#emc | Logs for 2005-08-07

[00:01:31] <robin_sz> hmm they quote 4 to 10 ns as typical ... thats several orders of magnitude better than wwv is it not?
[00:03:49] <Jymmm> till they reenable select service
[00:04:41] <robin_sz> doesnt make that much difference ... SA can be done by clock manipulation,. but also by modifying orbit data
[00:05:32] <robin_sz> goes up to about 20~100ns ...
[00:05:57] <Jymmm> if/when theres a national disaster, they could shut the whole thing down
[00:06:07] <robin_sz> sure ...
[00:06:14] <robin_sz> and they could turn off wwv too
[00:06:47] <Jymmm> but not likely
[00:07:28] <robin_sz> OK, well , say GPS and WWV, which one is more likely to suffer from say, hurricanes and earthquakes ;)
[00:07:58] <robin_sz> anyway
[00:08:08] <robin_sz> we have a similar thing, called rugby
[00:08:12] <robin_sz> on 60khz
[00:08:20] <robin_sz> but its days are numbered
[00:08:26] <Jymmm> I dont care about that, it's the disabiling of GPS as a time base
[00:08:43] <Jymmm> and having an alternative available
[00:08:47] <robin_sz> i think its will go in the next few years
[00:08:57] <robin_sz> sames as Loran did
[00:09:02] <Jymmm> nah, still used in calibration labs
[00:09:07] <robin_sz> wwv?
[00:09:10] <Jymmm> sure
[00:09:11] <robin_sz> no way jose
[00:09:45] <robin_sz> 10 years ago, maybe
[00:09:56] <robin_sz> now? nah.
[00:10:04] <robin_sz> eclipsed by GPS
[00:10:38] <robin_sz> and wwv is eclipsed by the intenet too
[00:17:50] <Yuga> hi all
[00:26:40] <Jacky^> usually should Z axis have less speed of X and Y ?
[00:27:21] <anonimasu> no
[00:27:36] <anonimasu> you dont need z to be as fast
[00:27:37] <wb9mjn> Hi All...I downloaded John Elson's jeupdate.tgz and copied in the /usr/local into my system...Do I need to compile , or is the binary updated as well?
[00:27:38] <Jacky^> can work at the same speed of others ?
[00:28:03] <anonimasu> work?
[00:28:07] <anonimasu> you mean plunge?
[00:28:10] <anonimasu> usually you dont
[00:28:12] <Jacky^> yeah
[00:28:13] <anonimasu> but for rapids..
[00:28:25] <anonimasu> going out off the work,, or doing interpolated stuff
[00:29:02] <anonimasu> :)
[00:29:16] <anonimasu> if you have the chance make your Z as fast as the other axis:es
[00:30:09] <Jacky^> i'm using 5 as value for all
[00:30:42] <Jacky^> i was looking at this: http://www.super-tech.com/root/rr.htm
[00:31:14] <Jacky^> Maximum Rapid Travel Speed for Z is half then XY
[00:32:14] <Jacky^> there's a reason ?
[00:32:27] <anonimasu> I dont know really motor size perhaps
[00:32:42] <anonimasu> and screw pitch
[00:32:55] <anonimasu> I find the stepper on Z to be the limiting factor when I do parts..
[00:33:16] <anonimasu> moves that should be fast(non cutting moves)
[00:33:41] <anonimasu> ends up long since I cant plunge/retract z fast enough
[00:33:55] <anonimasu> it's very noticeable when waitinf for cam generated programs..
[00:34:26] <Jacky^> understood :)
[00:34:35] <Jacky^> tnx
[00:34:58] <anonimasu> you might want to crank the Z axis speed down before you start making your first cuts ;)
[00:36:19] <Jacky^> right
[00:36:43] <anonimasu> or well your overall speed..
[00:36:51] <Jacky^> at the moment i'm working on wood, so i don't have too much problem
[00:36:54] <anonimasu> fast machine + slow brain = bad stuff ;)
[00:36:59] <Jacky^> depend on wood used
[00:37:02] <anonimasu> been there milled clamps ;)
[00:37:58] <anonimasu> hehe
[00:38:20] <Jacky^> i found very useful, sometimes, to roughing with a big endmill befor finish
[00:39:35] <Jacky^> anonimasu: how it seem to you the machine in the url i pasted ?
[00:40:28] <Jacky^> it seem to me not much expensive..
[00:40:42] <Jacky^> to homebuild I mean
[00:42:30] <Jacky^> I could trash my toy :))
[00:42:43] <anonimasu> ?
[00:42:44] <anonimasu> heh
[00:55:07] <Jymmm> bah... I REALLY dont want to run emc an an ext video card =(
[00:56:55] <Jacky^> Jymmm: I agree
[00:58:09] <Jymmm> ok, what do I need todo to start the nml server?
[01:02:22] <Jacky^> what is nml server ? :\
[01:03:33] <Jacky^> Jymmm: this ? http://www.isd.mel.nist.gov/projects/rcslib/nmlcfgsvr.html
[01:04:25] <Jacky^> you want to take control of remote pc ?
[01:04:45] <Jymmm> Jacky^ alex gave me a remote GUI that runs ion Windows to control EMC
[01:05:04] <Jymmm> I dont have to run X on the emc machine
[01:05:19] <Jymmm> it takes up too many resources.
[01:05:22] <Jacky^> ah.. ok
[01:06:13] <Jymmm> but he left before I had everything installed and I dont' know what exactly I need to startup on the emc box
[01:07:42] <Jacky^> no idea :\
[01:08:43] <Jymmm> me neither =)
[01:08:51] <Jacky^> it sure should have a conf file where to specify ip server adress on client side
[01:09:12] <Jacky^> probably you also need to set some permission on other side
[01:09:25] <Jacky^> there's no doc files ?
[01:09:33] <Jymmm> I *think* it's setup under sudo, but I could be wrong
[01:10:06] <Jacky^> why non by root ?
[01:10:38] <Jacky^> if sudo wont work..
[01:13:49] <Jacky^> Jymmm: http://www.yty.net/cnc/networking.html
[01:14:31] <Jacky^> you should use TkEMC i suppose
[01:18:10] <Jymmm> someone suggested running cygwin-x but damn, that'll take as along to install as emc did!
[01:18:37] <Jymmm> between putty and this thing alex gave me, should be fine
[01:19:53] <Jacky^> yeah..
[01:20:12] <Jacky^> the problem is that you've not X running..
[01:35:26] <Jacky^> bedtime..
[01:35:30] <Jacky^> night all
[05:58:59] <Jymmm> hey anonimasu
[07:52:55] <Jymmm> Boo!
[08:01:21] <Kn1ghtjp> gaaah!
[08:15:14] <lilo> [Global Notice] Hi all. There's a news item up on http://freenode.net/news.shtml .... we'll be experiencing some network disruption Monday morning, and we'll be doing our first test upgrade of the production system to Hyperion around that time. Please take a look at the news item for more detail, and have a great morning! :)
[10:02:05] <anonimasu> morning everyone
[10:02:14] <Kn1ghtjp> morning!
[10:53:12] <Jacky^> Gmorning
[11:37:25] <Jacky^> hi robin_sz
[12:04:33] <robin_sz> hi
[12:04:56] <CIA-9> 03paul_c * 10emc2/src/libnml/os_intf/_sem.c: Re-implement a semtimedop function for pre 2.4.22 systems.
[12:17:37] <anonimas1> hello
[12:18:38] <robin_sz> hi
[12:18:40] <robin_sz> so ...
[12:18:45] <robin_sz> imagine ai have a servo loop
[12:18:54] <anonimas1> Yes?
[12:19:14] <robin_sz> to control my rorch height ...
[12:19:17] <robin_sz> torch
[12:19:19] <anonimas1> yes
[12:19:33] <robin_sz> the voltage control bit is easy enough ..
[12:19:48] <robin_sz> just feed in target voltage, actual voltage
[12:19:59] <robin_sz> scale the result and let the PID drive it
[12:20:24] <robin_sz> I'll also have a encoder and another loop
[12:20:46] <robin_sz> for "retract" and "go down until you hit the sheet" etc
[12:21:00] <robin_sz> *anyway* ...
[12:21:13] <robin_sz> since I just want to retract ...
[12:21:28] <robin_sz> you think I can get away without trajectory planning?
[12:21:34] <les> ha
[12:21:36] <les> hi
[12:21:39] <anonimas1> heh
[12:21:45] <robin_sz> just "set desired poition to zero"
[12:21:50] <anonimas1> you are making this awffuly complicated..
[12:21:51] <robin_sz> and let the pid catch up?
[12:22:39] <anonimas1> you are not making large interpolated moevements with the torch anyway
[12:22:55] <robin_sz> no interpolated movements,
[12:23:01] <robin_sz> btu some large ones
[12:23:08] <anonimas1> oh, dosent matter..
[12:23:10] <robin_sz> retract will be 100mm or so
[12:23:25] <anonimas1> if you are running torch height indipendently you dont need interpolation..
[12:23:31] <robin_sz> quite
[12:23:44] <anonimas1> unless your cutting \ stuff ;)
[12:24:10] <robin_sz> well possibly, but the voltage tracking will follow it
[12:24:21] <anonimas1> ah yeah but
[12:24:26] <anonimas1> interpolated..
[12:24:33] <robin_sz> no, not interpolated
[12:24:34] <anonimas1> ;)
[12:24:38] <robin_sz> its 2.5 axis
[12:24:41] <robin_sz> not 3 axis
[12:24:50] <robin_sz> ah, les ...
[12:25:08] <robin_sz> you mentioned you had a cost effectice design for a small servo amp?
[12:25:23] <anonimas1> robin_sz: look at the one jon elson sells :)
[12:25:31] <robin_sz> yeah, no use sadly
[12:25:49] <les> I think I mentioned I NEEDED one
[12:25:54] <robin_sz> ahh, OK
[12:26:07] <les> and I see no big problems doing it
[12:26:15] <robin_sz> anonimas1 : it needs PWM input
[12:26:23] <les> oh btw....
[12:26:43] <les> had a breakthrough friday I think
[12:27:06] <robin_sz> les: probably use L6206 6203 then
[12:27:07] <anonimas1> erovin
[12:27:08] <anonimas1> bb
[12:27:08] <robin_sz> breakthrough? you hit oil?
[12:27:10] <anonimas1> fasdjlkmsadf
[12:27:13] <anonimas1> err look again.
[12:27:20] <anonimas1> he has a voltage input one
[12:27:48] <les> a little thing the size of a cigarette butt is turning a tiny bit of compressed air into 30 vrms 40 kHz
[12:28:13] <anonimas1> les: very cool
[12:28:14] <anonimas1> :)
[12:28:41] <les> I was about to give up and complete the research report saying it was not practical
[12:28:48] <les> and then....
[12:28:58] <anonimas1> did any of the ramblings we had help?
[12:29:16] <les> I thought the measuring equip was in error...
[12:29:30] <les> yeah kept my mind on it...
[12:29:41] <anonimas1> great!
[12:29:46] <les> patent patent patent
[12:29:52] <les> I called up the corp right away
[12:30:25] <robin_sz> anonimas1: the analogue one is a kit, and look more complex than i need by a factor of about 10 ...
[12:30:25] <les> Likely to get development contracts
[12:30:25] <robin_sz> wow
[12:30:25] <anonimas1> robin_sz: look again.
[12:30:25] <anonimas1> he has 0-10v amps.
[12:30:29] <anonimas1> :)
[12:30:39] <robin_sz> http://pico-systems.com/servo.html
[12:30:41] <robin_sz> this?
[12:30:52] <anonimas1> I cant check..
[12:31:13] <anonimas1> I will be playing with the USC later today
[12:31:21] <les> I see nothing particularly expensive on common servo amps like Copley...
[12:31:39] <les> Fets biggest cost I guess
[12:31:45] <robin_sz> les: so you ordering the 7 series BM yet?
[12:31:48] <anonimas1> bm?
[12:33:03] <les> haha
[12:33:03] <robin_sz> bmw
[12:33:03] <anonimas1> ah
[12:33:03] <anonimas1> hehe
[12:33:03] <les> still looking at "girl car" Audi cabriolet...
[12:33:03] <robin_sz> anonimas1: his analogu 0-10v amp is a kit, and it looks to be about 200 components, I just need someting small and cheap to drive a small z axis motor ...
[12:33:03] <robin_sz> les: the TT?
[12:33:03] <anonimas1> robin_sz: well, you can use a single FET..
[12:33:03] <anonimas1> and a micro to do the feedback loop..
[12:33:07] <les> A4 but like TT as well
[12:33:08] <Ron_B> les is going for a Maybach
[12:33:14] <anonimas1> * anonimas1 likes the TT lots
[12:33:16] <les> hi Ron
[12:33:23] <robin_sz> they seem quite quick ...
[12:33:24] <anonimas1> I hope the server didnt eat my raid array
[12:33:27] <anonimas1> yay
[12:33:29] <anonimas1> it lives
[12:33:47] <robin_sz> one managed to get away from me last week ... I was impressed
[12:33:52] <anonimas1> heh
[12:34:18] <Ron_B> congratulations to les on that breakthrough - hope the search goes well and the patent attorney is free
[12:34:29] <anonimas1> some kids tried to race me when I was trying to pass by them yesterday :)
[12:34:31] <les> thanks
[12:34:46] <robin_sz> patent attorney? free? yeah, like thats going to happen!
[12:34:51] <les> I still can hardly believe it
[12:34:53] <robin_sz> they charge just to answer the phoen
[12:35:02] <anonimas1> they drove 20kmh and floored it when I was at the side of them..
[12:35:08] <anonimas1> didnt work out too well :)
[12:35:20] <robin_sz> heh
[12:35:30] <anonimas1> then they drove like mad to catch up..
[12:35:43] <robin_sz> uh huh ...
[12:35:45] <les> I have ITW corp patent attorney assigned. Free.
[12:35:50] <anonimas1> then at the next intersection they took the opposite way of me..
[12:36:22] <robin_sz> les "hi, you are a patent attorney? can i ask what you charge?"
[12:36:22] <anonimas1> some kids in their parents audi A6 I think ;)
[12:36:36] <robin_sz> les "sure, its $2000 for 3 questions"
[12:36:53] <robin_sz> "that seems quite expensive, is that typical?"
[12:37:08] <robin_sz> "yes, thats standard for the industry, now whats your 3rd question?"
[12:37:18] <les> hahaha
[12:37:20] <anonimas1> hahaa
[12:37:50] <les> I have a new one...my old one that did all my patents retired
[12:38:17] <robin_sz> at the age of 27?
[12:38:28] <anonimas1> that sounds like a nice plan
[12:38:33] <les> heh naw he was old
[12:39:06] <robin_sz> im still worried about this Rabiit licensing thing ...
[12:39:15] <les> He kinda helped with getting my fellowship granted
[12:39:28] <robin_sz> nioce
[12:40:58] <robin_sz> I cant beleive jon elson really did that ... made a nice little servo amp .. and made it PWM input ...
[12:41:11] <robin_sz> instantly losing about 95% of the potential market
[12:41:20] <anonimas1> robin_sz: theres another one
[12:41:20] <anonimas1> w
[12:41:23] <Ron_B> I worked with one years ago who was a scoundrel
[12:41:35] <robin_sz> anonimas1: yes, but its a kit, and huge
[12:41:56] <les> PWM modulator is not much of a big deal
[12:42:06] <les> a dollar or so?
[12:42:07] <anonimas1> robin_sz: ill look them up later when I get home
[12:42:20] <robin_sz> no internent where you are?
[12:42:35] <anonimas1> oh, no web-browser
[12:42:40] <anonimas1> I am on a BSD console..
[12:42:48] <robin_sz> wow,
[12:42:53] <robin_sz> linx?
[12:42:53] <anonimas1> at a server room at work..
[12:42:57] <robin_sz> heh
[12:43:04] <anonimas1> waiting for a NT server to install
[12:43:12] <robin_sz> you brought food with you?
[12:44:00] <robin_sz> and something to sleep on?
[12:44:21] <Ron_B> * Ron_B thinks prayer and divine intervention may be a good idea
[12:44:36] <robin_sz> no, thats not it ..
[12:44:46] <robin_sz> it needs goats blood and chicken bones
[12:45:13] <robin_sz> divine intervention is for thngs from above, NT is from below :)
[12:45:22] <anonimas1> I dont have a firstborn.. so it's kind of hard to find good sacrifices
[12:45:22] <A-L-P-H-A> * A-L-P-H-A sells prayer beads wholesale. Gett'm while their hot!
[12:45:54] <robin_sz> A-L-P-H-A: I think I saw some pictures on the 'net .. the was this blonde chick ...
