wrong homing velocity...
[00:15:22] * JT-Shop really has to start dinner now... good luck
[00:15:33] <Tom_itx> JT-Shop
[00:15:37] <Tom_itx> did that help your PC?
[00:16:19] -!- factor has quit [Ping timeout: 260 seconds]
[00:17:14] <joe9> djdelorie: the limit switches are connected to the NC, correct?
[00:17:32] <joe9> closed when the switch is not activated.
[00:17:55] -!- cylly2 [cylly2!cylly@p54B1223C.dip.t-dialin.net] has joined #linuxcnc
[00:18:18] <djdelorie> doesn't matter as long as you set the invert or not correctly
[00:19:05] <joe9> net min-home-x <= parport.0.pin-10-in-not
[00:19:09] <joe9> i have that.
[00:19:13] <joe9> and it is NC.
[00:19:27] <joe9> I am suspecting that it should be "not inverted".
[00:19:35] <djdelorie> if you get a limit switch error when the axes are nowhere near the limit switches, you have at least one of them backwards
[00:20:20] <djdelorie> you should be able to jog the axes manually without errors as long as you stay away from the limit switches
[00:20:23] -!- Loetmichel has quit [Ping timeout: 260 seconds]
[00:21:18] -!- gambakufu has quit []
[00:22:04] -!- factor [factor!~factor@r74-193-21-187.cnrocmta01.conrtx.tl.dh.suddenlink.net] has joined #linuxcnc
[00:27:44] <joe9> removed the invert, now I can jog the axes manually as long as I stay away from the limit switches.
[00:28:28] <djdelorie> now, use the jogging controls to check each limit/home switch to see if it's in the positive direction or the negative one
[00:28:36] <joe9> djdelorie: why not just have a home switch? instead of a limit + home switch?
[00:28:58] <djdelorie> if you have another switch you can mount, and another pin (or three) on your parallel port, go for it
[00:29:27] <joe9> no, I do not. should there be a limit switch if I have a home switch?
[00:29:45] <joe9> can I not have no limit switches, just a home switch?
[00:30:03] <djdelorie> if you don't want the software to stop itself from smashing up your machine, go for it
[00:30:42] <djdelorie> in theory, you can have just one home swtich per axis, and trust the software to stay away from the limits, but the switches are there for when the software is wrong
[00:30:43] <joe9> but, I would have hard limits in the .ini file, that would prevent that.
[00:30:53] <joe9> oh, ok. got it.
[00:30:55] <joe9> thanks.
[00:30:56] <djdelorie> s/would/should/
[00:33:04] <djdelorie> on mine, the controllers also watch the limit switches, and won't travel past them even if the software says to
[00:33:29] -!- GeorgeH [GeorgeH!~George@c-69-141-105-145.hsd1.nj.comcast.net] has joined #linuxcnc
[00:36:59] -!- rob_h has quit [Ping timeout: 244 seconds]
[00:37:52] <Tom_itx> hard or soft limits or both?
[00:38:43] <joe9> djdelorie: just curious, do you have 2 limit switches per axis?
[00:39:00] <djdelorie> my controllers have separate home and limit configurations. The motors won't go past the limit switches, but they do keep track of where the motors *would* be
[00:39:07] <djdelorie> I have two per axis, one on each end
[00:39:38] <djdelorie> Tom_itx: so the switches are placed near the hard limits, but the PC software can treat them as soft limits
[00:40:00] -!- asdfasd has quit [Ping timeout: 265 seconds]
[00:40:23] <Tom_itx> how much do you back off of them ? .1" or so?
[00:40:35] <djdelorie> mine? I put them about 5mm from the end
[00:43:09] <Tom_itx> i just have one limit on z
[00:43:31] <djdelorie> in theory, you don't need *any* switches, if you manually home the machine and trust it to do the right thing.
[00:43:35] <Tom_itx> 2 on x and y but one of the y isn't hooked up atm
[00:43:51] <Tom_itx> i use the same switches for home
[00:44:11] <djdelorie> me too
[00:46:38] <joe9> djdelorie: are you using gecko G540?
[00:46:43] <joe9> oh, sorry
[00:46:50] <joe9> just realized that you were on servos.
[00:46:58] <djdelorie> yup, with custom control electronics
[00:47:19] <joe9> "Home switch inactive before start of backoff move" any thoughts on this error?
[00:47:59] <djdelorie> debounce
[00:55:43] <joe9> is there some way of just homing a particular axis?
[00:55:53] <joe9> don't bother.
[00:56:00] <joe9> homing sequence is what I need
[00:56:10] <djdelorie> you can home one axis in the machine menu, I think
[00:56:18] <Tom_itx> it's in the ini
[00:56:21] <Tom_itx> the sequence
[00:56:37] <Tom_itx> or the hal file
[00:56:39] <Tom_itx> one or the other
[00:58:53] <djdelorie> changing the homing sequence won't help with debouncing a mechanical switch
[01:03:40] -!- Valen1 [Valen1!~Valen@c122-108-45-139.blktn6.nsw.optusnet.com.au] has joined #linuxcnc
[01:05:21] Valen1 is now known as Valen
[01:12:52] -!- cmorley has quit [Ping timeout: 265 seconds]
[01:27:15] <joe9> djdelorie: reducing the search velocity to 0.05 helped with the issue
[01:27:19] -!- stevegt_ has quit [Ping timeout: 252 seconds]
[01:27:35] <djdelorie> still should debounce
[01:28:00] <joe9> don't know, it seems to work fine. atleast the other axis did.
[01:28:25] <jdhnc> debounce is just a few lines.
[01:28:43] <djdelorie> or a capacitor
[01:29:24] <jdhnc> software is easier to tweak
[01:30:14] <djdelorie> yeah, but I know where my capacitors are ;-)
[01:30:23] <jdhnc> I know where my custom.hal is
[01:30:28] <joe9> i can add a capacitor. any ideas on what size the capacitor should be?
[01:31:04] <joe9> jdhnc: how do you recommend adding a custom software debounce? any examples that can help me through.
[01:31:57] <jdhnc> let me see if I can get to my emc box from here.
[01:32:02] <djdelorie> there's a software debounce module, but it delays the edge by a little time. A capacitor across the switch contacts (assuming an N.O. switch) is immediate
[01:32:18] <djdelorie> or a different debounce routine that only debounces the release, not the hit
[01:33:00] <djdelorie> I went for an 0.1 second delay before de-activating the limit switch
[01:33:33] <jdhnc> are you actually worried about a few cycles of debounce delay for a mechanical switch?
[01:34:08] <djdelorie> it depends on how accurately you want to hit that position, and a mechanical switch can bounce for a remarkably long time
[01:34:24] <djdelorie> since mine double as limit switches, I want the "hit" to happen as soon as possible, too.
[01:34:50] <djdelorie> 0.1 seconds at full speed is plenty of time to smash things
[01:40:00] <Tom_itx> switches vary but 30ms should be enough
[01:41:18] <djdelorie> even at 30ms, I'd have to move my limit switches
[01:41:41] <djdelorie> hence, debounce only the release, hits happen immediately
[01:42:02] <djdelorie> well, only my Y switches
[01:42:43] <djdelorie> maybe I'm just paranoid, but it's a limit switch - an unusual situation, I don't want any delay in handling it
[01:42:57] <jdhnc> I think you are right
[01:43:18] <djdelorie> if you have n.o. switches, a cap across the contacts does the job anyway
[01:44:38] <Valen> software debounce should be instant if its written right
[01:45:04] <Valen> take action on the trigger then mask off the bounces for a set period
[01:45:31] <djdelorie> I have similar code in http://www.delorie.com/cnc/machineactive.comp
[01:45:51] <djdelorie> active turns on immediately, but there's a delay before it turns off
[01:46:27] <jdhnc> mine are all NC
[01:46:52] -!- Gast540 [Gast540!~Gast540@203.83.196.18] has joined #linuxcnc
[01:47:26] -!- Gast540 has quit [Client Quit]
[01:49:40] -!- sumpfralle has quit [Ping timeout: 272 seconds]
[01:51:35] <joe9> djdelorie: "if you have n.o. switches, a cap across the contacts does the job anyway" -- how about NC switches?
[01:51:51] <djdelorie> I can't think of an easy hardware solution for those
[01:52:21] <joe9> i have an NC switch now. I could change to NO and invert the switch in stepconf, correct?
[01:52:38] <djdelorie> with N.O. the switch immediately discharges the cap, and the pullup slowly recharges it once the switch opens
[01:52:43] <jdhnc> do you only have the one switch on that pin?
[01:52:51] <joe9> yes.
[01:53:17] <jdhnc> do the software thing first and see if you are happy.
[01:53:26] <jdhnc> then buy a better switch if not.
[01:54:14] <djdelorie> the built-in debounce module should be enough for now anyway
[01:54:32] <joe9> djdelorie: "or a different debounce routine that only debounces the release, not the hit" -- any thoughts on how to go about this?is
[01:54:34] -!- sumpfralle [sumpfralle!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[01:54:44] <djdelorie> joe9: look at the .comp URL I posted
[01:55:05] <joe9> i am, just wanted to check that it is what you were talking about.
[01:55:20] <djdelorie> it's a starting point
[01:55:25] <joe9> ok, thanks.
[01:55:42] <djdelorie> for now, just use the debounce that comes with linuxcnc. At least it will get your auto-homing working :-)
[01:56:57] <Tom_itx> http://www.all-electric.com/schematic/debounce.htm
[01:58:46] <joe9> one of the stepper motors at the reverse velocity of 0.025 has a screeching sound just when it is moving. is that common?
[01:58:53] <joe9> it is happening on only one motor.
[01:59:10] <Tom_itx> maybe the pulse is too narrow?
[01:59:50] <joe9> Tom_itx: that is a good article. thanks.
[02:33:18] -!- demacus [demacus!~demacus@erft-4d07d11d.pool.mediaWays.net] has joined #linuxcnc
[02:33:34] -!- capricorn_one [capricorn_one!~raffi@zima.linwin.com] has joined #linuxcnc
[02:35:59] -!- demacus_ has quit [Ping timeout: 245 seconds]
[02:47:07] -!- koax has quit [Ping timeout: 265 seconds]
[02:53:32] -!- kb8wmc has quit [Quit: ChatZilla 0.9.88.2 [Firefox 11.0/20120310193829]]
[03:14:01] -!- sumpfralle has quit [Ping timeout: 252 seconds]
[03:35:52] -!- ve7it has quit [Remote host closed the connection]
[03:47:18] -!- stevegt_1 [stevegt_1!~stevegt@64-130-244-2-gracehospitalitygroup.bwg.phx.sparkplugbb.net] has joined #linuxcnc
[03:52:08] <joe9> djdelorie: can you please post your axis_2 section of the .ini file?
[03:52:30] <joe9> the search and latch are working fine, but, the home is going in the opposite direction.
[03:52:35] <djdelorie> http://www.delorie.com/cnc/dj_cnc_1.ini
[03:53:39] <djdelorie> if your jogging needs to be reversed, invert the direction pin for that axis
[03:53:49] <joe9> the axis_2 home is closer to the max_limit than the min_limit.
[03:54:01] <djdelorie> if the jogs are right but the switch isn't where you think it is, move the home_offset to that end
[03:54:02] <joe9> ok, thanks. will do.
[03:54:20] <joe9> yes, the jogging is correct, it is just the switch.
[03:54:33] <djdelorie> mine home to the positive end
[03:55:16] <joe9> http://codepad.org/e59Vlukg is my z_axis
[03:55:30] <joe9> HOME = -100 disregard this
[03:55:40] <joe9> i am about to add this in to see how it works.
[03:59:13] <ssi> I've got a real bad chip stringing problem
[03:59:19] <ssi> damn machine's building birds nests :(
[03:59:39] <djdelorie> are you sure "chip" is the right word then? ;-)
[04:00:01] <ssi> what would you prefer I call them?
[04:01:00] <ssi> https://p.twimg.com/AqVTuk2CEAImzSI.jpg:large
[04:02:00] <djdelorie> shavings ? tinsel?
[04:03:15] <ssi> as you can imagine, it's somewhat scary when it does a tool change and whips that ball of razorwire around
[04:03:46] <djdelorie> you need a chipbreaker on whatever tool that was
[04:04:37] <ssi> it has a chipbreaker!
[04:04:48] <djdelorie> the photo says otherwise :-)
[04:05:16] <djdelorie> maybe you need a CNC shop vac ;-)
[04:06:04] <Valen> needs moar chip breaker
[04:06:31] <djdelorie> or a different cutting angle, so the shavings come off in a clean spiral and fall to the floor
[04:06:34] <Valen> need some of this http://newsliteimgs.s3.amazonaws.com/100430_hulk1.jpg
[04:07:11] <ssi> needs moar DOC I think
[04:09:02] <joe9> in this custom.hal file, shouldn't there be a reference to the limit too?
[04:09:09] <joe9> i see the reference to home-sw-in.
[04:09:19] <joe9> but, not to the min limit.
[04:09:28] <joe9> just wanted to check that it is needed.
[04:10:00] <djdelorie> mine does: http://www.delorie.com/cnc/dj_cnc_1.hal search for "lim-sw"
[04:11:38] <ssi> twice the DOC, same feed and speed, still stringing
[04:11:40] <ssi> FML :(
[04:12:50] -!- mhaberler [mhaberler!~mhaberler@195.191.253.94] has joined #linuxcnc
[04:15:39] -!- phantoxe has quit []
[04:23:35] <joe9> djdelorie: i notice that you have the zdir inverted. that is what I am missing. now, I think I have it mirrored to yours.
[04:23:50] * djdelorie thought I had Y inverted
[04:24:05] <joe9> djdelorie: the only issue I notice is that in the gui display, when the z axis is physically moving up, the cone is moving down.
[04:24:05] <djdelorie> inverting the DIR pin changes the *jog* direction also
[04:24:12] -!- stevegt_1 has quit [Ping timeout: 272 seconds]
[04:24:23] <djdelorie> for Z, positive should be UP
[04:24:29] <joe9> yes, i changed the search and latch velocity to negative.
[04:24:39] <djdelorie> check your Z jog accordingly. PgUp should move the Z axis up
[04:25:18] <djdelorie> no, get the regular travel correct first - jogging - *then* worry about which end the home switch is on
[04:25:34] <djdelorie> the gcode ignores the home direction, you want the travel directions to be correct
[04:25:55] <joe9> djdelorie: my travel directions are messed up both on Y and Z axis
[04:26:00] <joe9> i have them both inverted.
[04:26:08] <djdelorie> fix that first, then.
[04:26:22] <joe9> ok, will do.
[04:26:27] <joe9> thanks for that tip.
[04:29:49] <joe9> djdelorie: now, I have the jog directions correct.
[04:29:57] <joe9> the PgUp moves the Z-axis up
[04:30:04] <joe9> and PgDown moves the Z-axis down
[04:30:10] <djdelorie> ok, now for each axis, determine if the switch is hit with a positive jog or a negative one
[04:30:24] <djdelorie> the set the HOME_OFFSET, HOME, and homing speeds accordingly
[04:30:27] <joe9> and, that is possible without the invert on z-axis. invert is needed on y-axis though.
[04:31:10] <joe9> the switch for the z-axis is hit with a Pg-up (positive jog, correct?)
[04:33:00] <r00t4rd3d> http://i.imgur.com/Vm2UI.jpg
[04:33:17] <djdelorie> joe9: depends on where you put the switch ;-)
[04:33:27] <joe9> djdelorie: i put the switch on the top
[04:33:40] <djdelorie> then your Z is set up like mine; I home up also
[04:34:15] <r00t4rd3d> 3 of us building a house :/
[04:34:40] <djdelorie> you need a really big cnc pick and place machine there...
[04:35:58] <r00t4rd3d> we poured the footers today, sucked.
[04:36:05] <joe9> djdelorie: i do not understand one thing. the min_limit is -4 for you and the max limit is 0. I am assuming that your min limit is at the top.
[04:36:17] <djdelorie> -4 < 0
[04:36:25] <joe9> the search velocity should be negative, correct? going from 0 to -4
[04:36:33] <joe9> but, you have it at +ve
[04:36:35] <djdelorie> no, 0 is at the top, -4 is at the bottom
[04:36:46] <joe9> oh, gotcha.
