#emc | Logs for 2009-01-07

[01:13:37] <shrdlu-> how do I work out my lead screw pitch if I have a belt system?
[01:17:41] <eric_unterhause1> how many teeth on the belt pullesy
[01:19:14] <BigJohnT> how many turns does the motor go to go one inch/mm?
[01:19:42] <eric_unterhause1> 4
[01:20:24] <shrdlu-> * shrdlu- looks
[01:20:59] <eric_unterhause1> what's an inch per millimeter?
[01:21:12] <BigJohnT> inch or mm what ever you use
[01:22:07] <eric_unterhause1> pidgin has a "last said" feature, that's nice
[01:22:30] <shrdlu-> damn, no teeth on this end, the other end is kinda inaccessible
[01:22:45] <eric_unterhause1> itsa gonna slip
[01:23:23] <eric_unterhause1> eric_unterhause1 is now known as unterhausen
[01:23:52] <shrdlu-> it doesn't slip with the original driver board
[01:23:54] <BigJohnT> shrdlu-: you could take a SWAG at it and measure the actual distance moved vs the commanded distance then do the math
[01:24:01] <shrdlu-> I assume that wont change
[01:24:26] <shrdlu-> I will try that, thankyaw
[01:24:30] <unterhausen> shouldn't change
[01:24:41] <unterhausen> what were you using before?
[01:25:19] <shrdlu-> crappy plotter card that came with the laser
[01:26:07] <unterhausen> so it's a commercial laser?
[01:26:39] <shrdlu-> yeah, chinese cheapo 40W co2
[01:26:52] <unterhausen> did you retrofit with something better?
[01:28:13] <shrdlu-> just this card, so far
[01:28:30] <shrdlu-> although this is a cheap chinese driver card too ;)
[01:28:34] <shrdlu-> not crappy, however
[01:28:54] <shrdlu-> very nice, for the price, in fact
[01:33:29] <vlmarshall> hoorah.
[03:36:05] <LawrenceG> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=170291998953 anybody used this guys spindles for pcb work?
[03:36:25] <cradek> yes, they work great
[03:37:17] <LawrenceG> hi Chris... cool... I was hoping you would say that... I just bought one
[03:37:47] <unterhausen> what are you going to power it with?
[03:39:21] <LawrenceG> I have had my eye on them for a while.... to power it I have 2 possiblities... a small a/c universal motor from an old handheld kitchen blender or maybe use a small trim router running at 1/2 speed
[03:39:43] <LawrenceG> to keep the noise down
[03:40:44] <jepler> I'm glad to see he's got them listed again
[03:40:53] <LawrenceG> I got tired of trying to make pcb's with the 3000rpm shoptask spindle.... about 4"/min was top speed
[03:41:18] <jepler> LawrenceG: when I bought mine he was selling a small AC motor with pulley that he says runs the motor at about 20krpm; I haven't tested the speed. It *is* noisy.
[03:41:44] <LawrenceG> Hi Jeff... he is only making the top end ones now... 3 ceramic bearings
[03:41:57] <jepler> yes, I got a "triple bearing"
[03:42:01] <jepler> here's my writeup: http://axis.unpy.net/01218927215
[03:42:12] <unterhausen> http://www.youtube.com/watch?v=DT0xDY31ZX0
[03:42:48] <jepler> runout is very low so unlike with the dremel I could actually get narrow traces. and the comparison for cutting out the boards is pretty stark too, the wolfgang spindle gave a nice clean edge.
[03:42:52] <unterhausen> oh yeah
[03:43:18] <jepler> (er, narrow trace isolation I should say)
[03:45:23] <jepler> 'night all
[03:45:33] <LawrenceG> cheers
[03:47:16] <unterhausen> I have some spindles that would probably work great for this, but I don't know how to run them
[03:47:27] <LawrenceG> in that video, the guy is using a 0.005" endmill...
[03:47:30] <unterhausen> in fact, I have the spindle drive off of a ttech machine
[03:47:54] <unterhausen> what were you going to use?
