Re: [Emc-users] [Fwd: Jog wheel ?]
Hi, - Original Message - From: Stephen Wille Padnos [EMAIL PROTECTED] snip You're probably thinking of the Griffin PowerMate: http://www.griffintechnology.com/products/powermate/ It used to be available in black as well, but it looks like it isn't any more. Although it's probably a quadrature encoder inside, I think it's only 32 counts per rev (at 4x). I believe that X (Linux) gets events like a keyboard, so it may work with the input driver, possibly after some modifications. It has 5 basic operations: rotation CW or CCW, button press (and release of course), and rotation CW/CCW while the button is pressed. It's not RT in any case, so it could be used with halui or a GUI, but not as a jogwheel as Andy was wanting. and I believe the encoder is metallic contacts. If you want a very controllable movement for jogging to a touch, rather than one click is a thou, for manual machining it is worth looking at: http://www.contourdesign.com/shuttlepro/shuttlexpress.htm 5 buttons (axis selection?), a spring loaded ring with 7 steps each side of centre (continuous jog at 7 different speeds - make them geometric ratios), a centre jog wheel (coarse steps but fine for the last few thou). I removed my real MPG from a mill in preference for this $60 device . You can whip from and to end of end to end of an axis at full rapid but slow to a kissing touch with great safety. IMO a case of not having to emulate how handles connected to the old racks of TTL controlled machines :=) Not done it with EMC2, but device is a well behaved HID (Vendor Id 0x0B33, Product Id 0x33) so it might not be too hard. Don't bother with the driver that comes with it (Win only I think anyhow). I would send someone $60 to buy a wheel if that got us a HAL module for it (seriously). John Prentice ps its brother Shuttle Pro has more buttons but I don't think I would remember which does what and no easy way to label them :=) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] [Fwd: Jog wheel ?]
John, I can't make any promises, but I believe that hal_input, which will be a part of emc 2.2, will support most HID devices recognized by Linux. How's your Linux and emc expertise level? You should be able to try out this driver with emc 2.1 by copying a few files from the development version: hal_input.py (copy to /usr/bin/hal_input, and chmod +x): http://cvs.linuxcnc.org/cvs/emc2/src/hal/user_comps/hal_input.py?rev=1.11;content-type=text%2Fplain linux_input.py (copy to /usr/lib/python2.4/site-packages/): http://cvs.linuxcnc.org/cvs/emc2/lib/python/linux_event.py?rev=1.5content-type=text/plain There is a manual page here: http://linuxcnc.org/docs/devel/html/man/man1/hal_input.1.html Once you've installed the driver, the trick is to first find out how to give the device proper permissions (section PERMISSIONS AND UDEV in the manual page), and second to find out how to tell hal_input to use that device (section INPUT SPECIFICATION in the manual page) Jeff - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] [Fwd: Jog wheel ?]
On Friday 09 March 2007, John Prentice wrote: Hi, - Original Message - From: Stephen Wille Padnos [EMAIL PROTECTED] snip You're probably thinking of the Griffin PowerMate: http://www.griffintechnology.com/products/powermate/ It used to be available in black as well, but it looks like it isn't any more. Although it's probably a quadrature encoder inside, I think it's only 32 counts per rev (at 4x). I believe that X (Linux) gets events like a keyboard, so it may work with the input driver, possibly after some modifications. It has 5 basic operations: rotation CW or CCW, button press (and release of course), and rotation CW/CCW while the button is pressed. It's not RT in any case, so it could be used with halui or a GUI, but not as a jogwheel as Andy was wanting. and I believe the encoder is metallic contacts. If you want a very controllable movement for jogging to a touch, rather than one click is a thou, for manual machining it is worth looking at: http://www.contourdesign.com/shuttlepro/shuttlexpress.htm 5 buttons (axis selection?), a spring loaded ring with 7 steps each side of centre (continuous jog at 7 different speeds - make them geometric ratios), a centre jog wheel (coarse steps but fine for the last few thou). Now that I may have, I have an editing controller that matches that description, made by panasonic for the dvc-pro stuffs. Spring loaded, center off outside dial, and a continuous inner dial with a finger depression so you can spin it. NDI how to go about interfaceing it with emc though, hints welcome obviously. I removed my real MPG from a mill in preference for this $60 device . You can whip from and to end of end to end of an axis at full rapid but slow to a kissing touch with great safety. IMO a case of not having to emulate how handles connected to the old racks of TTL controlled machines :=) Not