[04:36:47] <djdelorie> a +ve home moves up towards 0
[04:37:14] <djdelorie> +ve Z travel moves up, remember?
[04:38:03] <ssi> problem solved
[04:38:06] <joe9> then it is not a minimum limit/home switch, but a maximum limit/home switch, correct?
[04:38:34] <djdelorie> right. My axes all home to the maximum limit switch
[04:39:10] <joe9> that is where I was different. I was homing to the minimum limit switch.
[04:39:43] <djdelorie> X and Y can home either way, Z should always home up, since that's usually the not-crashing way :-)
[04:40:24] <joe9> so, the only difference in my case is that Z should be a maximum limit/home switch. Thanks.
[04:55:04] <ReadError> wow joe9
[04:55:13] <ReadError> i thought this would be much more simple
[04:55:24] <ReadError> after seeing all you have been through i fearing the task
[04:55:30] <ReadError> but still must do it
[04:56:08] <joe9> it is a learning process. dj is pretty good at it and thank god that he is around.
[04:56:22] * djdelorie only learned a few weeks ago
[05:00:05] -!- asdfasd [asdfasd!~grgrgrgrg@87.194.58.242] has joined #linuxcnc
[05:00:05] -!- asdfasd has quit [Client Quit]
[05:00:37] -!- asdfasd [asdfasd!~grgrgrgrg@87.194.58.242] has joined #linuxcnc
[05:01:17] <joe9> djdelorie: does this look ok to you? http://codepad.org/MEf4lrxF it seems to work. just wanted to get your once-over.
[05:01:20] <joe9> if you do not mind.
[05:01:33] -!- ybit2 has quit [Changing host]
[05:01:33] -!- ybit2 [ybit2!~ybit@unaffiliated/ybit] has joined #linuxcnc
[05:02:05] <djdelorie> can you put <= and => in the same hal line?
[05:03:16] -!- stevegt_1 [stevegt_1!~stevegt@64-130-244-2-gracehospitalitygroup.bwg.phx.sparkplugbb.net] has joined #linuxcnc
[05:03:21] <djdelorie> hmmm... a custom hal... I'll have to see about adding one of those to mine instead of just editing the regular one...
[05:03:52] ybit2 is now known as ybit
[05:05:02] -!- ktchk [ktchk!~eddie6929@n058153016214.netvigator.com] has joined #linuxcnc
[05:09:11] <joe9> djdelorie: i got that from jdhnc and it seems to work (afaik).
[05:09:17] <djdelorie> ok then :-)
[05:16:08] -!- H264 has quit [Read error: Connection reset by peer]
[05:16:42] -!- WalterN [WalterN!~walter@host-174-45-122-145.bzm-mt.client.bresnan.net] has joined #linuxcnc
[05:18:01] <asdfasd> good morning guys
[05:19:56] <asdfasd> I think if debounce make any difference then the electronics need improvement
[05:21:24] <djdelorie> mechanical switches bounce
[05:21:47] <djdelorie> optical interrupters or hall switches don't
[05:22:36] <asdfasd> small capacitor parallel to the switch do much more than mechanical bounce
[05:22:53] <asdfasd> also filter electrical noise
[05:22:56] <Thetawaves> it is irresponsible from a cost/engineering perspective to hardware debounce since it can be done so effectively with software.
[05:23:30] <ssi> IRRESPONSIBLE
[05:23:54] <Thetawaves> :P
[05:24:23] * djdelorie used both hardware and software debounce :-)
[05:24:24] <joe9> djdelorie: how did you come up with the search velocity? stepconf?
[05:24:39] <asdfasd> it is irresponsible to solve hardware problems with software, hardware problems should be solved by hardware, my personal oppinion
[05:24:45] <djdelorie> I watched my machine travelling and sped it up until I started worrying what would happen if it missed...
[05:25:01] <djdelorie> most of the time, I manually jog it most of the way and let it home the rest, to save time
[05:25:34] <djdelorie> but I have a big machine, homing from the other end takes time
[05:26:01] <djdelorie> well, "relatively big" ;-)
[05:26:37] <djdelorie> it is irresponsible to assume one solution is enough, when you can choose between two or more... ;-)
[05:26:50] <ssi> hahaha
[05:27:05] <ssi> I'm trying out my new smoker controller
[05:27:09] <ssi> pid loops everywhere!
[05:27:52] <asdfasd> yes, you can do just in case both, but leaving one and making another is just not proffesional
[05:28:08] -!- stevegt_1 has quit [Ping timeout: 240 seconds]
[05:33:05] <Valen> asdfasd: in an embedded system hardware and software are pretty interchangable
[05:33:35] <Valen> you want to solve switch bouncing without introducing delay, so you use a bunch of flipflops and some RC components
[05:33:45] <Valen> somebody says hey thats a pain, and makes a hardware debounce IC
[05:33:58] <Valen> then they say well thats good but this switch bounces longer/shorter
[05:34:09] <Valen> so somebody makes an I2C debounce IC
[05:34:14] <joe9> djdelorie: noticed that you do not have volatile_HOME set.
[05:34:24] <djdelorie> servos
[05:34:26] <joe9> wouldn't it be a good idea to have it.
[05:34:29] <joe9> oh, got it.
[05:35:06] <Valen> you might as well put an attiny in there then because they are mass produced and hence cheaper and can debounce 5 channels
[05:35:23] <Valen> or just put that code into your mcu and get rid of all those other failure modes
[05:35:52] <Thetawaves> ^^^
[05:36:04] <Valen> I draw the line when the hardware means the software has to guess
[05:36:06] <djdelorie> I put a small debounce RC to handle EMI noise, and a software one to handle actual bouncing
[05:36:15] <asdfasd> valen: it is like to smoke a cigarette when you are hungry - you dont feel hungry any more but you still missed your lunch
[05:36:26] <djdelorie> the software one knows which way is "hit" too, so it can debounce in the right direction
[05:36:27] <Valen> no its not
[05:36:35] <Thetawaves> djdelorie, debounce rc or a low pass filter?
[05:36:37] <Valen> the goal is to read a switch into your software
[05:37:06] <djdelorie> there's a cap from signal to gnd, which is how most of my switches are setup (n.o. to gnd) with a weak pull-up
[05:37:08] <Valen> the issue is switches bounce and naive debounce methods add delay
[05:37:34] <Valen> if the switch line is long and you can't reliably determine a trip from noise, then you need to add hardware
[05:37:36] <djdelorie> Valen: mine assumes the first sign of "hit" is real, then just keeps it real until it stays "not hit" long enough. I don't care where the "not hit" time is
[05:38:04] <Thetawaves> djdelorie, how do you limit current?
[05:38:08] <Valen> djdelorie: i haven't needed to do that, I just take the hit as on then fixed delay ;->
[05:38:17] <Valen> before polling the switch again
[05:38:25] <djdelorie> the only things I cared about were: (1) the home "trip" is reliable, and (2) it stays hit long enough that the homing sequence doesn't give errors
[05:38:43] <djdelorie> Thetawaves: I have ADCs measuring the current in each motor lead, and reduce the PWM in software as needed
[05:39:00] <Thetawaves> your switches i mean
[05:39:09] <Valen> (in micros I have my buttons etc on interrupts, the ISR generally does the "button action (setting a var)" then masks off the interrupt
[05:39:10] <djdelorie> Thetawaves: I don't
[05:39:24] <Valen> after a set number of timer timeouts it re-enables the interrupt
[05:39:38] <djdelorie> Valen: I poll the switches in the pwm interrupt, every 50 uS. There's a timeout there for the debounce
[05:39:43] <asdfasd> many contemporary enginiers solve theyr hardware problems with software solutions, but that is just not serious, it is doind the job but not in the proper way
[05:39:46] <Thetawaves> the pros put a resistor in series to limit current
[05:40:11] <Valen> asdfasd: if you wanted to solve the problem the "propper" way go build a switch that doesn't bounce
[05:40:27] <Valen> thats the actual problem
[05:40:32] <djdelorie> might be a good idea, I have resistors elsewhere for that purpose (and for terminating)
[05:40:36] <Thetawaves> and some schottky diodes that activate before the pull up diodes in the mcu
[05:40:41] <Valen> everything else is just ways of mitigating the problem
[05:40:59] <Thetawaves> the totem pole diodes
[05:41:07] <Thetawaves> i don't know what they are called
[05:41:27] <djdelorie> protection diodes ?
[05:41:32] <Valen> my way has fewer failure modes and its cheaper than yours, given that everything is a compromise in engineering my solution is nominally better
[05:42:02] <djdelorie> I've got a full EMI stage on the wires on my furnace controller, didn't think I'd need them on the limit switches.
[05:42:30] <Thetawaves> you do if they are connected through a distance of wire
[05:42:39] <Thetawaves> it acts like a inductor
[05:43:15] <Thetawaves> but of course that is only if the switch is like 20 or more feet away
[05:43:36] <djdelorie> mine isn't, and the cap on the pcb will absorb some of that anyway
[05:43:40] <Thetawaves> i don't know why i mentioned it on the topic of limit switches
[05:43:52] * Valen is off to the pool\
[05:43:54] <djdelorie> I did put strong pull-downs on the wires coming from the *pc* though
[05:44:04] <djdelorie> those were picking up motor noise EMI
[05:44:30] <djdelorie> the limit switch wires are all within the machine itself, since they go to the controller boards, so the longest is only a few feet.
[05:44:44] <djdelorie> and the signal is in the middle, with power and gnd on the outsides
[05:44:57] <joe9> is there any way I can change the gui to reflect my home. I home to minimum limit on x and y axes and maximum on z axis.
[05:45:28] <djdelorie> what about the gui do you want to change?
[05:46:20] <asdfasd> djdelorie: on each wire I have capacitor and 5.1V zener diode, also on each calble one of these thick ferrite rings
[05:46:47] <djdelorie> I put the ferrites on the motor wires and power, haven't needed any on the limit switch wires yet
[05:47:40] <djdelorie> at least, the machine's been running reliably without them
[05:48:16] <asdfasd> I did a lots of noise thests I noticed that most of the noise distruption the board is because of negative current pulses in the ground
[05:49:17] <asdfasd> 5.1V zener cut them before they reach the processor or any other IC
[05:50:20] <djdelorie> there's plenty of USB protection chips, they work great in 5v systems
[05:51:51] e-ndy|afk is now known as e-ndy
[05:52:08] <asdfasd> I didnt try that
[05:53:55] <djdelorie> my furnace uses the SP724 chips, four 0-5v TVSs per sot-23
[05:54:55] <asdfasd> but when I decided that it is enough noise resistant and I can close the box, then mechanical switch bounce wasnt an issue any more
[05:55:18] <djdelorie> :-)
[05:57:11] <djdelorie> my furnace was "noise reistant enough" when nearly lightning strikes stopped killing my sensors
[05:57:18] <djdelorie> nearby
[05:57:24] <asdfasd> sp724 look nice
[05:57:39] <Thetawaves> djdelorie, :D
[05:58:07] <Thetawaves> what sensors?
[05:58:39] <Thetawaves> ir?
[05:59:08] <djdelorie> also used EXC-CET222U to block noise.
[05:59:38] <djdelorie> temperature and humidity sensors. The outdoor one kept dying during storms, and taking out parts of the controller board with it
[06:00:46] <Thetawaves> with sensors.. do you need to do emi protection on both sides of the link?
[06:01:01] <Thetawaves> (assuming there is actually a circuit..)
[06:01:17] <djdelorie> yes, because the wire itself picks up the EMI
[06:01:30] <asdfasd> EXC-CET222U cut too high frequency I think you need something with higher inductance
[06:01:47] <djdelorie> that depends on what the signal is :-)
[06:02:12] <djdelorie> and what the EMI is
[06:02:15] <asdfasd> of course but I guess you dont have 80 mhz signal :)
[06:02:35] <djdelorie> no, but close to the 1MHz range
[06:02:46] <djdelorie> there are other filters in there too
[06:03:20] <asdfasd> my tests wad for noise generated by other devices
[06:03:46] <Thetawaves> who wants to make chips in a thunderstorm anyway
[06:04:02] <djdelorie> that board was bolted to my furnace, so it had motors, fire, heat/cold, and an electrostatic air filter. I.e. I bolted it to an EMI generator
[06:04:10] <asdfasd> like VFD, brush motors like hoover, welding
[06:05:48] <asdfasd> nice challange :)
[06:07:32] <asdfasd> long time ago I did PI metal detector, the search coil was 10 turns square 1m by 1m, it pick a lightening strike from 50km :)
[06:14:52] <Thetawaves> nice
[06:26:05] -!- tiago [tiago!~tiago@62.12.175.50] has joined #linuxcnc
[06:40:48] <asdfasd> anyone tested hall sensors for home swithces? and specially accuracy and repeatibility?
[06:42:12] -!- DanteTDI [DanteTDI!~quassel@host70-241-dynamic.1-87-r.retail.telecomitalia.it] has joined #linuxcnc
[06:42:15] <alex4nder> I've used them in racecar engines
[06:42:21] <alex4nder> they're totally legit.
[06:44:06] <asdfasd> I doubt they may change position a little bit because of external magnetic field
[06:45:56] <alex4nder> I don't know how much you can get away with, but they work well in harsh environments.
[06:45:58] -!- DJ9DJ [DJ9DJ!~Deejay@unaffiliated/dj9dj] has joined #linuxcnc
[06:46:02] <DJ9DJ> moin
[06:46:05] <alex4nder> yoh
[06:46:16] <DJ9DJ> hi everyone
[07:13:19] -!- maximilian_h [maximilian_h!~bonsai@212.23.103.34] has joined #linuxcnc
[07:21:35] -!- mhaberler has quit [Ping timeout: 260 seconds]
[07:24:04] -!- capricorn_one has quit [Remote host closed the connection]
[07:26:43] -!- mhaberler [mhaberler!~mhaberler@195.191.253.94] has joined #linuxcnc
[07:34:13] -!- zlog has quit [Ping timeout: 265 seconds]
[07:35:36] -!- zlog [zlog!~zlog@ip68-102-193-26.ks.ok.cox.net] has joined #linuxcnc
[07:38:09] -!- mhaberler has quit [Quit: mhaberler]
[07:38:29] -!- mhaberler [mhaberler!~mhaberler@195.191.253.94] has joined #linuxcnc
[07:42:15] -!- ktchk has quit [Ping timeout: 260 seconds]
[07:46:32] -!- DanteTDI has quit [Remote host closed the connection]
[07:59:06] -!- DanteTDI [DanteTDI!~quassel@host70-241-dynamic.1-87-r.retail.telecomitalia.it] has joined #linuxcnc
[07:59:17] -!- bedah [bedah!~bedah@g228251227.adsl.alicedsl.de] has joined #linuxcnc
[08:05:38] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[08:05:45] -!- mhaberler has quit [Quit: mhaberler]
[08:12:08] -!- phantoxe [phantoxe!~destroy@a95-92-89-24.cpe.netcabo.pt] has joined #linuxcnc
[08:33:59] -!- logger[psha] [logger[psha]!~loggerpsh@195.135.238.205] has joined #linuxcnc
[08:37:45] -!- mk0 [mk0!~x-ray@fiztech.basnet.by] has joined #linuxcnc
[08:43:26] <DanteTDI> hi, is there some diy hardware that works good with linuxcnc?
[08:46:03] <DanteTDI> (electronic hardware)
[08:46:26] <Jymmm> budget?
[08:47:56] -!- ktchk [ktchk!~eddie6929@n058153016214.netvigator.com] has joined #linuxcnc
[08:48:08] -!- ktchk [ktchk!~eddie6929@n058153016214.netvigator.com] has parted #linuxcnc
[08:48:55] <asdfasd> usually DIY hardware stepper ot servo is based on step/dir control via LPT and it wirks fine with linuxcnc
[08:49:41] <Jymmm> DanteTDI: http://www.geckodrive.com/g540-digital-axis-motor-control-p-39.html Worth the money
[08:50:19] <DanteTDI> Jymmm: under 500€ for electronics...
