Posts by DS186
-
-
First you can make a quick TCP check with your old TCP data to verify if the TCP ist still good and if the enhanced mastering improved your reorientation accuracy.
-
Yes. For example NUT 000 where the 000 are the turn numbers. You will find more detailed information about the configuration and turn numbers in the basic operation manual.
-
-
Enhanced mastering is this what you mean?
Exactly.
-
Also, the manual you need is the Profinet Operators manual B-82864EN_08.
B-82864EN/08. Just a small correction because the search function of the FANUC Portal (at least the European) is a pain.
-
Since you have a PROFINET option already installed it would make sense to go with it.
With the Molex Dual Channel PROFINET your controller can act as a PROFINET Controller and also as a PROFINET Device (at the same time).
If you have a Wago IO system you can use the Wago 750-375 PROFINET IO coupler. You can also use any other standard IO modules and connect them to the controller. In the above cases the controller would be the PROFINET Controller. If you have a PLC running you can also use your robot as a PROFINET Device and send your IOs from the PLC to the robot. You could also use a combination of both.
-
Just FyI
Zero Position Master can be done in every position of your cell where it is easy to reach marks for you. I worked in many cells where the room is really small and the 0 position from FANUC (Marks on the Robot) are not always possible to reach. If you do this, make marks on the robot to find the positions later again. They are for quick Mastering.
Bei careful! Zero position mastering must be performed with all axes at 0 (witness marks). I guess you mean the Qick mastering which is possible on every (user defined) position. This requires the Qick master reference position to be set before mastering is required. The default setting for the reference position is set to the zero position (all axes 0).
-
There is no program running in any of those. And yes it is just a standard controller. North American.
This robot was purchased used. It was originally used at one of the auto plants.
Thanks for the information. Can you verify if the standard core software is running on the controller or if there is something special installed (due to its earlier usage in an automotive plant)? Do you know which automotive plant by the way? Maybe it has something to do with the described behavior.
Performing an init start as suggested by ROBOT_G could also be worth a try.
-
Which mastering procedure did you use (Quick, Zero position, Single axis, ...). Have you checked the zero position of the robot after mastering? Mastering position was correct?
Do you get the MOTN-017 alarm when trying to move to a position with a linear motion or when using SHIFT + MOVE_TO? Did you already set up any axis limits?
Is it possible to manually jog the robot around? Create a small program with a joint motion with a position or position register in joint representation and set all values to 0. Run the program in T1. The robot should move to its zero position.
-
I am interested in your solution, is it possible to you to share it when you get back to your office?
Thank you againYes this is possible, I will keep it in mind. But it will be end of September until I'm back.
-
Some programmers also using digital inputs and outputs to differntiate between positive and negative values. Below is a simplified example to show the basic principle:
Code! Send value from robot to PLC ; R[1]=1.23 ; R[1]=R[1]*100 ; IF R[1]>=0 THEN ; DO[1:Value is positive]=ON ; ELSE ; DO[2:Value is negative]=ON ; R[1]=R[1]*(-1) ; ENDIF ; GO[1]=R[1] ;
There is much more to do to make the conversion more robust, e.g. make sure that there are no decimal places left, make sure that the value is within the maximum range of the group input/output (the maximum is 16 bit), sign check, etc.
I'm using a whole different solution for the conversion of Real2Int and Int2Real on my FANUCs.
Unfortunately, I cannot provide you with my code because I'm out of the office and I don't know the whole code by heart.
-
You should then think about using the Enhanced Mastering suggested by HawkME.
PostRE: Problem with accuracy TCP point
It won't help.
There is a procedure called poor man's enhanced mastering you could try. You have to search for it on the forum.
Before you do anything, take a backup.HawkMESeptember 8, 2023 at 3:44 AM -
But I got "SYST-007 NETWORK is the master device error". Is this because that I tested the robot with ROBOGuide? Do I have to connect my robot to a real PLC to get rid of this error?
It has probably something to do with the value of the system variable $RMT_MASTER. Also verify the correct position of the remote switch (just in case you have one on your controller).
-
Due to the number of poles of your TBOP connectors I guess you have a R-30iA Mate Controller for Europe. Take a look into the maintenance manual B-82725EN-1/08. This should be the correct manual for your controller type.
-
So you said there is no BG Logic task and no Autoexec program running on the controller. I guess you already verified this but could you double check via MENU > SYSTEM > Config if there is really no program entered under COLD START Autoexec program and HOT START Autoexec program? Do the same for BG Logic under MENU> SETUP > BG Logic and check if there is a program entered in the list which is running.
Just for clarification, we are talking about a standard R-30iB controller? North American or European version?
-
You could perform a quick test before thinking about mastering the robot. Jog the robot in a straight line in X or Y direction relative to world or to a user frame which is already taught (if there is one). If the TCP deviates from the straight X or Y path then mastering could help. Check the position of the TCP also visual and not only by the position data.
-
-
-
No, only integer is possible. You have to convert the real value into an integer before sending it via PROFINET. There are different approaches on how to do that.