done it with EMC2, but device is a well behaved HID (Vendor Id 0x0B33, Product Id 0x33) so it might not be too hard. Don't bother with the driver that comes with it (Win only I think anyhow). I would send someone $60 to buy a wheel if that got us a HAL module for it (seriously). John Prentice And I'd go for another $50 myself. ps its brother Shuttle Pro has more buttons but I don't think I would remember which does what and no easy way to label them :=) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVD EV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users -- Cheers, Gene There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order. -Ed Howdershelt (Author) Pull the trigger and you're garbage. -- Lady Blue - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
[Emc-users] lead screw compensation
Gentlemen, I am nearing completion of the installation of EMC2 on a small three axis mill. I will shortly need to implement lead screw compensation. I have been unable to find much in reference to this. Where would I find out how to implement this? thanks Stuart - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Return of the wrong-way DIR signal
provides the ability to trigger on them I -think- this would work, but it assumes the position-cmd signal is correct... which may not be the case. Set up a g-code routine to G0 between Z=0 and Z=0.1, loop 10 times (or all afternoon, when it seems appropriate). Run stepgen.2.position through a d/dt block, say ddt.9.in. The output is 0 when the position is increasing, showing that Z is going up. Feed ddt.9.out and a constant 0.0 to comp.0, so comp.0.out is True if the d/dt 0, so Z is going up. If HAL has an XOR... Feed stepgen.2.dir and comp.0 to xor.0. If xor.0.out goes high, then the direction points one way and the change in position points the other way. HAL seems to lack an XOR, so... Feed stepgen.2.dir and comp.0.out to sum2.0; they're both boolean, so the sum will be 0, 1, 2 or 3. (I think, anyway). Feed sum2.0.out and stepgen.2.dir to wcomp.0, set wcomp.0.min = 0.5 and wcomp.0.max = 2.5. If wcomp.0.out goes high, then stepgen.2.dir and position-cmd point differently. Now, I can write that in English, but a bit of fumbling convinces me I can't write it in HAL. If you'll cook it up for me, I'll run it to see if it fails here -and- learn how to write this stuff; I can tweak it if you'll get me started. The critical assumption here is that the commanded position doesn't glitch backwards when the direction signal does, which seems entirely unfounded. However, I can test that by applying the output to my mill. If the HAL code doesn't trigger when the mill chokes, then we've learned something useful. -- Ed - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Return of the wrong-way DIR signal
Ed wrote: HAL seems to lack an XOR, so... Not any more ;-) (although the XOR component I just added is only in the CVS version of EMC, so it won't help much) If you aren't afraid of a bit of compiling, you can get the CVS version. That will give you the xor component, and more importantly will give you any fixes that I manage to come up with. See section 3 of this page: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2 In section 3.2, use the last example, to get the development version. The CVS version will happily co-exist with your installed version, as long as you use the --enable-run-in-place option mentioned in section 3.3 of the wiki page. To run your installed emc (2.1.1) use emc, to run the CVS version, cd to the checkout directory and use scripts/emc. More later, once I get home and can look deeper. Regards, John Kasunich - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] connection from display to hal
This sounds like a good idea. I have added this feature to the development version of AXIS, for eventual inclusion in emc 2.2. The new HAL pin will be named 'axisui.jog.increment'. Jeff - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] StepTimingCalculator.ods
Hi Andy. You should download that document and save it. OpenOffice documents are stored as zip archives of XML files, so that is the correct file, but something is identifying it as an archive for you and opening it with the wrong program (I think). - Steve Andy Ibbotson wrote: Hi All, Does anyone know where I can obtain a copy of StepTimingCalculator.ods? I've followed the link in TweakingSoftwareStepGeneration but only seems to down load a zip file full of .xml files. Thanks for your help. Regards Andy - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] StepTimingCalculator.ods
