#emc | Logs for 2006-03-07

Back
[00:05:33] <giacus> rap is nice
[00:05:57] <giacus> but.. also latest Eros ramazzotti- Anastasia song is nice
[00:07:19] <fenn> i have a feeling rap wouldn't go over so well
[00:08:21] <giacus> I've some cd (demo) taked in the streets of venice beach very nice
[00:09:33] <giacus> povia won sanrmo festival yesterday http://www.sanremo.rai.it/news/articolonews/0,9217,121444,00.html
[00:09:39] <giacus> sanremo *
[00:09:42] <giacus> great song
[00:10:32] <giacus> very nice guy
[00:10:42] <giacus> and simple guy
[00:10:56] <giacus> talk about family in his songs
[00:11:23] <giacus> and importance of family unit
[00:13:12] <giacus> he looks like an 'open souce' guy
[00:13:40] <giacus> I'm happy for him
[00:15:02] <fenn> "sorry due to the olympic winter games video rights, this content is reserved to the users in the italian country"
[00:15:35] <fenn> but i have no clue who povia is or what sanremo is or why i should
[00:16:01] <giacus> sanremo is an itali competiotion of singer
[00:16:06] <giacus> singers
[00:16:14] <giacus> knowed all over the world
[00:16:42] <giacus> great prestige
[00:16:50] <fenn> meh
[00:16:56] <giacus> yeah ..
[00:17:10] <giacus> infact
[00:17:29] <giacus> it was in eurovision
[00:17:43] <giacus> dunno in us
[00:18:06] <giacus> but is a veri important competition for singers
[00:18:49] <giacus> for you should be more familiar Eros-Anastasia song
[00:18:55] <giacus> the latest
[00:25:18] <giacus> Eros Ramazzotti & Anastacia. I Belong To You
[00:31:06] <lerman_> lerman_ is now known as lerman
[01:07:40] <giacus> night
[04:29:53] <CIA-8> 03cradek * 10emc2/src/Makefile: so configs can be copied around and still work
[04:31:07] <CIA-8> 03cradek * 10emc2/src/ (configure configure.in): a rip install should be able to run custom configs
[04:31:42] <CIA-8> 03jmkasunich * 10emc2/configs/ (19 files in 9 dirs):
[04:31:42] <CIA-8> a new approach to handling the configs/common/* files. The files will exist in
[04:31:42] <CIA-8> common, and make will copy them to the other sample configs. This means no
[04:31:42] <CIA-8> relative paths in the ini files, and config directories (including the sample
[04:31:42] <CIA-8> ones) can be copied using ordinary system tools
[06:47:54] <SWPadnos> .help log
[06:49:29] <A-L-P-H-A> SWPadnos. yup. help. log. cos. sin. tan.
[06:49:39] <SWPadnos> atan
[06:49:44] <A-L-P-H-A> damn. hungry now.
[06:49:52] <A-L-P-H-A> acos
[06:49:54] <A-L-P-H-A> asin
[06:50:00] <A-L-P-H-A> :)
[06:50:03] <SWPadnos> stupid chatzilla is logging poorly. it doesn't seem to corerctly substitute dates in the log file name
[06:50:08] <A-L-P-H-A> cos^2(x)
[06:50:21] <A-L-P-H-A> don't use chatzilla.
[06:50:22] <A-L-P-H-A> I used mirc.
[06:50:26] <A-L-P-H-A> windows machine.
[06:50:36] <SWPadnos> same here (Windows)
[06:51:19] <A-L-P-H-A> it's nearly 2am.
[06:51:34] <SWPadnos> there was supposedly a bugfix a couple of days ago that made it switch log files when the date changes (basically making it a nice logger)
[06:51:36] <SWPadnos> yes it is
[06:52:03] <A-L-P-H-A> brb... going to lookin the fridge for food.
[06:52:38] <SWPadnos> see ya. I'm priobably headed for bed
[06:52:45] <A-L-P-H-A> wimp. :)
[06:52:52] <A-L-P-H-A> lets play a game of chess or something instead.
[06:52:57] <SWPadnos> ok.
[06:53:00] <SWPadnos> I concede
[06:53:04] <A-L-P-H-A> :(
[06:53:06] <SWPadnos> night
[06:53:09] <SWPadnos> :)
[06:53:11] <A-L-P-H-A> later.
[07:02:19] <Jymmm> A-L-P-H-A funny videos the other night, well kinda stupid for RL =)
[07:02:56] <A-L-P-H-A> ? what video ?
[07:03:03] <A-L-P-H-A> shit I'm cold... brb.
[07:03:05] <Jymmm> fight club
[07:04:41] <A-L-P-H-A> oh. that. :) hehe
[07:04:42] <A-L-P-H-A> yeah.
[07:04:48] <A-L-P-H-A> I thought it was funny, so I shared.
[07:06:15] <Jymmm> heh
[07:46:43] <bill20r3> Zzzzz
[08:00:27] <alex_joni> g'night bill20r3
[08:15:10] <A-L-P-H-A> hey aj
[08:29:23] <alex_joni> hey A-L-P-H-A
[08:29:26] <A-L-P-H-A> sup?
[08:35:03] <alex_joni> work :)
[08:37:07] <A-L-P-H-A> fun
[08:37:10] <A-L-P-H-A> my school's on strike.
[08:37:15] <A-L-P-H-A> I thinkI don't get to go to school now
[08:39:20] <alex_joni> heh.. nice
[08:39:31] <A-L-P-H-A> not really.
[08:39:34] <A-L-P-H-A> delays me finishing school.
[08:40:08] <alex_joni> oh.. that's bad
[08:41:08] <A-L-P-H-A> yeah.
[10:09:32] <Bo-Dick> * Bo-Dick has a question to you guys who have built your own stepper driver or know how they work...
[10:10:19] <Bo-Dick> Can an adjustable current regulator be built with a single transistor and a potentiometer plus additional diodes/resistors?
[10:38:42] <A-L-P-H-A> Bo-Dick, no clue. i'm thinking you could do a voltage regulator with a diode and a transitor... but current... V = IR. I = V/R so I think you can.
[10:38:54] <A-L-P-H-A> but it wouldn't be smooth.
[10:49:55] <Bo-Dick> There are some information on the "jones on stepper motors" site but no clue how to make it adjustable.
[10:54:19] <A-L-P-H-A> best to check on yahoo groups. cad_cam_edm_dro
[12:09:19] <lerman_> lerman_ is now known as lerman
[13:17:20] <jepler> Simple single-transistor circuits. (1) Voltage regulator outputs 6-volts DC. (2) Constant-current output. (3) Simple audio amplifier. (4) Impedance-matching circuit. (5) LED-control circuits turn either off or on in the presence of a DC voltages from 5- to 9-volts. (6) Relay-driver circuit has a latching feature that keeps the relay on once energized.
