Print Head slamming into back right corner and unleveling

[jbothwell95 0]

My ultimaker 3 will start printing fine, but then when it goes to swap filament usage from 1 to 2, it first goes to slam itself into the back right corner. It attempts to push into the frame, making a loud clicking and banging sound. It then does actually swap the filament, but in the process the bed gets pushed down, unleveling it in the process. It will then start extruder from print core 2 but at about 6 mm (1/4inch) off the build plate, and then the print is ruined at that point.

I have tried releveling, changing slicing files, changing filaments, shrinking the file, and checking the belt tension, but nothing seems to fix it. 

It does work if I just use 1 printcore.

 

Does anybody have any experience with this issue?

[gr5 2,265]

This is an S3?

 

Can you show a video?  

 

It sounds like the collision isn't an end-of-travel in X or Y axis, right?  The collision is only nozzle to glass?  I guess I need to see it.

 

I think I would try to do a factory reset - I think that might be in the menus somewhere.

 

There are a TON of constants/values/parameters stored on the printer.  For example, position of the print core switching device.  Where to nozzle purge.  How high to move up after a print.  How much to extract after a print.  Temperature values.  And so on.

 

One of those values is probably where to go in Z axis for the print core switch step (or how much to move the bed down).  That value may be corrupt and may be negative slamming the nozzle into the bed.  This is one possible theory (but it would be helpful to see what you mean in a 5 second video).

 

One fix is to try "factory reset" if the S3 has that option (it might not - I forget).  The other fix would be to try reinstalling the firmware.  Again, assuming this is the issue.

 

Also be aware that the "hard drive" (it's bascially an SSD) on the S3 gets corrupted somewhat easily and I know people who had to re-install the firmware because of this.  Blocks of data just... fail occasionally.  I'm talking like, on average, one block of data per month.  90% of the time it doesn't matter.  But sometimes it corrupts something important - the block is automatically detected and walled off and never used again but it will read the wrong values until you re-install the firmware.

 

 

[gr5 2,265]

Oh - and it's not the active leveling.  Active leveling expects the right core to go down 1.5mm more than the left with a tolerance of 0.7mm.  Beyond that and Active Leveling fails.  It sounds like you are seeing errors of more than 0.7mm, lol.

[Enigma_M4 123]

Hi,

for me it sounds like the lift switch is out of calibration.

Did you try to (re)calibrate the lift switch?

Regards