Andy Ibbotson wrote: Hi All, Does anyone know where I can obtain a copy of StepTimingCalculator.ods? I've followed the link in TweakingSoftwareStepGeneration but only seems to down load a zip file full of .xml files. That collection of XML files _IS_ the spreadsheet. Unlike Microshaft, Office, Open Office uses open file storage protocols. In this case, that means standard compression technology (zip) and standard data storage (xml). The .ods extension tells the system that is it a spreadsheet. What kind of system are you downloading it with? On a Windoze box, it comes up as an archive full of XML files. This is because windoze is clueless about Open Office. On my Linux (Ubuntu 6.06) box with Open Office installed, the system understands that it is a spreadsheet, and opening it works perfectly. Regards, John Kasunich - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] StepTimingCalculator.ods
John Kasunich wrote: Andy Ibbotson wrote: Hi All, Does anyone know where I can obtain a copy of StepTimingCalculator.ods? I've followed the link in TweakingSoftwareStepGeneration but only seems to down load a zip file full of .xml files. [snip] What kind of system are you downloading it with? On a Windoze box, it comes up as an archive full of XML files. This is because windoze is clueless about Open Office. As is Linux, until OO is installed. On my Linux (Ubuntu 6.06) box with Open Office installed, the system understands that it is a spreadsheet, and opening it works perfectly. This also functions perfectly on a Windows system with OpenOffice installed, for what it's worth. - Steve - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] [Fwd: Jog wheel ?]
Jeff Thanks for this suggestion. It shows the power of such a modular system - I am conditioned by too many special cases :=) I was using EMC2 2.0 under Ubuntu so I decided it would be best to compile Trunk - just completed OK. less on /proc/bus/input/devices has the Shuttle as: I: Bus=0003 Vendor=0b33 Product=0020 Version=0117 N: Name=Contour Design ShuttleXPress P: Phys=usb-:00:1d.1-2/input0 S: Sysfs=/class/input/input6 H: Handlers=mouse2 event5 B: EV=7 B: KEY= 1fff 0 0 0 0 0 0 0 0 B: REL=180 I am struggling with the -KRAL switches and the Input Specification syntax. How does -B for buttons fit in? Could you give me an example of a suitable Input Spec to map my two HID axes and 5 buttons? Similary I am in trouble with the udev. The udev manual pages are too general for me and I cannot see how to relate your SYSFS example in hal_input doc. to the data I have. A bit of hand-holding would be much appreciated. I will post a wiki page for others who expressed interest when I get it going. Thanks John Prentice - Original Message - From: Jeff Epler [EMAIL PROTECTED] To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net Sent: Friday, March 09, 2007 1:56 PM Subject: Re: [Emc-users] [Fwd: Jog wheel ?] John, I can't make any promises, but I believe that hal_input, which will be a part of emc 2.2, will support most HID devices recognized by Linux. How's your Linux and emc expertise level? You should be able to try out this driver with emc 2.1 by copying a few files from the development version: hal_input.py (copy to /usr/bin/hal_input, and chmod +x): http://cvs.linuxcnc.org/cvs/emc2/src/hal/user_comps/hal_input.py?rev=1.11;content-type=text%2Fplain linux_input.py (copy to /usr/lib/python2.4/site-packages/): http://cvs.linuxcnc.org/cvs/emc2/lib/python/linux_event.py?rev=1.5content-type=text/plain There is a manual page here: http://linuxcnc.org/docs/devel/html/man/man1/hal_input.1.html Once you've installed the driver, the trick is to first find out how to give the device proper permissions (section PERMISSIONS AND UDEV in the manual page), and second to find out how to tell hal_input to use that device (section INPUT SPECIFICATION in the manual page) Jeff - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] StepTimingCalculator.ods
Ah ha, got it. Thanks for the guys. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Wille Padnos Sent: 09 March 2007 21:13 To: Enhanced Machine Controller (EMC) Subject: Re: [Emc-users] StepTimingCalculator.ods John Kasunich wrote: Andy Ibbotson wrote: Hi All, Does anyone know where I can obtain a copy of StepTimingCalculator.ods? I've followed the link in TweakingSoftwareStepGeneration but only seems to down load a zip file full of .xml files. [snip] What kind of system are you downloading it with? On a Windoze box, it comes up as an archive full of XML files. This is because windoze is clueless about Open Office. As is Linux, until OO is installed. On my Linux (Ubuntu 6.06) box with Open Office installed, the system understands that it is a spreadsheet, and opening it works perfectly. This also functions perfectly on a Windows system with OpenOffice installed, for what it's worth. - Steve - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDE V ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
[Emc-users] WIFI cards and Ubuntu 6.06