[12:46:20] <A-L-P-H-A> robin_sz, I think you're thinking anal beads...
[12:46:29] <robin_sz> ahh, theyre different?
[12:46:51] <A-L-P-H-A> yeah... prayer beads don't have such large gaps between beads.
[12:47:09] <robin_sz> * robin_sz phones the local convent .. "please do not open the gift pack I sent"
[12:47:52] <Ron_B> * Ron_B knew he would learn a lot just auditing this forum
[12:48:13] <A-L-P-H-A> Ron_B, if you use lube... it'd hurt less.
[12:48:45] <robin_sz> jons little amp looks neat ...
[12:49:07] <A-L-P-H-A> does it have allegator clips for some shocking entertainment?
[12:49:19] <robin_sz> mmmm, nope
[12:49:36] <A-L-P-H-A> electrical phylic probes?
[12:49:46] <robin_sz> mmmm .... nope.
[12:50:28] <A-L-P-H-A> I saw one fetish video that had a crank box, that created electrical shocks on the victim. It looked very very painful...
[12:51:25] <robin_sz> its amazing what people will do for 300 bucks
[12:53:26] <anonimas1> I wonder how hard it would be to designa servo amp
[12:53:52] <robin_sz> with all the PID and stuff?
[12:53:58] <robin_sz> or jsut the amp?
[12:54:12] <anonimas1> a amp doees not have PID..
[12:54:39] <robin_sz> well, easy I think .. I just want minimal component count
[12:54:55] <robin_sz> so a L6206 and a power stage will do it ..
[12:55:01] <robin_sz> or a TDA340
[13:03:28] <A-L-P-H-A> I think I scared Ron_B away.
[13:05:09] <Ron_B> * Ron_B nah ... just looking for a magento
[13:05:35] <anonimas1> log wiping time_
[13:05:36] <anonimas1> ?=
[13:05:41] <A-L-P-H-A> is that what kids are calling them now. <just some smart ass comment I could make>
[13:06:13] <A-L-P-H-A> 'ight. I'm off...sleep time.
[13:06:19] <robin_sz> night
[13:06:28] <anonimas1> night aloha
[13:06:30] <anonimas1> alpha..
[13:06:35] <A-L-P-H-A> it's actually 807am.
[13:06:37] <Ron_B> actually I ahve some old engine magnetos in the shop ... they easily can get one's attention
[13:06:56] <anonimas1> I am at a DEC keyboard ;)
[13:07:33] <A-L-P-H-A> what's the history of DEC?
[13:07:45] <A-L-P-H-A> dec merged with alpha, and alpha got bought by HP, which merged with Compaq?
[13:07:51] <anonimas1> yeah..
[13:07:59] <anonimas1> this is a digigtal"
[13:08:00] <anonimas1> "
[13:08:08] <anonimas1> branded as :)
[13:08:16] <A-L-P-H-A> old.
[13:08:27] <anonimas1> yeah
[13:08:32] <anonimas1> but it works nicely
[13:08:34] <anonimas1> but its awky to type on
[13:08:37] <A-L-P-H-A> damn... stop taunting me with tidbits of info which I then have to spew upon you all.
[13:08:40] <Ron_B> mine appears to be a ...."Made in China"
[13:09:27] <robin_sz> my fave was a SGI granit keyboard ..
[13:09:27] <anonimas1> A-L-P-H-A: just go to bed.
[13:09:31] <robin_sz> lovely feel to it
[13:11:38] <robin_sz> the ebay seeling price ofr an Octane seems to be �100
[13:11:58] <robin_sz> still, people putting htme up with buy it now at �1500 or more ...
[13:34:07] <anonimas1> hm
[13:34:20] <anonimas1> my hands are sunbbuurned..
[13:34:28] <anonimas1> err sunburned..
[13:34:29] <anonimas1> ;)
[13:34:38] <anonimas1> I tig welded some stuff and skipped the gloves
[13:45:41] <wb9mjn> Hi All...
[14:39:24] <jmk_away> jmk_away is now known as jmkasunich
[15:14:58] <jmkasunich> quiet today
[15:15:31] <robin_sz> NOT THAT QUIET!
[15:15:51] <jmkasunich> !?!
[15:17:13] <CIA-9> 03paul_c * 10rcslib/src/posemath/Makefile: Kick of the compile farm and see if this fixes it's 2.4.18-rtai problem.
[15:17:52] <jmkasunich> hi paul.... slot 2 was having problems, I just un-hung it and kicked it off manually about 5 mins ago
[15:18:51] <jmkasunich> expect results in about 15 mins, based on the code before your commit just now
[15:19:12] <jmkasunich> I'll kick it off again when it finishes, to see the results of your commit without waiting another hour
[15:22:32] <jmkasunich> oops... I didn't read the commit message closely enough.. your commit was to rcslib, so it will compile automagicaly after emc2 finishes
[15:22:38] <jmkasunich> no results until both are done
[15:23:51] <robin_sz> sigh ... pesky FPGA
[15:24:42] <jmkasunich> hi paul
[15:24:50] <gezr> jmkasunich : I didnt have any problems with that build after I correctly co the source
[15:24:52] <paul_c> Afternoon John
[15:25:00] <paul_c> ?
[15:25:02] <jmkasunich> gezr: good to hear it
[15:25:16] <jmkasunich> hi Paul
[15:25:31] <anonimasu> robin_sz: what are you having trouble with?
[15:25:31] <paul_c> Farm appears to be down.
[15:25:39] <jmkasunich> just kicked off the 2.4.18 slot, it is working on rcslib
[15:25:52] <robin_sz> anonimasu: the encoder inputs on the Gwhatever card
[15:26:10] <jmkasunich> slot 2 was having problems, those are fixed now, it is working on the latest emc2 build, to be followed by an rcslib build
[15:26:52] <robin_sz> getting busy all of a sudden
[15:27:33] <paul_c> Much feedback on the tp fix ?
[15:27:48] <anonimasu> paul_c: les was going to try it later tonight
[15:28:00] <robin_sz> ooh, its fixed?
[15:28:29] <gezr> there has been mention on the email about robin_sz . results are showing favoriable status
[15:28:30] <jmkasunich> robin: at least one long-standing problem is fixed
[15:28:32] <paul_c> Yes. Only took a couple of hours - Most of that was testing.
[15:29:08] <jmkasunich> I'm sure there are other issues - extremely large numbers of extremely short segments probably still cause probs
[15:29:19] <robin_sz> * robin_sz nods
[15:29:32] <paul_c> That's a comms issue.
[15:29:37] <robin_sz> but if the arc to line and line to line is fixed .. thats a good start
[15:29:38] <jmkasunich> but one major type of bad blends seems to be a thing of the past
[15:29:43] <jmkasunich> paul_c: right
[15:30:13] <wb9mjn> Hi all....I downloaded John Elsons jeupdate.tgz and installed it...Do I need to compile for the tp.c fix to be there ?
[15:30:25] <anonimasu> be right back
[15:30:26] <anonimasu> going to work
[15:30:44] <paul_c> wb9mjn: dunno - You need to talk to J.E. about it.
[15:30:54] <wb9mjn> Ok...
[15:30:58] <paul_c> (suspect it is precompiled)
[15:31:20] <wb9mjn> Me too...but wanted to ask anyway...
[15:31:25] <paul_c> anonimasu: Did you get a chance to test that last build ?
[15:31:36] <gezr> wb9mjn : was there a README in the tgz?
[15:32:01] <wb9mjn> Did not see one that related to the tgz...
[15:32:21] <wb9mjn> It had a /usr directory...
[15:32:33] <wb9mjn> Most of it was /usr/local....
[15:32:37] <paul_c> /usr/local/plat/*/bin ?
[15:33:01] <wb9mjn> I copied over the /usr/local...
[15:33:14] <wb9mjn> I believe it was a complete /local...
[15:33:32] <paul_c> was there a plat/ tree in there ?
[15:33:41] <wb9mjn> I ll go look....
[15:34:53] <gezr> is there a pci stg like card available now?
[15:35:04] <paul_c> yes.
[15:35:28] <paul_c> Vital Systems.
[15:35:31] <jmkasunich> vital systems MOTENC-100
[15:35:39] <jmkasunich> and MOTENC-Lite
[15:35:39] <gezr> okay, im going to finally break down and purchase a box to dedicate to emc its going to be fast
[15:35:40] <wb9mjn> It has a /usr/local/emc/plat/"variousplatforms"/bin directory...
[15:36:18] <paul_c> gezr: www.vsi99.com
[15:36:46] <paul_c> wb9mjn: Should be two subdirs in plat/
[15:36:50] <wb9mjn> In the linux_2_4_21/bin directory, it looks like all the files are dated 8/1/05 ...
[15:36:55] <gezr> and yes, im goign to try to and learn how to add the doxygen tags to the source code, Ill see if I can find someone who can help me learn how to do it, or get me started in the right direction
[15:36:58] <Ron_B> speaking of box for EMC - is there any particular thing one should look for - board, processor - enough RAM is a given -
[15:37:12] <wb9mjn> There are 4 or 5 paul_c ...
[15:37:19] <robin_sz> gezr: be sure you are sitting down before you see the price ...
[15:37:19] <jmkasunich> avoid intel 810 chipsets
[15:37:30] <gezr> robin_sz : im sitting
[15:37:37] <paul_c> and i815/845/865
[15:37:59] <jmkasunich> gezr: vital board with breakouts will use well over half of a $1000 bill
[15:38:08] <robin_sz> gezr: $765 for the 8 axis, $550 for the 4 axis
[15:38:09] <wb9mjn> Hi dave-e ...
[15:38:11] <gezr> better then the 12k starting price for fagor control
[15:38:16] <dave-e> hi
[15:38:45] <gezr> howdy dave-e
[15:38:54] <dave-e> mornin
[15:38:58] <gezr> jmkasunich : have you used a vital board?
[15:39:28] <wb9mjn> Updated with jeupdate.tgz ...Looks like it updated executable...Machine still works... got a glitch on the Z column upward movememt..
[15:39:49] <robin_sz> gezr: add another $250 for a pair of breakout boards
[15:39:50] <wb9mjn> Need to figure out it is mechanical...Right now think it is...
[15:40:01] <wb9mjn> is mechanical...
[15:40:22] <jmkasunich> gezr: I experimented with one at the CNC workshop, but I don't own one myself
[15:41:15] <jmkasunich> paul_c: the rcslib build on TNG worked, I manually triggered a build of EMC1 to see if that works now that rcslib is ok
[15:41:31] <robin_sz> jmkasunich: if the ethernet thing gets going with the G-Rex it will be interesting .. 4 channels of analogue out, 6 encoder inputs for $200 ... thats more like it.
[15:42:09] <paul_c> jmkasunich: Does the emc1 build not get triggered automatically ?
[15:42:11] <dave-e> nice if it can be done
[15:42:51] <jmkasunich> it should (just looked at the rcs build log which jogged my memory)
[15:42:57] <robin_sz> dave-e: I cant see why it cant, if steve hardy says he will .. he usually does
[15:43:00] <jmkasunich> but it didn't according to the farm log
[15:43:03] <jmkasunich> I'll investigate
[15:43:48] <dave-e> so the g-rex is like a kaluga/mauch but with dacs?
[15:44:31] <robin_sz> its an micro for comms, an FPGA for whatever you want and 4 A-D, 4 D-A and much IO
[15:44:49] <dave-e> hmmmmm
[15:44:55] <robin_sz> reading ancoders and spitting them down raw ethernet woudl eb a snap
[15:44:57] <dave-e> hope it is rugged
[15:45:02] <robin_sz> seems to be
[15:45:10] <paul_c> buggered more like...
[15:45:30] <paul_c> with a damned restrictive licence on the compiler.
[15:45:35] <wb9mjn> gezr, My machine uses the Motenc-100....Appears to control well....dave-e has not has a similar experience however....
[15:45:56] <wb9mjn> several others on the news group have apparently gotten machines working with the Motenc-100...
[15:46:12] <dave-e> I think a lot of mine was early software but I've not gotten back to it.
[15:46:38] <dave-e> had hoped the one at fest would get up and running, just not enough time
[15:47:39] <dave-e> plan to put it on my cincinatti...much like a BP
[15:47:54] <wb9mjn> My present problem is stopping in the +Z direction...there is some high vibration ....Might be a loose Z screw nut..and the momentum
[15:48:42] <robin_sz> we used a Baldor control on our bridgport in the end, but that has issues it seems
[15:48:56] <wb9mjn> of the milling head is making the screw jump...Only happens in movements more than 2 inches, at the full 240 in/min setting...
[15:48:59] <robin_sz> their gcode interp sucks :)
[15:49:14] <wb9mjn> Its about a 30 pound milling head....
[15:49:31] <robin_sz> the whole hea dmoves?
[15:49:40] <robin_sz> not just the quill?
[15:50:18] <wb9mjn> Attempts to retun do not have any effect, except if i drop the P gain down by a factor of ten...which of course limits acceleration...
[15:50:32] <wb9mjn> retune, that is...
[15:50:48] <robin_sz> with servos? or stepers and freqmod?
[15:51:16] <wb9mjn> The high vibration is when it comes to the stop...not during the travel...
[15:51:37] <wb9mjn> Yes...its using servos....
[15:51:41] <dave-e> vibration as in oscillation/
[15:51:45] <robin_sz> right
[15:51:57] <wb9mjn> Yes...it shutters to a stop....
[15:52:03] <robin_sz> sounds like D is too high
[15:52:13] <dave-e> or something is really loose
[15:52:16] <paul_c> increase deadband a tad.
[15:52:17] <robin_sz> yeah
[15:52:50] <dave-e> but it only does it at high speed?
[15:52:59] <wb9mjn> Its possible with the hot weather, the nut came loose...as this was assembled in the cold weather....
[15:53:11] <wb9mjn> Yes...below about 100 in/min it stops just fine...
[15:53:20] <paul_c> bung a scope on the DAC outputs.
[15:53:26] <dave-e> that does sound like tuning
[15:54:05] <paul_c> dave-e is the expert for tuning emc.
[15:54:32] <wb9mjn> It could also be that the lube is stickier now than then.....Yep, dave-e has helped me get it tuned up allot...
[15:54:43] <dave-e> ah yes, but you know the definition of expert
[15:55:02] <dave-e> anyone 20 mi from home with a brief case
[15:55:19] <wb9mjn> Its been sitting for the past couple months...
[15:56:23] <wb9mjn> I do not have a proper wrench for the Z screw nut, which is kind of a spanner arrangement....
[15:56:38] <dave-e> make one
[15:56:46] <paul_c> on a cnc mill
[15:56:47] <wb9mjn> Yep...going to have to...
[15:57:13] <wb9mjn> Hi...can do that now....turn the blank on the lathe, and then cut the slots and center relief on the mill...
[15:57:32] <wb9mjn> Actually, tines....not slots...
[15:57:46] <wb9mjn> To fit into the slots on the nut...
[15:58:58] <wb9mjn> I need to get an indicator on the head, and shove it around and see if it moves, too...
[15:59:16] <gezr> hey, why not, drill 2 offset holes, and put pins into the holes?
[15:59:55] <gezr> or just flat bar, with a U cut in it, and 2 holes drilled for pins
[16:00:37] <gezr> or the sicsor method, 2 flat bars, with holes in the end for the pins, then a screw holding the 2 flats together \/
[16:00:53] <wb9mjn> The screw shaft comes up through the nut.....So, it has to be hollow...
[16:01:57] <wb9mjn> I used two screw drivers originally....with some vibration it seemed to torque up nicely...but then it was 65 degrees...
[16:02:38] <wb9mjn> Its been up to 100 + since then....and the column/screw is about 36 inches long....and the nut is aluminum....Just has me
[16:03:03] <wb9mjn> thinking it could have cold flowed a little, lost its friction, and worked free....
[16:03:53] <wb9mjn> 36 inches of CTE could have loosened it up with the 40 degree F temperature variation...
[16:04:46] <wb9mjn> At the time, did not have the x and y setup...Now , I do , and can make the wrench....
[16:05:07] <wb9mjn> anyway....
[16:05:08] <dave-e> hi tim
[16:05:11] <dave-e> tom
[16:06:02] <wb9mjn> I will also try the D and deadband experiments to see if that effects it...thanks for the comments...
[16:06:12] <wb9mjn> Off for a walk...will leave this up...
[16:07:39] <staggerlytom> * staggerlytom Hi Dave, Hi Paul
[16:08:59] <dave-e> tom...thought you were off to tw
[16:12:24] <staggerlytom> TW off till new CNC ready. Spanish EDM, probs with Spanglish control & Yaskawa Amps.
