Magnetic balls detach frequently while printing


Hello, about a week ago my vertex delta arrived.
After building the printer following the manual closely, i was ready to make my first print. At the time I didn’t have the most ideal oil, but I figured it would work to at least test the basic functions.

With the g-code of the keychain stored on the sd-card, I completed my frist print almost perfectly. With exactly the same settings and a clean printbed I tried to make another. On the second print the head detached for the frist time.

To prevent this issue form happening again, I pantiently waited until I had the right oil, read the forums on almost every topic that could be related, and found some good advice on oiling the rods for every print, making sure you can push down the head without the magnets detaching and tightening the belts.

Since then i attempted to make the keychain about 20 more times, for the sake of letting the printer get a bit settled in.

The problem that I believe im still having is that the printhead is traveling over an already printed area and getting stuck there with the nozzle, causing a magnetic bearing to detach. Apearantly using cura software could minimalize this occurance, but I’d rather have a more fool proof solution if one is available.

Any advice is greatly appreciated


Couple of things to check.
Make sure the cable is routed like the one in the picture

Check the calibration sensitivity

Go through the calibration settings again.
Don’t forget to save the settings.


thanks for your advice Wrong_way, I made sure to have a good spiral as you (and the manual) documented. The sensitivity calibration has also been executed, and to make sure i got good calibrations I usualy do both the height and auto callibration (and save those) before the print.

I’ll redo the sensitvity calibration, and see if the glass plate is not to tight to give false measurements. Tomorrow I can give an update on any results. I dont have time to print tonight unfortunatly.


So the sesitivity callibration has been done correctly now I believe, altough It’s not clear how much presure you are allowed to tap with to get a good callibration. I gave it a little tap with my figernail in a few different locations and callibrated accordingly.

The problem however still occures. The printhead detached quite a few time now causing some damage to the makertask surface, makes me believe that perhaps a new surface might yield better results


One thing to try: flip the glass plate over and print on the bare glass side. You might need to apply a thin coat of glue-stick or some other bonding material for your first layer. The glass surface provides a nice clean “tap” for the calibration.

I do not use the BuildTak side any more. I do all of my printing directly on the glass.


that might be a good thing to try. I did order a new buildtask, wich I expect to arrive soon. I’ll try printing on the glass before I apply the new one, so I got a baseline to compair with.


altough printing on the bear glass without the sticky something you mentioned didnt go well (No idea what to use for that), it did give some better insight on what the problem might be. when I took the glass plate off with the partly failed first layer, one side had a thinner layer then the opposite side. perhaps a sensor malfunction causing a faulty bed leveling?

Im gonna do a few more tests tomorrow to try and reproduce this, and see if different sensitivity levels give a better result


After you flip the glass over you might need to do the calibration again.
Also make sure the printer is on a solid surface if it’s bouncing around this can cause problems.


Yes, definitely do a full height/auto-calibrate/save settings.

For adhesive, it depends what material you are printing. With PLA, smearing a glue stick over the glass works very well. You can also use a light coating of inexpensive hair spray. I have used Aqua Net in the purple can - “unscented extra super hold” with good results.