Hi, Another question - Has anyone successfully installed a WIFI card under Ubuntu 6.06. My CNC PC is out in the workshop and it's a pain to drag it in to the house and connect it to the router via a cable everytime I need to update files via the internet (not to mention the grief received from the Wife re. swarf on the carpet). Unfortunately my Linux skills are v. limited so the install must be fairly simple. Regards Andy -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Wille Padnos Sent: 09 March 2007 21:13 To: Enhanced Machine Controller (EMC) Subject: Re: [Emc-users] StepTimingCalculator.ods John Kasunich wrote: Andy Ibbotson wrote: Hi All, Does anyone know where I can obtain a copy of StepTimingCalculator.ods? I've followed the link in TweakingSoftwareStepGeneration but only seems to down load a zip file full of .xml files. [snip] What kind of system are you downloading it with? On a Windoze box, it comes up as an archive full of XML files. This is because windoze is clueless about Open Office. As is Linux, until OO is installed. On my Linux (Ubuntu 6.06) box with Open Office installed, the system understands that it is a spreadsheet, and opening it works perfectly. This also functions perfectly on a Windows system with OpenOffice installed, for what it's worth. - Steve - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDE V ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] WIFI cards and Ubuntu 6.06
andy wrote: Another question - Has anyone successfully installed a WIFI card under Ubuntu 6.06. My CNC PC is out in the workshop and it's a pain to drag it in to the house and connect it to the router via a cable everytime I do you already have wireless in the house? if so, another option (besides installing a wireless card in your emc linux pc) would be to install a wireless bridge in the shop. the ethernet on your PC connects to the bridge, which connects via wireless to your home network. the advantage would be that it avoids having to do any configuration on linux -- your linux box only has the ethernet. a friend has one of these and really likes it: http://www.newegg.com/Product/Product.asp?Item=N82E16833180021 paul =- paul fox, [EMAIL PROTECTED] (arlington, ma, where it's 25.3 degrees) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Return of the wrong-way DIR signal
Ed wrote: provides the ability to trigger on them I -think- this would work, but it assumes the position-cmd signal is correct... which may not be the case. Set up a g-code routine to G0 between Z=0 and Z=0.1, loop 10 times (or all afternoon, when it seems appropriate). Wrote a quick program like so: G0 Z0 G0 Z0.1 G0 Z0 G0 Z0.1 repeated the move ~80 times Run stepgen.2.position through a d/dt block, say ddt.9.in. The output is 0 when the position is increasing, showing that Z is going up. Feed ddt.9.out and a constant 0.0 to comp.0, so comp.0.out is True if the d/dt 0, so Z is going up. If HAL has an XOR... Actually what I did here was take Zvel (the ddt of the commanded Z position) and run it thru a scale block with a gain of -1, to make Zvel-inv. Then I ran Zdir to the select input of a 2 input mux. Zvel is hooked to one input of the mux, and Zvel-inv to the other. If all is well, then when the commanded velocity is positive, dir be false, and Zvel will be selected - yeilding a positive outpout. When Zvel is negative, dir should be true. That will select Zvel-inv, and yield a positive output. A negative output indicates a problem - dir not matching the commanded velocity. The bad news is that I've run that little program quite a few times, without getting a trigger. Can you suggest any hints to try to make the event happen? I'm going to be up for a couple more hours at least, if you can get on IRC that would be great. /join #emc or #emc-devel on irc.freenode.net. Regards, John Kasunich - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] WIFI cards and Ubuntu 6.06
Gene Heskett wrote: Everybody in the neighborhood now has a dsl modem with wifi in it and apparently enabled, I can see anywhere from 2 to 4 with a pocket sniffer here. Not trusting to have a wifi on my side of the firewall, I took the chicken way out and ran a hunk of cat5 to the shop, hanging overhead across the yard. Std, out of the un-real box indoor cable from Belden, its been hanging there for 3 years now, good as ever the last time I used adept to update that box. If the weather does kill it, well, cat5 is cheap, at least for me. This is really dangerous! You must not have lightning like we do! I have had the ethernet port blown out on a computer INSIDE my house. There was about 50 feet of cable, mostly in the walls, between the hub and the computer. No way would this last long here in MO, with several hundred feet hanging overhead. Most likely, whatever was at both ends would be reduced to a smoldering hulk the first time a storm came over. Jon - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] WIFI cards and Ubuntu 6.06
