Machsupport Forum
Third party software and hardware support forums. => Galil => Topic started by: capsteve on January 11, 2011, 10:36:54 PM
-
ok i have everything running pretty good except the machine will just hang .....randomly during a program.
i have added some debounce and no improvement.
i do not see any triggers from limits occuring, or any errors that are present
the drivers are all still energized when it stops
the spindle continues to turn, its the motion that stops at random line numbers during the program
can the galil card stop motion because of an error. if so can i poll the gaglilcard to find stop codes or history??
1 in 10 times the program will run completely with all the same settings
i have turned all the brains off and tried, i have no custom macros
the mach history is empty
im open to any ideas, ....please...
-
Steve,
Enable debug output in the plugin configuration and restart Mach. Then try running your program. If it fails, exit Mach and send me the GalilDebug.txt file in your plugins directory. You can attach it to a message in this forum as well.
We will then be able to see what is happening on the Galil (communications, etc...)
Steve
-
steve,
will do , thanks
-
steve,
there is 5 or 6 grabfiles in the zip all were recorded after mach3 hung up....
hopefully you see some thing its all greek to me...;)
it actually took a while to get it to act up, once it did, it became frequent.
steve
-
Steve,
What kind of machine are you running? (Processor size, memory, operating system, etc...) I don't believe this is a Galil issue. The movement queue is drained meaning no more movement information from Mach. So it looks like something is interfering with Mach at this point. I'm not ruling the plugin out, but if you could give me the lay of your machine, I'll dig some more.
Steve
-
I have a simular issue. Although, doesn't seem to be as sever as Steve's.
Its completely randum, and when the same code is run a second time, it runs without error.
It can happen happened 5 minutes into the code, or 5 hours into the code.
I think its a windows issue(win 7 32bit)
I have to close Mach3, and restart it when it happens.
I set Mach3's priority one step higher, and (knock on wood) haven't seen a reocurance yet.
-
pentium 4 pc 3ghz
1gig ram
xp pro
nvidia gforce mx 440 video
contec pio16(pci)
galil 1842 motion controller
power scheme is set to always on
no screen saver
no anti virus
most non used features of windows turned off
no remote assistance
no remote desktop
no autoupdate
system restore point on drive c IS on
paging file is 1536 mb
hid stuff usb keyboard, usb trackball mouse,
pc running xp, everything non essential is turned off.
-
when mine locks up it acts like an unregistered version of mach3, it gets to a line then just sits.....but it does go past 100 lines....
when it does lock up i hit the stop (alt-a) then hit spindle to restart spindle then cycle start and it will take off where it was at..
steve
-
i did get a 40 minute cycle to complete yesterday......followed by a 2.5 minute lockup..
steve
-
when it does lock up i hit the stop (alt-a) then hit spindle to restart spindle then cycle start and it will take off where it was at..
steve
Thanks, I will try that if it ever happens again.
Chris
-
Steve,
Your machine looks fine. Heh... with the exception of the trackball. :) I just can't get used to those things!
Steve and Chris,
I'll get you both a new plugin this evening. I have Steve's email. So Chris, send me a PM with your email address.
This new plugin is working toward release 4.5. It's pretty different as far as how it works internally and there may be bugs in it (MPG and homing code related), so I would not suggest running it full time. But I would like to see if any of you can duplicate the lockup with this new plugin. If you can duplicate the lockup, then it's a high probability that it's not the plugin and we need to start looking for things that might make Mach hiccup.
What versions of Mach are you guys running?
-
mine says R3.042.04
steve
-
Same version R3.042.40
-
Steve,
after trying the new plugin without success, i looked over the machine. i found the encoder position wires were shielded but not
grounded. i grounded them and tested further. i thought i had the problem beat , because the machine ran flawlessly for almost 2 hours running a 20 min program.
but sadly, the machine began to stop during the cycle randomly again. i called galil to get some info on how to check for errors logged in the control card.
Tc1,sc and a few others were mentioned, after looking i dont see any reason why it should stop. i noticed that the wdsk program used to be able to run along side the mach software, now mach has
to be shut down before wsdk can see the control card.
to keep it simple i used the wizzard to write a small keyway slot program, straight with no arcs. my machine even locked up during a liesurely 3ipm g1.
most frequently the machine just stops with the spindle still turning, less often the machine estops in the middle of a move, a very few times it stopped on a limit trigger(not sure which one)
i think im going to try another computer as i wanter to use a smaller form factor anyway.
is there anything i need to do to windows 7, other than the usual turning stuff off??
thanks steve
-
Steve,
I sent you a new plugin to try.