[08:50:20] -!- Tecan has quit [Quit: live long and phosphor]
[08:50:53] -!- mk0 has quit [Ping timeout: 245 seconds]
[08:51:29] <Jymmm> DanteTDI: The G540 has mid-band compensation, built in BOB, optical isolation, speed and relay control, etc
[08:52:04] <DanteTDI> it has got a position feedback?
[08:52:28] <Jymmm> you looking for servo?
[08:53:34] <asdfasd> if you dont have enough experience with electronics it will be cheaper for you to buy motor driver
[08:54:01] -!- vladimirek [vladimirek!~vladimire@adsl-dyn-153.95-102-102.t-com.sk] has joined #linuxcnc
[08:54:11] <DanteTDI> i'm able to weld comopnents and in the past made my own circuits(simple circuits)
[08:54:30] <DanteTDI> ii'm able to design a power driver
[08:54:53] <DanteTDI> but i havent got any idea for the controller
[08:55:23] <DanteTDI> Jymmm: wht's the difference between stepper and servo?
[08:55:49] <DanteTDI> not in general, but i can't understand how i can use servo on cnc...
[08:58:50] <asdfasd> https://www.google.co.uk/search?q=stepper+vs+servo&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
[09:01:10] <Valen> DanteTDI: servo is not those things you see in RC planes
[09:01:43] <DanteTDI> i see...
[09:02:20] <DanteTDI> i used servo rc with microchip pic in the past...now i see that your servo is different... :)
[09:03:26] <Valen> thats not a servo, this is a servo http://www.iowacitylist.com/Household-Goods-/Pets----no-live-animals-/New-indramat-10-kw-servo-motor-132B-1335OB1-BS03-A2N1.HTM
[09:03:54] -!- mk0 [mk0!~x-ray@fiztech.basnet.by] has joined #linuxcnc
[09:05:44] <DanteTDI> i will return in a minute...i have to reboot
[09:05:48] -!- DanteTDI has quit [Remote host closed the connection]
[09:10:56] <alex_joni> GM7221-4SB61
[09:11:03] <alex_joni> rated power: 100kW
[09:11:11] <alex_joni> rated current: 188 A
[09:11:18] <alex_joni> weight: 650kg
[09:11:24] <Jymmm> battery powered
[09:11:40] <alex_joni> 636 Nm
[09:12:03] <alex_joni> Inertia: 1.48 Kgm^2
[09:12:07] <alex_joni> now that's low
[09:12:30] -!- DanteTDI [DanteTDI!~quassel@host70-241-dynamic.1-87-r.retail.telecomitalia.it] has joined #linuxcnc
[09:13:09] <DanteTDI> how much for the gecko?
[09:13:32] <alex_joni> www.geckodrives.com
[09:14:13] <alex_joni> 299$ for a G540
[09:14:53] <alex_joni> hmmm.. seems prices are a bit up from what I remember
[09:15:03] <alex_joni> 81$ for a G250 is a bit pricey imo
[09:15:05] <Valen> for that big a motor its low
[09:15:08] <Jymmm> yeah, sounds like it to me too
[09:15:40] <Valen> with a weight of 600kg
[09:15:50] <Valen> 148 i'd believe lol
[09:16:07] <alex_joni> Valen: it's a servo
[09:16:20] <alex_joni> it's supposed to have low inertia, otherwise it's a regular AC motor
[09:16:41] <Valen> 1.4kg/m means its rotor is the same as 1.4kg sitting on a 1 meter stick
[09:17:07] <Valen> 110 CC's of copper
[09:19:05] -!- ktchk [ktchk!~eddie6929@n058153016214.netvigator.com] has joined #linuxcnc
[09:19:15] -!- vladimirek has quit [Remote host closed the connection]
[09:19:42] -!- rob_h [rob_h!~rob_h@5ace70c9.bb.sky.com] has joined #linuxcnc
[10:01:16] e-ndy is now known as e-ndy|afk
[10:05:45] e-ndy|afk is now known as e-ndy
[10:48:43] -!- DanteTDI has quit [Read error: Connection reset by peer]
[11:00:04] -!- bbarum [bbarum!5c3ce9f4@gateway/web/freenode/ip.92.60.233.244] has joined #linuxcnc
[11:14:06] e-ndy is now known as e-ndy|afk
[11:30:59] -!- bbarum has quit [Quit: Page closed]
[11:33:04] -!- engkur [engkur!~engkur@180.243.236.63] has joined #linuxcnc
[11:33:09] <engkur> halo
[11:46:33] -!- X704 has quit [Read error: Connection reset by peer]
[11:47:03] -!- X704 [X704!~walter@host-174-45-122-145.bzm-mt.client.bresnan.net] has joined #linuxcnc
[11:47:07] -!- H264 [H264!~walter@host-174-45-122-145.bzm-mt.client.bresnan.net] has joined #linuxcnc
[11:49:30] -!- syyl [syyl!~syyl@p4FD12E43.dip.t-dialin.net] has joined #linuxcnc
[11:50:05] -!- WalterN has quit [Ping timeout: 246 seconds]
[12:01:43] -!- bedah has quit [Quit: bye]
[12:07:08] -!- skunkworks__ has quit [Ping timeout: 240 seconds]
[12:45:00] e-ndy|afk is now known as e-ndy
[12:45:58] e-ndy is now known as e-ndy|afk
[12:48:21] -!- engkur has quit [Quit: Leaving]
[12:56:55] -!- phantoxe has quit [Remote host closed the connection]
[13:08:02] -!- syyl_ [syyl_!~syyl@p4FD143A7.dip.t-dialin.net] has joined #linuxcnc
[13:08:14] -!- mk0 has quit [Quit: LinuxCNC 2.5.0!]
[13:11:37] -!- syyl has quit [Ping timeout: 276 seconds]
[13:14:17] -!- phantoxe [phantoxe!~destroy@a95-92-89-24.cpe.netcabo.pt] has joined #linuxcnc
[13:14:59] e-ndy|afk is now known as e-ndy
[13:19:26] -!- howard_ [howard_!~howard@122-148-218-32.static.dsl.dodo.com.au] has joined #linuxcnc
[13:20:47] -!- howard_ has quit [Client Quit]
[13:29:57] -!- maximilian_h [maximilian_h!~bonsai@130.255.104.157] has joined #linuxcnc
[13:29:59] -!- maximilian_h [maximilian_h!~bonsai@130.255.104.157] has parted #linuxcnc
[13:30:22] -!- FinboySlick [FinboySlick!~shark@74.117.40.10] has joined #linuxcnc
[13:40:12] <JT-Shop> djdelorie: I tried all sorts of combinations for home but never could get the same error on my plasma... and I've set my limits properly now
[13:42:03] -!- ries [ries!~ries@200.125.129.54] has joined #linuxcnc
[13:44:20] <Jymmm> JT-Shop You just didn't try hard enough ;)
[13:44:26] <Jymmm> JT-Shop: How ya doin =)
[13:44:41] <JT-Shop> doin good how about you
[13:45:34] <Jymmm> JT-Shop: Not too shabby, fighting myself to do taxes =)
[13:46:06] <JT-Shop> yea, I need to do mine too...
[13:46:07] <Jymmm> I dont have a issue paying, just gimme a postcard and not 85,000 forms!
[13:48:34] <Jymmm> JT-Shop: Hey, I'm trying to heat seal the ends of plastic tubing, kinda like the end of a tube of toothpaste. You have any idea on how they do the 1/8" heat sealed crimp by chance?
[13:49:15] <Jymmm> without the tooling sticking to the material when squeezed together
[13:52:44] <JT-Shop> vibration usually
[13:53:28] <archivist> non stick cover to the heater
[13:53:39] <archivist> teflon
[13:55:31] <JT-Shop> archivist: would you have time to gander at a couple of steam engine drawings?
[13:55:49] <archivist> can do
[14:07:34] -!- mhaberler [mhaberler!~mhaberler@91.208.88.13] has joined #linuxcnc
[14:19:05] -!- DanteTDI [DanteTDI!~quassel@host70-241-dynamic.1-87-r.retail.telecomitalia.it] has joined #linuxcnc
[14:30:22] <ssi> what's a braking resistor for a vfd cost?
[14:32:10] <JT-Shop> depends on the vfd I'd assume
[14:32:19] <JT-Shop> automationdirect has them listed
[14:32:43] <awallin> something like this? http://www.ebay.com/itm/150-Ohms-3kW-CARBORUNDUM-WIRELESS-ABSORBER-OF-HIGH-POWER-RESISTOR-LOT-OF-1-/280851983843
[14:33:13] <JT-Shop> http://www.automationdirect.com/adc/Shopping/Catalog/Drives/GS2_%28115_-z-_230_-z-_460_-z-_575_VAC_V-z-Hz_Control%29/GS2_Braking_Options_%28Braking_Units_-a-_Resistors%29
[14:33:21] -!- maximilian_h [maximilian_h!~bonsai@anon-151-8.vpn.ipredator.se] has joined #linuxcnc
[14:35:01] -!- sumpfralle [sumpfralle!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[14:37:02] -!- maximilian_h [maximilian_h!~bonsai@anon-151-8.vpn.ipredator.se] has parted #linuxcnc
[14:37:39] -!- ysli has quit [Quit: Leaving]
[14:38:46] <ssi> JT-Shop: thanks
[14:38:55] <ssi> now I just have to figure out how exactly to size them
[14:39:29] <archivist> vfd manual should say
[14:39:35] <JT-Shop> yea, I was lucky and just needed the model number to match up to
[14:39:35] <ssi> yeah probably
[14:39:51] <ssi> so how much can you cut down on your decel time with the resistor?
[14:42:59] <archivist> those jt pointed at looked expensive http://uk.rs-online.com/web/p/high-power/0136339/
[14:44:12] <ssi> I'm tryin to find the manual so I can find out what exactly it wants
[14:47:45] <thinpete> You should be able to deccel as fast as you can accell if the load resistor can dissipate the peak power at the maximum DC bus voltage (and the brake MOSFET/IGBT can deliver the peak current)
[14:48:14] <ssi> I see
[14:48:19] <archivist> if your resistance is a bit high you just lose some breaking
[14:48:32] <ssi> I wish the accel/decel parameters were in rate units instead of total time
[14:48:45] <ssi> I'm having a lot of trouble with my machine faulting on decel overcurrent
[14:48:50] <ssi> just from M5 from high speed
[14:49:01] <ssi> I've started programming a release clutch then M5
[14:49:03] <ssi> which I'm not crazy about
[14:49:50] <joe9> djdelorie: do you have a touch-off switch?
[14:50:30] <joe9> djdelorie: jdhnc: just wanted to let you know that with software debouncing, everything works like a charm.
[14:50:38] <thinpete> When you brake you dump power back into the DC bus raising the bus voltage (destructively if theres no brake/PWM shutdown protection)
[14:50:50] <ssi> yeah I understand that
[14:51:20] <thinpete> so adding a brake resistor should help (weve used a electric stove element)
[14:51:33] <ssi> but I have like a 10 second decel time, and it seems like simply M5 shouldn't cause it to fault merely from spindle inertia
[14:51:48] <ssi> I mean it's a 2lb chunk of aluminum, not a 75lb workpiece (which happens to be the machine's rated workpiece load)
[14:52:42] <thinpete> whats the natural deccel time? if 10 seconds is any faster you will be pumping up the DC bus
[14:52:51] <ssi> not sure tbh
[14:53:16] <ssi> I'm wondering if I might be better off putting the mechanical brake back on
[14:54:13] <ssi> anyway, see the problem with trying to turn up the decel time
[14:54:30] <ssi> since it's a fixed decel time, it may solve the faults for high-speed to zero,
[14:54:37] <ssi> it also increases the overall decel for low-speed
[14:54:42] <ssi> which makes rigid tapping overshoot a lot more
[14:55:18] <thinpete> Yeah you definately need brake resistor
[14:56:06] <ssi> so the manual doesn't call out what to use for a resistor
[14:56:11] <ssi> instead it talks about a DC braking unit
[14:56:14] <ssi> and it can do DC injection
[14:56:21] <ssi> I'm wondering if that means you can't just use a simple resistor
[14:56:25] <ssi> or if you can, but they won't tell you how :P
[14:57:22] <thinpete> if it can do DC injection surely that is a setup option of some kind
[14:58:14] <ssi> it is
[14:59:10] <ssi> Braking units are required for applications where a load with high inertia needs to be stopped rapidly.
[14:59:13] <ssi> Use a power-matched braking unit and resistor to dissipate the energy generated by the load while stopping. Otherwise inverter will trip on over voltage.
[14:59:17] <ssi> that's from the manual
[14:59:30] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[15:00:15] <mazafaka> Used 30-mm in diameter End Mill to mill the sides of 1000 by 600 mm steel plate 50 mm thick. Everybody said I'm mad. But what can I do on a 2.5D machine?
[15:00:31] <mazafaka> 40 mm per minute
[15:00:41] <mazafaka> 120 RPM
[15:06:46] <ssi> aha
[15:06:50] -!- syyl_ws [syyl_ws!~sg@p4FD143A7.dip.t-dialin.net] has joined #linuxcnc
[15:07:19] <ssi> 220v class, 3hp, 260W 70R
[15:07:25] <ssi> then it says "max connectable quantity"
[15:07:28] <ssi> not exactly sure what that means :P
[15:08:43] <ssi> I guess with four I could do a 2S2P and end up with a 1kW 70R resistor
[15:08:52] <ssi> but why not just by a 1kW 70R resistor? :P
[15:12:00] -!- sumpfralle [sumpfralle!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[15:12:15] <mazafaka> ssi: I have only understood it like "My PlayStation? It sucks so I modify it, what casn I do with all that junk, all that junk inside the truck? I want-want want brand new games, brand new games to look damn smart!"
[15:13:56] <ssi> I...seee....
[15:14:06] -!- Thetawaves [Thetawaves!~Thetawave@7-139-42-72.gci.net] has joined #linuxcnc
[15:14:12] <mazafaka> Ha!
[15:14:32] <mazafaka> I sold that second bike today
[15:16:51] <mazafaka> Two different departments, the boss and the quality control department, have perplexed me quite a lot this wee
[15:16:59] <mazafaka> *this week
[15:18:34] <mazafaka> I'm 2.5D-milling on a coordinate-drilling machine which lacks rigidity. One need more details to be finished, others say this machine must be in a museum. That's how double standards work.
[15:19:24] -!- maximilian_h [maximilian_h!~bonsai@anon-137-180.vpn.ipredator.se] has joined #linuxcnc
[15:25:05] <thinpete> SSI i wonder if by "braking unit" they mean that they do not have the IGBT/comparator and leave that up to an external box
[15:25:37] <thinpete> are there setup options for the brake voltage?
[15:28:02] <ssi> DC braking time, start freq, voltage %
[15:30:32] <thinpete> Hmm
[15:31:18] <thinpete> no nice pictures of power resistor connected to two VFD pins?
[15:31:53] <ssi> nope
[15:31:59] <ssi> I searched for a "braking unit"
[15:32:02] <ssi> and I got a manual
[15:32:07] <ssi> http://www.tecowestinghouse.com/Manuals/Breaking%20Unit%207200PA-7200GA.pdf
[15:32:10] <ssi> misspelled :D
[15:32:18] <ssi> and it's scans of some chinglish translations
[15:32:32] <ssi> but there's a table in there that insinuates that for the lower powered vfds, you use resistors
[15:32:38] <thinpete> hope its not a breaking unit...
[15:32:40] <ssi> and the higher powered ones use this "breaking unit"
[15:34:19] <thinpete> "rubber bush" is nice
[15:34:25] <ssi> yeah
[15:34:28] <ssi> they couldn't afford a rubber tree
[15:35:10] <ssi> I may just set it to free coast and think about getting the mechanical brake back working and installed
[15:35:47] <thinpete> My all time favorite chinglish is "regal sticked rubbles" (for stick-on rubber feet)
[15:36:07] <ssi> hah
[15:36:38] -!- phantoxe has quit [Ping timeout: 240 seconds]
[15:36:51] <thinpete> So yeah its a box with the IGBT and comparator (and resistor)
[15:37:45] <ssi> haha from hsm:
[15:37:46] <ssi> My Teco EV does have an internal braking mechanism, and I have it enabled. The brake works, but trips occasionally, no matter how I tweak the configuration.