On Friday 09 March 2007, Jon Elson wrote: Gene Heskett wrote: Everybody in the neighborhood now has a dsl modem with wifi in it and apparently enabled, I can see anywhere from 2 to 4 with a pocket sniffer here. Not trusting to have a wifi on my side of the firewall, I took the chicken way out and ran a hunk of cat5 to the shop, hanging overhead across the yard. Std, out of the un-real box indoor cable from Belden, its been hanging there for 3 years now, good as ever the last time I used adept to update that box. If the weather does kill it, well, cat5 is cheap, at least for me. This is really dangerous! You must not have lightning like we do! I have had the ethernet port blown out on a computer INSIDE my house. There was about 50 feet of cable, mostly in the walls, between the hub and the computer. No way would this last long here in MO, with several hundred feet hanging overhead. Most likely, whatever was at both ends would be reduced to a smoldering hulk the first time a storm came over. Jon My day may well be coming I suppose. I did consider that when I strung it, but both ends of the system are bonded quite well to ground at the powerline interface. Here in the coyote.den, this whole room full of electronics all runs on a single wall plug, and all the phone lines etc go through a huge surge arrester on the wall, which is itself plugged into a 1500VA UPS. Having lost a couple of modems to lightning surges, I was bound and determined to make this whole room bounce in unison if a nearby strikes EMP induced a jolt. In 6 years now I've had NO further surge problems just because everything is bonded to that single plug, plugged into one duplex that I opened up when I set this up, and made sure all its joints clear back to the service were well soldered. The situation at the shop is similar, its about 65 or 70 feet from the 200 amp service which was new 3 years ago and whose grounding exceeds the NEC requirements by quite a bit. There at the shop, the last 10 feet of the buried 6/3 cable are inside a piece of 1 emt, and that emt is welded to the breaker box in the shop so it has a a decent real earth ground too. All the fancy stuff out there is plugged into a 6 plex plugged into one duplex, which as a 300 joule surge arrester in it. Again, the single common point for it to bounce from, all in unison even if its several thousand volts of bounce. I could get tagged by the open loop between the ground and the about 8 feet height of the cable at midpoint of a 45 foot span. Its been pretty noisy here several times, including one strike about 2 years ago that hit the top of the pole holding my service transformer, it was blinding and the crack was only a few milliseconds after the flash. They had to come and replace the street light fixture about 70 feet from my house, but I was home free. And I believe all 8 ports on my switch are still good after about 4 years of service. That cable is plugged into it. That thumping sort of a noise you hear Jon? That's me, knocking on wood. :) I do intend to go for an 802-11 connection eventually, but its going to have to be a heck of a lot more secure than the WEP keys my WAP-11 uses. -- Cheers, Gene There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order. -Ed Howdershelt (Author) User was distributing pornography on server; system seized by FBI. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Return of the wrong-way DIR signal
Repeated runs of: G0 Z0.1 G0 Z0 same thing 80 more times didn't do anything strange. So I looked for something with more random direction reversals. Flowsnake.ngc fits the bill, so I moved my glitch detection logic to X and tried it. (This is using your ini and hal files.) The result: http://jmkasunich.dyndns.org/pics/stepgen-dir-toggle-1.png The ramping trace is stepgen.0.frequency, an parameter that shows the frequency after all scaling, limiting, etc. It is in Hz. Your ini has a scale of 16000, a max stepgen velocity of 0.44 in/sec, and a max accel of 2.20 in/sec^2. That works out to 7040Hz and 35200Hz/sec respectively. The initial downward ramp drops 200Hz in 10 milliseconds, or 2Hz/sec, well withing the accel limit. The short upward ramp (which I agree shouldn't be there) rises from about -50Hz to +100Hz in about 4.5mS, which works out to 3Hz/sec. Again, within spec, but ugly. It is followed by a down ramp at about the same rate, until it reaches the original profile. I'm almost certain the glitch is because of granularity in the feedback, which currently is in whole counts. The last two steps before the first direction change are about 8mS apart. That's 125Hs, which matches up with the average value of the frequency ramp over that period. Very shortly after that, the frequency goes through zero. At that point, the stepgen has accumulated a very small fraction of a step. (The frequency at the last step was 40Hz at most, so the average between then and the zero crossing is only 20Hz. 2mS at 20Hz is 0.04 steps. So when the frequency crossed zero, it only took another 2mS to back out that 0.04 