Jump to content
OMRON Forums

JeffLowe

Members
  • Posts

    283
  • Joined

  • Last visited

  • Days Won

    1

JeffLowe last won the day on June 29 2022

JeffLowe had the most liked content!

2 Followers

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

JeffLowe's Achievements

Rising Star

Rising Star (9/14)

  • Dedicated Rare
  • First Post
  • Collaborator Rare
  • Week One Done
  • One Month Later

Recent Badges

1

Reputation

  1. Where does the Servo Analyzer inject the disturbance signal? Specifically is it before or after the trajectory filter? Similarly is the Motor[x]/CompDesPos before or after the trajectory filter?
  2. We continue to have amp enable failures, even with the ECO in place. It appears to be thermal in that after leaving the rack off over night it will often work again. This has been discussed with engineering and our understanding this is not the SSR that is failing, but rather the driver circuit supplying the LED in the SSR. Has there been any effort to replacing the driver with something more robust?
  3. I don't like seeing two PMAC devices 0 and 1 when there is only one PMAC in the system. Delete the second one. Also, please provide a list of I1..I99 as this is where the communications configuration is setup.
  4. This issue is again rearing its ugly head. our service department is reporting perhaps one to two failures in the field per week. These get returned to Axis New England for testing whereupon they return them to us as no problem found. The issue is thermal in nature where turning off for some time will restore normal function. Likewise, hitting the area of the board with canned air will also return functionality. The fault is more common with boards equipped with the ACI card. For reference look at either ECO 4042 or 5839.
  5. I am checking progress on this issue. Our application code also uses SSH and as such also will cause these errors. I have mostly stuck with defining the problem in terms of the IDE preempting servo and phase and generating errors that way, but the issue is larger than that. I would like to suggest that one approach might be to have a two tiered communications protocol so that normal communication (watch/position/unsolicited messages and so on would take place on a low priority basis on a separate port. In any case, this bug is important to us to have some sort of fix.
  6. Under prior versions of the firmware we were able to map to a shared folder on our PC by adding the following lines to the end of fstab: //192.168.0.100/share /mnt/win cifs credentials=/etc/credentials 0 1 //192.168.0.100/FTS /mnt/FTS cifs credentials=/etc/credentials 0 1 This has worked well for all revisions up to 2.6.0.1, however in the latest FW, 2.6.1.0, the first mount fails and only the second is successful. Adding the _netdev directive to the lines has no effect, however by duplicating the first line to: //192.168.0.100/share /mnt/win cifs credentials=/etc/credentials 0 0 //192.168.0.100/share /mnt/win cifs credentials=/etc/credentials 0 0 //192.168.0.100/FTS /mnt/FTS cifs credentials=/etc/credentials 0 0 Will now mount correctly, ignoring or erroring on the first mount entry and correctly picking up the next two. While this is an effective workaround, it does not seem like the correct approach and we were wondering why these changes are happening internally to the Linux configuration. It would be much better for us if we could rely on a stable configuration.
  7. Motor[x].CompDesPos is also an available node for injecting externally generated trajectory if not targeted by a CompTable.
  8. Test configuration: 24/12/6. (41.6 usec, 83.3 usec and 166.6 usec ) After resetting Sys.MinPhaseTime=0 and MaxServoTime=0 the immediatly revert to 7.6 and 37.96 respectivly. Opening and closing the Task Manager, the Sys.MinPhaseTime remains at 7.6 and the MaxServoTime Jumps to 107.24 which increases ServoErrorCtr to 33 and bymps RTIntrErrorCtr up to 2. Note there are 13 gpascii tasks. Closing and reopening the IDE MaxServoTime jumps to 151.12. At this point there are now 14 gpascii sessions along with now two instances of geterrors.
  9. For many years now we have been operating our 460 UMAC with Phase/Servo/RTI at 16/8/4 KHz with near flawless operation. Admittedly, on some certain complex part programs the 460 will throw an insufficient calculation time error. Since implementing the Quad core, the same settings have worked well and have eliminated all instances of insufficient calculation time errors. However in trying to increase performance I have increased the clocks to 32/16/8, but I now get a burst of errors on startup. These can be cleared and the system almost works, however opening the IDE (4.5.2.9) the machine crashes hard with Phase/Servo/RTI errors. Looking at task manager does not show any undue CPU utilization (I am running "Balanced"). Dropping down to 24/12/6 does not resolve this. It seems that the communications tasks are preempting the real time tasks and I do not understand this. I do note the ps -a command shows a bunch of gpascii sessions. Also, when opening or closing the apps within the IDE (Tuning/Plot/Watch) generates bursts of errors. Is there anything I can do to make the hard real time tasks more robust so that they are not preempted by the IDE? (FW is 2.6.0.,1)
  10. Is it possible to upgrade NC5 to operate under windows 10, 32 or 64 bit?
  11. The servo analyzer can inject noise/swept sine either at the desired position or at the force output nodes. Are these addresses documented? I'd like to add a dither to a valve. Also, is it possible to to inject the swept sign ahead of of the trajectory filter?
  12. We have two current development threads going, a production one based on the 460 with fw 2.1.1.6 and an arm project based on the latest FW. We have backward comparability issues with saved motor files due to all the new motor configuration variables added to the new firmware. I would like to get a copy of your internal C source for the restore motor so that I can add some firmware sensitivity to the restore process to suppress the inclusion of undefined configuration variables to the older firmware. IS this possible to get a copy?
  13. The syntax you are using is for calling gpascii from a c program. If you are in a script plc it is simply System"gpascii -2 -i/var/ftp/whateverfile.txt" gpascii is just a linux command that opens up a pmac terminal and can accept any PMAC command that can be typed into the terminal window in the IDE. the -2 switch tells it to use the project variable names such as those defined in Global Includes. The -i is for opening a file and reading and interpreting its content. At the end of a successful load of the file you will get an EOF line, or if errors it will flag the error and line number.
  14. Log in with putty and verify that gpascii correctly downloads your text file, and the syntax in the text file is correct.
  15. It should be possible to install the package, if you can find the right version. The 460 is on debian 5 (lenny) with PowerPC so it can be difficult to find a repository still up. If you can find the file I can share instructions. Here is an example file that should work for Dual/Quad Core ARM https://packages.debian.org/jessie/exfat-fuse Fat32 seems to always work. The fussy statement probably has more to do with me trying various sticks without paying attention to the formatting.
×
×
  • Create New...