Steve
-
any progress with this? we have one in uk doing exactly the same... pulling hair out trying to find it!
-
Hi, I'm having exactly the same problems on my router, I've got a 2160 with extended i/o running off an intel XP pro box with little else but Mach 3.042.40 installed.
It hangs in the middle of a program, spindle running, as though it's a feedhold, all servo drives enabled, sometimes Mach estops, sometimes just seems to hang, always needs a restart (otherwise next program run fails to start spindle). It can run a file for several hours before doing this, or sometimes 10 mins.
Get "Galil lost connection" message usually. As mentioned by capsteve earlier (too many steves in this thread already!), thought all our previous comms dropout problems had been sorted when we found a broken screen in an encoder cable (noise, apparently), but still happening, albeit much less frequently. Going to reinstall on a fresh PC tomorrow.
Will try the 'start spindle, cylce start' workaround tomorrow too. Had the odd little hiccup which might shed light on it, such as today when homing, one axis just ran over both the home and limit switches, then when almost beyond the back end of the switch ramps, suddenly Mach lit both LEDs, as though it had dozed off for a second..
Battwell has done all my Mach setup, so I'm sure he'll be asking much more in-depth questions, but I just saw this thread tonight after a frustrating day and thought I'd pitch in!
-
any progress with this? we have one in uk doing exactly the same... pulling hair out trying to find it!
oh hello, cross-posted you!
-
my machine doesnt seem to do this! im using a galil 2280 which is 10-100 network if that maight make a difference. (it might do it if i run a long enough file but i had it running a 3d file yesterday for an hour which it completed perfectly)
steves machine is on the same version of mach as mine .040
steves machine does act differently to mine on a few points, the main one being it requires sleep command after everything in vb scripts or it can miss doing them! mine didnt appear to be that fussy.
also of note on steves machine is explained below.
its a cnc router , ac servo driven with encoder pulses passed through amps to the galil. the machine is capable of 65 metre rapiding speeds no problem.
the test files have been cutting 3d reliefs of differing sizes, differing resolutions etc.
of note was the file seems to run fine until it stops! after this it can be restarted only by resetting mach. however unless the galil is hard reset the next file then doesnt run straight. seems to add "steps " at the end of each pass which causes the once square relief to start running at an angle of approx 20 degrees.
-
Any ideas on this random stopping problem yet? I was having several problems with my machine, so I tried replacing the 2160 with a 1750 I had on the shelf, and all my problems disappeared, apart from this one.
I previously wondered if it was an Ethernet issue, but it does the same thing with a bus controller. The 2160 was running off a custom
-made XP pc, non-onboard video gigabyte mobo, intel CPU, nothing installed except Mach and galil stuff. The 1750 is in a brand new Advantech industrial pc chassis, intel, 4gb ram, again nothing except XP, Mach and smarterm installed. In both cases everything turned off (I think)- auto updates, power management, screen saver etc.
Several people seem to be experiencing it, any thoughts from Kenny/Smurph? Did Smurph's new plugin help? ( could I get a copy to try too, I need to run a 130-hour program this week!) thanks.
-
I left the machine running last night, doing an air cut of a 35Mb relief, and it stopped after 47,000 lines... it had done 100,000 lines yesterday on a different test before I stopped it myself.
Mach hadn't crashed, but the Galil red LED was lit. Although all the screen buttons seemed to work as they should (ie they blinked as they were clicked), there was no movement because of the Galil fault. The program couldnt be restarted for the same reason.
So I've just tried again, with debug turned on, to see if it did it again.. and it did on the first move!
Here's the debug file:
10:22:42:953 |***** DMCDiagnosticsOn *****|
10:22:50:906 Entering DMCGetHandle.
10:22:50:906 Leaving DMCGetHandle.
10:22:50:906 Entering DMCVersion.
10:22:50:906 Entering DMCmCommand.
10:22:50:906 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:22:50:906 Entering DMCmCommand.
10:22:50:906 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:22:50:906 Could not parse version information <DMC1750 Rev 1.2n>.
