TCP shaking during oscillating motion

what is the robot model? did you enter correct load information (mass, CG and inertia) for tool?

Load information is probably wrong (mass, CG and inertia)
This is low payload robot and tool does not look too heavy (heavier the tool and closer to robot maximum payload - better the LDD results)
Therefore LLD information provided by robot are far from actual values.

Mass is the easiest as you can remove the tool and put it on a scale.
Not sure how to calculate correct CG but mechanical engineer probably can do this correctly.

mt thoughts exactly...

I have faced a similar situation when working with a robot used for high precision vision measurements. In reality, the vibrations come from a combination of robot performance, accel/deccel/gear jerk/vel setpoints, and path/motion data:

1) Robot performance. I know I am stating the obvious but in general, robots cannot perform (in terms of speed, repeatability, precision, strength, etc) at the same level as gantry or more conventional CNC machines (especially in terms path following and motion). I am am only stating this because at some point, we had to accept that a certain amount of the vibrations we were seeing were coming from "normal" robot performance.
2) Motion setpoints. To begin, spline motion is the best motion to use as the KRC4 itself will plan the path out ahead of time in order to achieve optimal motion (optimal being a combination of speed, smoothness, etc). If you are using the BAS function (BAS (#INITMOV,0)) to initiate/set your motion parameters then the parameters in the Config.dat file are the setpoints your motion will use. Changing the CP-Movement setpoints such as GEAR_JERK, VEL_CP, ACC_CP, and ACC_ORI can have a significant impact on the motion. Try reducing some of these parameters (especially gear jerk and acc_cp) and see if that helps.
3) Path/motion data. I'm not sure how you are obtaining your path data or NC points (from Catia or other software) but sometimes, smoothing the path data or adding more points could help your motion. This mostly comes into play when there is orientation changes of the endeffector from one point to the other. This does not seem to be the case for you (judging from the video); however, it could be possible that there are slight, unwanted motion between points (such as a drop of 0.1-2mm) between NC points that are close together. This would cause the robot to drop rapidly and could cause some of the vibrations you are seeing.

Again, I wish I had a more clear cut answer but experimenting a little with each option could help you achieve what you need.

Best regards!

Can you post the .src file?
As i agree that the load can cause that, especially with these short moves, the reason can be also the $APO.CDIS value and the acceleration/deceleration values

we have observed the same issue, on tightly curved paths at velocities of .05 - .1 TBH what you are seeing is not "that bad". I have seen it shake much worse... usually if you loosen the approximation distance values it will smooth out. But I have also noticed this seems to happen in certain "areas" of the work envelope. Of course if you are reaching a long way that makes it worse, but not always... and you aren't reaching that far. Also FWIW, I use the LDD software, and have a tool about 50% of max payload. Doesn't seem to affect things like this, though it certainly affects the robots ability to "maintain" accuracy as it reaches farther out. I tend to lean towards the "its not a CNC machine" explanation. You are weaving the TCP pretty quick. Also, how smooth is the path? That has a huge effect. If possible, use splines, they can definitely be smoother at higher speeds. Unfortunately, I can rarely use them, as you can't really go over 800-1000 positions in a block. If I had to bet on the way to fix it, check your $APO.CDIS....

Quote

So, as I can understand, incorrect mass and other parameters are not the reason of the problem.

If you also checked additional loads.

Fubini

And you programmed additional load to zero, because Default would be maximal addition load for A3 and only Zero for A1 and A2?

Fubini