Did you change robot and controller battery? What alarm do you keep getting. Try shift + reset. With the old RJ controllers I know you have to restart the controller
to clear the battery alarm but with RJ3 I'm not sure.
Posts by TRBe
-
-
What happens when you do try to jog an axis ? You should get an alarm if something is wrong.
-
Hi , it is possible there is a communication error between the panel board and main board. Is this the only error you get?
Is the green led on the panel board on? -
That is a strange situation, do you have programs running in the background? Background logic or is there a monitor instruction in your program?
If you can post an AOA backup I can have look in roboguide. -
FWD/BWD disabled means that your buttons forward and backward do not work. I guess to prevent people from running your programs in teach mode. You can enable/disable it by pressing fctn and then disable fwd/bwd.
You are losing programs, do you mean they are not called/used by the robot or they are just gone. I Always put the write protection on for each program to prevent accidental deleting/modifying.
-
Im not sure if it would help you in this case, It had to do with the look-ahead function. I was doing some calculations between 2 motions and the robot "paused". With the lock preg and unlock preg instruction the pause was gone.
Lock position registers means the position registers after the instruction can not be changed. So the motion behaves like a normal position motion.Here is a piece from the explanation in the manual:
Motion instructions that use position registers cannot be read in advance because the values in the
position registers could be changed by the program, data transfer function, and so forth.
If the robot reads a motion instruction that uses a position register prior to its execution, the value of
the position register might yet be changed by a program or another function (such as data transfer).
Such a change is not reflected in the motion instruction that has already been read by the robot.
Consequently, the robot’s operation might be unpredictable.You can find it in the setup and operations manual under advanced functions.
Hope it helps
-
You can use the instruction Lock Preg and Unlock Preg before and after your movement.
-
Maybe Nabtesco. They make the bearings for Yaskawa and Fanuc also I believe. But I dont think you can order it from them,
contract limitations and such. -
Where are you located, there are a lot s420 robots on the second hand market. 40mm deviation sounds like a lot! Are you sure it is the axis 1 reducer?
The cost and downtime to repair will maybe be more then a second hand reducer. Just to be sure did you check the tool and fixing of the robot? Does the robot "drop" when you release the servo's? Or can you really move the robot with the brake applied on axis 1 -
Can you post you programs? Maybe we can see something then.
-
Do you have the manuals ?
-
If you use pns program calls, you have the build in hand shake with the plc, so that is an advantage. I usually use one pns for each product variant. Check the start conditions and set other signals/registers for that product in the pns program. Then call a main program. When receiving group inputs or sending group outputs to the plc Always check it with a handshake.
Specially with older controllers I have had problems with the instruction R[1]=GI[1]. -
Hi, if I understand well you will only use the robot manually? In T1 and with deadmanswith en by pressing fwd?
-
Hi I understand what you mean but the problem is I can't find the macro_30 program anywhere, its not in the macro list.
I took the AOA backup from a working robot in production and made a roboguide cell.Is it possible its has something to do with karel? Before the macro_30 is a comment with start conditions so I would
like to know what is checked in this macro_30. -
Hi guys,
I have a macro I can't find in a existing program. Its called macro_30 and when I try to execute it in Roboguide I get the error INTP-277 :
INTP-277 (%s^4, %d^5) Internal MACRO EPT data mismatch
Cause: The EPT index in macro table doesn't point the program name defined in macro table. That is, the EPT index in macro
table is incorrect.
Remedy: Please set the correct EPT index for the program name defined in macro table.What does this mean?
-
We never got rid off it totally. We improved the error a lot by absolute calibrating the arm and tool frame calibration and calculations for tool centerpoint.
You have 6 axis linked so the tolerances add up. A robot is not a cnc machine. We made a test piece that is also used for cnc machine calibration and got a motoman 165N produce a piece accurate to +/- 0.2mm. For foam and soft materials. -
We have had the same issue, with different robots. With absolute mastering and calibrating your tool frame to perfection it helps but does not disappear. It is most visible when axis 1 changes direction. If you put a scharp pin in your toolholder and jog in different directions in slow speed close to a flat surface you can see the backlash.
To see improvement/tolerance you can program a surface under 4 angles. So milling the same block in the same userframe but under diffent angles. In theory it should be a nice flat surface but you will get 4 triangles.
-
So your pick position is Always the same and accurate if you pick from a cnc. If your bins are not the same a userframe will not help you much, only if your robot is not level with the floor. If you want to use 1 program to load your bins they have be positioned to perfection and the bins have to be the same. This is not the case here.
If you would place 6 inductive sensors on your bin positioning system you can count until 64. Then on your bins make plastic plate where you can turn in 6 screws. The sensors and screws have to align offcourse.
The first bin has only 1 screw, and 5 empty, the second bin has 1 screw in the second hole, the 3 bin has 2 screws in the first and second hole. Then you know which bin you have standing in front of the robot and you can give the load program a unique offset.I hope you understand what i mean. It will be some programming and hardware work though.
-
I have had one robot that switches the outputs to 0V. So you can try and measure between your outputs and the +24VDC pins.
-
What is the purpose of putting the rods in the bins like that? Is the robot unloading a saw machine or cnc? Your tolerance (1mm in fact 0.5mm) is too small to allow bins that are not the same. I have not worked with softfloat yet. My feeling tells me that it does not help you much in this case. Maybe touchsense would be better. Im guessing too expensive. If you are moving down in world frame and the robot and your bin have different angles it will not work. Can you make the holes 35mm?
If you could make sure which bin you have (use a binary code) and its Always centered in the same position you could make one program with 40 different offsets depending which bin you have.