steps, and generate the first reverse step. When the control loop saw a full step of reverse motion only 2mS after crossing zero, it said hey, you are going backwards too fast and corrected the other direction. The correction lasted until a forward step was issued, and then it recovered. I've been thinking about redoing the core of the stepgen anyway - my recent work with Peter Wallace on the mesa 5i20 FPGA stepgen resulted in an approach that provides sub-one-step feedback resolution by using the low order bits of the DDS accumulator. I think that will totally eliminate this problem. But I have to code it and test it first. Give me a couple of days, and it should be in CVS. I'm not sure whether we'll want to stick it in 2.1.2 unless it gets a lot of testing first. Regards, John Kasunich - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] lead screw compensation
Stuart Stevenson wrote: Gentlemen, I will have a machine running very shortly. I cannot find anything about lead screw compensation. How would I implement it? thanks It's 1am here, and I'm about to go to bed. There are others who are a little more familiar with screwcomp, but if nobody has replied by late tomorrow (later today I guess) I'll try to answer your questions. Please be patient. Regards, John Kasunich - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] lead screw compensation
I hope to get cheaper rolled lead screws and use lead screw compensation to get ground screw performance on my lathe project, so I will be watching this thread. Just in case it might help, there is below an excerpt from: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Kinematics On Sat, 2007-03-10 at 00:12 -0600, Stuart Stevenson wrote: Gentlemen, I will have a machine running very shortly. I cannot find anything about lead screw compensation. How would I implement it? thanks Stuart --- From KINEMATICS on Wiki --- Unlike programs that ignore the difference between part and machine coordinates, EMC always makes the distinction, even when it doesn't have to. EMC uses kinematics to translate from one to the other. For machines like Sherlines and Bridgeports, where the part and machine coordinates are the same, EMC uses a kinematics module called trivkins, which stands for trivial kinematics. It is definitely trivial - it simply copies the coordinates from one side to the other. However, trivkins can be replaced with other kinematics modules, that do more complicated translations. For something like our radial drill, it is pretty easy to do the geometry to convert from X and Y to pivot and radius. (The math is known as cylindrical coordinates.) For the robot arm it is more complicated, but still just a bunch of math. Once you understand that kinematics is nothing more than translation from part to machine coordinates, it becomes clear that kinematics can be used to compensate for leadscrew error, non-square axes, and many other types of machine errors. If the table travel is 89.5 degrees to the saddle instead of 90, then for every inch the table moves, the saddle needs to move 0.00872 inches to compensate for the misalignment. The kinematics module is the ideal place to make such a correction. Leadscrew Compensation Leadscrew correction could be done in the kinematics module as well, but EMC treats it as a special case. Since the error is parallel to the travel of the screw, the correction is applied directly to the motor in question, not to any other machine coordinate. That means it can be done downstream of the kinematics module, right before the commands go out to the motors. (On the part coordinates side of the kinematics, the coordinates are a single entity that consists of three linear and three angular values. But after the kinematics, each machine coordinate is an independent value and can be compensated individually.) EMC1 had leadscrew compensation downstream of the kinematics, so that even with trivkins you could still do screw compensation. EMC2 has hooks for the same thing, but the code is currently commented out. (The actual compensation code is simple, but the method used for getting the compensation data from a file on disk to the realtime code that does the work was rather messy. In the interest of getting things working, it was disabled. It needs to be fixed and enabled.) The leadscrew compensation algorithm lets you specify two different compensation profiles, one for each direction of travel. The difference between the two represents the backlash. Using this approach you can compensate for a screw that has more backlash near the center where the wear is bad, and less near the ends where there is less wear. For users who don't want to provide customized screw error and/or backlash compensation profiles, conventional backlash compensation can be used instead. This replaces the position dependent forward and reverse compensation curves (from disk) with a fixed correction that is either plus or minus half the configured backlash value, based on the direction of travel. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] lead screw compensation
