Joint motor mass

AmirH

Hi there,
I am programming the UR5 dynamics. i was able to find moment inertia and mass for links but could not find the joint motor masses. Is there ant specs sheet for it? Thank you in advance
Amir

matthewd92

I would say based on just looking at the UR documentation that there is something missing, the UR product page for example lists the weight of a UR10e as 33.5 kg but if you add up all of the weight on the kinematics page that UR provides the total mass of the links only adds up to a little over 29kg. so we are missing 3.5 kg of mass on the arm. Having replaced the joints on these robots I will say the bulk of the link mass is the joint, the actual tubes connecting the joints are very light weight.  I am not sure where the missing mass is between the product spec page and the kinematics page but I would assume that the masses listed on the kinematics page are the total mass of the link element including the motors.  I would start there and see what your control algorithm looks like.

jelms

The only specifications I'm aware of are here. If you need additional ones I'd reach out to UR support.

AmirH

Thanks. I have used the specs you addressed, which provide link masses. However, there are motors inside UR5 joint caps that carry extra weight; and I planning to consider that in my code. Please advise. Amir

jelms

I believe the UR specs in the link include the mass of all parts (including motors) of a unmodified arm.

AmirH

So I do not need to add any weights pertaining to the motors? Is there any resource you could refer me to please? The reason why I am unsure about this is, link masses are supposed to be taken as concentrated mass at the CG of the each link which I did so in my code. It does not appear to have motor weights included.

jelms

I think that makes sense. The 3.5kg are probably the portion of the base that doesn't move.

matthewd92

That makes sense, seems a little heavy but I bet it’s close. 

AmirH

matthewd92jelms , thank you. I have got two more questions:
1. I am using FT300 and TCP/IP Communication data stream socket 30003 to collect TCP force data. Data pertaining to the former ranges btw -20+20 and for the latter it ranges btw -600+600. I should say I just lay my right hand on the fixture kept by the gripper at the endeffector. That being said, outcome pertaining to FT300 makes much more sense given the Newton as the force unit. What about the data collected from socket 30003 ?

2. Again using socket 30003 , I have collected the joint current. In order to convert that to actual torque I need the Kt associated with the UR5. How can I find the Kt?

Thanks

jelms

AmirH, I don't have any direct experience with that. You may be able to compare using a script. The values returned from get_tcp_force (see this link) for force. For per joint torque you'll need to use get_joint_torques()