[16:13:02] <dave-e> well that should keep you off the streets for awile
[16:13:05] <staggerlytom> * staggerlytom Hello Matt
[16:13:20] <dave-e> morning Matt
[16:13:35] <staggerlytom> yep, for since Galesburg... & When is the re-union?
[16:14:17] <dave-e> someone said Sep but that gets really messy for me
[16:14:46] <staggerlytom> Sep means there should be more discussion soon.
[16:14:50] <dave-e> parts of Oct are also iffy
[16:15:13] <dave-e> I still don't have my docs back
[16:15:25] <dave-e> so if you happen to be going that way
[16:15:32] <mshaver> hey guys!
[16:15:42] <dave-e> ah Matt is awake!
[16:15:58] <staggerlytom> I >will< go out there & wiil get 'em, just being slow on this control wiring...
[16:16:41] <mshaver> I can go back too to help...
[16:17:20] <dave-e> I've gotten stalled on mine....I'm just trying to get so I can make parts
[16:17:58] <dave-e> i.e. I'm stalled on the plc/tool change
[16:17:58] <paul_c> \o/ All the farm slots are returning the expected results now..
[16:18:40] <dave-e> got to the point where i could read magazine position
[16:18:56] <dave-e> but not automatically select it...yet
[16:19:03] <staggerlytom> nice job Paul, want to compare the files to learn what you had to do.
[16:19:11] <jmkasunich> * jmkasunich is having fits trying to get ntp working on the farm
[16:19:18] <staggerlytom> Dave, what are you suing? ClassicLadder?
[16:19:33] <dave-e> just plain C at this point
[16:20:03] <staggerlytom> plain C and inbyte/outbyte or using any rtai/emc stuff?
[16:20:05] <dave-e> classicladder ... even standalone would be nice
[16:20:20] <dave-e> inb outb
[16:21:13] <staggerlytom> to make it more interactive, try TclTk.. not more interactive... less compiled :-)
[16:21:35] <dave-e> I don't do tktcl well
[16:22:17] <staggerlytom> hmmm, seemed like Ray showed me a tcltk widget that'd do inb/out over a wide range,
[16:22:55] <staggerlytom> like a while 8255, so lotsa i/o off a single pci card , might let you model stuff faster
[16:22:55] <dave-e> yeh, Ray has about half the tool change already in tkio
[16:23:27] <staggerlytom> i must miss a lot, where did you see Rays' 1/2 a tool changer?
[16:23:38] <dave-e> I have a couple of the cards like the one we used at Galesburg
[16:23:58] <dave-e> also picked up a Natl 96 i/o
[16:24:19] <dave-e> go look at the code in tkio
[16:24:34] <staggerlytom> WinFord? and Yeah I got a 96 too...
[16:25:03] <staggerlytom> tkio... was studying it prior to Galesberg, not since..
[16:25:34] <dave-e> If I end up using the natl...I need to get the cards that split it out into opto-22 compatible
[16:25:58] <staggerlytom> maybe use LabView on Win sys to model up what the Natl 96 will do ( freebee 30 day demo vrsn )
[16:26:07] <dave-e> hmmm
[16:26:40] <dave-e> I think it is 8255 so should not be too difficult
[16:27:07] <staggerlytom> I bought Natl's cable for the 96, it's a Y and splits out to 2 opto 22 hdrs ( a bundle with card cable & brkouts )
[16:27:12] <dave-e> gotta run to town and get a bolt...so i can finish a jig
[16:27:18] <staggerlytom> bye
[16:27:30] <dave-e> see ya later
[16:27:37] <steve_stallings> Any interest in ModBus? Does Classic Ladder talk to ModBus?
[16:28:05] <steve_stallings> dang, got to type faster
[16:28:50] <paul_c> mshaver needs ModBus
[16:29:37] <steve_stallings> what is he up to with ModBus?
[16:30:42] <paul_c> some fancy VFD on a Lathe
[16:31:41] <steve_stallings> OK, I was considering doing something to just supply lots of low speed I/O points via ModBus
[16:32:36] <paul_c> would the "something" also include PLC functionality ?
[16:32:51] <steve_stallings> idea being to avoid having yet another card to find a slot for and fat cables
[16:33:06] <robin_sz> is modbus an open standard?
[16:33:13] <steve_stallings> nope, just dumb I/O, that is why I asked about Classic Ladder
[16:33:16] <mshaver> or I just need to get that vfd working via rs-232, or give up and add a relay to turn it on & off... ;)
[16:33:35] <steve_stallings> ModBus is open standard, but you can pay for drivers if you like 8-)
[16:34:18] <robin_sz> ok, I didn't know if it was ... I saw somehting recently about CANopen over modbus ... might be interesting
[16:34:30] <staggerlytom> http://membres.lycos.fr/mavati/classicladder/ lists the Comedi drivers that Classic Ladder >could< use. I dont see any of the newer Bus-es
[16:34:47] <paul_c> steve_stallings: Got any links to docs for the low level protocols ?
[16:34:59] <steve_stallings> somewhere, let me look
[16:35:17] <robin_sz> I presume we are talking modbuss tcp here?
[16:35:52] <steve_stallings> http://www.modicon.com/techpubs/toc7.html
[16:36:34] <paul_c> CL has a client module for ModBus...
[16:36:43] <steve_stallings> tcp possibly, but I was mainly thinking simple serial because every PC has it
[16:37:37] <robin_sz> eh?
[16:37:47] <robin_sz> * robin_sz passes steve_stallings a typical laptop
[16:38:27] <robin_sz> serial seems to be disappearing fast
[16:38:37] <steve_stallings> USB to serial adapters could work with serial ModBus
[16:38:45] <robin_sz> thats why all these USB to serial dongles are appearing
[16:39:08] <steve_stallings> but I don't like using laptops for CNC controllers
[16:39:47] <robin_sz> well, possibly not .. but I suspect ethernet is going to be around longer than serial ports on things
[16:40:04] <paul_c> OK.. Looks like there might be enough there to write an NML<->ModBus connector.
[16:40:52] <steve_stallings> thoughts about using ModBus to access simple I/O points vs. external PLC functions ??
[16:41:33] <mshaver> wanna read something interesting & funny? (relates to plcs & modicon): http://www.barn.org/FILES/historyofplc.html
[16:42:50] <mshaver> modbus seems to be mach's way of expanding their hardware horizons beyound the parallel port
[16:43:39] <robin_sz> steve_stallings: it shouldn't make a whole lot of difference I guess . HAL should be able to interface to a PLC or a mdobus connected IO device and present a consistent interface to EMC, whether the PLC logic is odff baord or in HAL shouldne make much difference
[16:43:58] <mshaver> they're also (i think) relying on those little inexpensive hardware plcs to do things like toolchanger control (i really have to get mach up here to undrerstand it...)
[16:44:30] <robin_sz> that must be mach3, dont htink it was in mach2
[16:44:43] <mshaver> <nods> what robin said about HAL intf
[16:44:51] <mshaver> yep, m3
[16:45:04] <robin_sz> not tried m3 ...
[16:46:45] <staggerlytom> Paul, ModBus & CL been done, code freely available... http://lintouch.org/pipermail/lintouch-dev/2005-January/000608.html
[16:46:53] <mshaver> modbus would seem to give us a "no backplane card" way of getting more slow i/o
[16:47:08] <robin_sz> I hear they have some sort of click and drool "stability tester" now that tries to estimate how likely it is to run on a particular PC
[16:48:47] <mshaver> mach does?
[16:48:53] <robin_sz> yeah
[16:49:02] <robin_sz> again, not tried it
[16:49:26] <mshaver> probably tests APIC timer functionality (in part)
[16:49:41] <robin_sz> probably ...
[16:50:02] <robin_sz> but everytime I think about trying it .. I go sort of funny and start twitching
[16:51:35] <mshaver> i thought you were fairly happy with mach (as if anyone is really happy with any cnc control solution...)
[16:52:02] <robin_sz> fsck no
[16:52:12] <mshaver> oh
[16:52:48] <robin_sz> motion thats stops, waits 5 seconds, delivers all the missing pulses in one hit and carries on? I think not.
[16:53:09] <steve_stallings> 8-)
[16:53:31] <robin_sz> and the radnomenss of Arts changes, coupled with closed source ... nah. not going there again.
[16:53:41] <mshaver> really, it does that? i've never used it... not that emc is perfect...
[16:53:57] <robin_sz> both have there plus and minus points
[16:54:27] <steve_stallings> lets see now, file is named A, banner says B, about says C and nobody saved a copy of the previous version, which by the way had the same name....
[16:54:45] <robin_sz> thats part of it ...
[16:54:53] <robin_sz> if Art had CVS it would be a start
[16:55:06] <robin_sz> he has NO WAY of going back ... can you imahine?
[16:55:09] <robin_sz> imagine
[16:55:15] <paul_c> cough svn cough
[16:56:21] <robin_sz> his release numbering is totally random ... I have neve had any clue which is supposed to be stable
[16:56:51] <robin_sz> but .. the macros and screen designer rock.
[16:57:21] <staggerlytom> subversion , social venture network, signal versus noice... must be #1 or #3
[16:57:28] <robin_sz> subversion
[16:57:40] <robin_sz> theres tortoise SVN for windows
[16:57:59] <Jymmm> Morning Ladies
[16:58:00] <mshaver> sorry about the whole thing on the geckodrive list - saw your post to rabbit-semi - i would have expected a bigger reaction on that list - i'm curious to see what steve hardy digs up at work this coming week
[16:58:11] <robin_sz> sorry/ why?
[16:58:23] <robin_sz> nowt to be sorry about surely?
[16:58:53] <robin_sz> if it is how Dan said, then I'll send the whole ting back in a shoebox with a dog turd :)
[16:59:13] <robin_sz> right now, i'm assuming he is mistaken
[16:59:22] <mshaver> i don't want to come off as too tedious... no one else seems too concerned
[17:00:08] <jmkasunich> I'm concerned
[17:00:12] <robin_sz> it wold be a show stopper I think ... I'd point out all the code shared on the rabbit semi list and mention laches to their lawyers and see what happens
[17:00:50] <paul_c> Cooo... The lintouch message structures are not that much different to those of an NML message.
[17:00:59] <staggerlytom> nifty!
[17:01:35] <paul_c> A limited set of data types...
[17:01:50] <robin_sz> laches ... basically if you point out a breach and they dont do anything, then they lose the right to do anyting about it 9or similar events) later
[17:02:04] <robin_sz> kinda force their hand
[17:02:38] <robin_sz> mshaver: I just cant beleive anyone woudl construct a licence that fuckwitted
[17:03:02] <Jymmm> sounds like it's due to anti-hardware cloning
[17:03:28] <robin_sz> nah
[17:03:35] <mshaver> there must be some miscommunication between them & their lawyers who wrote the language
[17:03:45] <robin_sz> i hope so ...
[17:03:51] <Jymmm> mshaver: Dont you believe that
[17:04:05] <Jymmm> I've seen much worse EULA
[17:04:16] <Jymmm> Heh, look at XP's
[17:04:26] <mshaver> Jymm: I give everyone the benefit of the doubt
[17:04:37] <robin_sz> even XP doesnt say "anyting uou write on XP, you cant publish"
[17:04:52] <Jymmm> mshaver: Not when lawyers are around
[17:05:09] <mshaver> :)
[17:05:21] <Jymmm> robin_sz: No, but it does say they can go thru your entire hdd
[17:05:42] <robin_sz> it does of course mean no Dynamic C app can use any open source code
[17:06:13] <mshaver> well, gpl anyway...
[17:06:14] <paul_c> M$ also says you can not modify the behaviour of Windows... and what does M3 do ?
[17:06:20] <robin_sz> well, yeah GPL
[17:07:04] <Jymmm> M3?
[17:07:18] <mshaver> paul_c: that's an interesting thought - is that in their license?
[17:07:39] <paul_c> Last time I looked... Yes.
[17:07:44] <mshaver> wow
[17:08:21] <paul_c> Also a clause in there preventing you from projecting a M$ session vis vnc to another machine.
[17:09:35] <les> just got in from tpfix tests on the big machine
[17:09:39] <les> looks great
[17:09:42] <robin_sz> wow
[17:09:44] <mshaver> ok, but you can at least see that by prohibiting this they're trying to get you to buy their terminal server product...
[17:09:47] <mshaver> YEA!
[17:10:29] <les> had to come in to grab our spiral program...and run that
[17:10:51] <robin_sz> les so you can dispense with your modified code with inverse wiggles now?
[17:11:03] <les> yes
[17:11:22] <les> I will have to rem out the antivibration in the production code
[17:11:38] <les> i don't mind!
[17:11:49] <mshaver> because now it will probably ADD vibration!
[17:12:00] <les> it will!
[17:12:13] <robin_sz> amazing ...
[17:12:38] <les> I just now wanted to check the stutter bit with the spiral at high speed
[17:12:47] <les> that may be another issue
[17:12:49] <les> but
[17:13:13] <les> I'll bet this will help because i'll bet every segment was misblended
[17:13:21] <robin_sz> one bug at a time
[17:13:29] <les> bluh
[17:13:33] <les> yeah
[17:13:47] <les> high speed stutter may be a comms thing
[17:15:05] <anonimas1> les: : does it work?
[17:16:14] <anonimas1> mshaver: : actually the terminal server part comes with 2k server
[17:16:45] <anonimas1> every 2k/xp seat are already licensed to use the terminal server alreay..
[17:16:58] <anonimas1> :)
[17:17:04] <les> it works anders
[17:17:05] <Jymmm> anonimas1 : a SINGLE license to use it
[17:17:09] <les> I am happy!
[17:17:09] <anonimas1> YAY!
[17:17:15] <anonimas1> Jymmm: no.
[17:17:22] <Jymmm> anonimas1 Yes
[17:17:34] <anonimas1> Jymmm: it comes with 2 licencing options per client or per computer
[17:17:36] <mshaver> ok, didn't know that
[17:17:43] <Jymmm> anonimas1 for every connection to a terminal server you MUST have a client license
[17:18:04] <gezr> howdy les
[17:18:06] <Jymmm> anonimas1 same goes with file/print sharing
[17:19:01] <Jymmm> anonimas1 And if you really dig into the EULA, just run run IIS you have to have a client license for every connection that comes in.
[17:19:15] <anonimas1> Jymmm: yes but every 2k/xp license comes with a license to use the terminal server
[17:19:41] <Jymmm> anonimas1 to USE ther server, yes. but to have client CONNECT to it is another story.
[17:21:24] <anonimas1> Jymmm: I know their licensing that's why m$ havent raided the office yet
[17:21:39] <anonimas1> ;)
[17:21:43] <anonimas1> * anonimas1 kicks the server
[17:21:59] <les> well back to the machine...back in a bit to report more
[17:22:19] <anonimas1> ok great
[17:22:49] <anonimas1> les try blending something at a high speed
[17:22:50] <anonimas1> :)
[17:24:29] <Jymmm> les : Yeah, blend a Cadillac Margarita at high speed
[17:25:18] <anonimas1> heh
[17:25:41] <anonimas1> Jymmm: the nt server is eating away my USC playtime...
[17:25:46] <anonimas1> *cries*
[17:26:02] <Jymmm> anonimas1 then dont use nt server
[17:26:48] <anonimas1> Jymmm: the management software we use at work wont run at another kind of server
[17:27:03] <Jymmm> SMS ?
[17:27:10] <anonimas1> besides I need terminal services..
[17:27:13] <Jymmm> lol
[17:27:16] <Jymmm> sucker
[17:27:19] <anonimas1> yeah...
[17:27:29] <anonimas1> but there's no good way to have 2 places far away run the software
[17:27:32] <Jymmm> nobody NEEDS terminal services, not on M$ anyway
[17:27:46] <anonimas1> Jymmm: you are being childish..
[17:28:35] <anonimas1> Jymmm: I tried running the software over the VPN.. the place I tried to had a 4 mbit connection
[17:28:47] <anonimas1> searching through the customer db took 15 minutes..
[17:28:51] <anonimas1> instead of ~2secs
[17:30:39] <robin_sz> anonimas1: can you connect windows ot a network then?
[17:30:55] <anonimas1> robin_sz: network?
[17:31:06] <robin_sz> like the internet
[17:31:24] <anonimas1> robin_sz: yes, through the BSD firewall.. :)
[17:31:35] <robin_sz> ahh, they got that part to work then .. nice
[17:31:44] <anonimas1> actually it's a bridge..
[17:31:49] <anonimas1> that does filtering
[17:31:57] <anonimas1> and I do nat on the NT server..
[17:32:21] <Jymmm> anonimas1 how many servers are you dealing with?
[17:32:29] <anonimas1> just one..