[03:48:46] <unterhausen> jepler is using .03125
[03:48:49] <LawrenceG> I have had really good luck with a vbit from a guy in California... kind of a diamond shape tip
[03:49:41] <unterhausen> yeah, video guy mentioned he was too incompetent to set his height correctly, so an endmill was better
[03:52:37] <LawrenceG> I love being able to mill pcbs..... lots of microcontroller projects get built just because I can make a pcb
[03:53:22] <unterhausen> I'm trying to figure out a z axis since I can't find the little ballscrews I want to use :(
[03:53:36] <unterhausen> they're in the basement somewhere
[04:20:44] <unterhausen> LawrenceG: do you have build details about your pcb mill online?
[04:28:13] <LawrenceG> I think I have a pix... let me look for a link
[04:28:42] <LawrenceG> http://members.shaw.ca/swstuff/spindle-encoder.html shows the machine and some of its work
[04:31:35] <LawrenceG> http://imagebin.ca/view/8fgrLms.html is a better machine picture
[04:31:47] <unterhausen> I like that spindle encoder
[04:31:57] <cradek> yeah that is cool
[04:32:33] <cradek> that's the little spindle mounted? how do you run it then?
[04:32:42] <LawrenceG> It works well, but I havent really put the lathe config through its paces yet
[04:32:46] <unterhausen> he spins the shop task
[04:32:51] <cradek> ha
[04:33:06] <cradek> I thought briefly that you could hold the pulley and spin the body of the spindle ...
[04:33:27] <unterhausen> actually my joke didn't come out right
[04:33:31] <cradek> the spindle would be 'going', but not so useful for engraving
[04:33:55] <LawrenceG> chris... the pix shows an extenstion... I dont have the spindle yet.... thinking of maybing adding a second Z axis if the end of the existing one
[04:34:00] <cradek> oh if it has a draw bar, you could put the motor at top
[04:34:08] <cradek> oh, ok
[04:34:09] <unterhausen> the machinist at work has a high speed spindle that goes in some tooling in the main machine spindle
[04:34:26] <unterhausen> the high speed hangs off the side
[04:34:26] <cradek> I'd sure like to have that, but have not come up with a working design
[04:34:45] <cradek> yeah 'clamp to the quill' could work, but eats travel etc.
[04:34:54] <LawrenceG> yea.. its not easy to turn 3000rpm into 30000
[04:35:44] <LawrenceG> quil is pretty bad in the shoptask... I have another axis module I could mount as a new Z slide
[04:35:46] <unterhausen> don't recall the full details of the attachment
[04:36:29] <unterhausen> there is an anti-shoptask fan group at cnczone, I saw one of their threads get locked
[04:36:37] <cradek> heh
[04:36:51] <cradek> the 3" travel or whatever it is would sure be painful
[04:37:08] <cradek> with a knee that would be rough - without, jeez
[04:37:30] <unterhausen> never thought of that
[04:38:05] <LawrenceG> yea.... tools are always too short or too long.... the pix show several ways of building up the work so the tool can get to it
[04:38:11] <unterhausen> I don't really feel comfortable using a lot of quill travel though, I move the knee a lot
[04:38:23] <cradek> I've heard jmk complain about having to put everything up on stilts
[04:38:32] <LawrenceG> yup
[04:39:13] <cradek> unterhausen: I have never noticed a difference in rigidity with quill in/out on my bridgeport rigid-ram
[04:39:44] <LawrenceG> the mill head is just bolted to the top of the lathe spindle box.... need to fab up a knee for it...
[04:40:04] <cradek> there you go
[04:40:24] <LawrenceG> there is a tree cnc mill available withing driving distance, but my shop is already overflowing
[04:40:42] <unterhausen> tree would be nice though
[04:40:44] <cradek> would there be room if you pushed the shoptask out the door? :-)
[04:40:54] <LawrenceG> probably!
