Calibrating Towers and Z offset
swclark last edited by unlimitedbacon
I (think) am wanting to re-calibrate my end-stops on my Kossel Pro towers. I am noticing a difference in the height/offset of the first layer. One side of the printer prints a lot tidier on one side and it looks stringy (maybe as it's higher) on the other side. It seems to improve after the first layer is complete.
How can i calibrate the towers to fix this? I think it maybe the z tower that needs adjusting.
unlimitedbacon last edited by
There is no way to manually calibrate the endstops on the Kossel Pro. In theory, it should all be handled by the automatic calibration. However in practice this is not always the case. There are some lengthy posts over on the Kossel Pro forums about this issue. I will do my best to give an overview of the solutions here.
- Disable the probe and switch to MatterControl's software leveling. This will have you manually measure the height of the bed at various points using paper as a feeler gauge.
- Use shims to ensure the correct distance between the arm joints. The basic problem is that, due to the geometry of the delta system, if the arms are not perfectly parallel with each other it will cause the end effector to to tilt slightly as it moves from one place to another. This tilting means that the probe can move slightly up or down relative to the nozzle, which throws off the measurements it is taking. To take care of this problem, ZT Automations is now selling a set of shims you can use to ensure that the arm joints are evenly spaced.
- Upgrade to a better probing system, like Force Sensing Resistors (FSRs). These are much more accurate since they detect the nozzle itself pressing against the bed, instead of a seperate probe. Ultimachine sells a $40 FSR kit with all the electronics you need. You will also need to print some new bed mounts and remove the probe offset in the firmware configuration.
dthurbon last edited by dthurbon
Reviving an old thread...
The Kossel Pro calibration was always problematic. The designers somehow thought that they could build a system that would automatically calibrate using the probe. However, because of slight variations in the printer geometry, the effector would tilt during movement and the probe would move up and down relative to the nozzle and the calibration doesn't work properly. The simple solution to this would have been to get rid of the probe and put endstop adjustment screws on the carriages. I'm going to try and do this. The end effector was also not suited to anything but the hotend that came with the printer, which is no longer available. For this you can print or have made a E3D V6 effector V8 from Thingiverse. Kossel Pro E3D end effector v8 . Also, the Brainwave board was an attempt to move away fro the Arduino/RAMPS combination, but doesn't have the processing power to deal with the delta calculations at high speed. So swapping out the board for something else, perhaps an Azteeg board for example (be careful of the endstop wiring) could make this a workable printer.
For the most part the Kossel Pro was obsolete as soon as it appeared, there were some minor design flaws, and there has been no follow up on correcting these flaws, or for proving spare parts. Basically anyone who bought the Kossel Pro was left high and dry.
We do not have a problem with anyone reviving an old thread here, we welcome everyone and want to know their opinions. I am sorry that the machine was essentially DoA but hopefully some manufacturers were able to take note and possibly improve on the short comings of printers like this one.
Let us know if you have any more thoughts on the matter!