[17:32:56] <Jymmm> anonimas1 : when you get to admin 40 locally and 120 remote, let me know.
[17:33:18] <Jymmm> wait a sec... ONE? and you're running SMS ?!
[17:33:23] <Jymmm> WTF?!
[17:33:43] <anonimas1> no
[17:33:52] <anonimas1> where did the sms come from?
[17:33:59] <jmkasunich> hmmm... I thought BDI-4.20 had a 2.6 kernel?
[17:34:12] <Jymmm> anonimas1 scroll back
[17:34:36] <jmkasunich> _sem.c:188: warning: #warning Please consider upgrading your kernel to 2.4.22 or higher and installing a recent glibc to take advantage of more efficient system calls.
[17:35:09] <anonimas1> hm
[17:35:10] <Jymmm> jmkasunich gimme a sec
[17:35:15] <anonimas1> cant find it..
[17:35:35] <anonimas1> dosent matter really, I just want to get the updates installed and go home for dinner
[17:35:53] <Jymmm> anonimas1 which year! lol
[17:36:01] <anonimas1> perferably today :)
[17:36:17] <anonimas1> 7 minutes left on the service packs
[17:36:17] <Jymmm> anonimas1 and why only one? no other DC's?
[17:36:22] <paul_c> fsckit.... A define is missing..
[17:36:26] <anonimas1> Jymmm: there's no need for another DC..
[17:36:29] <anonimas1> :)
[17:36:33] <anonimas1> too few computers
[17:37:06] <Jymmm> anonimas1 : There's ALWAYS a need for a BDC, it IS M$ we're tlakign about
[17:37:27] <jmkasunich> yeah, it is... on a Linux/EMC channel
[17:37:41] <jmkasunich> how about /msg'ing that cruft
[17:38:35] <Jymmm> 2.6.10-adeos #1 Sat Mar 5 22:56:08 GMT 2005 i686 GNU/Linux
[17:38:53] <Jymmm> That's what I have for BDI 4.20
[17:39:06] <les> back
[17:39:12] <jmkasunich> yes, I know... I was commenting about the warning that happens during the compile, there is a problem somewhere, that shouldn't appear on BDI-4
[17:39:37] <Jymmm> jmkasunich Thre are lots of issues with bdi 4.20 =)
[17:39:40] <les> spiral is much smoother, but it still finally hits commschoke and stutters
[17:39:54] <les> until then it is dead smooth though
[17:39:58] <anonimas1> les: when is finally?
[17:40:13] <anonimas1> and how fast are you running?
[17:40:22] <les> let me check here..
[17:40:25] <anonimas1> ok
[17:40:25] <anonimas1> :)
[17:40:26] <paul_c> les: The last half turn ?
[17:41:19] <les> at 60 with the 4" redius spiral you mailed first half turn
[17:41:42] <les> at 20 almost to the end
[17:43:34] <les> so that's about .005" to .01" at 1 ips so
[17:44:01] <les> .005 to .01 seconds is the min segment period
[17:44:31] <paul_c> jmkasunich: fsckit.. Looks like one of the headers I was expecting to be included, isn't.
[17:44:56] <les> so max block processing/planning speed seems to be about 200/sec
[17:45:13] <les> not so great by today's standards
[17:45:15] <anonimas1> les: what would happen if you increased your period
[17:45:20] <jmkasunich> at least it happens on BDI-4, easier to test without waiting for the farm
[17:46:21] <jmkasunich> paul_c: is all this stuff just "ripple effect" from adding GNU_SOURCE?
[17:46:24] <les> well I am at 500 microseconds servo, 5000 trajectory
[17:46:46] <jmkasunich> 5000uS traj = 200 blocks/sec
[17:46:58] <les> exactly
[17:47:03] <anonimas1> maybe you should increase it..
[17:47:09] <les> world is round
[17:47:23] <anonimas1> heh
[17:47:39] <les> I can increase it....however cubic subinterpolation will be reduced
[17:47:45] <jmkasunich> upping the block rate with fixed traj period will require significant changes, not a simple bug fix
[17:48:05] <anonimas1> jmkasunich: lookahead?
[17:48:23] <Jymmm> paul_c: You know that tkemc that alex gave me yesterday? Do you knoww what I need to start on the server to run that?
[17:48:25] <jmkasunich> something like that, plus changes to the comms to allow more than one message per period
[17:48:27] <les> Well we had hoped it was just raw computing speed needed...but found that is not the case
[17:48:31] <anonimas1> yeah
[17:49:04] <les> about 400 microseconds is the brick wall now-with any machine
[17:49:10] <les> servo
[17:49:17] <anonimas1> jmkasunich: how much work is it to re-work it so you can cache the next move at the servo period..
[17:49:19] <paul_c> Jymmm: Did you get the note I sent you ?
[17:49:26] <Jymmm> paul_c no
[17:49:39] <jmkasunich> anon: no idea, I'm not intimate with that part of the code
[17:49:43] <anonimas1> is that the shmem issue?
[17:49:53] <paul_c> jmkasunich: No. the semtimedop is a result of a "bug"
[17:50:29] <paul_c> Jymmm: http://emc.sourceforge.net/Handbook/node111.html
[17:50:38] <jmkasunich> ok, just never showed up before GNU_SOURCE was added to the makefile
[17:50:59] <anonimas1> paul_c: are you more intimate with that part?
[17:53:07] <paul_c> jmkasunich: You're going to have to write an m4 macro for configure to test for semtimedop
[17:53:22] <jmkasunich> I am?
[17:56:01] <paul_c> If you want to use semtimedop, yes.
[17:56:27] <jmkasunich> I haven't the slightest idea what semtimedop is, that is part of libnml
[17:56:53] <paul_c> no. It's a libc/syscall
[17:57:11] <jmkasunich> nitpicking... it is called from part of libnml
[17:57:35] <paul_c> bigger nits - It's called from emc.
[17:57:54] <jmkasunich> and until recently, it worked...
[17:58:09] <paul_c> Part of the memory locking mechanism within libnml.
[17:58:10] <jmkasunich> so what changed in the last week that made it stop working?
[17:58:32] <paul_c> It worked most of the time.
[17:59:22] <paul_c> anonimas1 and a few others were experiencing certain problems with auto/manual mode changes.
[17:59:51] <anonimas1> it appears at higher period settings..
[17:59:54] <jmkasunich> ok, now I'm starting to understand... the probs were caused by poor locking, you added a bugfix to fix the locking
[18:00:19] <jmkasunich> the bugfix requires semtimedop ( I assume )
[18:00:21] <paul_c> either the changes you made to the low level motion are borked, or there is an occasional problem with a semaphore.
[18:00:40] <jmkasunich> dunno what changes you are referring to
[18:00:49] <jmkasunich> low level motion doesn't use libnml
[18:00:50] <paul_c> pfft.
[18:00:53] <anonimas1> it's quite a while since it stopped working..
[18:01:29] <jmkasunich> pfft is very helpfull
[18:04:02] <paul_c> So... Do you want me to try and fix the problem anonimas1 reported ?
[18:04:20] <jmkasunich> of course
[18:05:54] <jmkasunich> hi alex
[18:06:06] <les> Hey paul...I forgot our test stuuf...did we go faster without full debug?
[18:06:37] <alex_joni> hello
[18:06:46] <les> hi alex
[18:06:49] <paul_c> les: Yes.
[18:07:11] <les> ok. I'll crank that down
[18:07:19] <paul_c> set it to zero.
[18:07:24] <les> ok
[18:07:28] <alex_joni> what's up guys?
[18:07:51] <les> Moved big iron fast with tpfix. looks great
[18:08:01] <alex_joni> nice ;)
[18:08:06] <les> yeah!
[18:08:27] <alex_joni> your mill?
[18:08:37] <alex_joni> how fast?
[18:08:44] <les> Now I can buy the big spindle and double the speed
[18:08:59] <les> I ran at 200 or so
[18:09:19] <les> Have to change the ini
[18:09:52] <les> I can go up to about 600 if power supply doesn't bog
[18:10:34] <les> blended segment paths now seem very smooth as long as they are not too close
[18:11:19] <les> I will do zero debug and drop traj period a bit
[18:11:40] <les> but dropping traj period lessens cubic smoothing
[18:12:08] <les> we really need just faster overall servo times
[18:12:45] <les> Lots of comms overhaed according to paul
[18:12:52] <anonimas1> yep
[18:13:18] <anonimas1> 5m/min..
[18:13:23] <anonimas1> perfect! :)
[18:13:38] <les> yeah
[18:13:49] <anonimas1> I'll be able to crank my mill way up
[18:14:09] <anonimas1> les: are you going to be making turkey calls tonight ? ^_^
[18:14:15] <les> I can't use it without a bigger spindle though
[18:14:30] <anonimas1> les: it'll shorten your days still wont it?
[18:14:38] <les> We have to start next year's production soon
[18:15:00] <les> It will shorten the days...when I buy the bigger spindle
[18:15:26] <les> I did not before...because I could not run the machine fast due to the misblends
[18:15:55] <paul_c> les: Apologies to your sanding guy....
[18:16:15] <les> But I will buy a spindle in the 5- 7kW range as soon as I find a good deal
[18:16:27] <paul_c> He is going to have to spend more time chasing girls/"little white balls"/fish
[18:16:42] <anonimas1> paul_c: or do painting ;)
[18:17:08] <les> ha paul....here's the plan....our costs drop...so we will have more sales...so he will be sanding just as much....
[18:17:46] <les> No one wants to paint...it's not fun being in the booth with all that protective grear on
[18:18:08] <les> That's why I designed the paint robot
[18:18:33] <les> have not built it yet...have this research contract to finish
[18:18:51] <les> and as you know it looks like that is getting hot...red hot
[18:19:05] <Jymmm> les did you confirm your 1W ?
[18:19:43] <les> Nope...just calcs from the capacitance of the PZT device
[18:19:50] <les> but physics works
[18:20:01] <les> I will be on it tommorow though
[18:20:37] <les> It uses so little air....I have leaks that consume more air than that thing
[18:21:14] <paul_c> * paul_c is gone.
[18:21:53] <les> size of cigarette butt....tiny air consumption...no moving parts....enough power to run tronics
[18:22:07] <jmkasunich> cool...
[18:22:15] <jmkasunich> isolated power supply for remote sensing
[18:22:24] <les> john I am pretty excited
[18:22:33] <steve_stallings> duh! I missed something. What is Les working on?
[18:22:51] <les> I was about to give up, write the research report saying it was not practical
[18:23:07] <jmkasunich> just guessing... air blows over a piece of piezo material, makes it vibrate, extract power from that
[18:23:13] <anonimas1> yeah
[18:23:24] <jmkasunich> like a reed in an wind instrument
[18:23:26] <anonimas1> les: I am just curious, why havent that been done before?
[18:23:26] <les> Les is working with power scavenging and had a breakthrough result friday
[18:23:52] <gezr> les : its good to see you happy
[18:23:57] <anonimas1> isnt it the reason piezo materials were made..
[18:23:58] <anonimas1> :)
[18:24:07] <jmkasunich> les: how much pressure is needed?
[18:24:35] <les> haha....I don't know. But I have invented things for a living a good chunk of my career, and most better not be something already done
[18:24:46] <les> John I ran at 40 psi
[18:24:49] <jmkasunich> * jmkasunich sees an application for hot-stick mounted instruments, pressurize the stick
[18:25:22] <jmkasunich> shine a modulated LED back down the inside of the stick to a readout in the handle
[18:25:27] <les> The trick was to make a little precision lapped adjustable cavity resonator
[18:25:57] <les> I also changed some stuff to get a higher mach number
[18:26:20] <jmkasunich> now you need to make a highly efficient air pump to go with
[18:26:34] <Ron_B> will it have an aviation use?
[18:26:48] <les> it only uses perhaps .01-.1 cfm
[18:27:01] <alex_joni> jmkasunich: the air pump is there
[18:27:10] <alex_joni> remember this is foar coating ;)
[18:27:10] <anonimas1> ship les with the product ;)
[18:27:18] <les> Ron, I don't know. It could have many uses.
[18:27:25] <jmkasunich> ok, I didn't know the original application
[18:27:51] <les> I am still dumbfounded that I got that much signal. Checked it several times though.
[18:27:53] <Ron_B> just thinking aviation uis usuallly a high-pay project
[18:28:13] <Ron_B> and for small misiles might power a guadance system
[18:28:24] <anonimas1> yeah
[18:28:36] <anonimas1> no moving parts power generation..
[18:28:38] <Ron_B> figure the typo out
[18:28:52] <jmkasunich> the reed or whatever is a moving part
[18:29:15] <les> I have only done two aerospace products that I can recall, even though I am an aerospace engineer.
[18:29:19] <jmkasunich> for something like a missle, batteries are probably easier - limited duration load after all
[18:29:31] <Ron_B> engineering be engineering
[18:29:33] <anonimas1> jmkasunich: I think batteries would cost more.. maybe
[18:29:42] <les> yeah
[18:29:49] <anonimas1> the piezo stuff is dirt cheap :)
[18:29:54] <jmkasunich> but where are you gonna get the air?
[18:30:06] <anonimas1> jmkasunich: you launch missiles dont you?
[18:30:14] <les> I just had to tell someone....but patent lawyers will be shutting me up heh
[18:30:17] <Ron_B> in a missile?
[18:30:20] <jmkasunich> yeah, but you kinda loke to have them powered up first
[18:30:45] <Ron_B> at rest there is a battery pack
[18:31:04] <anonimas1> I hope your lawyers dont google :)
[18:31:10] <Ron_B> at speed there will benumerous "taps"
[18:31:23] <jmkasunich> ram air IOW
[18:31:28] <les> There are some military energy scavenging projects out there
[18:31:32] <anonimas1> you can even have several.. and generate lots of power
[18:31:37] <les> they are getting milliwatts
[18:31:40] <anonimas1> les: will that scale?
[18:31:54] <les> yes
[18:31:57] <anonimas1> :)
[18:32:02] <jmkasunich> lots is relative, missile guidance needs to drive fin actuators
[18:32:09] <anonimas1> generating KW's ;)
[18:32:25] <jmkasunich> using hurricanes ;-)
[18:32:31] <anonimas1> yeah
[18:32:32] <anonimas1> hehe
[18:32:34] <les> it's all in the impedance matching.....
[18:32:42] <anonimas1> or in airplanes..
[18:33:06] <les> although that is kind of a misnomer with dispersive shock wave propagation
[18:33:22] <les> impedance implies linearity
[18:33:23] <jmkasunich> actually it could be nice for missles, infinite shelf life compared to batteries
[18:33:51] <anonimas1> jmkasunich: yeah and cheap... :)
[18:34:01] <jmkasunich> power level probably an issue tho
[18:34:14] <les> I could get a few joules from a match head I guess
[18:34:19] <les> electrical
[18:34:28] <les> heh
[18:34:42] <jmkasunich> hmmm... for air-to-air missiles, it could be part of the safe/arm system... no 200knot airstream, no shoot missle
[18:35:01] <anonimas1> somehow I hope the US wont use this for war..
[18:35:25] <Ron_B> any development will be used for war
[18:35:36] <Ron_B> if not by the US, someone else -
[18:35:52] <anonimas1> * anonimas1 imagines emc on some unclear submarine
[18:36:03] <Ron_B> unless somehow a law can be passed toend human nature
[18:36:17] <les> I mostly make consumer products and the machines that make them
[18:36:18] <Ron_B> and be enforced
[18:36:58] <jmkasunich> yeah... the missile thing was speculation
[18:37:04] <les> Did those heated clothing research for the UK ministry of defense a while back.
[18:37:12] <anonimas1> jmkasunich: probable use..
[18:37:12] <jmkasunich> the hot stick application is more interesting to me
[18:37:17] <les> you know, the ptc stuff
[18:37:55] <jmkasunich> and other applications needing isolated power because of high voltage
[18:38:18] <jmkasunich> air hose makes a nice isolator ;-)
[18:38:41] <jmkasunich> I wonder if transparent hoze can be used as an optical fiber?
[18:38:49] <jmkasunich> power and signal over the same path ;-)
[18:39:13] <anonimas1> lol
[18:39:30] <steve_stallings> there is one company doing a shaft encoder that is entirely optical
[18:39:40] <jmkasunich> I'm always looking for better isolation for my day job
[18:39:41] <steve_stallings> no EMI issues
[18:40:01] <anonimas1> jmkasunich: what do you work with?
[18:40:08] <jmkasunich> need to get analog and digital signals across a 1KV or so barrier, with 50KV/uS slew rates
[18:40:14] <jmkasunich> large VFDs
[18:40:19] <anonimas1> nice
[18:40:22] <jmkasunich> 150kW and up
[18:40:29] <anonimas1> but very scary :)
[18:40:34] <jmkasunich> yeah...