[04:40:57] <unterhausen> you need a lathe
[04:43:15] <LawrenceG> Tree Journeyman 330 CNC for $2300 with a broken Dynapath s10 control, some erickson quickchange 40 taper tooling
[04:43:52] <unterhausen> does it have working motors?
[04:44:02] <cradek> is the spindle drilled for a drawbar? it would be nice to be able to use regular 40 tooling. (I bet QC40 is a lot rarer than QC30)
[04:44:33] <unterhausen> but the good news is there is no competition for it
[04:45:03] <LawrenceG> not sure... havent talked to the guy..... it may be gone
[04:45:13] <unterhausen> LawrenceG: I just emailed the guy, told him to deliver it to your house tomorrow or else
[04:45:27] <unterhausen> looks like lots of tooling
[04:45:49] <cradek> if I was doing it again, I'd try really hard to get a 40 taper spindle with power drawbar
[04:45:51] <unterhausen> http://vancouver.en.craigslist.ca/tls/961912329.html correct?
[04:46:03] <LawrenceG> thanks... will wait for the truck tomorrow!
[04:46:08] <unterhausen> what's wrong with 30 taper
[04:46:10] <unterhausen> ?
[04:46:35] <cradek> no cheap tooling, you need to use a wrench for the QC nut if doing anything heavy
[04:46:39] <LawrenceG> yes... thats the one.... they dont come up very often around here
[04:46:43] <cradek> (I've had it come loose a couple times)
[04:46:48] <unterhausen> look at the size of those control cabinets
[04:46:50] <cradek> a pull stud is a superior design
[04:47:39] <cradek> ah, plenty of tooling, that's nice
[04:47:53] <LawrenceG> he must be proud of the tool holders.... more pictures of them than the mill
[04:48:01] <unterhausen> almost looks like some kind of power drawbar thing going on there
[04:48:17] <unterhausen> the tooling might be worth more than the mill
[04:48:45] <cradek> yeah, odd, I don't see a QC nut (hard to see)
[04:49:21] <unterhausen> do you see the suspicious hump on top?
[04:49:23] <cradek> but there's the wrench for it I bet
[04:49:33] <unterhausen> true
[04:49:52] <cradek> yeah I don't know what the lump is
[04:50:06] <unterhausen> the tool that's on there right now has a tail
[04:50:40] <unterhausen> missing the x axis motor?
[04:50:56] <LawrenceG> 6000lbs would just work on a car trailer... I used to have a 7200gvw one... but it would be very dicely going past the weigh scales
[04:50:56] <cradek> yeah, missing something there
[04:51:36] <unterhausen> the cabinets just are mind boggling
[04:51:44] <cradek> neat - encoder (? tach ? resolver?) on the screws, not the motors
[04:51:52] <unterhausen> probably a resolver
[04:52:11] <unterhausen> seems like they had dualing feedback
[04:52:39] <cradek> tach in motors probably (extra brushes)
[04:53:14] <unterhausen> from before they figured out how to take a derivitive
[04:53:35] <cradek> from when they weren't scared of analog electronics :-)
[04:53:45] <LawrenceG> could probably put the shoptask in one of the cabinets
[04:53:52] <unterhausen> before they realized they should be afraid of them
[04:54:02] <cradek> nice 'pendant' though
[04:54:13] <cradek> remove cabinets, keep pendant
[04:54:25] <unterhausen> mpg is nice
[04:54:35] <LawrenceG> mount emc computer inside pendant
[04:55:07] <cradek> IMO, it's too expensive if it's not working and has motors etc missing
[04:55:08] <unterhausen> I wish I had that first cabinet on my BP, mine is not that nice
[04:55:31] <cradek> oh, BC, I have no idea of the market there
[04:55:46] <unterhausen> $1500 max I would think
[04:55:58] <unterhausen> but I'm an optimist about motors and controls
[04:56:04] <cradek> wait six months and offer $500
[04:56:15] <cradek> it's not going anywhere and it surely in the way