[15:37:49] <ssi> The EV has two terminals identified as a place to attach an external resistor. The external resistor setup does not work on mine, probably because I let the smoke out while experimenting (you have to experiment because the manual is as clear as mud). But, the fact that it has the terminals for an external resistor suggests that it does not require a separate extra-cost braking module
[15:38:35] -!- maximilian_h has quit [Ping timeout: 260 seconds]
[15:38:57] <thinpete> Yeah if the terminal are not simply DC bus then the switch is inside
[15:39:04] <thinpete> terminals
[15:39:33] <ssi> so if they don't include the switch, then the +/- terminals are just raw DC bus?
[15:39:44] <thinpete> Yep
[15:39:50] <thinpete> Dont get bit!
[15:40:07] <ssi> a little 230vdc never hurt anyone :D
[15:40:18] <thinpete> ~340 VDC
[15:40:23] -!- cevad has quit [Ping timeout: 246 seconds]
[15:40:26] <ssi> potato, tomato
[15:41:04] <ssi> I might oughta replace that VFD with one that is saner for braking
[15:41:07] <ssi> and use that one on my mill
[15:41:08] -!- phantoxe [phantoxe!~destroy@a95-92-89-24.cpe.netcabo.pt] has joined #linuxcnc
[15:41:20] <ssi> my manual bridgyclone is run off a static converter presently
[15:42:01] <thinpete> static meaning just 60 Hz?
[15:42:22] <ssi> static phase converter
[15:42:27] <ssi> so yeah, just 60hz
[15:42:31] <thinpete> Yeah
[15:42:32] <ssi> and 2/3 max power
[15:43:03] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc
[15:45:26] <ssi> anyway, so the gist of it is
[15:45:39] <ssi> randomly putting a resistor across those terminals might be a really really bad idea if it's raw dc bus
[15:45:42] <ssi> yes?
[15:46:20] <thinpete> Well a voltmeter will probably tell you whats there
[15:46:29] <ssi> what should I be looking for?
[15:46:34] <archivist> deterministicly work out which model you have and what it can handle
[15:46:39] <ssi> 0v during normal operation?
[15:46:46] <ssi> archivist: I know exactly what model I have
[15:47:12] <archivist> so you known which braking it can handle?
[15:47:16] <ssi> nope!
[15:47:16] <archivist> know
[15:47:19] <thinpete> there will be some leakage but if you get ~340V the its bus power
[15:47:26] <thinpete> then its
[15:47:34] <ssi> thinpete: gotcha... I can check that easily enough
[15:50:11] -!- sumpfralle1 [sumpfralle1!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[15:50:25] <thinpete> Be very careful not just high power/high voltage but high peak current/arc potential
[15:50:34] <ssi> understood
[15:52:49] -!- sumpfralle has quit [Ping timeout: 246 seconds]
[15:53:28] <alex4nder> hey
[15:54:28] <ssi> at any rate
[15:54:31] <ssi> HNC is makin parts
[15:54:31] <ssi> http://p.twimg.com/AqXqBSYCQAEyVvh.jpg:large
[15:54:32] <ssi> wooooo
[15:57:30] <alex4nder> werd
[15:58:25] -!- maximilian_h [maximilian_h!~bonsai@anon-148-56.vpn.ipredator.se] has joined #linuxcnc
[15:59:33] <thinpete> Neat! looks like the HNC was designed to crank that type of part out
[16:05:41] <JT-Shop> ssi: nice
[16:06:48] -!- syyl_ws has quit [Quit: Verlassend]
[16:09:06] <ssi> thinpete: that part is actually a little tricky just cause of the diameter
[16:09:18] <ssi> it's a 2"ø part, and I have to use a step collet
[16:09:24] <ssi> other than that, yeah it's really working well
[16:10:20] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc
[16:11:14] <ssi> those parts are worth $15 each to my customer, and he wants 30 of them a month to start
[16:11:25] <ssi> this machine is going to pay for itself!
[16:11:36] <ssi> which is probably a given for a lot of folks here, but this is a hobby for me :D
[16:13:18] -!- Guest83733 has quit [Quit: ZNC - http://znc.in]
[16:13:34] -!- mhaberler has quit [Quit: mhaberler]
[16:13:59] -!- DanteTDI has quit [Remote host closed the connection]
[16:14:08] <pjm> ssi http://www.postglover.com/dynamicbraking.html is worth a look
[16:14:14] <mazafaka> How did you find this job for your machine to work?
[16:14:22] <pjm> for my yakasawa vfd i used 150 ohm 100w loads
[16:14:39] <joe9> alex4nder: how do you do "touch off"? do you use a "touch off" switch?
[16:14:40] <ssi> mazafaka: through a friend... I have two friends of friends paying me to make stuff now actually
[16:14:50] <ssi> pjm: thanks!
[16:15:02] <mazafaka> But how have this all started?
[16:15:15] <ssi> mazafaka: I have ten machines, and everyone knows it
[16:15:17] <ssi> hell of a hobby I know
[16:15:51] <joe9> ssi, how is your machine coming along? is this the one that I was pitching in for?
[16:15:52] <mazafaka> I can imagine
[16:16:06] <ssi> joe9: pitching in?
[16:16:21] <alex4nder> joe9: I touch off the thing I'm machining
[16:16:24] <joe9> or, "planning to pitch in ofr".
[16:16:32] <ssi> hehe I don't recall
[16:16:34] <joe9> ssi, sorry, typing has been off
[16:16:36] <ssi> but yeah, it's mostly done
[16:16:41] <ssi> I need a panel and a pendant
[16:16:50] <joe9> ssi, no, the machine that we were discussing about.
[16:17:01] <joe9> ssi, oh, cool.
[16:17:07] <ssi> ohh you were wanting the little desktop mill
[16:17:21] <joe9> i bought a taig
[16:17:28] <ssi> I haven't done anything on that yet, but my next big project is to get my g0704 conversion finished
[16:17:34] <ssi> how's the taig working out?
[16:17:54] <joe9> it is well made, an awesome machine. but, I am still getting my feet wet on it.
[16:17:57] -!- tiago has quit [Remote host closed the connection]
[16:18:09] <joe9> added home/limit switches to it.
[16:18:31] <joe9> now, I am thinking about how to "touch off"
[16:18:55] <alex4nder> joe9: you should try milling something, touching off will become really clear
[16:21:03] <JT-Shop> joe9: you get the homing working?
[16:21:04] <alex4nder> you'll realize all of the measurement gear and random tools you need
[16:21:16] <joe9> JT-Shop: yes. it works like a charm now.
[16:21:21] <ssi> the way that I do it is get the tool close to a known Z position, then slip a .010 feeler between the tool and the surface, then switch to incremental mode and move .001 step at a time until the feeler gets tight
[16:21:23] <JT-Shop> what was it?
[16:21:31] <ssi> then I hit touch off and enter the known Z dimension + .010
[16:21:48] <joe9> JT-Shop: i think debouncing
[16:21:55] <joe9> switch was inverted
[16:22:02] <JT-Shop> ssi: I use a dowel and slowly raise the tool till it just rolls under
[16:22:12] <ssi> bear in mind that there are several coordinate systems in play and you need to make sure you know how they're set up and which you're affecting
[16:22:18] <alex4nder> yah
[16:22:25] <ssi> JT-Shop: that's not a bad way, but mostly I do this on the lathe in Z and X actually :)
[16:22:35] <alex4nder> I use a wiggler.
[16:22:48] <alex4nder> Fisher Machine makes cool small ones for the taig.
[16:22:53] <alex4nder> well, machines the size of the taig.
[16:22:59] <JT-Shop> I do that for the Z of lathe
[16:23:00] <joe9> i am thinking of a "touch off switch". takes the guess work out. is that a bad idea?
[16:23:23] <joe9> it takes care of the tool length and all that. doesn't it?
[16:23:48] <JT-Shop> works ok for setting up your tool changer but you still have to touch off the material
[16:24:04] <ssi> joe9: could get something like this: http://www.edgetechnologyproducts.com/pro-touch-off-gage.html
[16:24:20] <joe9> jt-shop. ok, thanks.
[16:24:40] <ssi> I actually want to buy their silly little tramming device
[16:24:45] <ssi> I was gonna MAKE one, because it's like $80
[16:24:46] <JT-Shop> ssi: I use that on the VMC
[16:24:49] <alex4nder> and you still need to know exactly where your material is in relation to whatever you touch off of.
[16:24:55] <ssi> but then I realized that I really don't need another project
[16:25:11] <JT-Shop> and the lathe tool height setting do-hickey
[16:25:27] <alex4nder> joe9: do you know what "touching off" means? it's telling the machine where your material is inside the machine's coordinate system. it sets up a reference to a relative position
[16:25:32] -!- maximilian_h has quit [Ping timeout: 246 seconds]
[16:25:34] <alex4nder> so you need to get near the material
[16:25:40] <ssi> JT-Shop: I have the little magic hardinge height gauge tool
[16:25:44] <ssi> that references it off the bed
[16:25:49] <JT-Shop> I don't have enough travel on Z on the knee mill to use the pro touch off gauge on it
[16:25:57] <alex4nder> you also need to position the material so that it's parallel with your axes
[16:26:00] <joe9> alex4nder: ok, thanks. next step is to start milling something.
[16:26:02] <JT-Shop> ssi: got a photo of it?
[16:26:19] <joe9> is there a "slow zone" in linuxcnc?
[16:26:38] <joe9> from mach-3: "You can also set up a slow zone, this is the distance that Mach will start to slow down so that it doesn’t over travel when reaching the soft limits, you will have to find out the perfect distance for your individual machine as all machines will vary due to differences in acceleration, velocity and even machine weight and motor size."
[16:26:47] <ssi> JT-Shop: not handy... I can take one tonight
[16:26:53] <joe9> curious, if that is possible with linuxcnc
[16:27:23] <ssi> JT-Shop: here you go
[16:27:23] <ssi> http://aafradio.org/garajmahal/Hardinge_L-2A_Tool_Setting_Gauge.html
[16:28:17] <cradek> ssi: not needed, because linuxcnc will decel in a controlled fashion to the stop point, never just blindly fly past it
[16:28:38] <ssi> cradek: aha
[16:28:47] <cradek> ssi: that "slow zone" sounds like a poor workaround for a mach bug
[16:29:04] <ssi> cradek: it seems like it overshoots, fwiw
[16:29:12] <cradek> what does?
[16:29:30] <ssi> cradek: I'm doing a rigid tap move starting 2.5mm from Z0, incremental -13mm
[16:29:37] <ssi> and it taps more than 11mm deep I'm pretty sure
[16:29:40] <ssi> I'll have to doublecheck
[16:29:45] <cradek> yes
[16:29:50] <cradek> that has nothing to do with soft limits
[16:29:58] <ssi> oh you're responding to somethign different
[16:29:59] <ssi> sorry
[16:30:03] <ssi> I think that was aimed at joe9
[16:30:07] <joe9> cradek: you are responding to me.
[16:30:13] <joe9> cradek, thanks for that.
[16:30:16] <cradek> oh sorry, yes
[16:30:21] <ssi> I thought you dug up my backscroll about deceleration in my VFD
[16:30:38] <cradek> about tapping: linuxcnc will continue to follow your spindle as it slows and reverses past the target - otherwise the tap would break
[16:30:42] <ssi> I was complaining that I get decel faults sometimes, but if i up my decel time, it makes my slow-speed decel so long that I'm afraid it'll screw up my tap
[16:30:58] -!- Farthen [Farthen!~Farthen@2a01:4f8:101:2a4:0:bc28:b2e1:9] has joined #linuxcnc
[16:30:59] <JT-Shop> ssi: that is better than what I use
[16:31:02] <ssi> I thought you were saying that linuxcnc would somehow plan for the long decel
[16:31:04] Farthen is now known as Guest45066
[16:31:10] <ssi> but really how could it know :)
[16:31:13] <joe9> alex4nder: i think my next step is to mill something. Any ideas? I have wood lying around and am thinking of starting with that. i remember you mentioned styrofoam.
[16:31:20] <cradek> ah no it doesn't know, just has to follow
[16:31:25] <ssi> JT-Shop: the guy that sold me the HNC sold me that tool for FORTYFIVE DOLLARS
[16:31:26] <cradek> add a resistor!
[16:31:37] <ssi> cradek: want to, but I think my VFD needs an external transistor with the resistor
[16:31:53] <cradek> my vfd has a regen-to-line setting for braking, and it has a "slow down as fast as possible without faulting" setting
[16:32:05] <JT-Shop> still better than putting a collet in and using the level thingy
[16:32:13] <cradek> did you study the manual very carefully? there might be a hundred braking options.
[16:32:39] <ssi> the latter would be nice
[16:32:41] <ssi> I'll have to look
[16:32:51] <ssi> I also thought about maybe putting the mechanical brake back in and setting vfd to coast
[16:32:55] <alex4nder> joe9: just mill a line in some wood manually.. then try running some g-code
[16:33:06] <joe9> ok, thanks.
[16:33:07] <cradek> sorry for the confusion up there
[16:33:10] <ssi> no problem :)
[16:33:18] Guest45066 is now known as Farthen_
[16:33:23] <alex4nder> joe9: I would crank up the spindle speed, and then do like 10 IPM feed rate, and just mess around
[16:33:30] <alex4nder> joe9: do you have a wet/dry vac?
[16:33:37] <ssi> JT-Shop: to be clear, the HNC (and probably chnc) version of that tool sits on the bedway, not on the carriage
[16:33:58] <joe9> alex4nder: vac == vacuum cleaner?
[16:34:04] <alex4nder> yup
[16:34:11] <alex4nder> joe9: also what endmills do you have?
[16:34:23] Farthen_ is now known as Farthen
[16:34:33] <joe9> i just have the normal home vac. it is a small dry one.
[16:34:48] <JT-Shop> I like this comment "The L-2A tool height setting tool is a quick way to set up the cutting height of various tool bits, especially if you have to change one in the middle of a job and don't want to break down the setup to do the usual experimentation. "
[16:35:10] <alex4nder> joe9: you remember that video I posted of me milling wood? it'll give you some good ideas of what you can do.. I had the feed set really low
[16:35:14] <ssi> heheh
[16:35:25] <ssi> JT-Shop: so I changed inserts mid job last night
[16:35:28] <ssi> not mid-cycle, between parts
[16:35:42] <ssi> I switched to a crown-type insert to try to get better chip control
[16:35:59] <JT-Shop> ah changed style... yep
[16:36:01] <ssi> and it ended up changing my finish diameter by almost .080
[16:36:03] <ssi> I was surprised
[16:36:07] <joe9> alex4nder: do you have the video still? would you mind passing it along?
[16:36:21] <alex4nder> joe9: http://www.andern.org/alexander/taig_milling_wood.m4v
[16:36:30] -!- syyl_ has quit [Quit: Leaving]
[16:36:44] <alex4nder> the feeds could be much higher
[16:36:46] -!- syyl [syyl!~syyl@p4FD143A7.dip.t-dialin.net] has joined #linuxcnc
[16:36:55] <alex4nder> the spindle is running at 10k
[16:37:01] <joe9> ok, will check it out.
[16:37:10] <joe9> actually checking it out now.
[16:37:28] -!- maximilian_h [maximilian_h!~bonsai@anon-153-15.vpn.ipredator.se] has joined #linuxcnc
[16:37:38] -!- ktchk [ktchk!~eddie6929@n058153016214.netvigator.com] has parted #linuxcnc
[16:37:53] <joe9> alex4nder: what do you call that bit? looks like a dremel cutter.
[16:38:29] <joe9> alex4nder: is it the vac that you are pointing at it?
[16:38:38] <alex4nder> joe9: it's an endmill.. and yah that's a vac.
[16:38:49] <alex4nder> the endmill is a 4 flute high speed steel 1/2 inch
[16:39:12] <joe9> ok, thanks.
