Linear movement not linear

Hello;

It seems wrong mastering or machine data. Please look in info menü robot type and compare real one on the manipülatör label.

If robot is wrong calibrated, positions is correct but linear and angular movements not going to parallel to world coordinate. You can make calibration again with EMT or Dial gauge.

Absolute Accuracy optional software for optimized accuracy. Not relevant to this problem.

Note: Be careful and check again. Machine data name is very similar

e.g. KR2210-2 SSCB Floor ZH 180 and KR2210 SSCB Floor ZH180 is different.

Hi ,
well I have an experience with something like that ,
well I want to get a linear trajectrory but as you said always there is a deviation !
in my case the robot was handling a tool and the trajectory is not straight .
So I checked the angle (A) of the first position of the robot before doing the linear trajectory in the config.dat it must be the same angle as the final position .
{x -241.914078,y 1108.01501,z 665.443054,a 74.5151672,b 0.0,c 0.0}
I don't know if you are in the same case as me .

That is also very common for tool calibration, if the tool point is not at the exact center of the robot arm line.
The robot will move in a linear line for sure, but that linear line might not be linear for your tool so you see it as moving at a radius from point A to point B.

Mastering pins not help to solve this problem. You can't see 1-4-6 Mastering pins because it is inside of axis. If you find EMT or Dial please calibrate again.

Quote from DaTa

First of all, thank you for your replies!

I have checked the Robot description again. It is the correct one as printed on the label of the robot:
#KR2150L110K S C2 FLR ZH150/180

I also performes a linear movement along the X-Axis and A, B, C Values did not change. That means, that the orientation of the tool does not change, correct? So the precise location of the TCP does not matter since the orientation of anything connected to the front plate should not change. Am I thinking correctly here? If so, then a new callibration of the TCP would not help.

So right now I will contact KUKA to see where I can get an EMT or a dial guage, but the home position seems to be right on the mastering pins. Would a slight error in the postiotion mess up the robot so badly?

In the mean time, does anyone have any other ideas what I could check to find the problem?

Display More

Hello DaTa,

Is your problem solved finally?

I have the same issue, even I started up two Agilus. One of them work properly but the other one has the same problem like yours.
The movement of the robot is not linear in any coordinate system nor along one axis.
For e.g. I move the robot along axis Y. Robot starts to move along Y but there is a little deviation along X, like it writes a small curve axis X away then return back to the right path. The deviation is really small, but eyes able to recognize. So, it is not accurate...

Does anybody have any idea?

Hi , i am ludo, i use a KRC4 and a KR16-3S

I have quite the same issue,
the robot didn't move in linear movement,
between 2 points, he goes quite straight but , it always go deeper (7mm for 1000mm straight line) along the Z axis and reach the final point.

Anyone has solve this issue ?

How is that motion programmed? What Tool and Base are active? Are you moving in Base or Tool coordinates? Have you confirmed that the axes of the coordinate system are parallel to the motion you want to make?

Okay...

First: what is this motion, exactly? What are the start and end coordinates?

Second: Does the robot arrive at the correct destination? That is, if you measure physically from the start to end points, has it moved by the commanded vector in each axis?

Third: How large is this "S-Curve" motion you're observing?

Dear Friend,

Few days a go some friends help to me , i has a similar problem like yours , check the version of KSS you are installing
read my post :

http://www.robot-forum.com/rob…tranges-movements-kr2210/

Hello
To skyfire I took some picture to show the issue.
The start point : X, Y, Z : 549.79, 603.44, 102.95
The end point x,y,z : -561.93,603.44,102.95

I do X axis movement in base 1
It's seem's that the vector mvt between start and end is correct.
The deviation is approx 10 mm for a displacement of 1200 mm
Yesterday i measure 8mm for 1000 mm displacement.

Thanks
Ludo

This suggests that the X axis of your Base 1 is not parallel to the piece of wood(?) you are using as a guide.

Also, does your LIN motion cross through any reversals of axis motion? This will definitely induce inaccuracy due to axis backlash.

Resolve.

I perform a calibration of the robot with the EMD device from Kuka
Now it's Ok, linear is linear.
I have had a collision few month ago, but i haven't notice this lack of accuracy before on my artwork
Now it's OK

Thanks
Ludo

Quote from DaTa

Not sure why the wrong mada got onto the robot in the first place... I am pretty sure it was installed by a Kuka technician. Interestingly the label is scratched in the exact place where the K should be.

label is not original, there are several decals on it. looks like robot went through few changes in its life and likely someone did not have right label at hand so they tried scratching the K out...