10:22:50:906 Leaving DMCVersion.
10:22:50:906 Entering DMCCommand.
10:22:50:906 Command <MG_XQ-1>.
10:22:50:906 Entering DMCmCommand.
10:22:50:906 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:22:50:921 Response < -2.0000 :>.
10:22:50:921 Leaving DMCCommand.
10:23:18:328 Enmtering DMCmSetTimeout.
10:23:18:328 Leaving DMCmSetTimeout.
10:23:18:328 Entering DMCClear.
10:23:18:328 Entering DMCmClear.
10:23:18:328 Entering DMCmCommand.
10:23:18:328 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:23:18:328 Interrupt Routine: received an interrupt.
10:23:18:328 Interrupt Routine: interrupt value <0>.
10:23:18:328 Interrupt Routine: controller index <0>.
10:23:18:328 Interrupt Routine: waiting for interrupt.
10:23:18:328 Entering ReadData.
10:23:18:328 ReadData: about to wait for g_hDMCCommandMutex.
10:23:18:328 readGalilString: Bytes read <0>.
10:23:18:328 readGalilString: Raw bytes <>.
10:23:18:328 Bytes read <0>.
10:23:18:328 Raw bytes <>.
10:23:18:328 Leaving ReadData.
10:23:18:328 Leaving DMCmClear.
10:23:18:328 Leaving DMCClear.
10:23:18:328 Enmtering DMCmSetTimeout.
10:23:18:328 Leaving DMCmSetTimeout.
10:23:23:265 Enmtering DMCmSetTimeout.
10:23:23:265 Leaving DMCmSetTimeout.
10:23:23:265 Entering DMCReset.
10:23:23:265 Entering DMCmCommand.
10:23:23:265 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:23:23:406 Entering DMCmCommand.
10:23:23:406 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:23:23:406 Entering DMCmCommand.
10:23:23:406 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:23:23:421 Entering DMCmCommand.
10:23:23:421 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:23:23:421 Leaving DMCReset.
10:23:23:421 Enmtering DMCmSetTimeout.
10:23:23:421 Leaving DMCmSetTimeout.
10:23:28:406 Entering DMCClose.
10:23:28:406 DMCClose: about to wait for g_hDMCOpenMutex.
10:23:28:406 readGalilString: Bytes read <0>.
10:23:28:406 readGalilString: Raw bytes <>.
10:23:28:406 readGalilString: Bytes read <0>.
10:23:28:406 readGalilString: Raw bytes <>.
10:23:28:421 readGalilString: Bytes read <1>.
10:23:28:421 readGalilString: Raw bytes <:>.
10:23:28:437 Entering DMCmClose.
10:23:28:437 Entering DMCmCommand.
10:23:28:437 DMCmCommand: about to wait for g_hDMCCommandMutex.
10:23:28:437 Entering Remove Interrupt Handlers.
10:23:28:437 Removing interrupt handler for controller <0>.
10:23:28:437 UnsolicitedMsgEventHandler: Posting WM_DMCTHREADCLOSING to window.
10:23:28:437 Leaving Interrupt Handler.
10:23:28:640 Thread shutdown complete.
10:23:28:843 Removing Unsolicited Msg handler for controller <0>.
10:23:28:843 UnsolicitedMsgEventHandler: Posting WM_DMCTHREADCLOSING to window.
10:23:29:062 Leaving UnsolicitedMsgEventHandler.
10:23:29:062 Thread shutdown complete.
10:23:29:265 Leaving Remove Interrupt Handler.
10:23:29:265 Disabling communication interrupts: calling IOCTL_GALIL_ENABLECMDDONEINTERRUPT.
10:23:29:281 Closing driver <320>.
10:23:29:296 Deleted index <0>.
10:23:29:296 Leaving DMCmClose.
10:23:29:296 Leaving DMCClose.
10:23:29:328 |***** DMCDiagnosticsOff *****|
-
im still having the same stoppage issues as well. i have replaced the computer with a with a new one containing a different brand processor and new ram. this time i used vista stripped to the bone.
i've tried setting the software priority higher in the op system as someone suggested. but it remains operationally the same as it has been, locking up randomly. the debug routine that runs in the background as a diagnostic tool always seem to show my problem occurring whenever a very specific memory address is used. don't know if its in use from other plug-ins (galil, and contec) or the op system.
steve
-
come to think of it, one of the uncertain areas i have.. does anybody know how the bios should be set as far as power management control? i get two choices neither which is off.
im thinking if the pci bus slows down or powers off their could be trouble. but i havnt seen any discussion about it so it may not matter.
steve
-
Since my last post I've been able to run the machine for several hours without a stop- over 7 hours in one case.
