My printer is Cartesian. This is a test rig to work the kinks out of my configuration before I deploy these sensors to my small (but growing) print farm. All of the farm printers are also Cartesian.

Changing the leveling method in Marlin was the ticket. I enabled linear leveling instead of bilinear and I got a perfect first layer! I went back and enabled 9 and 16 point bilinear leveling and tried it once more after having some success with linear bed leveling. Unfortunately, as soon as I ask my printer to calculate a complex mesh, things get wonky. So, I think we've solved the mystery. It was never the sensor's fault. The board wasn't performing the calculations correctly in the more advanced bed leveling options. Taking away most of the math and using only four points has enabled it to pull through and create a proper plane.

Thanks very much for your assistance. I will continue to buy and recommend Duet electronics. I can't wait to get my Duet Wi-Fi for my larger printer!

I just bought a mess of these Escher3D IR sensors after having fabulous results with them on my FT-5 build. I've installed one of these on an extra printer to workout the firmware before installing the whole lot of them… and it hasn't worked at all. However, the repeatability of the sensor checks out when tested, the trigger height is correct, endstoppullups are enabled per Escher3D's recommendation, and so on. Everything looks right on paper, but my end result is always wonky. Pictures below. I've tried a sheet of black glass, a piece of white paper, black buildtak, and masking tape. All the same result, despite these materials being officially supported by the sensor. After all this testing, I'm thinking this has to be a firmware issue. Right?

I'm open to suggestions, as I've tried everything I can think of. Post a reply and I'll get back to you as soon as I can try your suggestion.

Thanks in advance for your help.

Cheers!

Photos
https://ibb.co/i27Dfm
https://ibb.co/bBrH6R
UPDATE:

I've tried swapping the IR sensor with another one from my order in case I got a dud. It turns out, the results with the new IR sensor are exactly the opposite of the original sensor. This time, the back left is too high and the front right is too low.

https://ibb.co/jsdFBR