[13:17:25] <jepler> POPULAR ELECTRONICS [2] Dec 1995 (v.12#12) pg. 77
[13:17:25] <jepler> )
[13:17:28] <jepler> (no, I don't have a copy3$
[13:17:34] <jepler> http://www.nleindex.com/index.php?pID=HTDI&sID=BrowseIndex&tID=E/2590
[13:22:07] <jepler> but why do you say "single transistor"? You can use an adjustable regulator (which will come in the same sorts of packages) to get a current regulator.
[13:22:27] <jepler> http://emergent.unpy.net/projects/01111254927
[14:08:14] <Bo-Dick> Thanks jepler
[14:08:57] <giacus> hello
[14:09:03] <giacus> robin_sz: around ?
[14:11:22] <Bo-Dick> Its a shame the LM317:s only go up to 37 volts :(
[14:11:54] <giacus> hehe Bo-Dick
[14:11:59] <giacus> its hard :)
[14:12:09] <giacus> good luck ;P
[14:14:30] <Bo-Dick> What I want to do is a current regulator where the current can be regulated between at least 0-3 amps and with a voltage of at least 60 volts. Is that hard to do?
[14:15:20] <giacus> probably not
[14:16:05] <giacus> there is some projects online that may could help
[14:16:11] <giacus> for that section
[14:17:13] <Bo-Dick> It would also be nice if the maximum voltage could be adjusted as well
[14:17:54] <giacus> doesnt it depend on PS ?
[14:18:24] <Bo-Dick> Well if the power supply supports that or not.
[14:18:38] <giacus> oh.. that is simple to get I think
[14:18:54] <giacus> the issue could be tha range of work
[14:19:21] <giacus> for the final part of the circuit
[14:19:38] <Bo-Dick> There are roomers that a stepper driver drives the motors more powerful if the current is constant rather than if the voltage is constant. I have no idea why but thats what I've heard.
[14:20:10] <giacus> btw, usually the voltage isnt an issue
[14:20:59] <giacus> high voltage could be better then low
[14:21:42] <Bo-Dick> If the current is regulated the voltage will depend on the situation and adapt after the circumstances in order to keep the current constant right?
[14:21:44] <giacus> in the stepper, for what I know, voltage is relative
[14:22:38] <giacus> Bo-Dick: I think so ..
[14:23:23] <Bo-Dick> And if the voltage would have been regulated the current would adapt after the circumstances. For some unknown reason this is less effective.
[14:23:48] <giacus> steppers are drived in current not voltage
[14:24:28] <skunkworks> the higher the voltage - the faster it can stear the field in the coils
[14:24:58] <giacus> high voltage = better performance
[14:25:09] <skunkworks> for higher speeds
[14:26:03] <Bo-Dick> Is it the total power in the coils that determines the torque i.e. the product of current and voltage over the coils?
[14:28:52] <skunkworks> to some point - you almost have to look at steppers rpm/torque graphs to get an idea of what you will get in performance.
[14:37:13] <jepler> Bo-Dick: The reason that high-voltage supplies are used together with current regulators is this: The coils in steppers are inductors. Current rises slowly over an inductor with a time constant L/R (inductance over resistance). Current is related to Power is related to Torque.
[14:37:34] <jepler> Bo-Dick: take a look at the top graph on http://www.tpub.com/neets/book2/2d.htm
[14:40:38] <jepler> Bo-Dick: If you double the supply voltage, you double the maximum current, so that you'd actually reach the desired current before 1L/R, instead of only being 95% of the way there after 3L/R. At the point that the desired current is reached, a chopper-style drive turns off its transistors so that you don't violate the power dissipation limits of the motor. This gives a current graph with a fast rise and then a sawtooth appearance.
[14:41:28] <skunkworks> very nice jepler
[14:45:58] <alex_joni> hi guys
[14:46:11] <skunkworks> Hi alex
[14:46:29] <alex_joni> hi samco
[14:47:04] <skunkworks> have you ever hooked two wireless routers togather wirelessly?
[14:47:33] <skunkworks> so in effect one would be an access point?
[14:47:50] <alex_joni> you mean for crossing seamlessly between areas?
[14:47:55] <skunkworks> right
[14:48:02] <skunkworks> between the house and the shop
[14:48:09] <jepler> skunkworks: I tried once but I couldn't get it to work. I solved the problem by giving away the old router to a friend.
[14:48:38] <skunkworks> It may not be possible - I might have to take one of them back and get an actual access point.
[14:48:58] <skunkworks> d-link wbr-1310 $39
[14:49:30] <skunkworks> but I have some emails out - and maybe there is a hack or patch. I see nothing in the firmware.
[14:50:14] <Bo-Dick> I've tried a variable current regulation method called "current mirror".
[14:52:41] <Bo-Dick> It's described here: http://www.allaboutcircuits.com/vol_6/chpt_5/14.html
[14:53:53] <Bo-Dick> Can someone explain why the current mirror sucks?
[14:55:42] <Bo-Dick> The current level keeps changing all the time. That sucks.
[14:55:47] <jepler> well, it's linear, while most people use switching designs to drive steppers.
[14:56:44] <Bo-Dick> Ok so this is why it keeps changing. I thought there existed linear current regulators that works somewhat well. Am I wrong?
[14:57:12] <skunkworks> they work well - just waste power as heat.
[14:57:22] <jepler> http://www.piclist.com/images/member/RB-ezy-Q33/jelly01.jpg
[14:57:34] <jepler> how much heat? this photo might give you an idea.
[14:58:01] <skunkworks> ;)
[14:58:32] <Bo-Dick> I dont care about crappy efficiency as long as the complexity isnt extreme.
[14:59:18] <alex_joni> jepler: LOL
[14:59:44] <Bo-Dick> I'm gonna fan-cool my heatsinks too.
[15:00:55] <Bo-Dick> But seriously that heatsink was huuge!
[15:01:39] <alex_joni> Bo-Dick: water-cooling ;)
[15:02:50] <Bo-Dick> Besides the efficiency aspect does the PWM method give more performance in the motors than linear regs?
[15:03:33] <jepler> consider driving the xylotex stepper with a linear setup and commonly available ~6x supply voltage (12V). (http://www.xylotex.com/StepperMotor.htm, 2.8A, 1.13 ohm, nominal voltage 2V)
[15:04:18] <jepler> You have 2.8A * 3 through your linear regulator, and it has to drop 10V, so that's 84W of heat.
[15:04:35] <skunkworks> per axis
[15:04:57] <jepler> anyway, a lot of heat
[15:05:06] <jepler> skunkworks: My *3 was supposed to be the "per axis" part of the calculation
[15:05:23] <skunkworks> sorry - didn't see that.
[15:05:40] <Bo-Dick> Does it only drop 10 volts!?
[15:07:46] <Bo-Dick> Well if the supply voltage is only 12 volts then of course....
[15:07:56] <jepler> oops, nominal voltage is 2.8A * 1.13 ohm ~= 3V, not 2V. The drop in the linear regulator is (supply voltage)-(nominal voltage).
[15:08:18] <Bo-Dick> I've heard people running their motors at 25 times the rated voltage and you only 6 times.