[16:48:24] -!- cevad [cevad!~davec@host-174-45-229-40.bln-mt.client.bresnan.net] has joined #linuxcnc
[16:49:57] <ReadError> so i thought of a stupid question
[16:50:04] <ReadError> say i have a piece of metal
[16:50:09] <ReadError> and i want to cut a square out
[16:50:14] <ReadError> how would that be done?
[16:50:25] <ReadError> wouldnt the bit round the corners since its round?
[16:50:37] <alex4nder> yes.
[16:50:52] <alex4nder> finish with a smaller bit
[16:50:53] <archivist> drill then file or broach or use a slotter
[16:51:13] <archivist> or shaper
[16:51:41] <alex4nder> yup
[16:51:44] <alex4nder> depending on the need
[16:52:17] <archivist> you can make your own broaches but thats "interesting"
[16:53:46] <ssi> JT-Shop: http://www.flywheelmachinetools.com/TOOLHOLDINGHARDINGE.html
[16:53:52] <ssi> first one is the HNC/CHNC height gage
[17:00:25] -!- cevad has quit [Ping timeout: 265 seconds]
[17:03:07] -!- erasmo has quit [Ping timeout: 260 seconds]
[17:05:03] <archivist> ReadError, for brass and the square hole in a winding key, we manufactured them in qty, drill deep enough for the chips, hold in a rigid clamp, use a bit of HSS ground square(a degree or two relief) as a broach, used a fly press for the operation
[17:05:56] <archivist> the drilled hole a bit over sized
[17:08:19] -!- psha [psha!~psha@213.208.162.69] has joined #linuxcnc
[17:17:36] -!- phantoxe has quit [Read error: Connection reset by peer]
[17:29:07] <djdelorie> joe9: I also use feeler gauges for touching off Z, and just eyeballs for X and Y
[17:30:32] <alex4nder> I have been now to drop the endmill slowly towards the stock, and listen for the tick-tick-tick of the endmill licking the material
[17:35:59] -!- IchGuckLive [IchGuckLive!~chatzilla@95-89-236-191-dynip.superkabel.de] has joined #linuxcnc
[17:36:08] <IchGuckLive> hi all around the globe
[17:36:26] <joe9> djdelorie: alex4nder: ok, thanks.
[17:36:37] <alex4nder> IchGuckLive: hello
[17:40:39] <joe9> with the feeler gauge, do you put the feeler guage on the workpiece and slowly bring the tip to touch it?
[17:40:58] <joe9> djdelorie: any thoughts, please?
[17:41:11] <ssi> joe9: yes
[17:41:20] <djdelorie> in my case, the bits are too delicate for that. I bring the tool down, then see if the feeler fits under it
[17:41:34] <ssi> joe9: and when I'm close, I move the feeler back and forth, and I can feel if they start to drag
[17:41:42] <ssi> but yea for a tiny little pcb drill, that probably won't work
[17:42:26] <alex4nder> for a PCB maybe you want a vacuum table and a touch probe with a reference
[17:42:31] <JT-Shop> ssi: thanks for the link
[17:42:32] <alex4nder> (if you get serious)
[17:42:38] <djdelorie> also, I have a whole feeler set, so I can usually measure the actual distance that way in a few tries
[17:43:13] <joe9> djdelorie: i am thinking of buying this set: http://www.harborfreight.com/32-piece-sae-metric-feeler-gauge-32214.html
[17:43:15] <cradek> use a dowel pin or anything else of known diameter, like a tool shank. move low enough that it can't roll under. now, move UP slowly until it rolls under. this is much safer than moving down.
[17:43:22] <djdelorie> alex4nder: for pcb drilling, that kind of accuracy is not needed. In my case at least, the carriage needs time to stop vibrating after the XY travel, so I just use a longer drill path
[17:43:36] <alex4nder> djdelorie: this is more of a milling issue
[17:43:49] <cradek> also, because it's round, as you get close to it rolling under, you will feel the motion speed up
[17:43:55] <joe9> cradek: i have a punch set and a set of parallels. i am not sure if I can use them instead.
[17:44:03] <djdelorie> One of my "to do" projects is to mill a vacuum hold-down. I have the vacuum itself
[17:44:13] <djdelorie> joe9: use a drill bit
[17:44:20] <alex4nder> djdelorie: poul-henning kamp did a good write-up of h is technique
[17:44:21] <cradek> use a broken carbide pcb tool. the shanks will be very near .125
[17:44:47] <cradek> or whatever you have
[17:44:53] <djdelorie> if you have to be more accurate than any of these suggestions, you probably need a better mill anyway? :-)
[17:44:56] <cradek> if you have a hardware store you can get one dowel pin for a few cents
[17:45:54] <djdelorie> alex4nder: url?
[17:46:00] <alex4nder> hmm
[17:46:04] <alex4nder> djdelorie: http://phk.freebsd.dk/CncPcb/
[17:46:04] <joe9> cradek, ok. thanks. i will check for it.
[17:46:15] <djdelorie> thanks
[17:47:05] <IchGuckLive> PCB without a plan vaccuum is a miss special with V-shape cutters
[17:47:54] <djdelorie> yup, I have the vacuum system partly for pcb UV exposure too, to make sure the mask is tight. "flat" is not a description of blank PCBs :-)
[17:49:09] <IchGuckLive> UV within 60sec shoudt not be a miss
[17:49:18] <djdelorie> mine takes about 5 minutes
[17:49:35] <djdelorie> but hold-down is mostly to optimize the focus
[17:49:38] <IchGuckLive> get more UV sources
[17:49:46] <djdelorie> another "to do" project :-)
[17:50:15] <djdelorie> even so, the better the focus, the more reliable your results
[17:50:26] <IchGuckLive> id used a 100W lamp at 30cm/1ft distance that burned in folks
[17:50:29] * Thetawaves has been reading the epoxy/granite thread on the zone
[17:50:37] <alex4nder> thinpete: that's a good thread
[17:50:38] <IchGuckLive> 1000W O.O
[17:50:44] <djdelorie> I use 100 uv LEDs
[17:51:10] <IchGuckLive> a sun burner for broun skin is 15USD
[17:52:03] <djdelorie> yup, we've had this conversation before ;-)
[17:52:04] <IchGuckLive> djdelorie: do you use led shinings for more better resonance
[17:52:38] <djdelorie> I use LED because it was fun, a known technology, and only 8 degree dispersion
[17:52:46] <IchGuckLive> this gives you a strait uc sector
[17:53:12] -!- ries [ries!~ries@200.125.129.54] has parted #linuxcnc
[17:53:22] <IchGuckLive> 5minis realy mutch
[17:53:24] <Jymmm> Heh, I used to build 6ft long, 600Watt/inch UV Curing systems
[17:53:36] <djdelorie> the 8 degree spec means film will only get exposed by the LEDs directly above it, not some LED way off to the side
[17:53:46] <djdelorie> and I don't lose energy exposing the walls of the box :-)
[17:54:15] <IchGuckLive> agree i useed this 2
[17:54:35] <djdelorie> the whole box consumes 7 watts when running
[17:55:04] <IchGuckLive> Jymmm: for burning pictures into a female skin instat of tatoo
[17:55:07] <djdelorie> well, 8 including the current regulator losses
[17:55:19] <Jymmm> IchGuckLive: Printing presses
[17:55:20] <cylly2> *grrr*... just counted the wires on one iof the Steppers for the new CNC... on wire missing... *repair* ... -> http://www.cyrom.org/palbum/main.php?g2_itemId=13071 (click on tne pic to resize)
[17:55:35] <djdelorie> hey, there's a custom zervice - inverted "tan tatoos"
[17:56:11] <IchGuckLive> Jymmm: yeah but there are outsourced available on ebay for a view hundred dollars
[17:56:26] <Jymmm> 43K Watts per bulb + 5HP Blower to keep the thing cool enough not to melt the solid aluminum reflector
[17:56:58] <IchGuckLive> heavy duty
[17:56:58] cylly2 is now known as Loetmichel
[17:57:17] <djdelorie> "how long until it's... oh wait, it's done."
[17:57:42] <IchGuckLive> pizza by for me today
[17:57:47] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 11.0/20120310193829]]
[17:57:48] * djdelorie has a soldering iron like that
[17:57:57] <Jymmm> Yeah, I saw the blower go out once. the reflector looked like a melted candle
[17:58:06] -!- cevad [cevad!~davec@host-174-45-229-40.bln-mt.client.bresnan.net] has joined #linuxcnc
[17:58:45] <djdelorie> Jymmm: that's when you realize that redundancy is a good thing ;-)
[17:59:01] <djdelorie> water cooled FTW!
[17:59:40] <Jymmm> djdelorie: Oh yeah, water line break on a 43K Watt bulb, homie dont play that!
[18:00:07] <djdelorie> if you blow out the bulb, at least you won't melt the reflector ;-)
[18:00:35] <Jymmm> djdelorie: No, just make sure the client doesn't remove the hose clamp
[18:01:03] <djdelorie> or use non-removable (or hard-to-removable) connections
[18:01:08] <skunkworks> just heard a news clip about a guy that just got an portable artificial heart. The supply pack has 2 batteries that runs the heart (air pump) for 2 hours a piece.. That doesn't seem like enough time in an emergency...
[18:01:14] <Thetawaves> wth is this? http://moglice.com/index.php
[18:01:19] <Thetawaves> moldable bearing material
[18:01:19] <Jymmm> djdelorie: custom solid alumnum reflectors FAR cheaper than 60W/in UV bulbs
[18:01:21] <djdelorie> note: this is the same set of issues for watercooling thousand-dollar CPUs
[18:01:23] <Loetmichel> [20:00] <djdelorie> if you blow out the bulb, at least you won't melt the reflector ;-) <- no, instad the explosive vaporizing will cause it to Shatter all over the place ;-)
[18:02:11] <Loetmichel> Thetawaves: moglice is a GREAT way of refurbishing ways and glide bearings
[18:02:33] <Loetmichel> V-groove ways and flat ones
[18:02:38] <djdelorie> Thetawaves: in the olden days, liquid lead was poured in place to make custom bearings. Might be a similar application.
[18:02:44] <joe9> djdelorie: with using feeler gauges, don't the cutters grind out and have an incorrect depth after some time?
[18:02:49] <skunkworks> babot
[18:03:21] <djdelorie> joe9: I don't push them that hard; either the feeler fits without pressure, or doesn't fit at all. There's no "grinding" involved
[18:03:28] <Loetmichel> djdelorie: with the difference taht moglice is teflon-based and has a MUCH better glide factor AND is mor rigid than lead
[18:03:33] <djdelorie> and the feeler is brass, vs a solid carbide bit? No contest there...
[18:03:35] <Loetmichel> more
[18:03:45] <djdelorie> Loetmichel: and not toxic...
[18:03:54] <Loetmichel> lead isnt
[18:04:02] <Loetmichel> ... even if ingested
[18:04:03] <joe9> djdelorie: i meant, not while measuring. you measure it, run it and then while running, the cutter might lose some sharpness or edge
[18:04:18] <Loetmichel> ... only the oxides and sulfides of lead are dagerous
[18:04:19] <joe9> djdelorie: does that make sense? I am talking about after the measurement
[18:04:29] <joe9> djdelorie: after a few runs or so.
[18:04:35] <djdelorie> joe9: sure, but let's hope that's negligible over the length of the job
[18:04:35] <Loetmichel> because they are water soluble
[18:04:59] <djdelorie> Leotmichel: I know, try explaining that to the Eu RoHS committee
[18:05:29] <joe9> djdelorie: that is not a issue with feeler guage, in particular, that could happen with anything where the depth is measured before the job.
[18:05:40] * Loetmichel works for military... so i have the advantage to solder with the good old 60/40 ;-)
[18:05:41] <djdelorie> joe9: in commercial PCB shops, the computer keeps track of how many holes each bit has drilled, and swaps them out when they've "expired"
[18:05:51] <joe9> oh, ok.
[18:07:06] <alex4nder> joe9: you're going to want to touch off between a reference and the PCB, and then tool change and just use the reference
[18:07:10] <alex4nder> at least that's what I'd do
[18:07:29] <Thetawaves> Loetmichel, i'm thinking about using this to do the main slide in a lathe project
[18:07:57] <joe9> alex4nder: sorry if this is a dumb question. what do you mean by "reference"?
[18:08:13] <joe9> the taig table or something like that?
[18:08:14] <skunkworks> I use a round ground 'standard' a lot for touching off a tool. (set the jog to .001 -> jog in - roll round under tool -> repeat until the round doesn't go under the tool) (I use a 1/2 inch round so then my touchoff would be .5 for the surface to be 0)
[18:09:11] <alex4nder> joe9: something like the "pro touch off gage" that was posted earlier.
[18:09:13] <joe9> skunkworks: that is exactly what cradek recommends too. and it is sounds the easiest way to go.
[18:09:31] <joe9> alex4nder: ok, thanks. got it.
[18:09:41] <skunkworks> I think I picked it up from those guys at one of the fests... works very well :)
[18:09:42] <Loetmichel> Thetawaves: if you have low speed and high forces you will find noting better
[18:09:47] <alex4nder> if I changed tools more often on jobs, I'd get one.
[18:10:05] <Loetmichel> nit for high rpm Spindle bearings, though
[18:10:12] <joe9> alex4nder: just wanted to check, the reference also checks the tool tip length too, correct?
[18:10:25] <alex4nder> joe9: that's all it does.. it's for touching off Z
[18:10:44] <alex4nder> because with a taig you don't have a reasonable quick tool change option with fixed Z
[18:10:49] <Loetmichel> and it doesent spare you the scraping of the ways, the moglice goes just in the slide for a better fit
[18:11:08] <alex4nder> so you have to touchoff Z whenever you change the spindle's setup
[18:11:29] <alex4nder> i.e. unscrew collet nut, replace whatever, rescrew nut, touch off Z
[18:12:36] <joe9> alex4nder: do you have this? http://www.edgetechnologyproducts.com/pro-touch-off-gage.html or is it a different model?
[18:13:05] <alex4nder> I don't yet,. I don't change the spindle configuration enough during runs to need one.
[18:13:45] <joe9> alex4nder: ok, thanks. it is around $74
[18:13:53] <joe9> seems like a good buy.
[18:14:30] <ReadError> thats awesome
[18:14:32] <joe9> it is basically a dial indicator in a fitting.
[18:14:42] <ReadError> i was thinking i had to use a centering bit
[18:14:50] <ReadError> and put it on exactly 1 point
[18:14:58] <alex4nder> that won't replace a centering bit
[18:15:19] <alex4nder> centering is for X-Y
[18:16:46] <ReadError> oh thats for depth
[18:17:00] <ReadError> joe9: did you manage to get everything going?
[18:17:21] <joe9> i have the homing / limit switches working.
[18:17:40] <joe9> now, I am learning about "touch off".
[18:18:15] <djdelorie> joe9: the only trick to understanding "touch off" is realizing that there's a machine origin, and a part origin. Touching off is just how you tell linuxcnc "this point on the machine is this other point on the part"
[18:18:50] <djdelorie> so you can use a 1/4" drill bit to check the height of your milling bit, then tell linuxcnc "this Z position is 0.25 in part-space"
[18:19:46] <joe9> djdelorie: thanks.
[18:20:39] <joe9> djdelorie: i read how pcb's are bad on the cutters and I am thinking if the touching-off needs to be more often (in the middle of a job) to ensure that everything is fine.
[18:20:48] <joe9> and that the cuts are not too deep or too shallow.
[18:20:54] <djdelorie> so, for example, if you had g-code that carved your logo, you could re-use it on all your projects by just "touching off" where you want the logo to be
[18:21:32] <djdelorie> joe9: I would hope that once you start a pcb milling job, you wouldn't have to stop it and recalibrate partway through. If you find you're doing that a lot, add an automatic Z touch-off and code it ever so many minutes in the job file
[18:21:59] <djdelorie> but don't worry about stuff like that until it becomes a problem! I think you're just making more worry for yourself. Cut first, complain later :-)
[18:22:30] <joe9> djdelorie: yes, the "automatic Z touch-off" was what I was thinking. "but don't worry about stuff like that until it becomes a problem!" -- sound advice.
[18:22:50] <joe9> "Cut first, complain later" -- perfect sense.