[04:56:20] <unterhausen> I bet it would move tomorrow at $1k
[04:56:38] <unterhausen> or maybe not
[04:56:44] <cradek> hard to guess
[04:56:55] <cradek> depends if anyone wants it
[04:57:00] <unterhausen> problem with waiting 6 months is that it looks like parts might get thrown out
[04:58:09] <LawrenceG> the shop looks kind of clean.... I dont think that machine has ever run in that shop
[04:58:27] <unterhausen> the machine itself is pretty clean for a cnc
[05:00:01] <LawrenceG> thats the cool things with cnc... if they break early enough, they may never get fixed... or they get worked until they drop making 1 type of widget
[05:00:40] <LawrenceG> missing motor does make it look like a project machine that got dumped
[05:00:42] <unterhausen> I think those machines have an allergy to forklifts, that x axis motor gets broke
[05:01:11] <unterhausen> I keep reading that dc motors wear out, maybe it wore out
[05:02:53] <LawrenceG> could be.... I have 3 nice brushless 400w servos that would probably work, but drives are "to be developed"
[05:03:17] <unterhausen> seems small
[05:04:43] <unterhausen> I was spinning a brushless motor all day, got really tired of the lovejoy coupling falling off :\
[05:04:49] <LawrenceG> they are.... I bought them more for a gantry, not 6000lbs of cast iron
[05:05:42] <unterhausen> are those the sanyos from Nebraska ?
[05:07:15] <LawrenceG> I think so.... they had 300, 400 and 1kw at the time.... what a deal
[05:07:34] <LawrenceG> very good for holding down the work bench!
[05:07:49] <unterhausen> you ought to see my motor shelf in the storage room
[05:08:13] <LawrenceG> where are you?
[05:08:15] <unterhausen> I'm straightening out two warped 2x6"
[05:08:21] <unterhausen> pennsylvania
[05:08:29] <unterhausen> I guess I'll have to send pictures
[05:08:44] <LawrenceG> kind of far to drop by with a six pack
[05:11:55] <LawrenceG> hi ted
[11:27:53] <christel> [Global Notice] Hi all! I'm about to do a tiny bit of rehubbing to try sort out some desync issues with one of our client servers, affected users ~2500. It will be somewhat bumpy but should be quick! Apologies for the inconvenience and have a nice day.
[12:59:28] <flexmotioncnc> excuse my friends i want help for one problem on the emc2.2.8 is possible?
[13:00:24] <BigJohnT> ask away
[13:00:33] <archivist> yes just ask and if anyone can answer they will
[13:01:39] <flexmotioncnc> ok thanks
[13:01:47] <flexmotioncnc> the problem is:
[13:02:12] <flexmotioncnc> my configuration is debian etch 4.05
[13:02:23] <flexmotioncnc> rtai-3.5-cv patched
[13:02:33] <flexmotioncnc> emc2.2.8 compided ok
[13:03:14] <alex_joni> go on..
[13:03:18] <flexmotioncnc> hardware: mesa 5i20card whit analog interface
[13:04:20] <flexmotioncnc> when i start the emc go on no problem, the axis move ok and the scroll on screnn position axis increase or decreas with succes
[13:05:28] <flexmotioncnc> after the time the on screen posizion axis sometimes stop increase for 1 second and increase normaly after. because?
[13:06:07] <flexmotioncnc> the axis when the on screeen counter is stop move normaly
[13:06:34] <flexmotioncnc> the motion is ok , but the gui stop.
[13:06:48] <alex_joni> sounds like a GUi problem
[13:06:53] <alex_joni> did you try some other GUI?
[13:06:54] <flexmotioncnc> i have canche the video card, but the problem no canche
[13:07:23] <flexmotioncnc> yes the probles is on with teck and axis ecc...
[13:08:02] <alex_joni> so, you have the same problem with every GUI?
[13:09:25] <flexmotioncnc> yes the problem is only emc2 with gui for visualization axis position : X 20,445 Y34,5 ecc..