[15:08:37] <Bo-Dick> That would mean a lot bigger voltage drop wouldn't it?
[15:08:38] <alex_joni> Bo-Dick: no way you can run linear regulators at 25 times the voltage
[15:08:40] <alex_joni> that's insane
[15:09:10] <jepler> 24 * 3.164 * 2.8 * 3 == 640W of power dissipated by the regulator
[15:09:33] <alex_joni> you can drive HUGE motors with that power
[15:09:41] <Bo-Dick> Dont need a radiator in the room then :)
[15:09:42] <cradek> electricity is only what - 30c/kWh ?
[15:09:51] <alex_joni> and you still need to supply that voltage/power
[15:10:09] <alex_joni> so your PSU is ~1kW to drive 50W motors?
[15:10:33] <alex_joni> 10W .. sorry ;)
[15:10:34] <jepler> with a switching design you can use 25x voltage .. with a linear design, forget it.
[15:11:07] <alex_joni> cradek: you could use steam ..
[15:11:21] <alex_joni> and drive it mechanically
[15:11:22] <Bo-Dick> Has anyone of you guys installed a PWM current regulating system?
[15:11:24] <alex_joni> might be cheaper
[15:11:34] <alex_joni> Bo-Dick: I installed a few ;)
[15:11:38] <alex_joni> from geckodrive.com
[15:11:52] <alex_joni> you only need to connect wires
[15:12:45] <Bo-Dick> Theyre pretty small too.
[15:13:00] <Bo-Dick> But yet so powerful!
[15:13:27] <jepler> Bo-Dick: If you can make single-sided circuit boards, http://pminmo.com/l297-8/l297-8.htm is a good choice. I think it's under $25 in parts per axis, supply voltage up to 44V or so.
[15:14:36] <Bo-Dick> It looks like he's driving a floppy stepper motor in the photo.
[15:15:46] <jepler> Bo-Dick: cradek uses the design to drive some ~160 oz-in steppers
[15:17:34] <jepler> http://timeguy.com/cradek/cnc
[15:17:53] <jepler> cradek: you need to put a photo of your military surplus power supply on there :-P
[15:18:35] <cradek> jepler: yeah, it's pretty cool
[15:19:35] <Bo-Dick> The thing is that I've already implemented the polarity shifting bridges. I was suggested that design from a guy on ON-semiconductor. After I've built it I've been told that the design is crap compared to mosfet switches. Is this true? http://www.carmi.se/misterstarshine/Projects/polbridge.gif
[15:21:50] <Bo-Dick> * Bo-Dick wanna know the truth even if it hurts
[15:22:12] <jepler> I only know what's worked for me
[15:22:17] <jepler> other designs might be fine too
[15:22:41] <Bo-Dick> jepler: Have you the same design in your bridges?
[15:23:21] <jepler> Bo-Dick: I've only used integrated bridges, like the L298.
[15:23:30] <alex_joni> anyone fluent with wget?
[15:23:48] <jepler> alex_joni: I can tell you the same things you'll read in 'wget -help'
[15:23:57] <alex_joni> jepler: guess so ;)
[15:24:04] <alex_joni> * alex_joni needs to take a whole site
[15:24:08] <alex_joni> guess -r does that
[15:24:23] <jepler> -m, --mirror shortcut option equivalent to -r -N -l inf -nr.
[15:24:31] <jepler> I'd have to guesss that they chose some sensible defaults for --mirror ..
[15:24:32] <Bo-Dick> Hey the "HV" driver is tiny. What's the ratings for this little creature?
[15:25:04] <Bo-Dick> The thing takes less space than my polarity bridges alone!!
[15:26:16] <jepler> I think that the L298 is maximum 44V supply, 4A current with adequate heatsinking.
[15:27:13] <Bo-Dick> Like if it wasn't enough that my design is less efficient. It's bulky too :'(
[15:29:40] <Bo-Dick> Is PWM "built in" in the L298 bridge?
[15:33:03] <alex_joni> Bo-Dick: a lot if built in the L298, check the datasheet
[15:33:49] <alex_joni> s/if/is/
[15:34:37] <skunkworks> that is one of the nice things about pwm - It runs the componants pretty much on and off. there is no some on like in a linear supply. So there is very little heat disapation.
[15:36:05] <skunkworks> Plus the motors sing more. I like singing.
[15:36:10] <skunkworks> ;)
[15:36:17] <Bo-Dick> Whats the frequency of the PWM pulses?
[15:36:17] <alex_joni> get AC motors then
[15:36:25] <alex_joni> Bo-Dick: depends on the cost of the drive
[15:36:30] <alex_joni> good ones are above 20kHz
[15:37:07] <Bo-Dick> So if my BJT darlington bridges can handle that I dont have to be in dispare?
[15:37:55] <cradek> Bo-Dick: if you have H bridges built, and something to drive them, why not use them to get some motors moving, and worry about it later if you're not happy?
[15:38:51] <Bo-Dick> Well the motors are so weak I can stop them with my fingers when theyre moving.
[15:39:21] <cradek> even at low speed?
[15:40:12] <Bo-Dick> Yeah! Well my motors are from small 5.25" floppy drives and the coils are 75 ohms each. I guess that tells everything :(
[15:40:22] <cradek> hmm.
[15:41:40] <Bo-Dick> You know what?! The block diagram in the datasheet for the L298 tells that they are composed of four NPN transistors for each bridge!!?!
[15:43:34] <Bo-Dick> :S
[15:45:05] <Bo-Dick> I've neven been able to get a h-bridge to work with NPN transistors only! Well as you can see in http://www.carmi.se/misterstarshine/Projects/polbridge.gif
[15:45:29] <jepler> maybe the datasheet is wrong :-P
[15:46:10] <jepler> it doesn't make sense to me that they'd all be NPN
[15:46:42] <Bo-Dick> cradek: Does the "Zoli's board" support microstepping?
[15:47:23] <jepler> I agree the datasheet shows 4 identical transistors per bridge, though.
[15:47:23] <jepler> l297+l298 supports half and full stepping
[15:47:50] <Bo-Dick> Well I've heard the motors travels smoother with microstepping.
[15:48:04] <jepler> I've never used microstepping yet
[15:49:03] <Bo-Dick> * Bo-Dick wonders how halfstepping can be performed with a h-bridge
[15:49:35] <jepler> there are 2 bridges per stepper
[15:50:11] <Bo-Dick> i know...
[15:51:03] <jepler> this application note shows the waveforms on the l298's outputs: http://us.st.com/stonline/books/pdf/docs/1734.pdf
[15:51:06] <jepler> er, L297's outputs
[15:52:12] <jepler> you can do full, half, or microstepping with h-bridges
[15:56:38] <Bo-Dick> Yeah but the half step doesn't energize the coils at 100% as in full steps. One should loose some torque here right?
[16:00:44] <jepler> I think there's a clever way to get torque compensation with l297+l298 but I don't remember it. The pminmo board doesn't have it afair.
