1. Home
    1. Dashboard
    2. Search
  2. Forum
    1. Unresolved Threads
    2. Members
      1. Recent Activities
      2. Users Online
      3. Team Members
      4. Search Members
      5. Trophys
  3. Articles
  4. Blog
  5. Videos
  6. Jobs
  7. Shop
    1. Orders
  • Login or register
  • Search
This Thread
  • Everywhere
  • This Thread
  • This Forum
  • Articles
  • Pages
  • Forum
  • Blog Articles
  • Products
  • More Options
  1. Robotforum - Support and discussion community for industrial robots and cobots
  2. Forum
  3. Industrial Robot Support and Discussion Center
  4. Yaskawa Motoman Robot Forum
Your browser does not support videos RoboDK Software for simulation and programming
Visit our Mainsponsor
IRBCAM
Robotics Channel
Robotics Training
Advertise in robotics
Sponsored Ads

problems with robot positioning

  • A_ghost
  • June 12, 2024 at 7:20 AM
  • Thread is Unresolved
  • A_ghost
    Reactions Received
    1
    Trophies
    2
    Posts
    36
    • June 12, 2024 at 7:20 AM
    • #1

    We have a MA1900 robot with a DX100 controller. We can't /calibrate the tool.

    When checking the accuracy of the ruler, it turned out that the robot, when moving along the X or Y or Z axis, for example, at 500 mm in the current position, shows that it has passed 514 mm. We assume that this is why we cannot calibrate the tool

    What could be the problem of incorrect recalculation? The robot does not issue any errors and comes to zero positions.

  • ljuba
    Reactions Received
    33
    Trophies
    5
    Posts
    281
    • June 12, 2024 at 11:38 AM
    • #2

    Those are symptoms of bad calibration/home position.

  • 95devils
    Reactions Received
    231
    Trophies
    8
    Posts
    1,938
    • June 12, 2024 at 11:42 AM
    • #3

    Incorrect home position calibration.

    Controller believes different robot model is attached. Wrong -number robot. Incorrect link lengths.

    I know a thing or two, because I’ve seen a thing or two. Don't even ask about a third thing. I won't know it.

  • A_ghost
    Reactions Received
    1
    Trophies
    2
    Posts
    36
    • June 12, 2024 at 1:37 PM
    • #4
    Quote from 95devils

    Incorrect home position calibration.

    Controller believes different robot model is attached. Wrong -number robot. Incorrect link lengths.

    We checked the model on the nameplate of the manipulator and in the controller settings. They match.

    The second position also normally arrives at zero points.

  • ljuba
    Reactions Received
    33
    Trophies
    5
    Posts
    281
    • June 14, 2024 at 2:22 AM
    • #5

    To be precise, calibration has to be done with motocal.

  • A_ghost
    Reactions Received
    1
    Trophies
    2
    Posts
    36
    • June 14, 2024 at 5:55 AM
    • #6
    Quote from ljuba

    To be precise, calibration has to be done with motocal.

    can you tell us in more detail how or is there any instruction?

  • 95devils
    Reactions Received
    231
    Trophies
    8
    Posts
    1,938
    • June 14, 2024 at 1:54 PM
    • #7
    Quote from A_ghost

    can you tell us in more detail how or is there any instruction?

    MotoCal is either purely software or a combination of software and hardware. This is dependent on if MotoCal or MotoCal V is used. A job is taught, certain files and the job are brought into the software. The software calculates new correct abso data. The old MotoCal will also calculate new link length data.

    Being a robot that old, I'm sure its seen a crash or two. The mechanical home is probably off and doesn't match the kinematic model. MA 1400 and MA1900s are known to not get a good tcp, but that is extreme.

    I know a thing or two, because I’ve seen a thing or two. Don't even ask about a third thing. I won't know it.

  • A_ghost
    Reactions Received
    1
    Trophies
    2
    Posts
    36
    • June 19, 2024 at 6:33 AM
    • #8
    Quote from 95devils

    MotoCal is either purely software or a combination of software and hardware. This is dependent on if MotoCal or MotoCal V is used. A job is taught, certain files and the job are brought into the software. The software calculates new correct abso data. The old MotoCal will also calculate new link length data.

