Joint-2 of UX120FC21 robot oscillates after collision

[najafabadi]

Hi everybody,

Unfortunately after an operator's mistake and collision of an UX120FC21 with the fixture now the joint 2 of robots has some oscillation in each stop point. The robot application is spot welding. We decide to repair the robot but we don't know the origin of the oscillating exactly, is it from the cyclo-reduction gearbox of joint 2? or it's origin is the spring balancer?  or other thing?
I hope someone help me to find the exact location of the problem.
Many thanks in advance for your help.

[roboperson]

JT2 " oscillation" is actually an open wide statement when it comes to mechanical.Especially in a crash. Does it only do it in repeat, or is it prevelant in teach?What makes you believe it is coming from JT2?
Could it be that when it crashed, the end of arm tooling was damaged?
1. Go into AS location and make sure that the encoders are tracking. 2. stretch the arm out as far as possible and move JT2 positive and negative. Is there a jerk or bounce? If the cyclo or input gear has been damaged, it will jump when it passes the broken gear.3. Shake the arm, end of arm tool, try and move the weld gun side to side, (looking for damage in the wrist). Has the boom of the arm itself been damaged? If the counter balance is damaged, it would "clunck" all it is, is a big spring inside. It honestly isnt too common for a JT 2 to be taken out from a crash. the highest percentage is the end of arm tooling, or in your case, the weld gun. Is the weld gun misplacing when it it welds, or is it welding in correct position?

[najafabadi]

Dear Roboperson,

Thank you for your complete answer. Please see attached file. It is a very short movie of JT-2 oscillation in 3gp format.
The oscillation happens when the robot stops in each spot points. Actually when the robot holds and we force to the end of arm by hand then we can see the oscillation on joint 2 easily.
Also in check mode we can't see the oscillation untill robot stops!
The counter balance is not noisy.
During the crash robot tools (welding gun) broke and we changed it before the teaching. Also the welding spot point locations almost doesn't change during the welding!
What is your opinion? Is there any other thing that should check?
Do you have any recommendations if we should change the cyclo?
Many thanks in advance for your help.

Najafabadi

[roboperson]

Unfornunately, I can not view 3gp format, and am having problems locating a converter. I apologize I can not see your video. If you are convinced it is JT2, then there are 2 possible motions. If it is an movement up and down (or dropping down) does it "slip" down like it is missing a tooth on a gear? Have you set the robot on the Mechanical hard stop, pulled the motor to see if there is metal pieces in the grease from the cyclo being damaged, or the input shaft on the motor being broke? When check stepping, does the encoder counts show any movements while the JT2 is ocillating? Is it a side to side motion? possibly from the cross roller bearing? Did the crash put a strong side load to the JT2? If you have damaged the cross roller bearing, you pretty much need to remove the JT cyclo to replace it. JT2 is a pretty straight design though, Cyclo, cross roller bearing, shims, bolt. With out being able to see the video of movement, or the nature of the crash, it is hard to say. Sorry I can not offer much more help than that.

[najafabadi]

I change the video format to .mpg and now you can see it by using windows media player. It shows the oscillation of Jt2 when the robot stops for doing a spot weld.
I forgot to say that after crash we also changed the joint 2 servomotor and encoder and check its shaft. The motor shaft and the input shaft to cyclo aren't damaged.
I will check the encoder counting during oscillation and write the results in my next message.
By the way what is your opinion after seeing the movie?

[roboperson]

After seening the video, it does appear to be the cyclo if you are making that bouncing motor with breaks on (no servo power). As you know, the breaks hold the JT2 into postion when the breaks are not released. When the breaks are released, then the servo motor takes over. Since you stated that the Servo motor was changed, and assuming it was changed with a good motor, then the next logical component damaged would be the cyclo. From the video, it appears to be bouncing up and down in the positive and negative directions. Since your motor breaks should be holding this, then it is reasonable to assume the gearing has been damaged. Of course, that is just my opinion based on the video and what we have discussed. Best of luck to you.

[najafabadi]

The movie is relate to the continuesly running of the robots, when it stops to do a spot weld location.  But as I said before when in hold condition we force the weld gun (EOA tool) by hand, jt2 oscillates in a way that you can see it easily.
Now what is your opinion?
Many thanks in advance for your help.

[najafabadi]

Hi,

We changed the JT-2 cyclo reduction gearbox and the problem solved. 
As you can see in the attached photo (I show it by using a red arrow) the cyclo-reduction ball bearing was broken after collission in 2 parts (one part hasn't shown in the photo but you can see the balls that you can't see in a normall ball bearing from side view).
Also we have seen some loosing in the parts of damaged cyclo.
Many thanks  from Roboperson that help us to find the exact location of defect.

Najafabadi