[16:07:01] <Bo-Dick> One problem I've been having for quite a while is that how I can support the bases on the h-bridge transistors in order to make them switch fully. Assume that the supply voltage is 60 volts and the signal generated by the logics is 5v ttl. How can I boost up that voltage to drive the h-bridge? http://www.carmi.se/misterstarshine/Projects/polbridge.gif
[16:10:05] <jepler> figure 18 shows the circuitry to add to an l297+l298-based driver to get "half-step shaping": http://www.st.com/stonline/products/literature/an/1679.pdf
[16:10:17] <jepler> I've never built a discrete H-bridge, so I don't think I'm qualified to answer that question.
[16:11:45] <jepler> but I think that's why a lot of the very simple stepper driver boards are unipolar .. it's much eaiser to do
[16:12:31] <Bo-Dick> I quickly abandoned that unipolar concept. Not enough torque I'm afraid.
[16:16:44] <Bo-Dick> Since my motors are so small and weak I thought to get the absolute maximum from them. But I realize I wont prolly get near what I need.
[16:17:28] <Bo-Dick> The only mode where all coils work at an absolutely 100% all the time is the bipolar full step mode. Am I wrong?
[16:18:34] <Bo-Dick> This would then be the mode that uses the motors power to an absolute maximum. Besides that I'm gonna connect the coils in parallel. Its said this is more effective then in series.
[16:32:14] <CIA-8> 03cradek * 10emc2/src/emc/kinematics/tp.c: rotary axis support in tp, but accel handling in task is still wrong
[16:44:05] <cncuser> hi :)
[16:44:22] <cncuser> 2.4.26 is the magic number :)
[16:44:37] <Bo-Dick> Maybe you know about this SWPadnos? Assume I've fed my h-bridge with 60 volts and I wanna turn on and off the trannys with a 5v ttl signal. How can I boost up that ttl so that it can switch the bridge effectively? http://www.carmi.se/misterstarshine/Projects/polbridge.gif
[16:45:10] <SWPadnos> you need a "high side driver", which I have no idea how to build ;)
[16:45:29] <SWPadnos> there are ICs that will do this, but I'm not sure if they're rated for 60V
[16:46:06] <skunkworks> use higher voltage with oc ttl logic.
[16:46:28] <Bo-Dick> What is "oc ttl" logic?
[16:46:38] <skunkworks> open collector
[16:46:42] <SWPadnos> open collector TTL
[16:47:00] <Bo-Dick> Does it go as high as 60 volts?
[16:47:07] <skunkworks> no
[16:47:56] <Bo-Dick> So the question remains then...
[16:48:37] <cncuser> i do a new compile for puppysit
[16:48:38] <SWPadnos> yep. that's the big question when designing a high voltage bipolar driver
[16:48:44] <cncuser> is it good to check ozut testing now ?
[16:49:15] <SWPadnos> I think HEAD has the new TP, I'm not sure what's different with TESTING right now
[16:49:37] <SWPadnos> (not sure - there was discussion of putting the new TP in HEAD)
[16:49:42] <cncuser> hmmm
[16:50:03] <skunkworks> it is there (head) I played with it yesterday - Very Nice.
[16:50:05] <cncuser> else i recompile a 2 week old version
[16:50:25] <cncuser> skunkworks: so i should checkout: cvs -z3 -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/emc co emc2 ?
[16:50:40] <SWPadnos> that should work
[16:50:57] <skunkworks> ha ha ha. <- sorry not a linux person
[16:50:59] <Bo-Dick> Is it even possible to convert a 5v ttl signal to a 60v signal?
[16:51:00] <SWPadnos> give or take a cvs login
[16:51:35] <SWPadnos> I'm not sure how that part is done. That's whyu I bought Geckos
[16:52:13] <skunkworks> ;) I stopped trying to make my own sence ebay
[16:52:24] <SWPadnos> note that most of the low cost stepper drivers are limited to 30-40V supply voltage, which is the limitation of the various one-chip drivers
[16:52:32] <SWPadnos> (or lower)
[16:52:52] <cncuser> cvs [login aborted]: end of file from server (consult above messages if any)
[16:52:54] <cncuser> damit
[16:52:58] <cncuser> why why hwy
[16:53:11] <SWPadnos> you can try getting the .tar.gz from alex's site
[16:53:12] <skunkworks> try it again - usually 10 tries and it works ;)
[16:53:18] <cncuser> is this sf cvs so fucked up everytime i want to checkout something fast
[16:53:53] <jepler> cncuser: It's "****ed" up a lot of the time, seems like
[16:54:01] <SWPadnos> sign up as a sf user, and you can get non-anonymous access, which tends to work better
[16:54:15] <cncuser> SWPadnos: never, thats what they want
[16:54:23] <SWPadnos> well, then live with it ;)
[16:54:26] <jepler> SWPadnos: does any sf user get non-anonymous access, or do they have to be a developer?
[16:54:34] <SWPadnos> I'm not sure
[16:54:38] <jepler> I thought it was the latter.
[16:54:46] <SWPadnos> that would make sense
[16:55:14] <SWPadnos> unless they've given default read access to everyone, and only write access to developers
[16:56:05] <jepler> cncuser: maybe if you click on a few banner ads from your ip address they'll unblock you from cvs access
[16:56:08] <jepler> (j/k)
[16:56:14] <cncuser> harhar
[17:25:58] <cradek> setAccel(sqrt(0.23 + NaN + NaN));
[17:26:10] <cncuser> has anybody but may experienced the looping of Depending.... when tunning make with emc2 ?
[17:26:36] <alex_joni> cncuser: say what?
[17:26:58] <alex_joni> cradek: that's about -0.237
[17:27:05] <cncuser> alex_joni: hi :)
[17:27:06] <cncuser> yes
[17:27:09] <cncuser> the same thing
[17:27:19] <cradek> cncuser: try make clean?
[17:27:22] <cncuser> i know it must be some tiny tiny bug
[17:27:28] <cncuser> its a fresh checkout
[17:27:39] <cncuser> make clean also loops ;)
[17:28:00] <cncuser> it mayb do to emc2 using some thing it didnt use 2 months ago or so
[17:28:12] <cncuser> somthing my busibox doesnt emulate good enough ?
[17:28:31] <cncuser> well, i fixed my major problem, and now i cant get emc compiled
[17:28:34] <cradek> the makefiles are completely rewritten
[17:28:36] <SWPadnos> did you do a clean checkout, or an update?
[17:28:38] <cncuser> allways the same
[17:28:49] <cncuser> started from nothin, 1 hour ago
[17:29:08] <cncuser> ok
[17:29:10] <cncuser> hmmm
[17:29:14] <cncuser> got to think
[17:29:32] <cncuser> im up for almost 30 hours now
[17:29:37] <alex_joni> cncuser: ouch
[17:29:41] <cncuser> compiled 10 kernels :)
[17:29:54] <alex_joni> cncuser: what emc2 are you using? latest HEAD?
[17:29:57] <bill2or3> * bill2or3 got his rails.