    Being a robot that old, I'm sure its seen a crash or two. The mechanical home is probably off and doesn't match the kinematic model. MA 1400 and MA1900s are known to not get a good tcp, but that is extreme.

    The robot is old but has hardly been used, operating time is about 2000 hours.

    Are there any other ways to calibrate the robot? Now, when moving along the axis by 500 mm, the error is 3 mm. Moreover, on different axes X Y Z, the deviation in different directions is both plus and minus.

    Unfortunately, there is no Motocalc program and we don't know where to get it. Maybe at least a trial version.

    On the door of the controller, in addition to the factory absolute values of the encoders, there is a nameplate with parameters.

    I know that the FD parameters are responsible for the optionality of the controller.

    What are the S4D and MF1G parameters responsible for? When comparing, one of the numbers did not match.

  • ljuba
    Reactions Received
    33
    Trophies
    5
    Posts
    281
    • June 19, 2024 at 9:44 PM
    • #9

    For the MotoCal is necessary to have hardkey. It is not cheap software.

    In order to calibrate the robot, there is a procedure, where need to be done a job with at least 25 points, need to import robot parameters, execute the calibration in software, afterward insert absolute data to robot and do the judgment job.

    So online just need to be done the calibration job with 25 points, and judgment job.


    Not many integrators have this software.

  • 95devils
    Reactions Received
    231
    Trophies
    8
    Posts
    1,938
    • June 20, 2024 at 2:24 PM
    • #10

    The S1C is jogging in teach in joint low speed.


    The MF look like they are for the zeroing function.


    The S4D I’ll have to dig into.


    What parameter was different?

    I know a thing or two, because I’ve seen a thing or two. Don't even ask about a third thing. I won't know it.

  • A_ghost
    Reactions Received
    1
    Trophies
    2
    Posts
    36
    • June 21, 2024 at 9:18 AM
    • #11
    Quote from 95devils

    The S1C is jogging in teach in joint low speed.


    The MF look like they are for the zeroing function.


    The S4D I’ll have to dig into.


    What parameter was different?

    S4D0055 the parameter was different

  • 95devils
    Reactions Received
    231
    Trophies
    8
    Posts
    1,938
    • June 21, 2024 at 2:17 PM
    • #12
    Quote from A_ghost

    S4D0055 the parameter was different

    The S4D parameters have to do with job instructions.

    I know a thing or two, because I’ve seen a thing or two. Don't even ask about a third thing. I won't know it.

  • A_ghost
    Reactions Received
    1
    Trophies
    2
    Posts
    36
    • June 21, 2024 at 2:24 PM
    • #13
    Quote from ljuba

    For the MotoCal is necessary to have hardkey. It is not cheap software.

    In order to calibrate the robot, there is a procedure, where need to be done a job with at least 25 points, need to import robot parameters, execute the calibration in software, afterward insert absolute data to robot and do the judgment job.

    So online just need to be done the calibration job with 25 points, and judgment job.


    Not many integrators have this software.

    Quote from 95devils

    MotoCal is either purely software or a combination of software and hardware. This is dependent on if MotoCal or MotoCal V is used. A job is taught, certain files and the job are brought into the software. The software calculates new correct abso data. The old MotoCal will also calculate new link length data.

    Being a robot that old, I'm sure its seen a crash or two. The mechanical home is probably off and doesn't match the kinematic model. MA 1400 and MA1900s are known to not get a good tcp, but that is extreme.

    Thank you all for the answers, it is still unclear what to do. Most likely, you need to calibrate using MOTOCALV EG .

  • Mike01TJ
    Reactions Received
    3
    Trophies
    3
    Posts
    34
    • June 21, 2024 at 5:51 PM
    • #14

    Depending on where you are, you can ship the robot to Motoman for calibration. It costs less than the software but you have to ship the robot. They can also come to you, but that may be cost prohibitive depending on you location.

