1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Fix for R2 Connection Interrupted Error

Discussion in 'Troubleshooting' started by RoboticsRob, Dec 17, 2019.

  1. RoboticsRob

    RoboticsRob Member

    Joined:
    Oct 11, 2018
    Messages:
    60
    Likes Received:
    17
    Hello all,

    I've been working on solving the connection interrupted error displayed part way through a print that multiple of our R2's have been exhibiting, and I'm pretty sure I've come to a conclusion on at least one cause of this, and a fix for it.

    Symptom: Robo R2 fails mid-print and reports print head disconnect, "Error Detected Connection Interrupted"
    Suspected Cause: Early R2's (~pre-mid 2018, I believe) shipped with a PCBA that had an interference fit between the +5v 2-pin connector and the heatsink on the FET directly above the connector labeled "To Uptown Adapter". Over time and repeated temperature cycling, this interference fit causes the bond to fail between the heatsink and the indicated FET (see attached annotated photo). Removing the heatsink causes higher than expected FET temperature, which soaks into the board, causing the traces that connect the thermistor in the extruder to overheat and drop connection or report out of bounds readings to the mainboard. Over time, this trace will fail and cause the mainbaord to be scrap.
    Solution: Replace the heatsink on the mainboard that interferes with a larger 9x9x12mm heatsink, offset slightly to the right to prevent further interference. Fix tested on multiple R2 boards that were exhibiting disconnects, issue resolved.

    If your R2 shipped with color coded connectors for the X,Y, and Z steppers and limit switches, you likely have an R2 with a revised PCBA that shipped with a heatsink location shifted to the right, that does not exhibit this issue. Note that Robo changed the PCBA without changing the silkscreen on the PCB, so both the original and the revised PCBA says 2.1.5 on the board, though the shift in heatsink position is visible so you can identify which ones have this revision.

    Let me know if anyone else who has been seeing this issue found a different solution, or if you try this fix out and it solves your issues as well.
     

    Attached Files:

    WheresWaldo and mark tomlinson like this.
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,523
    Likes Received:
    7,305
    Interesting and good work digging into it.
    Most people that have had that issue in the early R2 sorted it by replacing the Pogo pins (bed connection) to the upgraded one.
    Not sure how that may play into this :) But likely the pogo pin failing increased the current demand (by becoming a higher-resistance connection and then eventually failing open)
     
  3. RoboticsRob

    RoboticsRob Member

    Joined:
    Oct 11, 2018
    Messages:
    60
    Likes Received:
    17
    Good point Mark. For sure the resistance would be increased as the Pogo pin board started to fail, increasing the overall current and temperature throughout the board. This would just have been the weakest connection and thus where we've been seeing the failure on these printers, all of which do have V1 print beds with pogo pins. We've actually soldered a number of the V1 print beds to the cable assembly connected to the mainboard, eliminating both the pogo pin board and the connector used in the V2 print bed assembly.

    Since V2 print beds are out of stock, and some are still using V1 beds, this may be a low cost solution for folks who did not upgrade theirs to a V2 bed and are having intermittent disconnects on the extruder thermistor as a result.
     
    WheresWaldo and mark tomlinson like this.

Share This Page