[17:29:58] <SWPadnos> that should only take 2.4 hours ;)
[17:29:58] <bill2or3> yay
[17:30:07] <cncuser> adeos intorduces a bug into 2.4.26 that keeps omn going to 2.4.32 wich leeds to my oopses
[17:30:11] <cncuser> aeh, 2.4.27
[17:30:18] <cncuser> 2,4,26 works
[17:30:31] <alex_joni> * alex_joni translates:
[17:30:38] <alex_joni> 2.4.27 - BAD, 2.4.26 - GOOD
[17:30:41] <cncuser> SWPadnos: well, i also had to reboot ;)
[17:30:49] <cncuser> yes :)
[17:31:09] <cncuser> now, how can i debug the make stuff with emc2 ?
[17:31:20] <alex_joni> did you complete ./configure ok ?
[17:31:22] <cncuser> a -x to the configurescript didnt help much :)
[17:31:25] <cncuser> yes
[17:31:39] <SWPadnos> heh
[17:32:07] <alex_joni> can you pastebin the Makefile.inc ?
[17:32:07] <cncuser> checking for kernel version string... 2.4.26-adeos
[17:32:07] <cncuser> checking for kernel module extension... .o
[17:32:07] <cncuser> checking kernel build system... normal
[17:32:07] <cncuser> checking for module installation directory... /usr/realtime/modules
[17:32:11] <cncuser> looks all right sofar
[17:32:19] <cncuser> hmm, where pastebin again ?
[17:32:28] <alex_joni> pastebin.com ?
[17:32:43] <jepler> cncuser: there's always config.log to look at after ./configure is done
[17:33:29] <cncuser> jepler: ahhh
[17:33:31] <cncuser> log
[17:33:31] <cncuser> ;)
[17:33:35] <alex_joni> jepler: it might not be configure related ;) .. it might be even the buildsys :D
[17:33:40] <cncuser> never seen that extension before :)
[17:34:16] <jepler> * jepler scrolls back a bit
[17:34:21] <cncuser> http://pastebin.com/589194
[17:34:29] <jepler> yeah, looking at configure is probably not the right thing
[17:34:42] <jepler> make sure the system clock is set right, and that no files are in the future
[17:34:49] <cncuser> hmmm
[17:35:21] <jepler> but when make sees files in the future, it prints "clock skew detected" and doesn't loop. at least this version of make (3.80)
[17:35:48] <cncuser> also it detects 2.4.26 and then writes 2.4.31 into some files...
[17:35:54] <cncuser> 2.4.31 was the old kernel
[17:36:08] <cncuser> uname -a saye 2.4.26, it should stick to that
[17:36:10] <jepler> I think the emc2 build may require that you have the kernel booted that you want to build for
[17:36:23] <jepler> I'm not sure where 2.4.31 might be coming from
[17:36:26] <cncuser> yes, it was allways that way ?... atrleast with rtai...
[17:36:30] <jepler> but that's probably not a cause of make looping.
[17:36:40] <cncuser> no, itas another thing
[17:38:02] <cncuser> i have no clou
[17:38:04] <cncuser> this loop
[17:38:07] <cncuser> hmmm
[17:38:36] <jepler> cncuser: if you type "make zzz" does it loop too?
[17:38:45] <cncuser> alex_joni: its head btw
[17:38:53] <cncuser> yes
[17:39:00] <cncuser> make zzz loops
[17:39:13] <cncuser> make anything loops
[17:39:47] <alex_joni> does it loop on dependencies?
[17:39:51] <alex_joni> or what does it output?
[17:39:52] <cncuser> yes
[17:39:58] <cncuser> Depending hal/utils/scope_rt.c
[17:39:58] <cncuser> Depending hal/hal_lib.c
[17:39:58] <cncuser> Depending hal/drivers/hal_vti.c
[17:39:58] <cncuser> Depending hal/drivers/hal_tiro.c
[17:39:58] <cncuser> Depending hal/drivers/hal_stg.c
[17:39:59] <cncuser> Depending hal/drivers/hal_skeleton.c
[17:40:00] <alex_joni> paste a few iterations ;)
[17:40:01] <cncuser> Depending hal/drivers/hal_ppmc.c
[17:40:03] <cncuser> Depending hal/drivers/hal_parport.c
[17:40:05] <cncuser> Depending hal/drivers/hal_motenc.c
[17:40:07] <cncuser> Depending hal/drivers/hal_m5i20.c
[17:40:09] <cncuser> make: *** [depends/rthal/drivers/hal_m5i20.d] Interrupt
[17:40:11] <cncuser> the interrupt was me
[17:40:13] <cncuser> harhar
[17:40:47] <alex_joni> don't see a loop in there..
[17:40:54] <cncuser> hehe
[17:41:10] <cncuser> but it loops
[17:41:26] <cncuser> i blame yetis
[17:41:51] <cncuser> damn yetis, allways fxxing around with my sourcedirs
[17:42:01] <alex_joni> lol
[17:42:11] <alex_joni> can you pastebin a full iteration?
[17:42:11] <cncuser> big feat, but good in hiding them yetis
[17:42:15] <jepler> cncuser: next to last line of Makefile, comment out the line "include $(RTDEPS)"
[17:42:18] <cncuser> hehe
[17:43:43] <jepler> cncuser: does it still loop when you comment that line out?
[17:44:00] <cncuser> # make
[17:44:00] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/demo_mazak
[17:44:00] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/common/core_stepper.hal ../configs/demo_step_cl
[17:44:00] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/common/core_servo.hal ../configs/m5i20
[17:44:00] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/common/core_servo.hal ../configs/motenc
[17:44:00] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/common/core_sim.hal ../configs/sim
[17:44:02] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/common/core_stepper.hal ../configs/stepper
[17:44:04] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/common/core_servo.hal ../configs/stg
[17:44:06] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/univpwm
[17:44:08] <cncuser> install -m 0644 ../configs/common/emc.nml ../configs/univstep
[17:44:10] <cncuser> thats what it does
[17:44:14] <alex_joni> that's ok
[17:44:22] <jepler> does it finish ?
[17:44:26] <cncuser> no thats all
[17:44:29] <alex_joni> it's stuff from last night
[17:44:33] <cncuser> its a little..hmm, little
[17:44:48] <cncuser> o k
[17:44:52] <cncuser> well
[17:45:07] <jepler> is there a ../bin/halscope ?
[17:45:18] <jepler> built freshly
[17:45:22] <cncuser> i cant say how much time this may take... but i dont intend to carry on more then 2 or 3 hours
[17:45:46] <cncuser> yes
[17:45:49] <cncuser> ok
[17:45:51] <cncuser> so its built
[17:45:53] <cncuser> ok
[17:45:54] <cncuser> hmmm
[17:46:09] <cncuser> ok
[17:46:32] <cncuser> so it builds all
[17:46:44] <cncuser> cant i scripot the rest ?
[17:46:56] <jepler> well there's still a problem.