Advertising from our partners

IRBCAM
Robotics Channel
Robotics Training
Advertise in robotics
Advertise in Robotics
Advertise in Robotics

Job Postings

  • Anyware Robotics is hiring!

    yzhou377 February 23, 2025 at 4:54 AM
  • How to see your Job Posting (search or recruit) here in Robot-Forum.com

    Werner Hampel November 18, 2021 at 3:44 PM
Your browser does not support videos RoboDK Software for simulation and programming

Tag Cloud

  • abb
  • Backup
  • calibration
  • Communication
  • CRX
  • DCS
  • dx100
  • dx200
  • error
  • Ethernet
  • Ethernet IP
  • external axis
  • Fanuc
  • help
  • hmi
  • I/O
  • irc5
  • IRVIsion
  • karel
  • kawasaki
  • KRC2
  • KRC4
  • KRC 4
  • KRL
  • KUKA
  • motoman
  • Offset
  • PLC
  • PROFINET
  • Program
  • Programming
  • RAPID
  • robodk
  • roboguide
  • robot
  • robotstudio
  • RSI
  • safety
  • Siemens
  • simulation
  • SPEED
  • staubli
  • tcp
  • TCP/IP
  • teach pendant
  • vision
  • Welding
  • workvisual
  • yaskawa
  • YRC1000

Thread Tag Cloud

  • abb
  • Backup
  • calibration
  • Communication
  • CRX
  • DCS
  • dx100
  • dx200
  • error
  • Ethernet
  • Ethernet IP
  • external axis
  • Fanuc
  • help
  • hmi
  • I/O
  • irc5
  • IRVIsion
  • karel
  • kawasaki
  • KRC2
  • KRC4
  • KRC 4
  • KRL
  • KUKA
  • motoman
  • Offset
  • PLC
  • PROFINET
  • Program
  • Programming
  • RAPID
  • robodk
  • roboguide
  • robot
  • robotstudio
  • RSI
  • safety
  • Siemens
  • simulation
  • SPEED
  • staubli
  • tcp
  • TCP/IP
  • teach pendant
  • vision
  • Welding
  • workvisual
  • yaskawa
  • YRC1000

Similar Threads

  • Shipmodel Pi based, the electrical and programming

    • Tweety777
    • February 2, 2024 at 4:39 PM
    • General Robotics Discussions - everything but NO Industrial Robots
  • Robot welding problems

    • mihnea0071
    • April 26, 2023 at 3:48 PM
    • KUKA Robot Forum
  • Kuka KR-200 Ackn regulator limit exceeded A6

    • Arwin
    • March 9, 2023 at 6:49 PM
    • KUKA Robot Forum
  • Abb 7th axis calibrating

    • Ivan Karlsen
    • December 8, 2022 at 7:07 AM
    • ABB Robot Forum
  • Tool positioning doesn't match virtual robot

    • cat91345
    • June 18, 2022 at 5:41 PM
    • Fanuc Robot Forum
  • Fanuc pallet tool question

    • Nicolas Romero
    • October 23, 2020 at 1:15 PM
    • Fanuc Robot Forum
  • I/O Communication problem wiith PLC

    • Dalin
    • September 27, 2018 at 9:23 PM
    • Kawasaki Robot Forum
  1. Privacy Policy
  2. Legal Notice
Powered by WoltLab Suite™
As a registered Member:
* You will see no Google advertising
* You can translate posts into your local language
* You can ask questions or help the community with your knowledge
* You can thank the authors for their help
* You can receive notifications of replies or new topics on request
* We do not sell your data - we promise

JOIN OUR GREAT ROBOTICS COMMUNITY.
Don’t have an account yet? Register yourself now and be a part of our community!
Register Yourself Lost Password
Robotforum - Support and discussion community for industrial robots and cobots in the WSC-Connect App on Google Play
Robotforum - Support and discussion community for industrial robots and cobots in the WSC-Connect App on the App Store
Download