[18:40:55] <alex_joni> anonimas1: how's an unclear submarine?
[18:40:57] <jmkasunich> my boss and I were talking the other day, wishing we worked on stuff that failed without blowing up
[18:40:57] <alex_joni> lol
[18:41:18] <anonimas1> alex_joni: make sense please?
[18:41:34] <alex_joni> anonimas1: read back on what you said
[18:41:45] <anonimas1> lol
[18:41:51] <anonimas1> I cant type :D
[18:41:55] <alex_joni> ;)
[18:45:35] <les> jmk with that high power stuff you work on failed components must be exciting
[18:45:43] <jmkasunich> yeah
[18:45:54] <jmkasunich> sounds like a 12GA shotgun going off
[18:45:59] <les> heh
[18:46:09] <wb9mjn> I blew up a 5 KW motor drive once....louder than a shot gun!
[18:46:18] <les> yikes
[18:46:28] <jmkasunich> we use laminated bus structures, two 1/8" copper plates separated by a 0.005 or so Valox insulator
[18:46:38] <wb9mjn> Can t imagine 125 KW ...
[18:46:43] <les> I have popped a few big caps
[18:46:46] <wb9mjn> 150...
[18:46:52] <jmkasunich> we hade a couple field failures where the valox punched thru
[18:47:10] <jmkasunich> the copper literelly had a hole blown in it
[18:47:16] <Jymmm> I blew up a 5KVA transformer once .... stunk the shop up for two weeks with all that burnt resin it was encased in.
[18:47:20] <les> must have been from abrasion?
[18:47:24] <jmkasunich> as if you had detonated a small quantity of explosives
[18:47:33] <wb9mjn> The electrician rewired the terminal block, and I wired it up to the labels...which he never changed....one of the screws was stripped out..
[18:47:51] <jmkasunich> dunno the root cause (of course the failure point was blown to hell)
[18:48:25] <les> This reminds me...now I can go ahead and buy the 5 or 10 kw 400 hz inverter....
[18:48:32] <jmkasunich> abrasion shouldn't happen, there is adhesive between cu and valox, everything is laminated to prevent relative motion
[18:48:34] <les> now that the cnc can go fast
[18:48:41] <wb9mjn> At work we mold combo Dsub insulators from Valox....
[18:48:50] <jmkasunich> I suspect a burr on the copper, poked thru the valox under heat and pressure
[18:49:22] <les> I use Valox a lot too...it's the phenolic of the 21 century
[18:49:22] <staggerlytom> thats EDM! the closest 2 points is where the dielectric is weakest
[18:50:00] <Jymmm> les valox is flame proof?
[18:50:00] <steve_stallings> Electrical Destruction Magnificant ?
[18:50:01] <wb9mjn> We keep getting underbid by people who regrind it,,,which turns it into stuff about as strong as phenolic....
[18:50:12] <jmkasunich> EDM with one fscking huge discharge instead of a lot of little ones
[18:50:25] <staggerlytom> yeh, just 1
[18:50:31] <les> yeah phenolic and fish paper used to be the rule
[18:50:50] <jmkasunich> approx 32,000uF at 700V
[18:50:51] <les> I don't remember if fish paper was UL 94V0
[18:51:01] <staggerlytom> wow fish paeper & fly dope
[18:51:24] <les> heh
[18:51:36] <wb9mjn> Nope...friend from India said they could never get phenolic to work for high tension insulators...too much water absorbtion...
[18:51:55] <jmkasunich> almost 8kJ
[18:51:58] <les> hmm
[18:52:00] <wb9mjn> He was in a company that did allot of Indian electrification equipment...
[18:52:33] <les> ach I will have to do some electrics rewiring I guess soon
[18:53:20] <les> nedd at least 10 kw in the little shop
[18:53:31] <les> bigger motors everywhere
[18:53:49] <les> now that I have production
[18:53:50] <gezr> hehehe
[18:54:06] <alex_joni> jmkasunich: finally the compile_farm is where it should be
[18:54:13] <anonimas1> jmkasunich: : di
[18:54:15] <jmkasunich> yep!
[18:54:19] <anonimas1> did you get ntp working?
[18:54:24] <jmkasunich> no
[18:54:33] <wb9mjn> BTW, I was a coinventor back in april....new microwave connector....
[18:54:34] <jmkasunich> works on this box, and the 4.20 farm slot
[18:54:51] <jmkasunich> downloaded and tried to build on the BDI-2.20 slot, failed miserably
[18:54:55] <jmkasunich> I said fsckit
[18:54:58] <wb9mjn> Company gave me a little plaque and a bonus...
[18:55:01] <anonimas1> jmkasunich: : strange
[18:55:24] <anonimas1> jmkasunich: probably no big deal..
[18:55:26] <anonimas1> :)
[18:55:45] <les> I only used to get the little plaques wb9.....no bonus haha
[18:55:54] <les> aw that's not quite true...
[18:55:59] <anonimas1> pfft..
[18:56:01] <jmkasunich> the 4.20 slot has the correct time, and I manually set the others to match (within a minute)
[18:56:05] <jmkasunich> close enough
[18:56:10] <wb9mjn> Its a little plaque..hi...they got the big one to put on the wall in the conference room...
[18:56:27] <anonimas1> les: didnt you buy a bmw ?
[18:56:29] <anonimas1> :D
[18:56:29] <gezr> les : I still plan on making a trip over to your shop someday, company wont let me take a vacation right now, but they did let me leave for 3 days to go to a funeral out in cali.
[18:56:35] <les> ha
[18:56:49] <les> anytime gezr.
[18:57:04] <gezr> we have been slamed at work, tons of orders
[18:57:16] <anonimas1> I nede to go for a bit going to have dinner
[18:57:18] <anonimas1> laters everyone
[18:57:21] <alex_joni> * alex_joni wonders
[18:57:24] <alex_joni> where's rayh?
[18:57:38] <jmkasunich> * jmkasunich is going sailing in a bit
[18:57:45] <steve_stallings> I think he is in Europe visiting family somewhere....
[18:57:46] <jmkasunich> dunno, haven't seen ray
[18:57:52] <alex_joni> jmkasunich: nice
[18:58:01] <alex_joni> steve_stallings: probably germany then
[18:58:13] <les> yes...things are sure brighter. This will be the first year I make more sitting on a farm in georgia in the sticks than I ever did as a corporate manager.
[18:58:46] <les> I was beginning to think I made a horrible mistake starting my own firm
[18:59:09] <alex_joni> les: that's nice
[18:59:39] <les> getting better...better than a couple years of only spending money
[19:00:46] <les> a lot of money...that cnc I made here was not inexpensive.
[19:04:26] <les> Seems like the larger machines are a collection of brackets.
[19:05:03] <wb9mjn> les...lets see if you have a take on the problem I am having on my mill...
[19:05:06] <les> That seems like it was all I did every day....machine brackets on the mill.
[19:05:13] <les> for months and months.
[19:05:36] <les> yes what is the problem?
[19:05:53] <wb9mjn> The Z axis comes to a stop with a shudder faster than about 150 in/min setting, only in the up direction, only after at least 3 inches of travel...
[19:06:16] <wb9mjn> below 3 inches of travel, or slower than 150 in per min , no shudder...
[19:06:21] <les> hmm
[19:06:26] <les> servo or step?
[19:06:31] <wb9mjn> servo...
[19:06:34] <les> k
[19:06:39] <les> stg?
[19:06:47] <wb9mjn> motenc-100...
[19:06:51] <les> ok
[19:07:04] <wb9mjn> When I first set this up, it did not do this....
[19:07:18] <les> no play in the components right?
[19:07:31] <wb9mjn> Well.....not confirmed yet....
[19:07:39] <wb9mjn> no following error....
[19:08:10] <les> tight versus loose could require different tuning
[19:08:52] <wb9mjn> it is timing belt 2:1 drive to a .2 inch pitch ball screw, and the kevlar belt is quite tight...
[19:09:09] <les> this is not a Bridgeport right?
[19:09:16] <wb9mjn> on this axis encoder is on the back on fhe motor...
[19:09:25] <les> ok
[19:09:40] <wb9mjn> No...Its a square column mill, with about a 30 pound milling head...
[19:10:02] <les> rong fu?
[19:10:26] <wb9mjn> No....its a D+M 6,,,but that probably does not mean much to you...
[19:10:38] <les> picture?
[19:11:01] <wb9mjn> let's see if I have one....
[19:11:31] <les> The column mills have a really big pole pair in z due to bending of the column
[19:11:59] <les> if your servo bandwidth reches that pole phase margin will be gone
[19:12:21] <les> emc does not have a lowpass/band reject in the servo loop
[19:12:49] <wb9mjn> Not sure what EMC has for feedback function...
[19:13:26] <les> just pidff...it does have a digital filter in the d section
[19:13:39] <les> not sure it works properly
[19:13:43] <wb9mjn> found a picture....
[19:13:55] <les> it's called SMOOTH_D or something
[19:14:35] <gezr> wb9mjn : does it stall if its already moving slowly then the speed is increased?
[19:15:43] <wb9mjn> gezr, I will write a little g-code to see...have not noticed it...
[19:16:07] <les> waiting for picture...is it a url?
[19:16:12] <wb9mjn> I did the EMC mascott program...
[19:16:45] <wb9mjn> probably could email it....do not have picture URL...
[19:17:01] <les> dcc transfer?
[19:17:10] <wb9mjn> how do you do that ?
[19:17:31] <wb9mjn> I m using Pirch...it has a thing in the tools DCC File sender....
[19:17:43] <les> on mirc just "send file" on a pull down box
[19:17:58] <les> give it a try
[19:19:18] <wb9mjn> did you get something ?
[19:20:11] <wb9mjn> One more try....should be getting something now...
[19:20:27] <les> no, I should see a request to send window pop up
[19:20:42] <jmkasunich> jmkasunich is now known as jmk_away
[19:24:14] <wb9mjn> I think I need more of an address than les ???
[19:24:36] <alex_joni> wb9mjn: usually you right click on the user you want to send to
[19:24:44] <alex_joni> and select the option DCC send from tehre
[19:24:45] <alex_joni> there
[19:24:57] <wb9mjn> Nope....it says it should work with the IRC nicknames....
[19:25:09] <alex_joni> mail it, it's way easier
[19:25:10] <alex_joni> :D
[19:25:30] <les> les should be enough
[19:25:43] <wb9mjn> There is a DCC tool....I selected the file, and put les in the send to box...and it appeared to go ...
[19:26:02] <les> yeah, leswatts@lmwatts.com
[19:26:48] <wb9mjn> Tried that...too, just now...will see if it works that way ?
[19:27:56] <les> ok
[19:28:32] <wb9mjn> the server says the action was cancelled by the freenode...I m looking at the instructions about how to let me send it ...
[19:28:45] <les> well anyway usually big resonace in z on column mills
[19:28:58] <Jymmm> if sender/receiver is running NAT you can't DCC
[19:29:09] <les> turn down gains a bit and see if it gets better
[19:29:21] <les> ferror will go up some
[19:29:56] <les> if that's it you can put a dynamic absorber on and run with higher gains
[19:30:14] <les> 20% effective mass ratio or so
[19:30:48] <wb9mjn> I think the problem is the file needs to be in the PIRCH directory...
[19:31:14] <les> could be...never tried it
[19:31:46] <wb9mjn> Yep...If I turn down the P to about 700, from 7000, leaving the I at 1000, and D at 30, it seems to be better....
[19:31:49] <les> what amps and motors are you running?
[19:32:38] <Jymmm> macintosh amps and bose 901's
[19:32:45] <les> haha
[19:33:03] <les> hey I have run servo motors from audio amps
[19:33:06] <wb9mjn> I m using the Galil MSA -12 - 80 amp with a
[19:33:32] <Jymmm> les I love 901's but nothing to truely drive them with
[19:34:11] <wb9mjn> brushless DC motor...can t find the page in the folder on it right now...
[19:34:24] <les> ok.
[19:34:35] <les> and you are running in torque mode?
[19:34:55] <wb9mjn> Do not think so...
[19:35:22] <wb9mjn> Although the amp can be set that way...
[19:36:03] <les> hmmm that could be an issue...if you are running velocity mode with the amps and also using d in emc you might have a marginal stability fourth order ststem
[19:36:46] <wb9mjn> Motor is a Poly Sci C23 L50 W30 (might be W40) ...
[19:37:09] <wb9mjn> Ooops...should say BRUSHED DC motor above...
[19:37:20] <les> ohhhh
[19:37:24] <les> no tachs?
[19:37:32] <wb9mjn> W designation is for the winding...A W30 winding is for 36 volt nominal...
[19:37:36] <wb9mjn> No tach...
[19:37:53] <les> ok. You are running torque mode. that's fine.
[19:37:56] <wb9mjn> 500 line encoder, on motor....
[19:38:02] <anonimasu> les: I was thinking, maybe you should drop a mail about your TP findings to the list :)
[19:38:17] <les> ok anders I will do that
[19:38:41] <anonimasu> les: great!
[19:39:09] <les> Yeah wb9 I think high loop gain and the inherent column bending mode might be an issue for you
[19:39:19] <anonimasu> hm, I need to head back to work and finish..
[19:39:24] <les> k
[19:39:27] <wb9mjn> As best I can remember, its set to voltage feedback...
[19:39:33] <wb9mjn> the amps that is..
[19:39:45] <les> hmm
[19:39:59] <les> that's just a rebadged copley...let me check
[19:40:50] <wb9mjn> Ok...that is another thing to look at...I am going to have to double check the screw nut too...to be sure its not loose....Along the column issue, the loaded column
[19:41:35] <wb9mjn> will bend and then when the upward motion slows to a stop it recoils....which would increase friction on the dove tails...and that all works into the system
[19:41:39] <wb9mjn> response ?
[19:42:31] <wb9mjn> I can try an experiment where I go from the lowest height, up 3 inches, then from 3 inches below the max height to the max height...and see if
[19:42:34] <les> looking at the amp specs
[19:42:36] <wb9mjn> shudder is different...
[19:42:55] <les> well it could be some mechanical lash
[19:43:00] <les> but
[19:43:11] <wb9mjn> The mass higher up the collumn increase the moement ...
[19:43:15] <les> if you are running voltage mode....
[19:43:27] <wb9mjn> Well, it did not do this previously, and now it is....
[19:43:32] <les> I would suggest running current mode instead
[19:43:46] <les> it is more stable
[19:43:51] <wb9mjn> Ok...
[19:44:18] <les> normally with voltage mode you do not use a d term
[19:44:53] <wb9mjn> Well...without the D term, it definately is unstable! It jumps off the floor...
[19:44:56] <les> emc's pid loop will run voltage, velocity, or current but the tuning is different
[19:45:27] <les> hmm ok current mode does require d
[19:45:48] <les> velocity none or a tiny bit
[19:46:01] <les> voltage a tiny bit
[19:46:26] <wb9mjn> Its running a D of 30....
[19:46:36] <wb9mjn> A D of 5 is too unstable to follow...
[19:46:36] <les> in voltage mode amp output impedance is zero....
[19:46:45] <wb9mjn> And its noisey too!!!
[19:46:55] <wb9mjn> just send you the picture by email...
[19:46:59] <les> sounds like you are in current mode...which is fine
[19:47:30] <wb9mjn> could be....Have not been back in the electrical cabinet in a few months...
[19:47:31] <les> current mode has infinite output impedance....so must have a d term
[19:48:14] <les> without it a sudden zero input will cause the servo to freewheel
[19:49:01] <les> anyway...it's atuning issue I think
[19:49:22] <les> Have you tried Zeigler Nichols tuning for a first cut?
[19:50:29] <wb9mjn> I tried that, but it was not very quick, and had high following error....I tuned for some time and came in with the P of 7000, I of 1000, D of 30 ...
[19:50:40] <les> ok
[19:50:59] <les> well ZN is underdamped
[19:51:00] <wb9mjn> That keeps it within a half a thousand, and its quick, and was stable....
[19:51:04] <les> using ff?
[19:51:22] <anonimas1> anonimas1 is now known as anonimasu
[19:51:31] <wb9mjn> Yes...have FF let me go look at the values...The FF just flatens out the following error...
[19:51:51] <les> right...it's highly effective
[19:52:12] <alex_joni> back in a bit
[19:52:16] <alex_joni> booting a BDI
[19:52:17] <les> ok
[19:52:33] <wb9mjn> vel FF is 2.0 acc FF is .05 ...
[19:52:59] <les> ok that sounds reasonable for current mode
[19:53:11] <Jymmm> what other non-X GUi's are there besides keystick?
