R-1000iA/100F programs faster in linear than joint inside Roboguide

  • More a question of curiosity than anything else.


    I received a Roboguide cell, containing a R-1000iA/100F, with R-30iB Plus controller (software V9.30) mounted upside down between two small presses. This robot will perform simple material handling, taking a 30kg part from one press and leaving it on another press.


    The robot didn't arrived yet, so the options were selected based on robot data sheet sent from FANUC.


    But what intrigued me as well the technician who originally built the cell and made the program is the fact, contrary to common sense, the same path, with same CNT and ACC combination, is little bit faster if we use linear movements instead joint movements.


    I noticed this particular model accepts 3000 mm/s speed on linear instructions, different from the majority of FANUC robots, that accept 2000 mm/s max, even the smaller ones.


    If I try to do the same path with, lets say, a R-2000iC/165F or a M900iB/280, joint program is faster than linear program. This also happen with smaller robots, like M10 series.


    I found some exceptions, like a M-410iC/110 that accept 4000 mm/s. In this case, linear program is also faster than joint program. Some LR Mate robots accepts 4000 mm/s as well. But even in these cases, robot should fly faster in joint than in linear.


    I'm intrigued about this.

  • hi

    When you did this

    " If I try to do the same path with, lets say, a R-2000iC/165F or a M900iB/280, joint program is faster than linear program. This also happen with smaller robots, like M10 series."

    where the robots upside down also ?

    Retired but still helping

  • Hi, Fabian.


    With M900, I put it upside down also. R-2000 isn't supposed to work mounted inverted, so I skipped this one in this scenario.


    But I also tried with R-1000iA mounted on floor, with R-2000, M-900 and other robots mounted on floor as well, and the results were the same.

  • I've seen this before as well. In some cases, linear will be faster than joint. In my case, I was seeing it on a R2000iB/165F running v7.70 software.

    Check out the Fanuc position converter I wrote here! Now open source!

    Check out my example Fanuc Ethernet/IP Explicit Messaging program here!

  • I've seen this before as well. In some cases, linear will be faster than joint. In my case, I was seeing it on a R2000iB/165F running v7.70 software.

    Interesting, Nation.


    With R-2000iC/165F, robot behaved as expected. But I didn't tried with R-2000iB/165F with R-30iA controller.


    I will test this later this week.

  • Interesting. Is it because of the turn number?


    Make sure the turn numbers are the same at the start and end of the movement and try again.

    All program points are with the same CONFIG : 'F U T, 0, 0, 0', HawkME. Joints and Linears


    I will try them with NUT, for curiosity.

Advertising from our partners