[18:24:43] <djdelorie> hmmm... cnc drill sharpener... DAMN YOU OVERKILL GENE!!!
[18:26:56] <Jymmm> not as ad as cnc pencil sharpener
[18:27:09] <Jymmm> s/ad/bad/
[18:32:56] <Jymmm> or as scary as a cnc toilet!
[18:33:51] <djdelorie> touch off... he he he... feed rate SLOWER THIS TIME... flood coolant... ESTOP! ESTOP! ESTOP!
[18:35:12] <Jymmm> Well, to be honest, the Japanese got the robotic toilet down to a science
[18:35:24] <Jymmm> $6000 toilet
[18:35:49] <djdelorie> oripoomi
[18:36:20] <Jymmm> heated seat, no touch lid raise/lower, then warm water jets for cleaning, and warm air for drying, etc
[18:38:06] -!- sumpfralle1 has quit [Read error: Operation timed out]
[18:38:31] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[18:38:38] <djdelorie> reminds me of the scene in Cars 2 where Mater tries to figure out all the Japanese button labels...
[18:38:41] <FinboySlick> Jymmm: You should be a salesman, now I want one.
[18:38:50] <Jymmm> FinboySlick: lol
[18:39:49] <Jymmm> http://www.youtube.com/watch?v=KU_vNb_XkTM
[18:39:50] <djdelorie> make sure it has an e-stop button
[18:40:47] <Jymmm> http://www.youtube.com/watch?v=gJazvumHNyk&feature=fvwp&NR=1
[18:41:40] <isssy> http://www.youtube.com/watch?v=bJVC7cctiVk&feature=endscreen&NR=1
[18:41:44] <isssy> german one
[18:41:46] <djdelorie> http://www.youtube.com/watch?v=q9-2pvy5mM8
[18:43:42] -!- bedah [bedah!~bedah@g228251227.adsl.alicedsl.de] has joined #linuxcnc
[18:45:28] <joe9> any one has any thoughts on this set of videos http://www.youtube.com/watch?v=U99asuDT97I&feature=related
[18:46:18] <Jymmm> no soap or towels in jp toilets?!
[18:46:23] <alex4nder> joe9: they're cool
[18:46:40] <alex4nder> joe9: just remember that your machine isn't as rigid as those
[18:46:51] <alex4nder> don't load up any steel and expect 30 IPM straight cuts
[18:48:15] -!- isssy has quit [Quit: Bye Bye]
[18:48:26] <joe9> ok, thanks.
[18:49:25] <JT-Shop> seems to be some good general info
[18:52:27] <djdelorie> JT-Shop: what *are* the expectations wrt milling steel in a "hobby" cnc machine?
[18:52:51] <djdelorie> is it just a matter of reducing the chipload until your machine can handle it?
[18:53:44] <JT-Shop> I don't have a hobby cnc machine but yea that makes sense to reduce chipload until the spindle can handle it, the only problem then becomes heat buildup in the part when you don't remove enough material
[18:54:18] <archivist> and watch for part bending too
[18:54:40] <joe9> harbor freight has this for $2.79: http://www.harborfreight.com/32-piece-sae-metric-feeler-gauge-32214.html
[18:54:49] <JT-Shop> I've machined RC~55 steel dry and as long as I pulled a big enough chip the material stayed cool
[18:55:45] <djdelorie> so better to try it with my 1.5hp router head, than the 55kRPM air spindle head?
[18:55:52] <ds3> JT-Shop: did you get to clean up those chips by hand?
[18:55:56] <djdelorie> big slow cuts better than small fast cuts?
[18:56:22] <archivist> djdelorie, yes, get under the skin
[18:56:24] <JT-Shop> well the SFM needs to be in the range for the material if possible
[18:56:51] <JT-Shop> ds3: no, the chips were too hot!
[18:56:56] * djdelorie wonders if my speed control fob would make things better, or worse, for the poor router... ;-)
[18:57:14] <djdelorie> too many projects, not enough time! :-P
[18:57:21] <archivist> djdelorie, work hardening materials need care
[18:57:22] <JT-Shop> joe9: do you have an edge finder?
[18:57:27] <ds3> JT-Shop: aren't they also sharper then glass?
[18:57:30] <joe9> JT-Shop: no.
[18:58:08] <JT-Shop> are you going to have a vise on there?
[18:59:05] <joe9> JT-Shop: not sure yet. I am thinking of mounting a wood piece almost the size of the table. mill the top to make sure that it is not slanted or horizontal.
[18:59:32] <joe9> then drill 4 holes and use them to mount my pcb piece on it.
[19:00:06] <JT-Shop> ok, then you won't need an edge finder
[19:00:24] <JT-Shop> do you have a dowel pin of any size?
[19:00:41] <djdelorie> joe9: that's pretty much what I did for pcb drilling, but for milling a vacuum setup might be better. Again, try first, see if you need it, it depends on how flat your pcb stock is
[19:00:59] <joe9> JT-Shop: i have a punch set from Harborfreight, which, I can use as a dowel pin.
[19:01:11] <cradek> I flatten my pcb stock before use, just by checking with a straightedge and bending it
[19:01:17] <joe9> JT-Shop: the feeler gauge seems quicker than using a dowel pin.
[19:01:24] <cradek> then it's close enough for double sided tape to hold it flat to the table
[19:01:30] <joe9> cradek: how do you flatten it?
[19:01:36] <cradek> I bend it
[19:01:49] <cradek> it's surprisingly flexible and it holds a bend
[19:01:57] <JT-Shop> the dowel is faster and safer as your not moving the tool down against an immovable object
[19:02:02] <ds3> cradek: so you machine pcbs dry?
[19:02:12] <cradek> yes
[19:02:22] <cradek> would have never thought of wetting them
[19:02:27] <ds3> cradek: how do you prevent fiberglass dust from getting all over?
[19:02:36] <djdelorie> don't cut the fiberglass part :-)
[19:02:45] <cradek> yeah you're cutting way too deep if you're getting a lot of it
[19:02:52] <JT-Shop> http://www.youtube.com/watch?v=VmRUJk0ptN4
[19:03:04] <ds3> doesn't a layer of fiberglass crumble when you take off the copper?
[19:03:14] <cradek> what do you mean crumble?
[19:03:18] <joe9> with any special equipment. will
[19:03:33] <ds3> as in the cutter will cut strands of glass stuck to the back of the copper
[19:03:43] <ds3> thus generating fiberglass dust
[19:03:50] <cradek> sure, there's some dust
[19:04:01] <cradek> machining is messy, vacuum it up when you're done
[19:04:01] <joe9> jt-shop: re "the dowel is faster and safer as your not moving the tool down against an immovable object" -- why not just bring the tool tip close enough and use the feeler guage to figure out which one slides through?
[19:04:13] <archivist> if worried use a vacuum cleaner
[19:04:17] <cradek> joe9: you should try both methods and use what you think is best
[19:04:32] <ds3> oh so it won't grind your leadscrews to obvillian in one session?
[19:04:54] <JT-Shop> joe9: did you watch the video
[19:05:08] <joe9> JT-Shop: yes, I just saw that.
[19:05:21] <JT-Shop> because it just takes longer to try a bunch of feeler gauges
[19:05:27] <ds3> cradek: what about when you are drilling it?
[19:05:29] <joe9> were you jogging the machine by hand?
[19:05:43] <JT-Shop> and it is prone to mistakes as your always having to enter a different number
[19:05:49] <JT-Shop> yes with a MPG
[19:05:57] <joe9> jt-shop: were you jogging the machine by hand?
[19:06:08] <JT-Shop> joe9: yes, with a MPG
[19:06:17] <joe9> oh, what is a MPG?
[19:06:22] <cradek> JT-Shop: nice video. about 10 seconds to measure a tool very precisely.
[19:06:32] <JT-Shop> Manual Pulse Generator
[19:06:38] <joe9> jt-shop, nice video. thanks.
[19:07:03] <JT-Shop> yes, you taught me well cradek
[19:07:03] <cradek> at the end you can see the dowel move more and more each increment
[19:07:24] <cradek> it's even easier by feel
[19:07:30] <JT-Shop> joe9: cradek taught me how to do that
[19:07:30] <joe9> jt-shop. the MPG thing seems to help. When I jog manually using the PgUp/PgDown, I am not able to get it to work correctly.
[19:07:42] <joe9> it either overshoots or does not move.
[19:07:49] <JT-Shop> joe9: set the increment to 0.001"
[19:07:55] <joe9> the response makes it worthless to try what you were doing.
[19:08:04] <cradek> either that or use a VERY slow continuous jog
[19:08:30] <JT-Shop> set the increment so each press gives you a specific amount of movement
[19:09:36] <archivist> I often measure a machined part to set offsets
[19:09:36] <joe9> jt-shop: my problem is when I press Pg-Up/Pg-Down, sometimes I get movement and sometimes I don't.
[19:10:05] <joe9> not sure if that has anything to do with using linuxcnc over ssh. but, it is pretty unresponsive.
[19:10:06] <JT-Shop> joe9: when my MPG on the BP knee mill has a headache I just use the buttons and 0.010 for rough position and 0.001 for final position
[19:10:07] <archivist> as long as its one step or less
[19:10:52] <archivist> why are you using ssh
[19:10:54] <joe9> cradek: "either that or use a VERY slow continuous jog" -- jog using PgUp and PgDown correct?
[19:11:06] <cradek> yes
[19:11:14] <JT-Shop> why are you using ssh?
[19:11:17] <FinboySlick> joe9: What's your front-end for using it over ssh?
[19:11:28] <joe9> archivist: because I have better monitors on my main machine
[19:11:43] <joe9> FinboySlick: what do you mean ssh -X
[19:11:52] <joe9> FinboySlick: what do you mean? ssh -X
[19:11:56] <FinboySlick> joe9: So you use Axis?
[19:12:00] <joe9> yes.
[19:12:12] <djdelorie> joe9: you're just redirecting the X display to a different X server, right?
[19:12:16] <joe9> yes.
[19:12:25] <djdelorie> yeah, that shouldn't change anything
[19:12:35] <FinboySlick> joe9: Did you have to do any special steps for that. I never could get it to work.
[19:12:41] <djdelorie> but you probably do want to change your jog mode from continuous to one of the steps
[19:12:56] <djdelorie> FinboySlick: just "ssh -X some.machine" and run axis
[19:13:11] <joe9> ok, will try the jog mode of "steps". thanks.
[19:13:33] <djdelorie> with step mode, you pick a distance, and each time you hit the pgup/pgdown key, it moves that amount,
[19:13:42] <FinboySlick> djdelorie: I'm familiar with remote X displays. It just didn't seem to like the small realtime window that pops up when you start AXIS for some reason.
[19:13:46] <djdelorie> so it becomes "tap tap tap tap tap" instead of "tap hold release"
[19:13:59] <joe9> FinboySlick: ssh -Y -p sshport -l user machineip
[19:14:13] <archivist> even with continuous you can tap tap tap
[19:14:20] <FinboySlick> joe9: Yeah, that's what I used.
[19:14:22] <joe9> i wrapped it up in an expect script so it does all that automatically.
[19:14:38] <joe9> export DISPLAY=...
[19:14:57] <djdelorie> the -X is supposed to set $DISPLAY for you
[19:15:01] <JT-Shop> so your keyboard is not connected to the machine PC?
[19:15:13] <joe9> instead of the above, try this: export XAUTHORITY=/../.Xauthority
[19:15:29] <joe9> JT-Shop: yes.
[19:15:34] <FinboySlick> joe9: Yes, that too. As I said, I've used remote X displays before... It's just AXIS that didn't seem to work. I thought it was related to some RTAI stuff.
[19:15:52] <joe9> FinboySlick: i could not get stepconf to work. but axis worked
[19:16:14] <joe9> FinboySlick: and, you did the ulimit too?
[19:16:21] <joe9> ulimit -l 1024000
[19:16:33] <Jymmm> Different http://www.youtube.com/watch?v=H3WdFTf2kBM&feature=related
[19:16:43] <joe9> and the rip-environment
[19:17:41] <djdelorie> oh great, cnc radios now...
[19:20:17] <FinboySlick> joe9: Maybe I was missing ulimit, it was a while back.
[19:20:39] <joe9> and, the whole thing works in bash.
[19:20:45] <JT-Shop> joe9: do you have enough hardware inputs to have a jog button in halui?
[19:21:06] <joe9> and, ssh does not use login. so, after ssh'ing, do a su user again.
[19:21:36] <joe9> JT-Shop: i have one input on the Gecko G540 that I am not using.
[19:21:43] <joe9> JT-Shop: does that help?
[19:22:09] <JT-Shop> iirc you could select the axis in the gui and use one button to jog
[19:23:13] <joe9> jt-shop, that would be cool. will check it out. let me try the step mode again. from what I recall, it was working pretty flakily too.
[19:23:52] <JT-Shop> but don't know how well that would work without an up down selector of some sort... maybe a pyvcp button
[19:24:09] <JT-Shop> http://linuxcnc.org/docview/html/man/man1/halui.1.html
[19:24:18] <joe9> i think i will just set it "continuous mode" with a velocity of 0.05 mm /sec
[19:24:40] <joe9> and use the E-stop button to stop as soon as the dowel pin rolls under it.
[19:25:24] <joe9> and, instead of jogging it, I will use MDI to start it up.
[19:25:39] <joe9> the response is pretty cool with MDI. no issues there.
[19:29:33] <JT-Shop> there you go
[19:32:58] -!- kb8wmc [kb8wmc!~chatzilla@nat.mtp.cmsinter.net] has joined #linuxcnc
[19:42:22] <Loetmichel> sooo, y axis done... and runs... now i will turn the mill over and doing x and z ;-) -> http://www.cyrom.org/palbum/main.php?g2_itemId=13074
[19:43:14] -!- psha has quit [Quit: Lost terminal]
[19:44:28] <JT-Shop> cool
[19:56:24] -!- maximilian_h has quit [Ping timeout: 245 seconds]
[20:01:13] -!- phantoxe [phantoxe!~destroy@a95-92-89-24.cpe.netcabo.pt] has joined #linuxcnc
[20:02:09] <mrsun> Loetmichel, motor not strong enough for direct drive?
[20:02:15] <mrsun> that looks like a backlash nightmare =)
[20:02:35] <Loetmichel> mrsun: no backlash
[20:02:41] <mrsun> just wait ;PO
[20:02:44] <Loetmichel> i.e less than a step
[20:03:00] <mrsun> looks very nice =)
[20:03:09] <Loetmichel> the motr mount is adjustable
[20:03:23] <Loetmichel> so if the gears wear down just adjust it ;-)
[20:03:36] <Loetmichel> and being them Polycarbonate that will not happen sooon
[20:04:29] <Jymmm> Would heat cause SS to rust?
[20:04:31] <Loetmichel> no, the direct drive would be less powerful AND the resolution will suffer
[20:04:44] <Loetmichel> Jymmm: whats ss?
[20:04:54] <Jymmm> Stainless steel
[20:05:07] <Loetmichel> o, just change colour
[20:05:11] <Loetmichel> no
[20:05:19] <mrsun> Jymmm, theres no such thing as rust free =)
[20:05:23] -!- FinboySlick has quit [Quit: Leaving.]
[20:05:32] <mrsun> its just more or less resitant to corrosion =)
[20:05:35] <Jymmm> mrsun: This is significant rust
[20:05:42] <mrsun> Jymmm, hmm, some kind of acid maybE?
[20:05:47] <Tom_itx> Jymmm it was also made in china
[20:05:56] <mrsun> playing with sulfuric acid or something? :)
[20:05:57] <Jymmm> not that I"m aware of, just water in them
[20:06:09] <Tom_itx> Jymmm, it may just be stainless 'color'
[20:06:14] <mrsun> Jymmm, hmm, sounds strange .. should not rust like that =)
[20:06:31] <Jymmm> Tom_itx: No, they say 18/8 SS
[20:06:38] <Tom_itx> yeah well
[20:06:44] <Tom_itx> we're talking about china here
[20:07:12] <Jymmm> Sure, I just wanted to make sure, as I've had food service SS, but never rusted that I saw
[20:07:53] <Jymmm> But I usually dont toss into 1300F coals either so I didn't know
[20:08:19] <mrsun> Jymmm, using it in fires? :)
[20:08:22] <Jymmm> heat marks, yes. but not rust.