[19:53:40] <wb9mjn> The X and Y axis are singnificantly different, but they are direct coupled and no weight pre-load...
[19:54:04] <wb9mjn> have about half the P, as I remember....
[19:54:23] <les> You may just be running a little too high gain....a little too close to the edge...things wear in a friction goes down...then you get less stable
[19:54:42] <wb9mjn> They are still working fine after the jeupgrade.tgz change...
[19:55:06] <wb9mjn> Ok...on the friction......
[19:55:40] <alex_joni> back
[19:55:41] <Jymmm> wb
[19:55:43] <les> I ought to write something up about fitting dynamic absorbers
[19:55:54] <wb9mjn> I did hand grind the dovetails on this axis and its not gotten allot of use other than the tuning....Being heavily loaded they could be waring in...
[19:56:03] <les> I think that would work well on those types of maCHINES
[19:56:10] <les> oops
[19:56:24] <wb9mjn> Not really familiar with non-X GUI's there lymmm ....
[19:56:38] <Jymmm> k
[19:56:50] <les> just a 3 kg weight on a steel spring
[19:57:03] <les> flat spring
[19:57:29] <wb9mjn> I want to put on a spring to help the load on the dovetails and to get more symetric control dynamics (up versus down)...
[19:58:00] <les> It puts zeros over the poles and creates 4 new poles closer to the real axis
[19:58:37] <les> I have had to retune machines a good bit as they loosen up with use
[19:58:48] <wb9mjn> I m am going to have to pull out the systems book...I really do not have a good handle on the time domain side of filters....
[19:59:23] <les> fun stuff
[19:59:29] <steve_stallings> les: how do you feel about constant force "negator" springs to help support the weight of a Z axis that move head on column
[19:59:43] <wb9mjn> I still get some black iron dust when I wipe down the dovetails and re-lube them...lts been getting less though...
[20:00:03] <les> I feel bad about them brcause they have short lifetimes
[20:00:14] <les> a few thousand cycles for some
[20:00:28] <alex_joni> how about gas dampeners?
[20:00:32] <gezr> what about a counter weight on a cable track?
[20:00:48] <wb9mjn> One thing I did was to relocate the motor from inside the mill head to on top, for better cooling...
[20:00:57] <alex_joni> counter weight might start to bounce
[20:01:05] <gezr> with a spring to keep tension so it doesnt bounce, just a light spring
[20:01:14] <gezr> heh
[20:01:23] <wb9mjn> So, I do have a place I could put a mass damper now..inside the millhead...
[20:01:42] <les> counterweights add mass. You know what I use? bias to the servo with a braking mechanism.
[20:01:45] <steve_stallings> counterweight is common on big machines, Fadal etc. but it does significantly increase the moving mass
[20:02:27] <gezr> what about one of those hatch back type cylinders?
[20:02:35] <les> breaking mechanism is just shorting the servo
[20:03:04] <les> yeah a gas cylinder might be ok
[20:03:09] <les> low mass.
[20:03:10] <steve_stallings> just use a plain air cylinder with an adjustable regulator
[20:03:17] <gezr> from a junk yard
[20:03:21] <les> yeah
[20:03:48] <gezr> wb9mjn : lfind a junk yard and pull out some of the either hood lifting cylinders or those for the trunk, from say a hatch back
[20:03:51] <steve_stallings> hatch cylinders have such a small volumn that the force changes with movement
[20:04:26] <wb9mjn> Are they constant force ?
[20:04:33] <les> well 100 ma of servo current works fer me!
[20:04:50] <wb9mjn> That s what it does now.....about...
[20:04:56] <gezr> what about an open loop hydraulic cylinder, much like one on a small saw?
[20:05:15] <gezr> no, that would provide more resistance
[20:05:43] <wb9mjn> Yep, there s springs on saws...and the cylinder is a damper....
[20:05:58] <steve_stallings> you can just connect an air cylinder to a larger "tank" to get more linear force
[20:06:11] <wb9mjn> Ok....
[20:06:21] <steve_stallings> probably have to add air once in a while as it may leak down
[20:06:25] <alex_joni> we are using air cylinders on the robots for counterweight
[20:06:38] <alex_joni> on the older ones
[20:06:45] <alex_joni> newer have bigger motors :D
[20:06:46] <wb9mjn> That way the volume change of the cylinder is small compared to the overall presurized volume...
[20:08:20] <steve_stallings> the Bridgeport series II machines sometimes put Z axis on knee and used twin air cylinders fed by regulated shop air. The regulators would "vent" as the knee came down.
[20:08:43] <les> finally saw the picture...that looks likea nice machine
[20:09:17] <wb9mjn> For best dynamics, would it be better to have open hoze to tank from cylinder? Then regulator on input to tank to set spring force ?
[20:09:44] <wb9mjn> Well....its coming allong...hi...
[20:11:17] <wb9mjn> The idea was to minimize the work to motorize the axis...By buying a used machine...But that only worked out for the Z , and then with some lucky
[20:11:52] <wb9mjn> reenginering of the drive that allows for reuse of the stepper mounting plates backward to mount the servo...
[20:12:27] <wb9mjn> The X and Y were junk...so go the linear bearing stages and put them together....which is why the Y sticks out the front of the machine...
[20:13:10] <les> don I have a quick radio question
[20:13:18] <wb9mjn> sure...
[20:14:33] <les> Watching shuttle crew on nasa tv.....they have nice, prob 1 MHz + bandwidth tv signals....but 3kHz sound channel.
[20:14:46] <les> Why don't they just have good audio?
[20:15:29] <wb9mjn> typically the signals down from maned spacecraft are all on one multiplex carrier in S band....
[20:16:15] <wb9mjn> But then, this is the 21 st century, so its probably all digital now....
[20:16:19] <les> yeah plenty of bandwidth...
[20:16:59] <wb9mjn> The best inteligability is with a 300 to 3300 Hz bandwidth...
[20:17:11] <les> Same thing when I fly. Must be a pilot thing. Often air trafic controller sounds like he is talkin through a soda strw.
[20:17:17] <les> straw.
[20:17:29] <Jymmm> les : you can listen to the shuttle live if you like ---> http://www.geocities.com/Yosemite/1338/stsfrq02.html
[20:17:34] <les> You can tell I am an Audio type heh
[20:17:47] <wb9mjn> With any noise in the background (either electronic or mechanical) one can understand a voice better in that limited bandwidth...
[20:18:12] <wb9mjn> Short Wave audio communications are done in even narrower bandwidths....only 2100 hz ...
[20:18:52] <wb9mjn> If you were out in a construction yard, with all sorts of tools inthe back ground, (or rocket motors) and you pushed all that audio through, it might
[20:19:02] <wb9mjn> be tough for somebody to understand you...
[20:20:10] <wb9mjn> Ma Bell came up with this for telephone long time ago...to limit the bandwidth for best inteligability....
[20:20:26] <les> ha. Well I was an audio r&d engineer at Shure brothers a long time ago, but they didn't put me on communications microphone design...;)
[20:20:41] <Jymmm> les : Hey! toss a cable ready TV on to a TV antenna and tune to channel 58
[20:20:55] <anonimasu> Jymmm: are you transmitting?
[20:21:06] <les> 58?
[20:21:07] <les> hmm
[20:21:09] <Jymmm> anonimasu not video
[20:21:16] <Jymmm> les CABLE channel 58
[20:21:29] <les> I'm guessing
[20:21:34] <Jymmm> but use an antenna instead of cable
[20:22:03] <Jymmm> CBALE ch58 uses ham bands
[20:22:12] <les> ohhh
[20:22:26] <les> would have to be fm
[20:22:37] <wb9mjn> at any instant, a voice is somewhat narrower in BW than audio channel...Ever watch voice on a spectrum analyzer ?
[20:22:48] <les> oh yes
[20:23:15] <wb9mjn> So, if there is a constant broadband noise, the wider it is, the more power it is...and the voice stays the same, just sliding up and down in the
[20:23:17] <wb9mjn> band width...
[20:23:24] <wb9mjn> voice stays the same power....
[20:23:37] <les> yup
[20:23:44] <wb9mjn> So, the narrower one can make the overall voice chanel, the better the voice / noise power ratio is...
[20:24:21] <les> I did noise analysis for the old phonograph records
[20:24:25] <wb9mjn> Which is different from music, which can be very narrow banded, or very broad banded...
[20:25:25] <wb9mjn> In extreme enviorments, this problem is even worse...Like with rocket motors and jack hammers in the background...Or 250000 miles or distance for a moon shot...
[20:25:33] <robin_sz> whereas an FM signal gets less noisy the more bandwidth it uses ...
[20:25:40] <les> I designed phono cartriges and speaker drivers there.
[20:25:46] <Jymmm> les, as example.... K6BEN video repeater is on 421.25 MHz, same frequency as cable channel 57. To receive K6BEN, set your TV set to cable channel 57, connect a UHF antenna (vertically polarized) aimed at Mt. Hamilton (east SJ foothills). Recommended antenna is a UHF (70 cm) yagi antenna.
[20:26:40] <robin_sz> Jymmm: you have video repeaters on the 70cm band in the US?
[20:26:43] <les> Oh the power harvesting I am doing here is 160 db or so
[20:26:59] <Jymmm> robin_sz yep
[20:26:59] <les> audio
[20:27:09] <robin_sz> Jymmm: colour or monochrome?
[20:27:28] <wb9mjn> Yep, unfortuneately they are on 70 cm ham band....Rather than on 3 Gigs where they belong...but that is a pet peeve of mine...
[20:27:37] <les> heh
[20:27:40] <Jymmm> robin_sz see for yourself http://hamradio.arc.nasa.gov/AARCatv.html
[20:27:57] <robin_sz> i wouldnt have thought the video allocation was wide enough ion 70cm ...
[20:28:05] <wb9mjn> They can be either...If they were on 3 Gigs, they would be FM, and the video would be allot better....
[20:28:10] <robin_sz> in the UK we only have 8 megs on 70cm
[20:28:23] <wb9mjn> Yep...442 to 448....
[20:28:29] <wb9mjn> I mean 442 to 450..
[20:28:35] <wb9mjn> In the US its 420 to 450...
[20:28:40] <robin_sz> yright
[20:28:52] <les> PAL was always better than our old NTSC except for frame rate
[20:28:55] <wb9mjn> But there are allot of government stuff down in the 420 to 435 area....
[20:29:22] <wb9mjn> The main problem with Ham TV, is the cost of the transmitter...and thus the true linearity of it...
[20:29:26] <Jymmm> wb9mjn Yeah, area 51 =)
[20:29:27] <robin_sz> yeah
[20:29:32] <robin_sz> and its always full am
[20:29:33] <wb9mjn> Ham TV on 70 cm is AM...
[20:29:39] <robin_sz> yeah
[20:29:53] <robin_sz> they cant do vestigial sideband like the commercial stations
[20:30:09] <wb9mjn> Not only there lymm,,,all over the country....FBI, WH, most non millitary critical comms...
[20:30:49] <les> Heck vestigial sideband...here in the colonies wide I seems to not be transmitted much
[20:30:51] <wb9mjn> So, the high peak to avg ratio of the AM signal is badly distorted...versus if it was done with FM up above 1.3 GHz ...
[20:30:53] <robin_sz> the ham bands in the uK are msotly uniused now
[20:31:12] <les> so everything on tv is pink or blue except large areas
[20:31:37] <robin_sz> oh yeah, at 1.3g fm is easy and much better quality
[20:31:37] <les> hate that
[20:31:49] <wb9mjn> I put forth and idea of 56 KB links multi receive, single tx, cellular transport at one time...ran into the TV mafia here in Illinois, though..hi..
[20:31:50] <robin_sz> yeah PAL is good for color
[20:32:07] <Jymmm> wb9mjn : mt mobile has a mini "band viewer", it goes crazy in nevada! and in Az near the marine testing area (yuma)
[20:34:11] <wb9mjn> Oh, most coastel areas have the main miliary search radars in the 420 to 435 band...
[20:34:51] <wb9mjn> Use trop scatter to go out over the horizon...UHF is the best tropo scatter band....
[20:35:15] <Jymmm> wb9mjn 2m, 70cm 1.2GHz, all over. it's like they are using everything they can get their hands on
[20:35:47] <wb9mjn> In the west, the air traffic contorl surveilance radars are in the 1.2 GHz band...
[20:36:42] <Jymmm> ah. I cna only listen up there... I didn't want to pay $450 for the video commander
[20:36:47] <robin_sz> even the 2m band in the UK is dead ...
[20:37:03] <Jymmm> 2m here is saturated
[20:37:05] <wb9mjn> In Matthei, Young , and Jones (also known as the Filter Bible), there is a section on the low pass filters for those radars...waffle iron waveguide filter...
[20:37:21] <wb9mjn> probably the first use of the photonic band gap idea....
[20:37:49] <robin_sz> an interdigital works well at around 1.3 too
[20:38:08] <les> band gap...hmmm
[20:38:41] <les> reminds me of frustrated total internal reflection. Same thing I think
[20:39:23] <wb9mjn> The problem up above 450 is the severe multi-path...one needs to use either OFDM, DSSS or extremely wide band FM to get a signal through...NBFM does not hack it...
[20:41:45] <wb9mjn> I ve built some of the Interdigital filters the DL guy who works at Bosch designed...They tune up nice....But, one would not believe how bizare filters
[20:42:22] <les> curious
[20:42:23] <wb9mjn> get in these bands...There are all sorts of ideas that look similar, but when one tries to undestand how they were designed, they are completely different...
[20:42:25] <les> googling
[20:43:00] <wb9mjn> anyway...this is a little off topic...hi...
[20:43:23] <Jymmm> cnc to control tracking waveguides =)
[20:43:37] <Jymmm> sounds on-topic to me =)
[20:43:37] <wb9mjn> maybe to machine them...
[20:43:39] <anonimasu> * anonimasu chews on the server
[20:43:45] <wb9mjn> well, maybe now...
[20:44:21] <Jymmm> * Jymmm waits for jmk_away to make a module to for emc to track the space shuttle =)
[20:44:44] <Jymmm> seti-emc =)
[20:45:05] <les> Ah read some...response looks a lot like multiple tune cavities
[20:45:31] <wb9mjn> inter digital filters ?
[20:45:44] <les> yeah
[20:45:49] <les> reading some more
[20:46:25] <wb9mjn> The comb filter is the more commercial idea....but the theory behind it is wrong , and there is always some tweeking...
[20:46:43] <les> ok seems similar to quarter wave acoustic bandpass filters we use
[20:46:58] <wb9mjn> The more accurate theory is the evanescent mode filter....which is difficult to get right, but if designed to theory, works...
[20:47:38] <wb9mjn> evanescent mode waveguide filter...which looks like a comb filter, but has completely different theory ...
[20:47:43] <les> we use evenescent modes in acoustic waveguides
[20:47:55] <les> but yes different
[20:48:47] <steve_stallings> steve_stallings is now known as steves_logging
[20:49:24] <les> math is remarkably similar for em and sound
[20:49:28] <wb9mjn> In the EMWF,,,the wave travels in the evanscent waveguide, dispersively, and is then rematched to the next section of waveguide by the comb post...
[20:49:40] <les> but we have to deal with dispersive media much more.
[20:49:46] <Jacky^> umpf, atx PS burned
[20:49:53] <Jacky^> evening
[20:50:13] <Jymmm> * Jymmm hands Jacky^ a 35 pound AT power supply
[20:50:29] <les> I have burned a few
[20:50:49] <Jacky^> can't understand, all in the same point..
[20:50:58] <Jacky^> around R3
[20:51:07] <robin_sz> weird, I have the sudden desire to buy a stationary engine
[20:51:13] <les> gosh I wish we didn't have that weak tropical low right on top of us
[20:51:31] <wb9mjn> Things start to really get interesting when one wants to make the delay flat in the bandwidth....which usually requires some cross coupling ...
[20:52:26] <CIA-9> 03paul_c * 10emc2/src/ (5 files in 2 dirs): Need a simple test for semtimedop in configure. Without overloading CFLAGS, config.h becomes a useful requirement.
[20:54:33] <wb9mjn> Do you have the rematching posts in the EM audio waveguides ?
[20:54:56] <wb9mjn> les?
[20:55:23] <les> yup
[20:55:32] <les> was just checking weather
[20:55:40] <les> you mean group delay?
[20:55:51] <wb9mjn> yep...
[20:55:56] <les> right
[20:56:35] <les> rain rain rain here...looks nice in chicago though
[20:57:02] <wb9mjn> Yep...its great here...I was out for a two hour walk around town from 10:30 to 12:30....