[17:47:06] <jepler> commenting out that include means the dependency information is wrong
[17:47:26] <jepler> is there a file depends/rthal/drivers/hal_parport.d ?
[17:48:34] <cncuser> depends/rtapi/rtai_ulapi.d
[17:48:37] <cncuser> thats all
[17:48:42] <jepler> that's the only file in depends/ ?
[17:48:57] <cncuser> a sorry
[17:48:58] <cncuser> hal
[17:49:27] <cncuser> depends/hal/drivers/hal_skeleton.d
[17:49:27] <cncuser> no
[17:49:29] <cncuser> no parport
[17:49:37] <jepler> there's no depends/rthal directory?
[17:49:37] <SWPadnos> I think that one time cradek suggested deleting the depends dir tree, has that been tried? (and might it be useful?)
[17:49:41] <cncuser> find . | grep parport
[17:49:41] <cncuser> ./hal/drivers/hal_parport.c
[17:50:00] <cncuser> hmm ?
[17:50:02] <cncuser> depends ?
[17:50:15] <SWPadnos> don't do that yet - wait for a more expert answer ;)
[17:50:39] <cncuser> allready did ;)
[17:50:42] <cncuser> but..hmm
[17:50:49] <cncuser> well, didnt change for th better
[17:50:53] <SWPadnos> ok then
[17:51:32] <cncuser> jo, masters of desaster :)
[17:52:14] <jepler> cncuser: I don't understand what's wrong, but by using a 2.4 kernel you're using a part of the Makefile that I don't.
[17:52:28] <jepler> starting around 424, 'ifeq ($(BUILD_SYS),normal)' ... 'endif'
[17:52:47] <cncuser> hmmm
[17:52:50] <jepler> I tested it once soon after changing build systems, on a bdi-live system, but either I've broken it since then or it never really worked
[17:53:42] <jepler> for each file dir/name.c that is part of a kernel module, it should be creating depends/rtdir/name.d and objects/rtdir/name.o
[17:53:51] <cncuser> im stuck with 2.4.26 :) till i feel funny enough to look whats the difference of 2.4.26 and 2.4.27 and the adeospatches and the squashfspatches is :)
[17:54:03] <jepler> if it worked with 2.4.27 I'm even more confused.
[17:54:36] <cncuser> dont know
[17:54:40] <cncuser> it worked with 2.4.31 :)
[17:54:51] <jepler> but that was with the old makefiles?
[17:55:03] <cncuser> i didnt compile emc/rtai on the kernels for i was hunting a systemreboot bug :)
[17:55:08] <cncuser> hmmm
[17:55:47] <CIA-8> 03jepler * 10emc2/src/Makefile: another case of objs/ instead of objects/
[17:56:07] <jepler> I don't think it'll cure your current problem, but this patch (which I checked in above) fixes a bug in the makefile that would affect you.
[17:56:10] <jepler> http://emergent.unpy.net/index.cgi-files/sandbox/makefile.patch
[17:57:40] <jepler> :q
[17:57:41] <jepler> oops
[17:58:03] <cncuser> hmmm :((((
[17:58:26] <cncuser> where was that nightly tarball again ?
[17:58:37] <cncuser> and should i really use head ?
[17:59:25] <alex_joni> cncuser: get the testing tarball from my site
[17:59:31] <alex_joni> http://solaris.cs.utt.ro/~emc/
[17:59:48] <alex_joni> but that won't have jepler's fix
[17:59:57] <cncuser> yes
[18:00:02] <cncuser> one at a time
[18:00:11] <alex_joni> http://solaris.cs.utt.ro/~emc/emc2-testing-2006-03-02.tar.gz
[18:01:13] <cncuser> tar: emc2-testing/include/CVS/Repository: time stamp 2006-02-20 06:56:02 is 1079911 s in the future
[18:01:20] <cncuser> # date
[18:01:20] <cncuser> Tue Feb 7 18:57:40 PUP 2006
[18:01:32] <cncuser> i dont get it
[18:01:42] <SWPadnos> it's March 7
[18:01:45] <SWPadnos> not Feb
[18:01:49] <cncuser> args
[18:03:00] <alex_joni> that might even cure your loop
[18:03:09] <cncuser> yes
[18:03:25] <cncuser> damit, make didnt say nothing about it.
[18:05:50] <cncuser> hmm, seems to work
[18:05:52] <cncuser> :)
[18:05:54] <cncuser> omg
[18:06:18] <SWPadnos> funny that there was a date problem (and that I actually know what day it is)
[18:06:45] <cncuser> feb mar : the sest was right :) even the Tue matched :)
[18:06:59] <SWPadnos> yeah- that darned 4.000 week month ;)
[18:07:02] <cncuser> hmmm
[18:07:04] <cncuser> ok
[18:07:14] <cncuser> whats that patch about jeppler ?
[18:07:19] <cncuser> i didnt aply but it worked
[18:07:31] <cncuser> hmm, i may have the one thing commentet still
[18:07:49] <jepler> cncuser: Without that patch, a kernel module might not get rebuild after changing a header file.
[18:07:58] <alex_joni> cncuser: you should only worry if changing stuff
[18:08:01] <jepler> cncuser: same for commenting out that line
[18:08:08] <cncuser> ok
[18:08:14] <cncuser> so this build should work :)
[18:09:33] <cncuser> install -m 0644 -o root /usr/local/etc/emc2/sample-configs/ppmc
[18:09:33] <cncuser> install: too few arguments
[18:09:33] <cncuser> Try `install --help' for more information.
[18:09:33] <cncuser> make: *** [install] Error 1
[18:09:33] <cncuser> #
[18:09:37] <cncuser> after a make install
[18:12:29] <bill2or3> can you mix steppers/servos on one machine?
[18:12:33] <SWPadnos> yes
[18:13:11] <bill2or3> cool.
[18:13:17] <cncuser> damn :(
[18:13:41] <cncuser> i had a old emc2 install there installed so it maybe mixed upt
[18:13:51] <SWPadnos> at least with emc2. I'm not sure how it would work for emc1, unless you use step-to-servo drives (like Geckos)
[18:14:14] <bill2or3> it's a theoretical question anyway. :-)
[18:14:26] <bill2or3> I got those thomson rails, they look pretty decent.
[18:14:38] <bill2or3> certainly better than I could homebrew.
[18:15:09] <Jymmm> bill20r3 any slides yet?
[18:15:51] <SWPadnos> cncuser, not sure if you noticed, but there should have been a -d on that install line (I think)
[18:15:55] <bill2or3> not yet.
[18:16:05] <alex_joni> bill2or3: you can mix on emc2, not on emc1
[18:16:21] <bill2or3> I'll probablly end up making some DIY slides, I doubt I'll find any that match my rails, that I can afford.
