I have a "three story" Lack table enclosure for an Anycubic Kossel Linear+ running a duet. The board (and power supply) is mounted below the lowest table, outside the enclosure. The duet is mounted vertically, to encourage convective circulation, with no fans.
Has worked great for months (maybe over a year, I just don't remember).
@tgarr No worries about the thread digging, altough quite old, it is still as relevant as it was. In fact I just started a print with the exact same effector design as posted here. The 4th axis attachements are still evolving and I do have some minor changes planned to the effector design but noting major.
I don't use any special washers or something like that. Just the 6 screws (with a regular washer on top) that thread into brass heat set inserts.
In terms of calibration results here's the full DWC output string 🙂 . That is for a 180mm probe radius on a 400mm diameter bed.
Calibrated 8 factors using 18 points, (mean, deviation) before (0.018, 0.046) after (-0.000, 0.043)
The results are certainly more consistent across the entire bed compared to the FSR's I used to use. I am honestly not sure if the 0.043mm deviation is considered good for a 400mm diameter bed.