Kossel 2020 printer suddenly can't print correctly <Updated>

I finally got my Folger Tech Kossel 2020 delta printer tweaked in. bed flat, zero offset z probe, auto leveling working right, cal cube nearly perfect, lots of prints, including some that use about 75% of the bed working fine, part fan added. Bliss.
And then BANG! Overnight, everything went south.

It now does not reach the bed, first layer isn't properly "sqashed", subsequent layers look like they are too high off of the last layer, dogs and cats sleeping together, mass hysteria... (nod to Ghostbusters).
I checked my leveling, everything had moved "up" about .3mm. OK, I releveled, modified the offset from z probe in configuration.h, no good, still the same problem.

Has anyone else seen this? I am using the same firmware, the same slicer/printer (Simplify3D), the same PLA, the same bed and hot end settings.
What gives?
Help?


Update of details:
Until yesterday, my delta printer was "dialed in" and printing whatever I threw at it. My bed is flat, auto-leveling was working like a charm. I have a part fan that is off for the first layer and set to 50% at layer 4. My bed is at 60 C and my hot end is at 210 C for the first layer and backs off to 200 C for all layers after that. I work in PLA.
Then, total dysfunction.

Problem one:
I found that suddenly all my endstops, leveling data changed. My bed used to measure out (X,Y,Z,0) at 1.1, 1.1, 1.1, 1.0. I am using a zero-offset magnetically attached mechanical switch to do the auto bed-leveling. I then measured the "new" bed level stats at 1.4, 1.4, 1.4,1.3, and the Z offset of the probe obviously moved.
As I re-leveled everything to find the new settings, they changed and finally settled to 1.2, 1.2, 1.2, 1.1.
Huh? Why did these all change, and why did they settle back down (most of the way)?
I updated my configuration.h file and kicked some prints off. This almost worked. If I have flat objects with large contact area-to-height ratios, the print will work. Things that have multiple, smaller contact points will not print, by about the 20th layer the print head will snag on something and pull it off the bed.

This brings up problem 2:
Even flat models now are festooned with "cobwebs". Gunk gathers and creates "zits" in places and more scars than usual on surfaces. The plastic may warp on a layer and when the print head comes along may hit it and knock it off. Also, it may be dragged around by drools from the hot end. I now see dark sections that look like the plastic got burned while coming out.

OK. I am clueless about Problem 1, I checked my mechanics, no loose or broken rods, my bed springs didn't suddenly relax and come back, my carriages are solid, no loose screws, my limit switches didn't move (they'd all have to move at once since the error distribution was consistent at all towers). The belts are the same tension and nothing that I could find was changed on the end effector. I

Problem 2 looks like a classic case of the PLA getting too hot: strings, warping, burned spots, drool dragging. But why all of the sudden? My settings over the last two weeks, when I got my printer working, have not changed. Is it possible that the thermistor is failing and the temperature of the hot end is getting hotter than what is being reported?

I will keep looking and testing, and will report what I find for posterity, but if anyone out there has had this kind of thing happen, I'd sure love to read what you found!

DLC

Edited 2 time(s). Last edit at 05/03/2017 10:39AM by dlc60.

Mystery solved. This mess was caused, through an innocent series of events that I didn't connect until now.
A couple of days ago my spool jammed during an unattended print job and the filament extruder snapped the filament. To get it out of the Bowden tube I needed to pull the tube from the extruder box. It didn't want to come out and during that process I yanked a bit on the tube. Since the tube was zip-tied to the wire cluster attached to the hot end, I ended up putting tension on the wires. The next time that I used the printer this tension pulled on the heater block causing it to unscrew about 15 degrees from its normal place. This created a leak between the top of the heater block and the cooled filament tube which allowed melted plastic to be pushed out the top, sit there a while to burn and then drool down the hot end onto the print job. This also reduced the amount of PLA coming out of the nozzle. Since the heater block was unscrewed a bit, it was a little closer to the bed now, which threw off all of the slices being printed.

So. Let my experience show others with inexplicable problems that something seemingly unrelated to your eventual problem may well be the cause of your problem - You just have to keep looking!

whew!
DLC