[18:16:21] <SWPadnos> unless you use step-to-servo drives, like Geckos or Rutexes
[18:16:43] <cncuser> motenc/README /usr/local/etc/emc2/sample-configs/motenc
[18:16:44] <cncuser> install -d -m 0755 -o root /usr/local/etc/emc2/sample-configs/ppmc
[18:16:44] <cncuser> install -m 0644 -o root /usr/local/etc/emc2/sample-configs/ppmc
[18:16:44] <cncuser> install: too few arguments
[18:16:45] <SWPadnos> (or a voltage -> step generator)
[18:16:50] <cncuser> itsalso with d
[18:17:21] <SWPadnos> it looks like the first one worked, and the second didn't
[18:17:43] <cncuser> i dont know
[18:18:02] <cncuser> i dont intend ever to know what Make does
[18:18:08] <SWPadnos> just looking at the manpage, you use -d to sepcify that it's creating a dirextory, rather than copying a file
[18:18:28] <cncuser> i dont get why anyone is using this :)
[18:18:40] <SWPadnos> heh
[18:18:49] <cncuser> really make is magic
[18:19:03] <SWPadnos> it certainly is
[18:19:08] <cncuser> to me... most of the time good magic, sometimes bad magic
[18:19:14] <cncuser> right now
[18:19:16] <cncuser> bad
[18:19:46] <alex_joni> cncuser: where did those 2 come from?
[18:19:48] <alex_joni> make install?
[18:20:00] <cncuser> yes
[18:20:01] <cncuser> but
[18:20:12] <cncuser> i try with the sourcepacke now. maybe it works with it
[18:20:41] <SWPadnos> let's hope so
[18:20:48] <cncuser> yes
[18:20:54] <cncuser> for i dont know what im doing anymore
[18:21:10] <SWPadnos> heh. maybe you should get some sleep ;)
[18:21:20] <alex_joni> that is cradek's magic in there :
[18:21:26] <alex_joni> define INSTALL_CONFIG
[18:21:26] <alex_joni> $(DIR) $(DESTDIR)$(sysconfdir)/emc2/sample-configs/$(1)
[18:21:26] <alex_joni> $(FILE) $(call GLOB,../configs/$(1)/*) $(DESTDIR)$(sysconfdir)/emc2/sample-configs/$(1)
[18:21:32] <cncuser> but dont want to continue this after i finally get ba sleep in a few hours. i dont want to see a computer for at least 2 days.
[18:22:18] <cncuser> install -d -m 0755 -o root /usr/local/etc/emc2/sample-configs/ppmc
[18:22:18] <cncuser> install -m 0644 -o root /usr/local/etc/emc2/sample-configs/ppmc
[18:22:18] <cncuser> install: too few arguments
[18:22:18] <cncuser> Try `install --help' for more information.
[18:22:18] <cncuser> make: *** [install] Error 1
[18:22:19] <cncuser> ok...
[18:22:33] <cncuser> anyone got a working build ? what happens after those lines ? something serious ?
[18:22:46] <cncuser> i dont see any way out of this make hell
[18:23:01] <cncuser> or just sampleconfigs ?
[18:23:21] <alex_joni> there is a line $(INSTALL_CONFIGS)
[18:23:26] <alex_joni> in the install: target
[18:23:28] <alex_joni> rip it out
[18:27:47] <cncuser> find: /usr/local/lib/libXfont.so.0: Input/output error
[18:27:47] <cncuser> find: /usr/local/lib/libXau.so.0: Input/output error
[18:27:48] <cncuser> hahaha
[18:27:51] <cncuser> im so lucky
[18:28:00] <alex_joni> huh?
[18:28:15] <cncuser> thats what happens if you play to much with unstable unionfs :)
[18:28:52] <cncuser> but it was really very old :)
[18:29:33] <cncuser> if your computer hangs while you delete something or add soemthing above another stack.... you get ext2 corrution
[18:30:16] <cncuser> ok, need to reboot
[18:30:19] <cncuser> cu later
[18:40:37] <cncuser> hmm
[18:42:03] <Jymmm> Fucking SBC...
[18:43:39] <Jymmm> Our DSL contract is up, was paying $25/mo for 1500/128. Non-contract price is $50/mo. But they have a online special for the SAME service of only $13/mo. or upgrade to 3000/384 for $18/mo.
[18:44:30] <Jymmm> They would only offer me yesterday $30/mo for the same service. Now when I call to cancel DSL, they offer me $20/$25 respectively.
[18:45:48] <skunkworks> sounds like something aol would do ;)
[18:45:52] <Jymmm> Yesterday hit Circuit City and got a cable modem ($0.01 after rebate) and a 6mo promo of $20/mo.
[18:46:39] <cncuser> i pay 50$/month of 2048kbit down / 512kbit up
[18:46:45] <cncuser> s/of/for/
[18:47:11] <Jymmm> The 40/mo 6mo promo is for 6000/384, then goes to $40/mo
[18:47:20] <Jymmm> The 20/mo 6mo promo is for 6000/384, then goes to $40/mo
[18:47:28] <cncuser> but i get part of the freen wireless project here when i have time and money
[18:47:53] <Jymmm> freen ?
[18:48:11] <cncuser> + i got a fairuse 20gb means if i have more then 25gb traffic during a month they cut me down to double isdn :)
[18:48:32] <cncuser> inode/upc ... bad companys
[18:48:38] <Jymmm> ah
[18:49:10] <Jymmm> Neither of these have a bandwidth limit. But they dont gaurntee the bw either
[18:49:38] <cncuser> nore does mine
[18:50:18] <Jymmm> ya know the really PITA part is why didn't they just offer the price and not have me go thru these headaches. bastards
[18:50:37] <cncuser> :)
[18:51:49] <cncuser> alex_joni: ok, i fixed my filoesystem now. what was i going to do about the install ?
[18:52:23] <alex_joni> take out the config stuff for now,
[18:52:27] <cncuser> i got the source freh from cvs 8and made a backup this time)
[18:52:27] <alex_joni> then worry later ;)
[18:52:39] <cncuser> hmmm
[18:53:07] <alex_joni> cncuser: comment out $(INSTALL_CONFIGS)
[18:53:17] <alex_joni> it's by itself on one line, try that
[18:54:07] <cncuser> ok
[18:54:13] <cncuser> it compiles
[18:54:19] <cncuser> now i copy the samples by hand
[18:58:29] <CIA-8> 03cradek * 10emc2/src/emc/task/emccanon.cc:
[18:58:29] <CIA-8> fix some very old bugs related to different units/maxvel/maxaccel
[18:58:29] <CIA-8> on rotary axes. now I think it does the things promised in the
[18:58:29] <CIA-8> Kramer document.
[19:03:01] <cncuser> alex_joni: ok, it built :)
[19:03:14] <cncuser> or,hmmm
[19:03:14] <alex_joni> cncuser: nice
[19:03:19] <cncuser> ok, i get the newest axis too :)
[19:04:19] <cncuser> stamp 2006-03-08 03:03:41 is 25393 s in the future
[19:04:20] <cncuser> harhar
[19:04:26] <cncuser> timezones ;)
[19:05:38] <cncuser> aeh, where should the emcroot be ?
[19:05:51] <cncuser> ../usr ?