[13:10:03] <flexmotioncnc> the couter position visualizated sometimes stop moving, but the axis move ok
[13:10:13] <jepler> a few days ago I made some suggestions on how you could troubleshoot this: use a "sim" configuration, and try to determine where between realtime and user interface the position is getting stuck. Did you do any of these things? http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2009-01-05.txt
[13:11:05] <flexmotioncnc> thank for you answwrs jepler.I have use the sim conf. , the problem not is present.
[13:12:11] <flexmotioncnc> the problem is on only the real machine use
[13:13:15] <flexmotioncnc> the raltime is ok because the test latency minlat=-1000 maxlat=1455 ed overrun =0
[13:14:31] <alex_joni> what is your BASE_THREAD?
[13:14:43] <jepler> bbl
[13:15:42] <flexmotioncnc> i don,t remebber , but it's the defaul value in m5i20.ini file.i thing 0,05 ms
[13:16:30] <alex_joni> flexmotioncnc: try running both AXIS and TkEMC, and see if both suffer from the same issue
[13:16:47] <jepler> alex_joni: he said it did
[13:16:55] <jepler> (bbl, really this time)
[13:17:02] <alex_joni> I still didn't understand if the problem you reported is only regarding AXIS or all GUI's
[13:17:28] <alex_joni> 15:05 < flexmotioncnc> yes the problem is only emc2 with gui for visualization <- I am not sure what that means
[13:17:57] <flexmotioncnc> the proble is only tthe axis visualization position.The gui fuction ok and move the windows and the menu normallly
[13:18:21] <alex_joni> can you please try to express it differently?
[13:18:59] <flexmotioncnc> i don't understating
[13:21:13] <alex_joni> what do you undestand under "axis visualisation" ?
[13:21:20] <alex_joni> what do you undestand under "axis visualisation position" ?
[13:21:29] <alex_joni> (make a screenshot, and put a pointer)
[13:23:20] <flexmotioncnc> the axis visualization position on tkec interface is the big on screen X 20,55 y40,5 z6,7 axis position
[13:24:53] <flexmotioncnc> i save one screenshot and put a pointer for this problem
[13:28:33] <alex_joni> ok, so the axis positions you mean
[13:29:03] <alex_joni> if you run a GUI with a backplot, does the backplot also freeze?
[13:29:22] <alex_joni> (for example if you run AXIS instead of TkEMC, you have a 3D preview)
[13:35:47] <flexmotioncnc> yes i have one 3d preview
[13:36:26] <flexmotioncnc> excuse i don,t read correct
[13:36:59] <flexmotioncnc> when the axis instead the 3d preview instead
[13:38:03] <flexmotioncnc> example in the AXIS interface the tool visualizzation in preview stop move when the position axis stop, but the motor move normally
[13:41:04] <skunkworks_> does this make sense? http://www.cnczone.com/forums/showthread.php?p=548400#post548400
[13:42:11] <alex_joni> skunkworks_: yes
[13:42:18] <skunkworks_> alex_joni: thanks :)
[13:42:55] <skunkworks_> good morning.
[13:44:25] <flexmotioncnc> good morning
[13:44:41] <alex_joni> flexmotioncnc: do you have some power saving enabled?
[13:46:02] <flexmotioncnc> in bios yes, but in kernel no.when i patch the kernel 2.8.17 with rtai-3.5-cv i have deleòection the apm and frequency scalable
[13:48:13] <flexmotioncnc> tommorro i compile the emc2.2.6 version and i want see if the problem present only 2.2.8.i remember on the ubuntu live 8.04 the problem not is present .the ardware is equivalent
[13:48:33] <alex_joni> flexmotioncnc: you can try the liveCD and update to 2.2.8
[13:48:46] <alex_joni> if that works, then probably there's a problem with some kernel options
[13:49:08] <alex_joni> (an emc2 update doesn't require a reboot.., so you can do it from the LiveCD)
[13:49:27] <flexmotioncnc> wath is the procedure for upgrade?
