RobotStudio event

Communication with the SMB

Hello,

I have an ABB IRB 6700 running RW 5.15 and IRC5, the cell has been running about 10 hours a day, more or less every day since 2014. In the past year or so, I've been noticing the robot losing its connection with the SMB quite often (38103: Lost communication with the SMB) which sends the controller into a System Failure state. At first, the fix was a simple battery replacement, but I have been consistently getting this error every 3 days or so. Power cycling the controller always fixes the issue, but what I want to figure out is why is this happening? I have checked the cable for kinks or damage but it seems perfectly fine. My only thought is that the SMB has likely reached the end of its life. I've also noticed this particular issue most commonly appears on Mondays, usually when the machine has sat idle for 48 hours. Has anyone else experienced this sort of issue? What was the fix?

Thanks

Comments

  • With the power off the system check the plugs at each end of the SMB cable.
    Oxidation can build up over time giving a higher resistance - some times just the act of unplugging and plugging back in can make a better connection.
  • Thanks, I will try this 
  • After replacing the SMB, the cable running to the SMB, and the SMB battery, the issue still persists. Anyone have an idea to what is causing this? 
  • Try running a cable to the SMB cable as per below photo's.
    Wrap the wire round the plug and secure with a hose clamp.
    wire runs from plug to plug and is earthed at each end also.


  • Correct me if I am wrong, but wouldn't this create a grounding loop because there are two points that are grounded, causing the cable to act as a sort of antenna? I was thinking we should avoid this as the SMB connector is signal cable, and interference may worsen the situation. 
  • graemepaulin
    edited May 2021
    I think that is an issue with the internal cable shield.

    This is an action from ABB that may help where there are issues of this nature, it is more like an additional ground/earth bonding between the robot and controller. I am not sure the exact reason it is supposed to help.

  • hi, i have same issue - did earth wire fix the issue ?  other issue maybe - smb board, R1-smb internal cable, robot to controller cable, internal controller cable, axis computer ?
  • is there a recommended fix ?