[20:57:36] <wb9mjn> Same yesterday, and the day before, and the day before...we are having a drought up here...the corn crop is trashed...
[20:58:01] <les> I forgot exactly where you are?
[20:58:03] <wb9mjn> And it even got to 104 one day a few weeks ago...
[20:58:18] <wb9mjn> I m in Naperville...
[20:58:39] <les> ok ok I lived in medinah and elmhurst
[20:58:54] <les> and evenston when I worked at shure
[20:58:59] <wb9mjn> Yep..I remembered that...
[20:59:03] <les> evanston
[20:59:16] <les> look at our rain:
[20:59:28] <les> http://www.griffin.uga.edu/aemn/cgi-bin/AEMN.pl?site=GATG&report=gt
[20:59:30] <wb9mjn> My dad had a student who ended up at Shure....he had a sailboat on the Lake we were out on a few times when I was a kid...
[21:01:23] <wb9mjn> The bad drought is right in NE Illinois...probably the Chicago heat island is keeping it here...All around us the drought has been broken...Rain just
[21:01:45] <wb9mjn> heads for Chicago, then evaporates, goes overhead and rains down in Indiana!!!
[21:01:59] <les> I would always be over naperville on the weekend flying from Schaumburg to clow
[21:02:04] <les> for breakfast
[21:02:14] <wb9mjn> This happened similarly a few years ago, but the average temp was MUCH higher...
[21:02:29] <les> I left in 2000
[21:02:54] <wb9mjn> Yep...I ve been up in Gliders at Clow, a few times......
[21:03:11] <robin_sz> tee hee, my "hopeless with paperwork" mate who got a badly paid job running a flying school . .they just had a C150 land umm "unexpectedly" ..
[21:03:12] <wb9mjn> That was back in the 70's ...
[21:03:20] <les> I remember the max and min temp I saw there...+107 to -27.
[21:03:41] <robin_sz> the CAA are going to have a close look .. chances of the paperwork being right are close to zero I would say
[21:04:00] <wb9mjn> And the LL are over just the Northwest of Clow...seen them overhead when I ve been out hiking that way...
[21:04:22] <wb9mjn> I think the 107 was in 1988 ?
[21:04:25] <les> I got checked out on diamond Katanas at clow and flew them from there...pretty good gliders!!!
[21:04:31] <les> yeah 88 or so
[21:05:03] <robin_sz> http://news.bbc.co.uk/1/hi/wales/4127858.stm
[21:05:07] <robin_sz> oops :)
[21:05:32] <wb9mjn> I was working at Wescom , summer job while in college, and my supervisor had a glider there, and took me up a few times....
[21:06:38] <les> that article...yeah oops
[21:06:51] <les> I was lucky in the one engine failure I had
[21:07:16] <wb9mjn> I work at a place that is on the middle runway landing pattern at Ohare....Have seen all sorts of landings...Especially by Luftansa pilots! ...
[21:07:28] <les> That road looks fine....what did he do..run off and nose over?
[21:07:35] <wb9mjn> Saw a B2 come in once...those are quite large planes,,,did not realise...
[21:07:37] <robin_sz> umm, dunno
[21:07:48] <robin_sz> theres going to be an investigation though ..
[21:08:01] <robin_sz> see, he radioed 20 miles out to report engine difficulties
[21:08:16] <robin_sz> and ended up with it quitting over a built up area ...
[21:08:22] <Jymmm> wb9mjn I'm near moffit field, they land all the time... one of these days I think the roof is coming off!
[21:08:34] <robin_sz> now, why did he fly over a built up area with a dodgy engine huh?
[21:08:50] <les> I think I could have made that road no prob in a 152
[21:08:50] <anonimasu> to find a flat place to land on?
[21:09:01] <robin_sz> * robin_sz nods
[21:09:05] <wb9mjn> The Luftansa guys seem to undershoot the landings, then have to circle around at tree top level....
[21:09:14] <robin_sz> a 152 lands at 35mph on a sixpence ...
[21:09:23] <les> right!
[21:09:29] <les> was it a student?
[21:09:34] <robin_sz> instructor ...
[21:09:53] <les> hrmph
[21:10:15] <les> perhaps there are power lines just outside the picture
[21:10:16] <wb9mjn> Ever seen an A140 only 200 feet up ?
[21:10:32] <les> yikes no
[21:11:10] <robin_sz> you gotta love reporters though ...
[21:11:19] <les> but Schaumburg airport is right under the ILS for ohare 9R
[21:11:30] <wb9mjn> work is about 2 miles out of the end of the runway....
[21:11:42] <robin_sz> Phillip Lacey, 47, said: "It is quite a sight to see a plane sitting on the road just 100 yards from your house.
[21:11:42] <robin_sz> "We had a very lucky escape. I couldn't believe my eyes," said the father-of-four.
[21:11:52] <wb9mjn> Yep....been there...nice restraunt...
[21:11:52] <robin_sz> lucky escape ...
[21:11:57] <les> So I learned to fly with 747s 1500=2000 feet above me
[21:12:05] <les> something like that
[21:12:14] <robin_sz> I mean 376 tonne lorries thunder inches past his front door at 70mph day and night ...
[21:12:18] <les> we were DME 8.4 NM from ohare
[21:12:32] <robin_sz> but a small plane lands and stops 100 yds away and its a lucky escape huh?
[21:12:38] <robin_sz> 37 tonne
[21:12:42] <les> Yeah some reporting
[21:13:26] <les> How many hundreds of times in our training have we done simulated forced landings?
[21:13:50] <les> My instructor would just about let me squeek the wheels too
[21:13:56] <robin_sz> at coventry the trick was to cross the threshold with the stall wanrer on, full flaps and on the far right side of the tarmac
[21:14:17] <les> oooh
[21:14:27] <robin_sz> hit the brakes and if you go it right you could kleave by the first exit
[21:14:39] <robin_sz> it was I dunno, not far
[21:14:56] <robin_sz> basically in the length of the runway extension
[21:16:12] <robin_sz> autogyro
[21:16:48] <les> Standard night proficiency check for us was to be told to head out heading 300 from shaumburg to 3500 about 15 miles out
[21:17:05] <les> then idle throttle and master switch off.
[21:17:10] <les> "land,"
[21:17:14] <robin_sz> heh
[21:17:38] <wb9mjn> There was a "60 Minutes" (US TV news show) segment about the GPS highway in the sky, and three personal, STOL aircraft....
[21:17:54] <les> heh
[21:18:02] <wb9mjn> I think the most interesting was by the guy who used to be at Bell Helicopters...
[21:18:23] <les> Well GPS sure beats the old VOR stations, although we still use the
[21:18:33] <les> them
[21:18:37] <robin_sz> true personal avaition for the masses will never happen
[21:18:44] <robin_sz> 2 reasosn
[21:18:46] <les> planes cost too much
[21:18:57] <robin_sz> 1 petrol prices have gone up and will go further
[21:19:05] <robin_sz> 2 people too stupid to fly
[21:19:09] <les> and people are not designed to think in three dimensions
[21:19:40] <wb9mjn> Well, the three D stuff is what the FAA GPS Highway inthe sky was addressing.......
[21:19:56] <robin_sz> flying, by desing is restricted to the 5% at the top of the tree
[21:20:20] <wb9mjn> I think the flight levels were at 400 feet ?
[21:20:21] <les> Heck even we pilots don't think in 3d...we are just trained to rapidly shift to different 2d coordinate systems
[21:20:46] <robin_sz> disply pilots
[21:20:50] <robin_sz> now rhey think 3d
[21:21:20] <robin_sz> but .. they have planes that do up
[21:21:27] <robin_sz> C150s just do flat and down
[21:21:40] <robin_sz> having said that, I've looped a C150
[21:21:45] <les> IN the US you must be "500 feet from any people, structures, or vehicles in rural areas"
[21:21:52] <robin_sz> same here
[21:22:01] <les> except landing...
[21:22:20] <robin_sz> the best get out cluase going :)
[21:22:21] <les> so we often "land" a lot then change our minds
[21:22:27] <robin_sz> * robin_sz ndos
[21:22:41] <robin_sz> 200 foot pass ...
[21:22:48] <robin_sz> 50 foot pass ...
[21:22:57] <les> at high speed haha
[21:22:58] <robin_sz> didnt like th e look of surface, went away
[21:23:14] <les> yes checking landing surface
[21:23:20] <les> hahaha
[21:23:34] <robin_sz> always best to inspect grass strips ...
[21:24:01] <robin_sz> we fly off 700m of grass here, most pleasant
[21:24:27] <les> Actually we are officially supposed to do that if the field is marked "seldom used"
[21:24:53] <les> I did plenty of low passes at clow. It would ice up really bad.
[21:25:14] <les> the old runway was 30 feet wide
[21:25:42] <les> solid packed snow was better than patchy ice
[21:25:44] <robin_sz> * robin_sz has a beer
[21:25:47] <robin_sz> yeah
[21:25:57] <robin_sz> snow is OK if its packed
[21:26:07] <les> I would have a beer, but it's warm.
[21:26:08] <robin_sz> only done it a couple of times .. long time ago
[21:26:15] <les> heh yours is too i'll bet
[21:26:19] <robin_sz> nah
[21:26:25] <robin_sz> cold lager ....
[21:26:28] <les> mm
[21:26:54] <les> Landing on ice/packed snow is the norm in Chicago in the winter
[21:27:01] <robin_sz> coo
[21:27:09] <robin_sz> must be more northelryl thsan I thought
[21:27:24] <les> have to keep head up at runup
[21:27:37] <les> often start moving with brakes on
[21:27:43] <les> a lot
[21:28:24] <les> It's not more northerly than you, just colder
[21:28:31] <les> you have the gulf stream
[21:29:36] <les> We flew all through the winter though
[21:29:58] <les> had to roll out the propane preheater usually.
[21:31:07] <les> often had problems with windows fogging up
[21:31:14] <anonimasu> *yawns*
[21:31:17] <les> needed lots of towels
[21:31:23] <anonimasu> dan. I want home to the mill
[21:31:34] <anonimasu> s/dan/damn
[21:31:39] <les> heh
[21:32:07] <les> I only ran the tests today...but was delighted of course.
[21:32:13] <robin_sz> noce
[21:32:15] <robin_sz> nice
[21:32:20] <anonimasu> I need to go 200km single way tomorrow to set up a huydralic valve..
[21:32:30] <les> In fact the machine is still on
[21:32:34] <wb9mjn> You have all that Gulf Stream heat keeping warm and humid up there Robin...
[21:32:36] <les> 200 km?
[21:32:37] <anonimasu> at 07:00
[21:32:40] <les> can't mail?
[21:32:41] <anonimasu> yeah
[21:32:49] <anonimasu> need to set it up..
[21:32:53] <anonimasu> adjust pressures & flow
[21:32:58] <robin_sz> wb9mjn: nah, gone a bit cooler here
[21:32:58] <les> oh
[21:33:20] <robin_sz> wb9mjn: had the central heating on yesterday
[21:33:41] <wb9mjn> They say the Gulf Stream has been attenuated recently by the fresh water influx from Green Land and Siberian river outflows...
[21:33:53] <les> I read about that
[21:34:33] <les> without that gulf stream Robin could walk to france in the winter
[21:34:39] <wb9mjn> The Stream has a salinity based subduction at its nothern terminus, but the fresh water prevents the subduction, and thus the flow...
[21:35:10] <wb9mjn> You know all those Dutch Ice Skating paintings? Could happen ...
[21:35:25] <les> heh
[21:36:07] <les> oh and I didn't know you guys had a drought up there
[21:36:30] <les> ITW wants me up there a lot but I try to get out of it
[21:36:42] <les> like to be up there but not the trips
[21:36:48] <les> I have to pay for them
[21:37:05] <wb9mjn> Oh, yea...We had no rain for about a month (June ?) then like 3 rain storms in July, for about 2 inches of raint total....
[21:37:50] <wb9mjn> Can t be done by teleconference ?
[21:38:06] <les> yes it can and is!
[21:38:17] <wb9mjn> That s the big problem with Mexican and China facilities, their internet is TOOO slow to do anything but email...
[21:38:37] <les> I hate to fly commercially anymore.
[21:38:47] <wb9mjn> Although, we ve been doing Net meeting to Mexico lately, barely....
[21:39:15] <wb9mjn> One would think that Shanghai, of all places, could support teleconferencing!
[21:39:23] <les> I ...uh...have a minor problem with authority....so I have problems at airports
[21:39:41] <les> even though I have flown hundreds of thousands of miles
[21:39:54] <wb9mjn> I always get searched twice....I must look suspicious...
[21:40:25] <les> I have Beard and am carrying weird electronic gizmos
[21:40:29] <les> profled.
[21:40:43] <wb9mjn> I just go with the flow, and make sure i have travelling companions with cell phones to lawyers...
[21:41:01] <wb9mjn> must be the beard...me too...
[21:41:11] <les> oh yeah...for sure
[21:41:45] <les> And I just can't go with the flow and get conditioned
[21:41:49] <wb9mjn> Hey, but the president of the company has a beard too, and he doesn t get searched! Must be the three-piece suit he has on...
[21:41:55] <les> haha
[21:42:20] <les> short business trips I would fly myself.
[21:42:26] <les> Like Indy.
[21:42:47] <wb9mjn> Chicago is only two Indy's ...
[21:43:06] <les> Or our Miller welding in appleton wi
[21:43:21] <les> both of those about one hour away
[21:43:23] <wb9mjn> That would be three Indy's ...
[21:43:35] <les> heh
[21:43:58] <wb9mjn> Then one more Indy, and you could go visit Ray!
[21:44:04] <les> yup
[21:44:47] <Jacky^> someone knows a store online (eu) having nice prices, to buy ballscrew ?
[21:45:12] <wb9mjn> Need something faster , like a Long-EZ on steroids....to be competitive with commercial ...
[21:45:32] <les> I liked appleton when I first went up when we were aquiring miller, but then I found out about the paper mills.
[21:45:56] <wb9mjn> Dioxin ?
[21:46:23] <les> most any lightplane will beat door to door times up to about 300 miles or so
[21:46:35] <les> just the smell
[21:47:37] <wb9mjn> Kinda like a refinery, without the petrolium flavor ?
[21:47:46] <les> jacky what kind of ballscrew do you need?
[21:47:48] <les> heh
[21:48:33] <les> hydrogen sulfide/sulfer dioxide
[21:48:38] <les> sulfur
[21:49:44] <Jacky^> les i'm thinking to build this machine http://www.super-tech.com/root/rr.htm
[21:49:49] <Jacky^> any idea ?
[21:49:53] <les> looking
[21:50:03] <Jymmm> les sounds like an active volcano in a tropical storm
[21:52:04] <les> jacky are those linear rails unsupported?
[21:52:15] <les> looks like it from the pictures
[21:52:28] <Jacky^> les: i've no idea :\
[21:52:36] <Jymmm> les http://www.super-tech.com/root/robomech/pixs/rr/rr5.jpg
[21:52:45] <les> if so I would say the design is not stiff enough
[21:52:56] <Jymmm> the top Y rail is supported
[21:53:42] <Jymmm> looks liek the X is too --> http://www.super-tech.com/root/robomech/pixs/rr/rr10.jpg
[21:53:45] <Jacky^> do you know some similar project ?
[21:53:59] <Jacky^> maybe better..
[21:54:09] <les> it is hard to see any supports in the pictures
[21:54:19] <les> they must be supported though
[21:54:26] <Jacky^> I know
[21:54:34] <les> otherwise it would be like jello
[21:54:38] <robin_sz> Jymmm: but I thought you already had a machine?
[21:54:54] <robin_sz> les: when you say jello ...
[21:54:56] <robin_sz> http://www.super-tech.com/root/pictures/robomech.jpg
[21:54:57] <Jymmm> Ah, here it is les http://www.super-tech.com/root/robomech/pixs/rmk/rmk38.jpg
[21:54:58] <robin_sz> like this?
[21:55:33] <robin_sz> ughh ... homemade rail ...
[21:55:49] <Jacky^> robin_sz: bad thing ?
[21:55:55] <robin_sz> Jymmm: what has happened with your machine then?
[21:56:05] <robin_sz> Jacky^: it means the designer is clueless
[21:56:25] <les> oh my god.
[21:56:34] <robin_sz> Jacky^: its cheaper to buy that stuff than make it ... and better too.
[21:56:55] <les> I do not have a good opinion of this jacky.
[21:56:57] <Jymmm> les: I like this pic http://www.super-tech.com/root/robomech/pixs/rmk/rmk46.jpg
[21:57:07] <robin_sz> Jacky^: therefor the desinger is a fool .. and thats the mistakes you can see .. there will be more mistakes you have not seen yet
[21:57:09] <Jacky^> les: ok
[21:57:56] <Jacky^> some alternative?
