3D printer - installing a BLTouch clone - part 1


Installation and firmware

I’ve been reflashing my Ender 3’s firmware recently. I’m using TH3D’s unified firmware package. I use OctoPrint and I’ve been editing the firmware recently to improve the serial performance on this poor little 8bit microcontroller.

I tried to install an old BLTouch clone. It had a metal sensing pin that it failed to lift - seems like the transformer wasn’t wound well enough. Instead of trying to fix it, I purchased another cheap clone with an Ender 3 length cable, apparently. It’s not quite long enough, but instead of fixing it, I’m rolling with it.

At first, I printed this without the 3mm drop, but it didn’t work, I immediately drove the head into the bed. I also managed to drive the head into the bed printing the 3mm drop version. Installing it seemed to work well. I used Octoprint to issue the XY home and Z home commands and the head didn’t crash. I tried a few things before removing the Z limit switch, to avoid it interfering with the Z axis or getting damaged by a lower than expected bed.

Next came bed levelling. For a little while, I had some strange calibration problems, before I tightened my X carriage wheels on the right hand side, forcing the bar to stay upright. I then could repeatedly issue G28 and G29 D1 to relevel XYZ and start Automatic Bed Levelling in Dry mode. Now I’m adjusting my bed screws to get it as level as possible. I’m doing so to reduce the bed wear and the least physical warping.

I pasted the output of Bilinear Levelling Grid into a spreadsheet, but I quickly realised that I would be levelling around a constantly moving Z index - the middle of the bed would rise and lower with each adjustment, making for a constantly shifting middle point. This might mean that it’s sort of pointless to over adjust, and reaching a happy medium where you’re not off by more than 34 of your nozzle width, you’ll be okay. (0.3mm should be fine for 0.4mm)

This process took me a few hours, where the installation of the BLTouch was around 1 hour and the remaining time was spent trying to level the bed further. I am getting excited to see if my first print is significantly superior after applying the final bilinear bed levelling grid with G29 (no D1 this time). A few more test probes and I’ll be on my way.