I can confirm the "double homing" is a result of attempting to print bad gcode (in my case, I was pulling my USB stick out before gcode file was finished saving).
Effectively, the printer will be set to "printing" but nothing will happen. As soon as you press stop I assume it does the initial homing it should have done for the printing stage, and then another for the post-print stage.
So this does bring up the question, in the instance the file is bad, and the printer will "hang", is there no way to identify and tell the user this?
I also had a print "air print" yesterday. I'm not sure what it was but heard the printer make an unusual noise from down stairs so checked up on it and found it was printing in the air.
I stopped the print and just checked that the offset was properly recorded by being able to move the Z axis towards to bed as per normal. restarted print and all was OK. No turn off was done.
I believe this was a fresh power-up, but could have been after the bad gCode mentioned above ^.
On a side note Sunger, I haven't been to happy with a noise on my Y axis and it is hard to describe the noise it makes, I thought it might have been the leadscrew not being true and grinding only at a particular position giving a inconsistent grinding noise, especially as I could see a bit of "wobble" on the open-ended tip end when it was running .
I've checked over all my axis (dissasembled, cleaned and relubricated and adjusted the leadscrew) so it runs smooth, and I swapped my Y axis for one of my Z axis - when it moves down everything is fine, but when it moves up it does click a few times.
Do you have any insight on this? I'm wondering if the linear slide is bad or maybe bent from overrunning? as i recall my Y axis did make a few awful overruns grinds on earlier firmware from just "stopping" a print so don't know if this has the ability to mess anything up?