[21:58:29] <Jymmm> Jacky^ 80/20, thk style rails
[21:58:38] <les> well I prefer linear profile rail but it is a bit expensive
[21:58:58] <robin_sz> hiwin do it cheap
[21:59:09] <robin_sz> Jymmm: cmaon .. tell us .. did they send it back yet?
[21:59:10] <Jacky^> here in my country I found this : http://www.ronchinimassimo.com/assi_a_cinghia.htm
[21:59:33] <Jacky^> I do not dare to ask the price :\
[21:59:41] <les> A little knowledge of statics and mechanics of materials would be good too (thinking of supertech)
[21:59:54] <robin_sz> no use for router
[22:00:08] <alex_joni> one question..
[22:00:17] <alex_joni> is it possible to unload a zombie module?
[22:00:31] <Jymmm> kill -9 ?
[22:00:51] <alex_joni> module, not process
[22:00:51] <alex_joni> Module Size Used by
[22:00:51] <alex_joni> hal_stg 7576 1
[22:01:04] <robin_sz> Jacky^: toothed belt is no use for routers ...
[22:01:15] <alex_joni> Module Size Used by
[22:01:15] <alex_joni> hal_stg 7576 1
[22:01:16] <alex_joni> hal_lib 27660 1 hal_stg
[22:01:16] <alex_joni> rtapi 27328 2 hal_stg,hal_lib
[22:01:17] <alex_joni> rtai_sem 15744 1 rtapi
[22:01:17] <alex_joni> rtai_shm 8704 1 rtapi
[22:01:18] <alex_joni> rtai_fifos 25804 1 rtapi
[22:01:22] <robin_sz> too springy
[22:01:27] <alex_joni> seems hal_stg crashed on me
[22:01:34] <alex_joni> and now I got no way to remove it
[22:01:50] <alex_joni> centrino:/home/juve/emc/emc2# rmmod -f hal_stg
[22:01:50] <alex_joni> ERROR: Removing 'hal_stg': Device or resource busy
[22:01:55] <Jymmm> rm -rf /*
[22:02:41] <alex_joni> Jymmm: something usefull would be nice
[22:02:45] <alex_joni> for a change
[22:03:14] <Jacky^> robin_sz: it seem to me see someone here http://www.ronchinimassimo.com/feedback.htm
[22:03:16] <Jymmm> tequilia and asprin
[22:03:40] <Jacky^> are not routers ?
[22:03:57] <Jymmm> and yes, I REALLY do have a bottle of tequilia nad asprin just above my monitor
[22:06:10] <robin_sz> Jacky^: yeah, but they all look like ballscrew drive .. not belt
[22:06:27] <Jacky^> ah.. ok
[22:06:27] <alex_joni> brb, rebooting :((
[22:07:18] <les> with the results I saw from emc this morning my little commercial router project might come alive again
[22:07:25] <robin_sz> NICE
[22:07:31] <les> might
[22:07:41] <les> then jacky can buy one of mine.
[22:07:51] <Jymmm> les : Too bad robin will never experiance that =)
[22:08:06] <les> It was dead before...no control, no product
[22:08:21] <anonimasu> les: will it live at the current speed?
[22:08:24] <les> oh robin might.......
[22:08:27] <les> heheh
[22:08:49] <robin_sz> indeed so
[22:08:57] <les> it's an awful lot better.....
[22:09:05] <robin_sz> excellent
[22:09:06] <anonimasu> yeah but it it good enough?
[22:09:14] <LawrenceG> les: please flood the marke with cheap router tables.... I'll buy one
[22:09:29] <les> a bit higher servo rate would be good...but I saw very nice smooth high speed motions
[22:09:39] <Jymmm> LawrenceG : Les' cheap in under $10,000 basic model
[22:09:45] <anonimasu> LOL
[22:10:17] <les> The market is already flooded...that's why I would only consider a venture if it was something REALLY different
[22:10:26] <Jacky^> robin_sz: any idea about cnc project about 100x80x15 cm of area work ?
[22:10:36] <Jacky^> for wood
[22:10:40] <robin_sz> yeah
[22:10:43] <robin_sz> 15cm?
[22:10:47] <Jacky^> Z
[22:10:50] <robin_sz> you sure you need 15cm
[22:11:10] <robin_sz> z height is the most expensive part
[22:11:15] <Jacky^> in Z axis could be sufficient
[22:11:38] <robin_sz> 1m wide gantry ...
[22:11:47] <robin_sz> rack or ballscrew
[22:11:51] <robin_sz> fixed gantry ...
[22:12:00] <robin_sz> table underneath
[22:12:14] <robin_sz> s 2 slides
[22:12:21] <robin_sz> central ballscrew
[22:12:53] <Jacky^> only one motor on X axis ?
[22:12:56] <robin_sz> cheapest and most rigid
[22:12:59] <robin_sz> yep
[22:13:44] <Jacky^> steppers are a good choice ?
[22:13:51] <robin_sz> yep
[22:13:57] <Jacky^> ok :)
[22:14:21] <robin_sz> well within the power range of a Nema 34 modern stepper
[22:14:38] <Jacky^> never eard
[22:15:01] <robin_sz> of Nema34 size?
[22:15:05] <robin_sz> google
[22:15:25] <Jacky^> * Jacky^ googling Nema 34
[22:16:16] <robin_sz> its justa size of motor flange
[22:16:30] <robin_sz> http://www.imshome.com/mdrive34.html
[22:16:37] <robin_sz> those are neet too ...
[22:16:41] <robin_sz> drive and motor all in one
[22:18:07] <Jacky^> bookmarked :P
[22:21:33] <Jacky^> internal encoder is optional
[22:21:58] <Jacky^> nice motor
[22:24:31] <Jymmm> is there somthing within emc to estimate how long a program will take to complete?
[22:25:17] <cradek> emcplot3d will display the total lengths of jogs and cuts
[22:25:46] <cradek> if you know something about your average feed and cut rate, you can make a good guess using that
[22:26:33] <les> my cam estimates the time.
[22:26:42] <Jymmm> cradek this isn't part of bdi live, is ot?
[22:26:48] <Yuga> can some one here please c if they can open http://www.frogproductions.co.za?
[22:26:49] <cradek> Jymmm: no idea
[22:26:57] <les> oh Chris, very smooth high speed motions on the big iron today with tpfix
[22:27:05] <cradek> wonderful
[22:27:14] <cradek> did you try colinear segments?
[22:27:34] <cradek> it still seems wrong (but probably less wrong?)
[22:27:46] <les> yes I did
[22:27:53] <les> seemed very smooth
[22:28:01] <les> I did not graph anything
[22:28:23] <Yuga> any one manage to open that page?
[22:28:30] <les> it might be still a little wrong...but it is about 1000% better
[22:28:41] <les> yuga the page was dead for me
[22:28:42] <cradek> that's great news
[22:28:50] <les> sure is!
[22:29:51] <les> It is good enough that I will go ahead and up the spindle hp while retaining emc as the control
[22:30:06] <cradek> great
[22:30:10] <cradek> I'm glad you don't have to ditch emc
[22:30:27] <les> me too
[22:31:03] <les> now ahem...if we can just get that servo rate up
[22:31:11] <les> always more ;)
[22:31:14] <Jymmm> Um, does the mouse get disabled during program run?
[22:31:15] <cradek> get a faster machine!
[22:31:23] <anonimasu> lol
[22:31:27] <anonimasu> or a usc
[22:31:27] <les> doesn't help Chris
[22:31:28] <anonimasu> :)
[22:31:40] <cradek> les: what's the limitation?
[22:31:42] <anonimasu> if he gets a faster machine he will increase peroid ;)
[22:31:44] <les> 400 us brick wall due to comms
[22:31:56] <cradek> isa bus?
[22:32:01] <anonimasu> no
[22:32:10] <anonimasu> userspace/realtiem comms
[22:32:16] <anonimasu> realtime
[22:32:50] <les> We did not test all machines, but paul thinks 400 us is it....150 MHz or 3 GHz
[22:32:57] <Jymmm> nm... stupid KVM switch!
[22:33:02] <les> user/kernel comms
[22:33:09] <anonimasu> need to add lookahead.. so you can keep the next move in memory already
[22:33:33] <anonimasu> and well pass several messages at once..
[22:33:44] <les> yes that's the idea
[22:34:15] <anonimasu> if we could do that we wouldnt haev any brick wall..
[22:34:30] <anonimasu> not anywhere near where we wille running though..
[22:34:32] <les> I don't know how all that works....seems you would be just passing pointers
[22:34:45] <anonimasu> les: it's to do with shmem :/
[22:34:58] <les> so I have been told
[22:35:12] <les> but I do not understand why it must be so slow
[22:35:21] <cradek> has anyone asked jmk about it? It seems like he's our expert
[22:35:27] <anonimasu> he didnt know..
[22:35:43] <les> 400 us is geological time in a processor!
[22:35:52] <anonimasu> les: probably because you have one query per loop
[22:36:06] <les> and many many loops?
[22:36:17] <anonimasu> err every period
[22:36:29] <les> of the rt thread?
[22:36:40] <anonimasu> yeah
[22:37:00] <les> seems strange to me
[22:37:31] <anonimasu> or well
[22:37:35] <anonimasu> userspace <> rt
[22:37:43] <les> yeah
[22:37:52] <anonimasu> so you end up with lots of queries and little data
[22:38:36] <anonimasu> I dont understand why they designed this stuff as periodic..
[22:38:37] <les> and I remember running many point FIR digital filters in early pentiums at 100 kHz in dos no problem long ago
[22:39:11] <anonimasu> why dosent the motion part grab a new segment into memory before it's done with the previous one?
[22:39:18] <anonimasu> err servo..
[22:39:42] <les> need one query witha LOT of data passed per period?
[22:40:14] <les> anders one point ahead and one behind is required for trapezoidal
[22:45:45] <alex_joni> couple reboots later ;)
[22:45:57] <Jymmm> couple hundered =)
[22:46:07] <alex_joni> nah.. about 4 or so
[22:46:10] <Jymmm> heh
[22:46:18] <alex_joni> till I figured out wtf this driver crashed ;)
[22:46:22] <Jymmm> alex_joni BDI Live lieks my laptop it seems
[22:46:25] <Jymmm> likes
[22:46:30] <alex_joni> ditto
[22:47:41] <Jymmm> now, if I could only figure out why the axis position keeps flickering (like it was connected to encoders)
[22:49:09] <Jymmm> keeps jumping from 56.1086 to 56.1340
[22:49:10] <Yuga> *cough*fucked*cough*
[22:49:24] <alex_joni> that might be ok
[22:49:24] <alex_joni> if it jumps
[22:49:32] <alex_joni> switch tkemc to commanded position
[22:49:33] <Jymmm> Yuga Theres no mahcine connected to it, so HA! bite me!
[22:49:38] <alex_joni> and see if that moves
[22:49:40] <Yuga> hehe
[22:49:53] <robin_sz> * robin_sz nods
[22:50:01] <Yuga> Jymmm... atleast u got a machine
[22:50:07] <alex_joni> it happens that the commanded position is between steps
[22:50:15] <Jymmm> Yuga you building one?
[22:50:22] <Jymmm> alex_joni ah
[22:50:25] <robin_sz> Jymmm: I asked, you didnt asnwer ... what has happened to your, returned yet
[22:50:26] <alex_joni> and the feedback position keeps jumping from a step to the next and back
[22:50:28] <Yuga> yep
[22:50:39] <Yuga> well trying atleast...
[22:50:43] <alex_joni> robin_sz: this module stuff is dull
[22:50:49] <Jymmm> robin_sz still waiting for it to be shipped abck to me.
[22:50:54] <alex_joni> if a module crashes during load.. you cannot remove it
[22:51:04] <robin_sz> Jymmm: got a due date yet?
[22:51:08] <robin_sz> thought you said 3rd?
[22:51:09] <Jymmm> Yuga router? mill? lathe?
[22:51:11] <alex_joni> next life
[22:51:22] <Yuga> router... u Jymmm?
[22:51:37] <Jymmm> robin_sz Yep, Aug 3rd. their shipment of rails was late.
[22:51:52] <robin_sz> * robin_sz writes that one down
[22:52:11] <Jymmm> Yuga ditto, http://k2cnc.com/CNC-router-KS-2525_detail.asp
[22:52:27] <Yuga> cool... what size?
[22:52:44] <robin_sz> Jymmm: can I guess the next one ... too much snow to get it onto the truck?
[22:52:50] <Jymmm> I told them to get it right this time, so if it take em a lil longer, fine.
[22:53:43] <Yuga> Jymmm... u cheater!!!! build your one :P
[22:54:32] <les> bet you are hot to get that in jymmm
[22:55:19] <les> When I was a kid I would send off cereal box tops for secret spy ring
[22:55:45] <les> would check the mail box every day right after the mailman came
[22:55:49] <les> for six weeks
[22:56:17] <alex_joni> lol
[22:56:37] <robin_sz> les: six weeks? oh, Jymmm will be waiting longer than that ;)
[22:57:03] <les> have the same thing now....600 lb disk plow to be delivered when dealer "gets around to it"
[22:57:06] <les> hahaha
[22:57:20] <robin_sz> paid him?
[22:57:36] <les> cod!
[22:57:39] <les> of course
[22:58:03] <robin_sz> ahh, that should encourage him to ship I guess
[22:58:07] <Yuga> some one recomend me cheap servo's plz... web addy
[22:58:32] <les> and I really need to take the tractortoy and basically level the town and make it look like mars
[22:59:07] <les> yuga where are you?
[22:59:36] <Yuga> * Yuga check's... infront of my pc
[22:59:39] <Yuga> :)
[22:59:43] <les> haha
[22:59:46] <les> come on
[23:00:00] <Yuga> south africa... but more than willing to import stuff
[23:00:10] <les> oh yeah...forgot
[23:00:16] <les> hmmm
[23:00:54] <Yuga> yep... i'm the chap with lions in my back yard
[23:01:07] <les> right
[23:01:12] <Jacky^> Yuga: :)
[23:01:13] <les> what size servo?
[23:01:29] <Yuga> les... not a clue... i dont realy know what i need
[23:01:46] <les> what type machine?
[23:02:00] <Yuga> making a big router... 2x3m
[23:02:16] <les> pity I sold all of mine
[23:02:25] <les> checking
[23:02:36] <Yuga> thanks
[23:03:38] <Jacky^> Yuga: what you must machine ? the entire forest ? O_o
[23:03:48] <Jacky^> 3 mt ??
[23:03:59] <Yuga> Jacky^... the size of the sheets of wood
[23:04:01] <alex_joni> thought there are bigger trees than that
[23:04:07] <Yuga> i dont realy have a choice
[23:04:08] <alex_joni> over there
[23:04:13] <alex_joni> ;)
[23:04:26] <Yuga> yes... in africa we do shit big :P
[23:04:31] <les> http://www.servosystems.com/servo_motors_surplus2.htm
[23:04:40] <les> I have bought from them
[23:04:50] <les> they tried to buy my stock too
[23:04:57] <Yuga> what size u but? and for what size machine?
[23:05:14] <Jymmm> les: They are going to pallet it back to me fully assumbled and test it 100%
[23:05:54] <les> I had 110 large sem 1200 in oz peak in stock...just right for routers
[23:06:06] <les> great jymmm
[23:06:23] <les> I sold them very quickly.
[23:07:38] <Jymmm> les and you didn't save three for me?!
[23:08:40] <anonimasu> Jymmm: he didnt save any for me either :(
[23:09:39] <Jymmm> damn gnuploy is missing from bdi live
[23:09:43] <Jymmm> plot
[23:14:18] <Jymmm> hi JymmmEMC
[23:14:26] <JymmmEMC> hi Jymmm
[23:16:14] <Jacky^> ^0-0^
[23:16:39] <JymmmEMC> Jacky^: Since I cant route anthing right now, I cloned myself instead
[23:16:50] <Jacky^> lol
[23:18:34] <Jacky^> JymmmEMC: how is the weather there, hot ?
[23:19:09] <JymmmEMC> Jacky^: hot enough to have the A/C on low
[23:19:21] <Jacky^> also here..
[23:19:33] <Jacky^> latest days are better
[23:19:36] <JymmmEMC> rel humid today, making it really mugging
[23:21:36] <Jacky^> mmm
[23:22:28] <Jacky^> my zone is ventilated enough
[23:23:13] <Jacky^> they call the zone hit from all winds
[23:23:49] <Jacky^> winds are playng to cards here
[23:24:07] <Jacky^> shooting I shooting you
[23:24:10] <Jacky^> :)