[13:52:28] <alex_joni> flexmotioncnc: put the LiveCD in, boot from it, connect to internet
[13:52:32] <alex_joni> sudo apt-get update
[13:52:36] <alex_joni> sudo apt-get install emc2
[13:52:44] <alex_joni> (that will install the newest emc2 package)
[14:01:26] <flexmotioncnc> ok thanks for all.another question.For compile emc2.2.8 is possible use ubuntu or is possible only debian?
[14:03:57] <jepler> of course it is possible.
[14:04:13] <jepler> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2#On_Ubuntu_6_06_or_8_04_from_source
[14:05:30] <flexmotioncnc> I have install on debian more library for possible compile the emc2.2.2.On ununtu live this library not present.The debian etch 4.05 have 3 dvd data, unubutu only one
[14:06:20] <alex_joni> flexmotioncnc: you can probably download a more complete ubuntu dvd which you can use along with the emc2 liveCD
[14:06:24] <alex_joni> (for installing packages)
[14:08:38] <flexmotioncnc> ok thanks for all at tomorrow.
[14:17:54] <skunkworks_> cradek: http://www.machsupport.com/forum/index.php/topic,9915.0.html
[14:17:55] <skunkworks_> :)
[14:33:39] <pjm__> afternoon all ,if i use the ubuntu update tool to upgrade 2.2.7 to 2.2.8 i take it my curretn config files wont get over written, i think not but just wanted to check wiht the experts
[14:34:23] <pjm__> sorry for my crappy typing
[14:34:41] <eric_unterhausen> you mean the ones in your directory, or the sample config files?
[14:34:58] <pjm__> my existing config files
[14:35:23] <eric_unterhausen> if you have modified one of the installed config files in the installation directory, it will possibly be over written
[14:35:29] <jymm> sss
[14:35:49] <pjm__> yeah fair play, i'll take backups though
[14:35:52] <pjm__> thanks
[14:57:11] <cradek_> cradek_ is now known as cradek
[16:22:27] <cradek_> cradek_ is now known as cradek
[18:19:37] <shrldlu-> does anyone have any recommendations for 2d laser cutting cam type software?
[18:25:18] <archivist> depends what you are looking for, we have a user in here that does a lot of laser cutting
[18:26:34] <archivist> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Cam is our page of things to try
[19:15:13] <alex_joni> shrldlu-: look at sheetcam, although not open source/free
[19:25:11] <jymm> shrldlu-: Many use HPGL format - http://www.google.com/search?q=hpgl+to+gcode
[19:37:08] <shrldlu-> oops
[19:37:24] <shrldlu-> sheetcam looks ok
[19:38:14] <shrldlu-> I hate it when people ask a question then don't listen to the reply, sorry ;)
[19:38:45] <shrldlu-> yah, I've looked at the wiki, didn't see anything that had lasers in mind specifically
[19:39:39] <shrldlu-> I'd like something artworky, like inkscape or corel with some kind of ability to modify speed and pwm on the laser
[19:39:44] <shrldlu-> guess that's asking a bit much
[19:40:12] <jymm> I know that a lot of the commercial laser engraversuse HPGL so that you can draw in Corel, then it outputs HPGL and uses the colors for laser power levels
[19:40:21] <archivist> you need to speak to micges when he is in
[19:41:04] <shrldlu-> The card this machine came with used some limited set of HPGL, but it offered very little control over the speed and no PWM
[19:41:18] <shrldlu-> micges, hokay
[19:41:22] <shrldlu-> thanks
[19:42:39] <archivist> he runs 3 machines and has done a page on the wiki
[19:43:07] <shrldlu-> based on lasers?
[19:43:13] <archivist> yes
[19:43:43] <shrldlu-> woah, that's a big laser
[19:44:35] <shrldlu-> dynamic beam modulation, nice
[19:44:40] <shrldlu-> that was the kind of thing I had in mind
[19:45:47] <alex_joni> this is wild: http://www.reghardware.co.uk/2009/01/07/asus_eee_keyboard/
[19:47:02] <shrldlu-> pretty damn cool
[21:03:25] <alex_joni> good night all
[21:41:36] <cradek> does someone have some nontrivial gcode handy that uses inverse time mode? preferably also cutter comp?