[20:08:34] <Jymmm> mrsun: Yeah, to boil water in
[20:08:52] <mrsun> remember water is alot more "reactive" when its hot also =)
[20:09:17] <mrsun> but still shouldnt rust just like that
[20:09:27] <Jymmm> Sure, I expect some deposits, but these botles were never stored that way and never for long term
[20:09:36] <mrsun> have some SS bars laying around in the yard, been there for 3 winters now and not a sign of rust on them =)
[20:09:44] <mrsun> tho thats swedish steel :P
[20:10:04] <Jymmm> mrsun: swedish fire steel?
[20:10:13] <mrsun> huh ? :P
[20:10:18] <Jymmm> mrsun: if so, I'll take 10 feet please =)
[20:10:18] <mrsun> the bars is just ordenary bars :P
[20:10:20] <mrsun> laying around :P
[20:10:26] <Jymmm> bah
[20:10:55] <mrsun> got some acid resistant tubing that i got for free but dont wnt to send stuffa round the world ;P
[20:11:25] <Tom_itx> there isn't a 18/8 stainless those are percentages of composition
[20:11:53] <Tom_itx> stainless it typed from 201 to 384 depending on the composition
[20:12:23] <Tom_itx> there may be more since this book was published
[20:12:24] <Jymmm> Nah, just need to be able to boil water in. They are "subero" brand SS bottles. the MFR is replacing them on the photos I sent, but her boss now wants the bottles and is sending a calltag for them. At least they care and are looking into it.
[20:13:04] <Jymmm> Tom_itx: Yes, but 18/8 is the nomincalture they use in the food service industry is all
[20:13:14] <Tom_itx> the '18' is likely Cr and the 8 is likely Ni
[20:13:15] <Jymmm> also the NSF I believe
[20:13:21] <mrsun> bottles ? :) sure it isnt just ordenary steel with a plastic coating? :P
[20:13:23] <Jymmm> Tom_itx: Yes,
[20:13:30] <Tom_itx> what do they know about metal composition?
[20:13:39] <Jymmm> mrsun: No, I shoved them into 1300F coals
[20:13:51] <mrsun> then the plastic is gone then? ;PO
[20:14:01] <mrsun> and voila, rust ;P
[20:14:20] <Tom_itx> coal is carbon isn't it?
[20:14:26] <Loetmichel> and being a pvc coating: some chlorine to speed up the rust ;.-)
[20:14:30] <mrsun> Tom_itx, yes
[20:14:39] <Jymmm> the rust ins't on the outside, it's on the INSIDE only
[20:14:43] <Tom_itx> maybe you altered the metal composition
[20:14:51] <mrsun> Jymmm, well isnt the inside where the water is? :)
[20:14:55] <Tom_itx> i'm gonna call the mfg and tell em what you did to their product
[20:15:10] <Jymmm> Tom_itx: want the number?
[20:15:15] <Tom_itx> not really
[20:15:37] -!- maximilian_h [maximilian_h!~bonsai@anon-190-67.vpn.ipredator.se] has joined #linuxcnc
[20:15:42] <Jymmm> It's not uncommon to do this, actually other mfr talk about doing this all the time.
[20:15:57] <Tom_itx> heat them to 1300F with some other heat source and see if they rust as much
[20:16:51] <Jymmm> all i have is a proane torch
[20:17:08] <mrsun> Jymmm, what kind coal is it ?
[20:17:16] <mrsun> "stone" coal ?
[20:17:17] <Jymmm> mrsun: bbq
[20:17:19] <Tom_itx> you should talk to JT-Shop about getting you some bigger tools to work with
[20:17:27] -!- erasmo has quit [Remote host closed the connection]
[20:17:44] <Jymmm> Tom_itx: no place for them =)
[20:17:56] <Tom_itx> oh he can fix you up with that too
[20:17:57] <mrsun> Jymmm, so charcoal ? ... if its not fully destilled it can contain some very corrosive stuff =)
[20:18:14] <mrsun> and i think it still does even when its well made
[20:18:25] <Tom_itx> the truth finally comes out
[20:18:44] <Jymmm> sure, but only the INSIDE rusted, not the outside which it perfectly fine actually
[20:18:51] <Tom_itx> you likely altered the chemical makeup
[20:19:03] <mrsun> Jymmm, ahh... im tired sorry =)
[20:19:35] <Jymmm> Tom_itx: I'm not the first to toss a SS container in a fire
[20:20:12] <Tom_itx> and likely not the first to have one corrode
[20:20:25] <Jymmm> It's one of the min reasons ppl are buying them now over aluminum ones
[20:20:30] <Jymmm> main
[20:21:01] <Jymmm> Actually Kleen Kateen talks about doing this, but with no issues.
[20:22:35] <Jymmm> Klean Kanteens® are constructed from high-quality, 18/8, food-grade stainless steel. The numbers stand for the percentages of chromium and nickel in the steel, 18% and 8% respectively. Chromium increases the metal's hardness and nickel gives it strength. Together they also make stainless steel highly resistant to stain or rust.
[20:23:31] <Jymmm> In a pinch, single-walled Kanteen® Classic and Wide bottles with the Brushed Stainless finish can be placed on a fire or a stove as long as you take the cap off. Do not do this with a colored bottle, as it will damage the paint.
[20:24:13] <Jymmm> Never heard of this....
[20:24:24] <Jymmm> The food-grade stainless steel in a Klean Kanteen® is highly rust resistant. If your bottle appears to have rusted, it’s probably just a mark left behind from the electropolishing process, which smooths the surface of the metal and makes it shiny. Sometimes electrolytes left on the surface turn a rust color, but it’s harmless.
[20:33:56] e-ndy is now known as e-ndy|afk
[20:38:25] e-ndy|afk is now known as e-ndy
[20:39:51] -!- cra_ [cra_!56b358b3@gateway/web/freenode/ip.86.179.88.179] has joined #linuxcnc
[20:40:50] <cra_> Bit of a wierd question, but if you had an unlabeled BOB and couldn`t check for continuity electrically, how could you test to determine the number of a certain pin
[20:42:48] <cra_> basically, I have a wierd TB6560 board and expect two pins to be the unused pin 16 and 17. There is no continuity but I`m almost certain this is due to opto-isolators and other circuitary. I need a test using emc by setting something in the software and measuring an output at the pin to check it is the pin that has been set.
[20:42:51] <cra_> any thoughts?
[20:43:21] <cradek> if you think it's outputs, you can just turn one at a time on. there's not many outputs to try.
[20:44:03] <cra_> I`m quite a noob really...when you just say "turn them on" ... that sounds perfect, and just test when it comes on. But how and what exactly am I turning on.
[20:44:18] <cra_> You see what I mean, how do I actually tell output pin 16 to come on?
[20:45:59] <cradek> run linuxcnc with a parallel port based config. while it's running, run at a shell halcmd -kf, to get a hal prompt. then setp parport.0.pin-14-out 1, etc etc
[20:46:31] <cra_> :-S shell. .... halcmd !!!
[20:47:08] <cra_> you going to be still on in 5 mins, I`ll just get to my machine
[20:47:17] e-ndy is now known as e-ndy|afk
[20:52:12] -!- mhaberler [mhaberler!~mhaberler@BSN-77-151-107.dsl.siol.net] has joined #linuxcnc
[20:54:23] <cra_> ok, so I`m in linuxcnc... can you please explain "run at a shell halcmd -kf" please?
[20:57:33] <cra_> I see show hal config, hal meter, hal scope ???
[20:57:44] <JT-Shop> cra_: there is a parallel port tester on the forum
[20:59:35] <JT-Shop> cra_: if your interested http://linuxcnc.org/index.php/english/component/kunena/?func=view&id=4743&catid=18#4743
[20:59:45] <skunkworks> cra_: try ALT-T
[21:00:29] <cra_> JT - will try the PP tester - that looks fairly simple!
[21:02:20] <cra_> OK...sorrry!! I really am a linux noob. when it says put them in the same directory and then run a terminal... ?? What any directory and I don`t need to point to them with the terminal code?
[21:03:17] <DJ9DJ> gn8
[21:03:58] -!- DJ9DJ has quit [Quit: bye]
[21:04:12] -!- mhaberler has quit [Quit: mhaberler]
[21:04:49] <cra_> when I type in terminal, it says invalid option --'l'
[21:05:15] <cra_> halrun -I -f ptest.hal is what it tells me to write
[21:06:42] -!- user_reader [user_reader!~user_read@84-166.192-178.cust.bluewin.ch] has joined #linuxcnc
[21:07:12] <JT-Shop> cra_: put both files in any directory and cd to that directory the do the halrun part
[21:07:24] <cra_> cd ???
[21:07:27] <cra_> :( sorry!
[21:07:37] <JT-Shop> http://linuxcnc.org/docview/html/common/Linux_FAQ.html#_terminal_commands_a_id_sec_terminal_commands_a
[21:07:38] <Jymmm> cd == change directory
[21:07:46] <cra_> ahh!
[21:08:10] <JT-Shop> cra_: not a problem if you don't know something just ask
[21:08:20] <cra_> so I`ve put them in my home directory
[21:08:28] <JT-Shop> that makes it easy
[21:08:29] <cra_> cd home/
[21:08:33] <JT-Shop> open a terminal
[21:08:38] <Jymmm> cd ~
[21:08:45] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[21:09:08] -!- isssy has quit [Client Quit]
[21:09:19] <cra_> terminal is open
[21:09:25] <JT-Shop> do a ls
[21:09:32] <JT-Shop> this will list the files
[21:09:34] <cra_> chris@chirs-desktop:-$
[21:10:20] <cra_> yes, I see both ptest.hal and ptest.xml files
[21:11:16] <JT-Shop> ok then the halrun command
[21:11:29] <cra_> WAHOO!
[21:11:54] -!- maximilian_h1 [maximilian_h1!~bonsai@130.255.104.157] has joined #linuxcnc
[21:12:01] -!- maximilian_h1 [maximilian_h1!~bonsai@130.255.104.157] has parted #linuxcnc
[21:12:33] -!- maximilian_h has quit [Ping timeout: 245 seconds]
[21:13:40] <JT-Shop> I take it that it worked...
[21:15:17] <cra_> yes.... but unfortunately, I don`t seem to see any output from these random pins on the TB6560!!
[21:15:23] <cra_> I HATE this board.
[21:17:57] <cra_> Can I please clarify something with you guys who clearly know what you are doing. I trying to control my spindle so I need a spare output. On the tB6560, pin 16 and 17 are not used. Using this schematic here: http://www.drkfs.net/REVERSESTEPPER.pdf there is a section towards the top left, that shows pin 16 and 17 in a 4 block connector labeled "PL3 4WP DSC".... it shows 16 and 17 on the two ends with two grounds in the centre.
[21:18:44] -!- gambakufu [gambakufu!~mooznach@bzq-84-110-5-111.red.bezeqint.net] has joined #linuxcnc
[21:18:55] <cra_> So I`m stood here now with my multimeter, I`ve hooked into a ground and connected to one of the outer pins and I`m right in thinking that when I hit either the 16 or 17 output on this PP tester so it goes green, I should see it jump to +5v ???
[21:18:59] <cra_> ....and I`m not!
[21:19:12] <djdelorie> at least 2.2v
[21:19:20] <djdelorie> on the parallel port
[21:19:39] <cra_> BUT...there is continuity between the the two centre pins and ground! ...so clearly the schematic must be right in some ways!
[21:20:30] <djdelorie> the schematic shows output pin 16 tied to ground...
[21:20:50] <djdelorie> or are you talking about the parallel port pin 16?
[21:21:27] <cra_> does it... ? I`m talking about the parallel port pin 16
[21:21:45] <cra_> I want to use the parallel port 16, as it is unused on this board, to control my spindle.
[21:22:47] <djdelorie> based on that schematic, pp pins 16 and 17 are pull-up-only for "PL3" pins A and D
[21:22:58] <djdelorie> or they're inputs :-)
[21:23:09] <djdelorie> nope, outputs
[21:23:42] <cra_> so does this mean I can tap onto them for an output?
[21:24:48] <djdelorie> the 1N4136 looks like a *91* volt zener. Wow.
[21:25:01] <djdelorie> do you have anything plugged in to PL3 ?
[21:25:10] <cra_> that doesn`t mean a lot to me???!
[21:25:38] <djdelorie> which that?
[21:26:22] <cra_> PL3.... one sec
[21:26:38] -!- bootnecklad [bootnecklad!~bootneckl@host-2-97-39-101.as13285.net] has joined #linuxcnc
[21:26:39] <cra_> PL3 - is this the 4 pins socket?
[21:26:43] <djdelorie> yes
[21:27:13] <cra_> No,,,,this is a totally mystical socket that is not discussed in the manual or any info on the net...
[21:27:52] <djdelorie> based on the schematic, you could connect two opto's to that socket, one on pins A-B and one on pints D-C (B and C are ground), with a suitable series resistor.
[21:27:55] <cra_> I presumed if I connected to one of the end pins on PL3 I could take an output for my spindle from the parallel port 16 or 17, depending on which outer one I used.
[21:28:13] <djdelorie> to test, connect a resistor across A-B or D-C, and watch the voltage across it with a multimeter
[21:29:05] <djdelorie> you *should* be able to test for continuity from PP pins 16/17 to PL3 pins A/D, but only in one direction - if you swap the two leads, you'll get nothing
[21:29:07] <cra_> OK, I can try that....I was just simply using the parallel port tester before and presumed I should see a voltage when I took 16 or 17 high?
[21:29:23] <djdelorie> it looks that way, yes.
[21:29:59] <djdelorie> assuming the schematic is correct, D5 and D19 are installed, and they're not blown
[21:31:17] <cra_> AHHH! I`m sorry, I`m fairly useless at all this. OK, just tested and I`ve discovered that I get a 4.5v output on this pin but it raises to 7.5v when I turn on pin 16!!
[21:31:21] <cra_> WOOP!
[21:31:44] <djdelorie> hmm.... that shouldn't happen, what are you using for a ground reference?
[21:31:55] <cra_> the ground pin next to it
[21:32:05] <djdelorie> and you might need a resistor across those too, to get a reliable reading
[21:32:37] <cra_> Please can I ask you a question based on what I`m even trying to use this for?
[21:32:38] <djdelorie> 270 ohms or so
[21:32:45] <djdelorie> sure
[21:34:34] -!- bootnecklad has quit [Quit: BRB WIREWRAPPING]
[21:34:51] <cra_> I`ve made myself a DC brushless motor which is working really well... I`m currently powering it with a ATX PC supply and that runs with an ESC driver. Into the driver is a servo tester, the servo tester has a pot on in and turning the pot changes the speed of the motor. Only three lines go from the ESC driver to the servo tester +/ - to give it power and then a signal line. I`ve been told that if I can pull off an output from 16 or 1
[21:34:57] <ReadError> OMG OMG OMG
[21:34:59] <ReadError> IT SHIPPED!
[21:35:02] <ReadError> *dances*
[21:35:25] <djdelorie> cra_: your sentence ended with "16 or 1"
[21:35:34] <cra_> spindle, I can get linuxcnc to generate the speed control for the motor automatically rather than using servo.
[21:35:46] <cra_> djelorie....huh?
[21:35:54] <djdelorie> there's a length limit to how much you can type at once in IRC
[21:36:05] <cra_> Try again....
[21:36:08] <cra_> I`ve made myself a DC brushless motor which is working really well... I`m currently powering it with a ATX PC supply and that runs with an ESC driver. Into the driver is a servo tester, the servo tester has a pot on in and turning the pot changes the speed of the motor.
[21:36:22] <cra_> Only three lines go from the ESC driver to the servo tester +/ - to give it power and then a signal line. I`ve been told that if I can pull off an output from 16 or 17 and set one of these to....
[21:36:24] <Jymmm> ReadError: Yes, to zipcode 00805
[21:36:34] <cra_> spindle, I can get linuxcnc to generate the speed control for the motor automatically rather than using servo.
