Okay, so the problem with my delta is apparently the OrangePi Zero 3 somehow getting overloaded doing delta kinematics math that the 100mhz printer board could do fine if I put Marlin on it.

It is strangely only happening with specific directions of "diagonal" movement. I know as a delta, all directions (except straight line Z axis moves) should be pretty much the same in terms of kinematics/processor load, which is part of why this was confounding me for like a full week.

Specifically, it'll print a cube aligned with the Cartesian coordinate axes perfectly, but if I rotate it more than about 15 degrees it either loses its ability to hold the effector's Z position, or it starts drawing curvy lines while maintining Z.

It has similar problems if i try to print a cylinder, but much less pronounced.

With the current prices of Rasberry Pis, I'm really hoping I can get my hands on a used 4B since my single surviving full size RPi (i have a drawer full of zeros, but just the one RPi 5) which is currently doing triple duty as a print server, file server, and home automation host.

#3dPrinting #RaspberryPi #OrangePi #DeltaPrinter

@EUCWolf

Tried printing a Delta calibration object on it yet?

https://www.thingiverse.com/thing:745523

(Tbh, That SOUNDS like a arm/trolly/joint problem I had a while back ?)

Advanced Delta Printer Calibration (tower position + individual diagonal rod) by dolpin

This calibration object will help you to calibrate angle and size errors on your delta 3d printer These errors are typically caused by tower position errors and different diagonal rod lengths. This calibration object will show if your printer suffers from sizing and angle errors and the instructions will help to solve these errors. My Achatz easyDelta printer had a 1 degree error in angles and over 1mm error in sizing in certain directions. This is caused by the towers not being positioned at a perfect 120 degree angle around the center, and small individual size errors between the rods. First errors must be fixed on the printer itself, but on some printers this might not be possible. I used a jig to glue the joints to the diagonal rods, but small errors (1/10 mm) result in bigger errors in the print. The basic delta printer calibration steps are: calibrate the print bed on each tower position by adjusting the z-height for each carriage correct for bowl/dome shape by adjusting the DELTA_RADIUS parameter tune the print size by adjusting the DIAGONAL_ROD parameter After the basic calibration you might experience that the actual print is not equal sized in all directions and the angles in all directions are not as they should be. Resulting in for instance trapezoid objects instead of square-angle objects. The next calibration step is to detect these errors and calibrate your printer. These errors are typically caused by tower misplacement and differences in diagonal rod length. After calibrating the tower angles and individual diagonal rod lengths using the marlin firmware, I managed to achieve 0 degree angle error and a maximum of 0.1 mm size error in any direction in the print ! The calibration object is printed after the basic calibration. Print out the calibration sheet and use this sheet to measure the angles between the X-Y-Z towers. Use a slide caliper to measure the print size in the X-Y-Z tower direction and calibrate your printer in the Marlin firmware. Check the instructions for the calibration steps and how to correct this.

Thingiverse - Digital Designs for Physical Objects