[21:41:58] <SWPadnos> that's a great question. we'll get back to you on that
[21:42:00] <SWPadnos> :)
[21:42:19] <cradek> bah
[21:44:27] <archivist> * archivist looks in and runs as well
[21:45:31] <archivist> I hope to not need inverse time mode one day and for the radius of the work to be taken into account
[21:45:35] <cradek> oh well, it's probably right anyway
[21:46:11] <archivist> currently I hack values till it works
[22:09:47] <maddash> 'hack' as in 'trial-and-error'?
[22:10:20] <archivist> yes
[22:15:53] <Paragon> Happy New Year All....
[22:16:36] <Paragon> It's been a while.... Work commitments and all :-(
[22:17:32] <archivist> at least you have work
[22:18:22] <Paragon> archivist: Have you finished up now then?
[22:18:30] <archivist> no not yet
[22:18:46] <archivist> but a lot are losing jobs
[22:19:14] <Paragon> Tell me about it.... It's dire times.
[22:20:51] <Paragon> archivist: You wouldnt happen to have any idea what the max freq in khz that can be obtained from an average pc paraport using EMC?
[22:20:58] <archivist> remember that 3 phase motor local to me, I never got it, seller never replied to emails oakestrees is a bad seller
[22:21:26] <Paragon> Got my converter all working :-)
[22:21:33] <archivist> that depends on mode the par port is in
[22:21:51] <Paragon> Mode?
[22:22:48] <archivist> yes , bog standard out only, epp and another one or two
[22:23:43] <Paragon> Could it reach say 170Khz - 200Khz or am I way off?
[22:23:51] <skunkworks_> no
[22:24:19] <acemi> Paragon: ~ max 100khz
[22:24:32] <skunkworks_> I would say average computers could do 30-40khz. if your lucky >50khz
[22:25:09] <archivist> and some chips fail on certain modes
[22:25:38] <Paragon> Right... but I could put a multiplier in line to reach that right? Or even use my little pluto_p?
[22:25:44] <skunkworks_> Paragon: you're looking at hardware step generation..
[22:25:55] <skunkworks_> pluto will do step generation
[22:26:09] <Paragon> OK here what I have in mind... Bear with me...
[22:26:27] <skunkworks_> http://www.linuxcnc.org/docs/2.2/html/man/man9/pluto_step.9.html
[22:29:07] <Paragon> I was playing with a homemade induction heater / furnace sometime back using an LCLR network parrallel tank circuit driven by a PLL and half bridge. The PLL and originaly a TL464 was used to genrate the wave form to switch the half bridge which fed the tank circuit which heated the work piece. OK so far?
[22:30:02] <archivist> yup
[22:30:28] <Paragon> See here for a better discription http://www.richieburnett.co.uk/indheat.html
[22:30:44] <Paragon> Well it got me thinking...
[22:32:07] <archivist> re program the pluto as needed
[22:32:22] <Paragon> The problem I had was tracking the resonant freq of the tank circuit Oh and also working out the L-match network but thats another matter. I thought EMC or the realtime tech behind it could be used to control and monitor tank freq?
[22:34:10] <Paragon> Ie use the pc to caculate and track frequency and then shift to correct frequency as needed. I figure this would be easier (?) to program than to do in hardware?
[22:35:02] <Paragon> Also software to monitor temprature etc depending on what comes into the para port?
[22:37:17] <Paragon> Is this pratical aproach?
[22:37:19] <archivist> hmm I have a toy in the car on the way back home that could produce the graphs/plots lower down the page
[22:37:50] <skunkworks_> Sounds like a job for hal :)
[22:41:14] <Paragon> According to the site mentioned Richie sugests using a high freq voltage transformer to monitor the phase and to insure that it lags 90 degrees behind the inverter voltage as this will facilitate soft switching (ZVS) of the fets in the half bridge. What sort of circuitry would one employ to facilitate this? Shmitt triggers ?