[21:37:12] <djdelorie> cra_: see http://www.delorie.com/tmp/irc-cra.png
[21:37:50] <djdelorie> yup, you need an "RC servo pulse" output I guess?
[21:37:58] <cra_> djelorie: I don`t understand? I haven`t copied to an image.
[21:38:29] <djdelorie> did you look at the url I posted? It's just an example of your sentence getting cut off by IRC
[21:38:36] <djdelorie> nothing to do with your servo tester ;-)
[21:38:48] <cra_> ahhh!
[21:39:13] <ReadError> its coming in 2 boxes
[21:39:16] <ReadError> hope thats normal ;/
[21:39:42] <cra_> so it should be possible. Can I ask then...do I need any other components or can I simply take a wire from pin 16 to my ESC driver?
[21:39:46] <Jymmm> ReadError: common for oversized thngs
[21:39:56] <Jymmm> cheaper too
[21:40:07] <Jymmm> no oversized/wight fees
[21:40:13] <Jymmm> weight
[21:40:15] <djdelorie> if your server driver is a normal RC servo driver, it wants volts near 0 and near 5v. So, since you're not quite getting that, try putting a resistor across the pins and check again
[21:40:22] <djdelorie> the resistor will help you get closer to 0 volts
[21:40:31] <Thetawaves> cra_, cool, i want to build a motor
[21:40:42] <djdelorie> ReadError: lots of my stuff comes in two boxes, if one of the parts is big and heavy and needs special padding
[21:41:06] <ReadError> too bad a2z kinda screwed me ;(
[21:41:09] <ReadError> ordered handles
[21:41:11] <ReadError> wait a week
[21:41:12] <Jymmm> djdelorie: That's a LOT of pr0n!
[21:41:13] <cra_> Thetawaves: I didn`t build the motor, I converted it to a spindle.
[21:41:20] <ReadError> they say "ohai, yea we dont have those"
[21:41:29] <Thetawaves> cra_, what motor?
[21:41:29] <ReadError> even though ebay said they had 2 ;(
[21:41:42] <djdelorie> cra_: the other alternative is to open up your box with PL3, take out those two diodes, and replace them with wires. Then PL3 is a direct connection to the pp pins
[21:41:49] -!- skunkworks has quit []
[21:41:57] <djdelorie> Jymmm: ?
[21:42:04] -!- Nick001-Shop [Nick001-Shop!~chatzilla@69.72.53.110] has joined #linuxcnc
[21:42:14] <Jymmm> djdelorie: dual boxes
[21:42:36] <cra_> what are the two diodes there for?
[21:42:41] <djdelorie> jointers, lathes, etc. "over the years" :-)
[21:42:47] <djdelorie> cra_: no idea
[21:43:09] <cra_> ThetaWaves: am I allowed to post a link to the forum I wrote up my motor build on, on here?
[21:43:22] <Thetawaves> i will murder you if you don't
[21:43:28] <JT-Shop> jyyyyyyyyyyyyyyyyym http://imagebin.org/207979
[21:43:43] <cra_> djdelorie: what are they doing there in terms of electronically....is there any harm leaving them in?
[21:43:54] <Jymmm> JT-Shop: WOOHOO!
[21:44:10] <djdelorie> if you leave them in, you'll need a resistor to pull the signal back down to zero. It looks like they're set up for optos but not logic output
[21:44:12] <Jymmm> JT-Shop: I only see diesel, where be the powder???
[21:44:33] <djdelorie> JT-Shop: that's the weirdest looking CNC machine I've ever seen. Cool, though :-)
[21:44:36] <Jymmm> JT-Shop: And YOU HAVE NO BALLS!!!
[21:44:42] <cra_> http://www.raynerd.co.uk/?p=1562
[21:44:55] <Jymmm> djdelorie: It's a CNC PUNCH!
[21:45:33] <cra_> OK...I`m sorry, you keep telling me I need a resistor there, I`ll listen! OK, so about 280ohm and just literally between the pin and the ground next to it?
[21:46:16] <djdelorie> for testing, yes. 270 ohms is about right to simulate an optocoupler. For logic, you could get by with a larger value. The diode prevents the PP pin from pulling the signal back down, so you need something else to do that
[21:46:23] <cra_> Thetawaves - got some pics of it in pieces here so you can see in more detail how it is together.
[21:46:49] <cra_> AHHHHH! ...my little bit of electronics knowledge, that makes sense.
[21:47:33] <djdelorie> the diode also drops 0.7 volts or so, reducing the "high" voltage, if that's a problem, replacing the diode with a wire might help
[21:49:03] <cra_> It seems quite intrecate in there and I wouldn`t like to remove it. Would it work if I just jump wires over the top of it and soldered a wire to either leg of the diode that I can see.
[21:49:12] <cra_> I don`t want to damage the board pulling it out.
[21:49:18] <djdelorie> sure
[21:49:29] <JT-Shop> Jymmm: getting close man
[21:49:37] <Thetawaves> cra_, that thing is sweet
[21:49:41] <djdelorie> just solder jumpers across each diode, shorting them out
[21:49:55] <Thetawaves> did you fab the spindle bearing/housing?
[21:50:04] <cra_> yes.
[21:50:11] <Jymmm> JT-Shop: Cool beans!
[21:50:23] <djdelorie> ok you, how do you make your own collet? Or do you buy that part?
[21:50:35] <cra_> Thetawaves - it works very well so far!
[21:50:51] <Thetawaves> what rpms?
[21:51:09] <Tom_itx> JT-Shop that looks like it's about time to rock the neighborhood
[21:51:35] <djdelorie> cra_: you're using a "hobby" ESC to run that?
[21:52:03] <cra_> djdelorie - I have made my own er32 collet chuck in the past..... BUT for £16 I couldn`t beat a chinese straight shank one!!
[21:52:35] <djdelorie> the three wires coming from the ESC include power and ground, you could insert an opto-isolator in there if you wanted
[21:52:46] <cra_> t is not a hobby brand- some chip copy off ebay
[21:53:00] <djdelorie> I mean, it's for radio controlled airplanes
[21:53:28] <cra_> oh I see. Yes, that is right. Outrunner, brushless DC, yes
[21:54:25] <JT-Shop> Tom_itx: it's getting close, today is the first time of fitting the barrel on :-)
[21:54:30] <djdelorie> does the ESC provide the +5v or does your servo tester provide it?
[21:55:03] <cra_> the ESC provides it.
[21:55:18] <ssi> I have a spindle made from a bldc + ESC
[21:55:36] <cra_> bldc?
[21:55:41] <djdelorie> good, you'll need that if you put an opto in it
[21:55:42] <ssi> brushless dc motor
[21:55:45] <cra_> burshless!!
[21:55:53] <cra_> yes, sorry, got it!
[21:56:15] <cra_> can I be honest, what the hell does an opto look like... what would I even buy?
[21:56:31] <ssi> like a little dip or soic usually
[21:56:35] <ssi> they can come in lots of packages
[21:56:55] <ssi> single optos are often in a package the size of an 8 pin dip or soic, but with only 4 pins
[21:57:02] <djdelorie> random example: http://search.digikey.com/us/en/products/FOD817A300W/FOD817A300W-ND/1049597
[21:57:10] <cra_> can you recommend anything from ebay UK and I`ll order one now....
[21:57:15] <cra_> I presume I need one?
[21:57:20] <djdelorie> you don't *need* one
[21:57:27] <cra_> and would this reduce the need for me to short the diode?
[21:57:33] <cra_> reduce = remove
[21:57:38] <djdelorie> they provide electrical isolation between your PC and the high voltage motor stuff
[21:57:49] <djdelorie> I don't have any in my CNC machine
[21:58:13] <cra_> why not?
[21:58:21] <djdelorie> In that schematic you posted, all the PC817 things are optos. LED on one side shines on optotransistor on the other side
[21:58:37] <djdelorie> they're for paranoia, not functionality
[21:58:52] <cra_> lol
[21:58:54] <djdelorie> my paranoia went elsewhere that day
[21:59:31] <djdelorie> my advice: go without them for now, add them later if you feel like it
[21:59:38] <cra_> but in some ways, with these diodes in place and in an ideal world, me not wanting to solder anything direct to the board, they might be quite handy....
[21:59:39] <djdelorie> cut first, complain later! :-)
[21:59:48] <djdelorie> you'd just have to solder the optos
[22:00:08] <cra_> well if I don`t use them, I`ll have to short those diodes and I don`t like soldering in tight spaces....
[22:00:12] <djdelorie> with what you have now, the easiest path is a resistor across A-B and wire that right into your ESC
[22:00:41] <cra_> infact, I don`t even know what socket that even is!! I might have to solder directly to those pins anyway~!
[22:01:07] <djdelorie> in that case, just solder to the PP side of the diodes and be done with it :-)
[22:01:43] <cra_> lol, that is true!!
[22:03:06] <cra_> so actually....I can just solder a wire directly from the PP side of the diode and take that wire right the way to me ESC where the signal is currently coming in from the servo tester
[22:03:25] <djdelorie> sure. And ground.
[22:05:45] <cra_> ground??
[22:05:56] <djdelorie> you need more than *one* wire to make a circuit
[22:06:04] <cra_> ooops.
[22:06:30] -!- Tecan [Tecan!~fsadf@unaffiliated/unit41] has joined #linuxcnc
[22:06:31] <djdelorie> two wires; one goes to the PP side of the diode, for the "signal" to your ESC, and the other goes to ground, for the "gnd" to your ESC
[22:06:42] <ssi> does anyone know anything about the DC servo motors that keling sells?
[22:06:57] <cra_> So if I then take a wire from the ground next to it to the black ground wire coming from the esc. However, the red +5v won`t be connected to anything.
[22:07:06] <djdelorie> right
[22:07:15] <cra_> you`ve been a god..nice one.
[22:07:33] <djdelorie> just make sure all your "grounds" are the same potential before plugging it all together :-)
[22:07:41] <djdelorie> (which they SHOULD BE)
[22:09:32] <cra_> well I`m just thinking, the ground from the esc is from the "esc" ...then I`ll be linking that to a ground from the driver ..... !
[22:17:06] -!- iwoj [iwoj!~iwoj@d23-16-234-107.bchsia.telus.net] has joined #linuxcnc
[22:18:31] -!- stevegt_1 [stevegt_1!~stevegt@64-130-244-2-gracehospitalitygroup.bwg.phx.sparkplugbb.net] has joined #linuxcnc
[22:22:40] -!- bedah has quit [Quit: bye]
[22:32:08] <ReadError> ssi
[22:32:13] <ReadError> you been to their new site?
[22:32:14] <ReadError> much nicer
[22:32:24] <ssi> no?
[22:32:26] <ReadError> http://www.automationtechnologiesinc.com/
[22:32:28] <ReadError> same place
[22:32:30] <ReadError> much better site
[22:32:41] <ssi> oh neat
[22:32:51] <ssi> that is in fact better :)
[22:33:01] <ReadError> We have changed its name to Automation Technology Inc
[22:33:01] <ReadError> Please Check Out Our New Website:
[22:33:01] <ReadError> www.automationtechnologiesinc.com
[22:33:06] <ReadError> ya
[22:33:11] <ReadError> much easier to navigate
[22:34:04] <ssi> much slower though
[22:34:04] <ssi> hahah
[22:34:06] -!- pfred1 [pfred1!~pfred1@unaffiliated/pfred1] has joined #linuxcnc
[22:34:08] <Thadius1> does anyone have any experience with "pycam"??
[22:34:30] <ssi> like, MUCH slower
[22:34:32] <ssi> painfully slow
[22:42:22] -!- stevegt_1 has quit [Read error: Connection reset by peer]
[22:46:25] -!- JT-Shop has quit [Read error: Connection reset by peer]
[22:47:01] -!- cra_ has quit [Quit: Page closed]
[22:56:36] -!- JT-Shop [JT-Shop!~John@184.20.140.167] has joined #linuxcnc
[22:58:49] -!- stevegt_1 [stevegt_1!~stevegt@64-130-244-2-gracehospitalitygroup.bwg.phx.sparkplugbb.net] has joined #linuxcnc
[23:02:31] -!- WalterN [WalterN!~walter@host-174-45-122-145.bzm-mt.client.bresnan.net] has joined #linuxcnc
[23:03:11] -!- asdfasd has quit [Ping timeout: 265 seconds]
[23:05:05] -!- fenn_ [fenn_!~fenn@131.252.130.248] has joined #linuxcnc
[23:05:25] -!- rob__H [rob__H!~rob_h@5ace70c9.bb.sky.com] has joined #linuxcnc
[23:05:29] -!- H264 has quit [Ping timeout: 244 seconds]
[23:10:20] -!- Tugge_ [Tugge_!~tugge@a91-153-45-7.elisa-laajakaista.fi] has joined #linuxcnc
[23:11:05] -!- kent_ [kent_!~kent@90-227-210-39-no122.tbcn.telia.com] has joined #linuxcnc
[23:11:06] -!- uwe__ [uwe__!~uwe_@dslb-088-066-186-052.pools.arcor-ip.net] has joined #linuxcnc
[23:14:19] -!- Connor1 [Connor1!~Connor@75.76.30.113] has joined #linuxcnc
[23:14:38] -!- mazafaka1 [mazafaka1!~ig@unaffiliated/mazafaka] has joined #linuxcnc
[23:16:49] -!- rob_h has quit [*.net *.split]
[23:16:49] -!- fenn has quit [*.net *.split]
[23:16:49] -!- Connor has quit [*.net *.split]
[23:16:50] -!- mazafaka has quit [*.net *.split]
[23:16:50] -!- uwe_ has quit [*.net *.split]
[23:16:50] -!- kent has quit [*.net *.split]
[23:16:50] -!- jtr has quit [*.net *.split]
[23:16:50] -!- Tugge has quit [*.net *.split]
[23:17:40] -!- jtr [jtr!~jtr@cpe-071-076-196-103.triad.res.rr.com] has joined #linuxcnc
[23:20:31] -!- kb8wmc has quit [Ping timeout: 264 seconds]
[23:23:45] -!- sumpfralle [sumpfralle!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[23:25:46] -!- JT-Shop has quit [Read error: Connection reset by peer]
[23:28:34] -!- sumpfralle1 [sumpfralle1!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[23:28:35] -!- sumpfralle has quit [Read error: Connection reset by peer]
[23:33:30] -!- JT-Shop [JT-Shop!~John@184.20.140.167] has joined #linuxcnc
[23:35:27] -!- kb8wmc [kb8wmc!~chatzilla@nat.mtp.cmsinter.net] has joined #linuxcnc
[23:35:30] -!- n2diy has quit [Quit: Ex-Chat]
[23:36:39] -!- sumpfralle1 has quit [Ping timeout: 252 seconds]
[23:36:41] <Thetawaves> ...they were right when they said IGBTs fail shorted
[23:37:18] <pfred1> do they go snap crackle and pop?
[23:37:19] -!- Himanshu [Himanshu!74caa09a@gateway/web/freenode/ip.116.202.160.154] has joined #linuxcnc
[23:37:36] <Thetawaves> mine didn't
[23:37:43] <Thetawaves> just switched 120v ac into my motor
[23:38:04] <pfred1> throw another shrimp on the barbie mate!
[23:39:16] <pfred1> it is times like those that fuses start looking a lot better
[23:39:40] <Thetawaves> yeah
[23:42:11] <Valen> fuses are good mmmkay
[23:43:37] <pfred1> this is making me dizzy http://www.youtube.com/watch?feature=endscreen&v=EdWWvnrFA9k&NR=1
[23:43:47] <Tecan> (EdWWvnrFA9k) "5 Axis Machining" by "gerrygprs" is "Howto" - Length: 0:02:18
[23:45:11] <Valen> turbines are always cool
[23:45:22] <Valen> I'd hate to think the price of the CAM program they use
[23:45:48] <pfred1> Valen you use blender don't you?
[23:46:03] <Valen> rhino and rhinocam
[23:46:12] <pfred1> oh who in here used blender?
[23:52:05] -!- sumpfralle [sumpfralle!~lars@31-16-21-63-dynip.superkabel.de] has joined #linuxcnc
[23:57:34] <Jymmm> me!
[23:57:46] <Jymmm> and mixer too