[19:06:07] <alex_joni> no, where you put the sources
[19:07:17] <cncuser> ok
[19:07:32] <ValarQ> 'lo folks
[19:07:46] <cncuser> hi varaq
[19:07:51] <cncuser> valarq
[19:08:03] <alex_joni> hi captain
[19:08:28] <ValarQ> does anyone here knows if i can get feedback/correction with the standard hal stepper component?
[19:08:44] <ValarQ> s/ows/ow/
[19:08:46] <alex_joni> ValarQ: sorry.. what?
[19:09:37] <ValarQ> alex_joni: i want to put a small sensor on my machine that tells me if my steppermotor fails
[19:10:15] <Jymmm> ValarQ: define "fails"? Do you mean stalls?
[19:10:27] <ValarQ> Jymmm: yeah
[19:10:40] <alex_joni> ValarQ: basicly a signal? (bit?)
[19:10:49] <ValarQ> Jymmm: i can handle the "catches fire and explodes" part in a higher level :)
[19:10:56] <Jymmm> ValarQ: I guess you could toss on encoders.
[19:11:13] <ValarQ> alex_joni: doesn't matter, i'm just looking on the possibilities
[19:11:29] <Jymmm> ValarQ: My experiance is that when the stepper fails , that part is already ruind.
[19:11:38] <Jymmm> err stalls I mean
[19:11:42] <ValarQ> :)
[19:12:14] <ValarQ> yeah, thats probably correct
[19:12:30] <alex_joni> ValarQ: endless possibilities
[19:12:49] <alex_joni> 1. have encoders for feedback (not for PID, as that's bad for steppers)
[19:12:56] <alex_joni> when ferror is too big: halt
[19:13:06] <ValarQ> alex_joni: i know, but there is a how-many-hal-components-you-must-hack factor
[19:13:20] <alex_joni> 2. have an external something that detects the stalling, and it sends an estop
[19:13:26] <ValarQ> encoders?
[19:13:44] <alex_joni> what I'm talking about is only connecting hal components, minimum coding
[19:13:53] <ValarQ> great
[19:13:55] <Jymmm> ValarQ: lil round disc's
[19:14:14] <ValarQ> Jymmm: like them in my pointing device?
[19:14:39] <Jymmm> ValarQ pretty much... http://usdigital.com/ iirc
[19:14:51] <ValarQ> (i got it for plaing xbill and manuevering the emc frontend ;)
[19:15:03] <ValarQ> s/plaing/playing/
[19:16:16] <ValarQ> what component should take the encoder input?
[19:17:14] <ValarQ> * ValarQ discovers hal/components/encoder.c
[19:32:00] <Jymmm> ValarQ: Curious, what drivers are you using?
[19:33:34] <CIA-8> 03cradek * 10emc2/configs/stepper-xyza/ (5 files):
[19:33:57] <cradek> oops
[19:34:03] <cradek> oh well...
[19:34:34] <alex_joni> forgot something?
[19:34:39] <cradek> the log message
[19:34:55] <alex_joni> that's what I meant ;)
[19:35:10] <CIA-8> 03cradek * 10emc2/src/Makefile: new four-axis (rotary A axis) sample stepper config
[19:38:29] <ValarQ> Jymmm: drivers?
[19:38:42] <Jymmm> stepper motor driver (geckos)
[19:39:08] <alex_joni> Jymmm: since when?
[19:39:54] <Jymmm> alex_joni what are you babbling about now?
[19:40:13] <alex_joni> < Jymmm> stepper motor driver (geckos)
[19:40:24] <Jymmm> alex_joni and your point?
[19:40:31] <ValarQ> Jymmm: i have been using my own crap to test the motors until now
[19:40:39] <alex_joni> oh, thought you were using them
[19:41:37] <Jymmm> ValarQ: Ah. Well I had a LONG conversation with Mariss about stalling and least with my motors
[19:41:39] <skunkworks> where is this kramer document?
[19:41:51] <cradek> skunkworks: google for rs274ngc, first hit
[19:41:55] <lilo> [Global Notice] Hi all. If you are running a Netgear model WGT624 (the white enclosure) and you would like to test their fix for the recent DCC exploit, the one that's probably knocking you offline when some childish person sends it to you, please send me your email address and I'll forward it to them. Thanks!
[19:42:00] <cradek> skunkworks: iirc it's a terrible long url
[19:42:42] <ValarQ> Jymmm: ok
[19:43:02] <ValarQ> Jymmm: i don't have them here at home but i believe it's this one: http://www.allegromicro.com/sf/3955/
[19:43:23] <ValarQ> Jymmm: haven't got it working well yet :/
[19:44:20] <Jymmm> ValarQ (mine --> http://www.xylotex.com/StepperMotor.htm) Now, when Mariss looked that over, and using his drives @ 48VDC he said they should never stall. He has circuitry to prevent stalling. If you goto the yahoo group you can read the whole thing.
[19:44:30] <lilo> [Global Notice] In the meantime, please remember, if you're getting hit, disconnect from freenode and reconnect to chat.freenode.net, port 8001. If you use port 8001, rather than port 6667, it should render the attack ineffective. Thanks!
[19:44:33] <bill2or3> this looks interesting, and cheap: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=7595274576
[19:44:47] <bill2or3> a linear servo
[19:50:29] <bill2or3> guh, it's just the rail.
[20:12:37] <cncuser> hi
[20:13:22] <cncuser> jepler: what was that again with pyc and pyo ? i made all to pyc and python barfs... i may be missing files again
[20:15:01] <sed> anyone know where I can find a good double depth thread chart?
[20:19:28] <K4ts> hello
[20:20:11] <sed> starrett had a good one..
[21:11:55] <K4ts> night
[22:13:49] <cncuser> ok
[22:14:04] <cncuser> now it works, im just finishing the Thank yous ;)
[22:14:46] <alex_joni> cncuser: you can skip us ;)
[22:14:51] <alex_joni> cncuser: glad it does
[22:14:58] <cncuser> noway
[22:15:06] <alex_joni> cncuser: say when you want to be mirrored :D
[22:26:47] <cncuser> last testreboot ;) hopefully
[22:39:43] <Jymmm> Jymmm is now known as Jymm
[22:47:05] <Jymm> Jymm is now known as Jymmmm
[23:05:00] <alex_joni> night all
[23:05:55] <bill2or3> good night.
[23:56:38] <cncuser> hi
[23:56:41] <cncuser> finally
[23:56:44] <cncuser> im uploading
[23:57:38] <cncuser> cooltool.he.fdread.org/cncforum/coolcnc/coolcncb04.iso
[23:57:43] <cncuser> finished in 10 minutes
[23:58:05] <cncuser> well, im of for a beer and then gonna sleep at least 20 hours :)
[23:58:11] <cncuser> please test this amnyone
[23:58:21] <cncuser> bye
[23:59:35] <cncuser> http://cooltool.he.fdread.org/cncforum/coolcnc/coolcncb04.iso for better recognition ;)