Hi Kirk, thanks for pointing out the stale docs. Lead screw compensation is fully enabled in emc2 (I think since 2.1.0 surely, not sure about 2.0.x though). You only need the following in your [AXIS_*] section [AXIS_0] .. COMP_FILE = joint0.cmp (you can also specify COMP_FILE_TYPE != 0 (default is 0) to make it use trim's, see below) .. (and of course the same for the other ones you want). The file joint0.cmp will be of the form: 0.00 0.00 -0.001279 0.10 0.098742 0.051632 0.20 0.171529 0.194216 etc /* Loads pairs of comp from the compensation file. The default way is to specify nominal, forward reverse triplets in the file However if type != 0, it expects nominal, forward_trim reverse_trim (where forward_trim = nominal - forward reverse_trim = nominal - reverse) */ Regards, Alex - Original Message - From: Kirk Wallace [EMAIL PROTECTED] To: Enhanced Machine Controller (EMC) emc-users@lists.sourceforge.net Sent: Saturday, March 10, 2007 9:14 AM Subject: Re: [Emc-users] lead screw compensation I hope to get cheaper rolled lead screws and use lead screw compensation to get ground screw performance on my lathe project, so I will be watching this thread. Just in case it might help, there is below an excerpt from: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Kinematics On Sat, 2007-03-10 at 00:12 -0600, Stuart Stevenson wrote: Gentlemen, I will have a machine running very shortly. I cannot find anything about lead screw compensation. How would I implement it? thanks Stuart --- From KINEMATICS on Wiki --- Unlike programs that ignore the difference between part and machine coordinates, EMC always makes the distinction, even when it doesn't have to. EMC uses kinematics to translate from one to the other. For machines like Sherlines and Bridgeports, where the part and machine coordinates are the same, EMC uses a kinematics module called trivkins, which stands for trivial kinematics. It is definitely trivial - it simply copies the coordinates from one side to the other. However, trivkins can be replaced with other kinematics modules, that do more complicated translations. For something like our radial drill, it is pretty easy to do the geometry to convert from X and Y to pivot and radius. (The math is known as cylindrical coordinates.) For the robot arm it is more complicated, but still just a bunch of math. Once you understand that kinematics is nothing more than translation from part to machine coordinates, it becomes clear that kinematics can be used to compensate for leadscrew error, non-square axes, and many other types of machine errors. If the table travel is 89.5 degrees to the saddle instead of 90, then for every inch the table moves, the saddle needs to move 0.00872 inches to compensate for the misalignment. The kinematics module is the ideal place to make such a correction. Leadscrew Compensation Leadscrew correction could be done in the kinematics module as well, but EMC treats it as a special case. Since the error is parallel to the travel of the screw, the correction is applied directly to the motor in question, not to any other machine coordinate. That means it can be done downstream of the kinematics module, right before the commands go out to the motors. (On the part coordinates side of the kinematics, the coordinates are a single entity that consists of three linear and three angular values. But after the kinematics, each machine coordinate is an independent value and can be compensated individually.) EMC1 had leadscrew compensation downstream of the kinematics, so that even with trivkins you could still do screw compensation. EMC2 has hooks for the same thing, but the code is currently commented out. (The actual compensation code is simple, but the method used for getting the compensation data from a file on disk to the realtime code that does the work was rather messy. In the interest of getting things working, it was disabled. It needs to be fixed and enabled.) The leadscrew compensation algorithm lets you specify two different compensation profiles, one for each direction of travel. The difference between the two represents the backlash. Using this approach you can compensate for a screw that has more backlash near the center where the wear is bad, and less near the ends where there is less wear. For users who don't want to provide customized screw error and/or backlash compensation profiles, conventional backlash compensation can be used instead. This replaces the position dependent forward and reverse compensation curves (from disk) with a fixed correction that is either plus or minus half the configured backlash value, based on the direction of travel. - Take Surveys. Earn Cash. Influence the Future of IT