However, last night it stopped after about half an hour.. ???
Here's the debug file from the mach folder, there are hundreds of lines where it seems to be doing g code ok, and this section is the last bit where it seems to have reset. I'm not sure whether the section beginning with the QZ command was added immediately, or during shutdown, or whatever- basically, I'd left the machine running code and gone home, but a colleague noticed that it had stopped, so he pressed reset in mach, closed mach, shut down windows and the pc, then turned off power to the machine, if that helps explain anything..
I'd be really interested to hear if any of the guys who worked on the plugin have any ideas or can suggest things to try..?
GCode ==> Y443.05 Z8.84
LI 0,78,-53,0<5892 Slots=477 Response=":" <MoveRelative()>
LI 0,84,-67,0<6714 Slots=474 Response=":" <MoveRelative()>
LI 0,95,-76,0<7602 Slots=474 Response=":" <MoveRelative()>
LI 0,104,-82,0<8276 Slots=474 Response=":" <MoveRelative()>
LI 0,107,-85,0<8540 Slots=474 Response=":" <MoveRelative()>
LI 0,107,-85,0<8540 Slots=474 Response=":" <MoveRelative()>
Do4MsGCode(): M=16, T=0.3280
LI 0,107,-78,0<8274 Slots=485 Response=":" <MoveRelative()>
LI 0,108,-68,0<7976 Slots=485 Response=":" <MoveRelative()>
LI 0,108,-58,0<7660 Slots=485 Response=":" <MoveRelative()>
LI 0,109,-48,0<7442 Slots=485 Response=":" <MoveRelative()>
LI 0,109,-37,0<7194 Slots=485 Response=":" <MoveRelative()>
LI 0,109,-27,0<7018 Slots=485 Response=":" <MoveRelative()>
LI 0,109,-16,0<6884 Slots=485 Response=":" <MoveRelative()>
LI 0,109,-7,0<6826 Slots=485 Response=":" <MoveRelative()>
LI 0,109,3,0<6814 Slots=485 Response=":" <MoveRelative()>
LI 0,110,15,0<6938 Slots=485 Response=":" <MoveRelative()>
GCode ==> Y439.93 Z8.68
LI 0,108,21,0<6876 Slots=479 Response=":" <MoveRelative()>
LI 0,108,36,0<7114 Slots=479 Response=":" <MoveRelative()>
LI 0,107,68,0<7922 Slots=479 Response=":" <MoveRelative()>
LI 0,102,106,0<9194 Slots=479 Response=":" <MoveRelative()>
LI 0,98,145,0<10938 Slots=479 Response=":" <MoveRelative()>
LI 0,94,183,0<12858 Slots=479 Response=":" <MoveRelative()>
Do4MsGCode(): M=16, T=0.1090
LI 0,90,222,0<14970 Slots=506 Response=":" <MoveRelative()>
LI 0,86,260,0<17114 Slots=506 Response=":" <MoveRelative()>
LI 0,80,293,0<18982 Slots=506 Response=":" <MoveRelative()>
LI 0,75,319,0<20480 Slots=506 Response=":" <MoveRelative()>
LI 0,68,342,0<21792 Slots=506 Response=":" <MoveRelative()>
GCode ==> Y436.11 Z9.39
LI 0,60,323,0<20532 Slots=505 Response=":" <MoveRelative()>
GCode ==> Y438.54 Z6.44
LI 0,58,259,0<16588 Slots=496 Response=":" <MoveRelative()>
LI 0,59,191,0<12494 Slots=496 Response=":" <MoveRelative()>
LI 0,60,123,0<8552 Slots=496 Response=":" <MoveRelative()>
LI 0,61,55,0<5132 Slots=496 Response=":" <MoveRelative()>
LI 0,62,-12,0<3946 Slots=496 Response=":" <MoveRelative()>
LI 0,64,-81,0<6452 Slots=496 Response=":" <MoveRelative()>
LI 0,64,-149,0<10134 Slots=496 Response=":" <MoveRelative()>
LI 0,66,-217,0<14174 Slots=496 Response=":" <MoveRelative()>
