I set up the mini yesterday and printed the test and one other item no problem. Started it up today and setup in the same way was before. This time when it starts printing it is scoring down into the bed and leaving marks. It also prints the first layer badly and unevenly as it printed before it paused.
It’s clear that the head is starting too low but the settings are the same and I tried loading up the sample piece and clearing the board from there before opening the new file in case there was a setting I had missed.
It’s been cleaned, rethreaded and turned off and on again several time but that’s where the limits of my knowledge end.
Usually I get that when the auto-leveling didn’t quite work correctly due to junk on the tip of the hotend.
I can tell when it’s going to happen because the auto-leveling process actually deflects the print bed a noticeable amount when the hotend touches the washers in the four corners. I’ll cancel the print and clean the hotend and usually that does the trick.
I had noticed it picking up a little piece of fluff from the cleaning pad but as it was a new pad and the head was already hot I wasn’t sure how to deal with it. I shall change the pad and see if that helps.
Thanks
Yeah, when my mini was new, fuzz from the cleaning pad kept getting caught on the hotend after the automatic cleaning. I stood by so I could wipe it off with a clean rag or paper towel before it started to auto-level. After some use, that stopped happening.
Wiper changed and no fluff picked up, I was ready with a cloth!
Works great again now, thank you very much for the quick answer.
No problem, glad it’s working now!
I frequently had this problem as well; the wiper would get all the plastic cleaned off the head, but sometimes leave a bit of felt “fluff” that would prevent proper leveling. I got in the habit of manually wiping the head with a green, non-scratch scotchbrite pad when the head was on its way to probe the first washer. That always gave me a good probing sequence.
So I decided to automate that – by designing a new “dual wiper” and updating the script to drag the head across a non-fiber pad after the normal cleaning on the felt pad was finished. Details can be found here:
http://www.thingiverse.com/thing:1976831
The new 1.1.0.10 firmware is also helpful. It is better at recognizing a failed leveling sequence, whereupon it re-wipes the head and re-starts the probing again. (I also modified the 1.1.0.10 firmware to use the dual-wiper pad, but it never happens because the initial use of both pads has given me 100% reliable probing on the first try!)
Thanks ScottW, that dual wiper is awesome! I always try to wipe the head once more after the programmed wiping too. This will make it it way easier! 
Addendum… As mentioned earlier, I did update the 1.1.0.10 firmware to include dual-pad support on the “rewipe” that occurs when probing fails. This isn’t really needed, because I have never seen the probe fail since implementing the dual-wipe in the startup script.
But for the sake of completeness… If anyone wants to add the dual-pad support for the firmware’s G29 “rewipe” as well as having it in the startup script, here are the changes I made before recompile.
Note: CHANGES ARE SHOWN IN RED - OTHER LINES FOR CONTEXT ONLY).
configuration.h
marlinmain.cpp