[22:42:16] <Paragon> s/shmitt/schmitt/
[22:42:37] <archivist> or comparators depends on circuit needs
[22:45:10] <Paragon> Right... I need to do some reading on these devices as I am not that familiar with there working and applications.
[22:45:28] <archivist> I used to work for an electronics company that made a phasemeter, input was a couple of LM3xx comparators
[22:46:19] <archivist> we were testing the phase from two coil sets on a rotating mass
[22:47:06] <Paragon> maybe using XOR to match or something?
[22:48:08] <Paragon> Just found this http://eshop.engineering.uiowa.edu/NI/pdfs/00/59/DS005921.pdf
[22:48:45] <archivist> after the comparators it went to a flip flop then that gated a counter
[22:49:50] <archivist> we were not controlling, yes that looks a useful chip
[22:50:20] <Paragon> right...
[22:50:41] <archivist> it was a torque meter
[22:50:53] <Paragon> OH I see
[22:51:14] <archivist> non contact of a rotating shaft
[22:51:48] <archivist> patents should have run out by now
[22:54:14] <Paragon> A while ago then... :-)
[22:55:00] <archivist> I worked there for 20 years and left over 10 years ago
[22:59:15] <archivist> yup 20 years max
[22:59:45] <Paragon> Thats some going archivist :-)
[22:59:59] <archivist> we can use the method now :)
[23:02:28] <archivist> static coil with a toothed core (bit like a motor field) inject dc and have a gear form on the shaft, use resultant ac on circuit for measurements add another for phase
[23:03:10] <archivist> spaced apart at some distance to measure shaft twist
[23:03:44] <archivist> rotate outer assembly if shaft is slow or stationary
[23:04:20] <archivist> I may still have code :)
[23:04:30] <Paragon> lol...
[23:04:48] <archivist> but that wouldnt be clean
[23:05:20] <Paragon> archivist: gotta shoot as I have real early start tomorrow :-( Catch up with you in the evening....!
[23:05:24] <archivist> description above is as its probably similar to patent
[23:05:36] <archivist> time I went home
[23:06:07] <Paragon> Thanks for the info ... CU soon!
[23:15:57] <archivist> what I said above basicly on their web page
[23:33:25] <DaViruz_> what do you call the tool used to measure the tpi of a thread? "gängtolk" in swedish, but i can't seem to remember the english name
[23:34:16] <archivist> thread gauge or a measuring microscope
[23:35:12] <DaViruz_> thanks
[23:36:56] <archivist> I use a traveling microscope with cross hairs
[23:38:16] <archivist> I have to replicate old odd threads sometimes, I have an angle cross hair in the light path as well so can measure the flank angle
[23:38:45] <DaViruz_> that sounds like a nice method.
[23:39:01] <DaViruz_> something to think about since i need to get a nice microscope anyway
[23:39:06] <archivist> http://en.wikipedia.org/wiki/Traveling_microscope
[23:39:47] <DaViruz_> oh, that simple
[23:40:44] <archivist> and see http://www.educationscience.com/microscope/physical.htm that has graticules
[23:40:46] <DaViruz_> i'm thinking ccd camera, linear encoder and computer interface.. :-)
[23:41:05] <archivist> no old steam methods :)
[23:44:12] <archivist> also there are projection and shadow graph methods
[23:44:39] <DaViruz_> a friend of mine asked me to make an adapter for an oxygen regulator, and one of the fittings only has two turns of threads, kind of hard to make out the pitch witout proper tools
[23:48:52] <archivist> also you could use a gear micrometer with points
[23:49:57] <archivist> for measuring a hole i screw wood in and measure the marks when removed
[23:50:51] <archivist> but Im only working on old clocks, oxygen stuff if modern should be to some standard