LI 0,67,-285,0<18298 Slots=496 Response=":" <MoveRelative()>
GCode ==> Y436.11 Z9.39
LI 0,68,-330,0<21058 Slots=483 Response=":" <MoveRelative()>
Do4MsGCode(): M=16, T=0.0940
LI 0,69,-337,0<21498 Slots=483 Response=":" <MoveRelative()>
LI 0,70,-338,0<21572 Slots=483 Response=":" <MoveRelative()>
LI 0,69,-337,0<21498 Slots=483 Response=":" <MoveRelative()>
LI 0,73,-325,0<20818 Slots=483 Response=":" <MoveRelative()>
LI 0,79,-295,0<19086 Slots=483 Response=":" <MoveRelative()>
LI 0,84,-259,0<17016 Slots=483 Response=":" <MoveRelative()>
LI 0,90,-222,0<14970 Slots=483 Response=":" <MoveRelative()>
LI 0,95,-187,0<13108 Slots=483 Response=":" <MoveRelative()>
LI 0,100,-150,0<11266 Slots=483 Response=":" <MoveRelative()>
LI 0,105,-115,0<9732 Slots=483 Response=":" <MoveRelative()>
LI 0,107,-79,0<8312 Slots=483 Response=":" <MoveRelative()>
LI 0,109,-44,0<7346 Slots=483 Response=":" <MoveRelative()>
LI 0,109,-10,0<6840 Slots=483 Response=":" <MoveRelative()>
GCode ==> Y432.64 Z9.12
LI 0,108,22,0<6888 Slots=482 Response=":" <MoveRelative()>
LI 0,109,34,0<7136 Slots=482 Response=":" <MoveRelative()>
LI 0,109,28,0<7032 Slots=482 Response=":" <MoveRelative()>
QZ Slots=479 Response=" 5, 24, 16, 28:" <ResetControl()>
MG_BAA Slots=479 Response=" 0.0000:" <ResetControl()>
MG_BAB Slots=479 Response=" 0.0000:" <ResetControl()>
MG_BAC Slots=479 Response=" 0.0000:" <ResetControl()>
MG_BAD Slots=479 Response=" 0.0000:" <ResetControl()>
MG_BAE Slots=479 Response=" 0.0000:" <ResetControl()>
SP 3000000,3000000,3000000,3000000 Slots=479 Response=":" <ResetControl()>
FL 2147483647,2147483647,2147483647,2147483647 Slots=479 Response=":" <DisableSoftLimits() State.nAixs = 5>
BL -2147483648,-2147483648,-2147483648,-2147483648 Slots=479 Response=":" <DisableSoftLimits() State.nAixs = 5>
GM 0,0,0,0 Slots=479 Response=":" <SlaveALL(false)>
GR 0,0,0,0 Slots=479 Response=":" <SlaveALL(false)>
DP 81813,256449,-26492,99061 Slots=479 Response=":" <ResetControl()>
MT -1.0,-1.0,-1.0,-1.0,-1.0 Slots=479 Response=":" <ResetControl()>
AB 1 Slots=479 Response=":" <ResetControl()>
MO ABCD Slots=479 Response=":" <ResetControl()>
CN 1 Slots=479 Response=":" <ResetControl() Limits low checked!>
CN ,1 Slots=479 Response=":" <ResetControl() Homes low checked!>
KPE=0 Slots=479 Response=":" <ResetControl()>
KIE=0 Slots=479 Response=":" <ResetControl()>
KDE=0 Slots=479 Response=":" <ResetControl()>
TM 1000 Slots=479 Response=":" <ResetControl()!>
OP 0 Slots=479 Response=":" <UpdateOutputs()>
-
my machine just did the unexplained stop after running a 3D file for 40 hours (it was only half way through though..!)
No Galil error LED, elapsed time still running in mach, but g code window halted.
I've noticed that after this happens and I hit stop (tried feedhold then start- doesn't do anything), mach won't let me jog even though I can click the jog button to make it green. I have to do a reset of mach before I can jog again. It seems like the user interface of mach is fine, but it's not talking to the galil any longer, although there doesn't seem to be any fault in the galil..?