Should zero really be zero?

Hi, recently I have been having problems with "banjo strings" on the first layer when printing white .3mm PLA. I am running the bed at 65C and the hot end at 210C. In addition much of the first layer outline was not laying down on the bed very well.


After doing an auto bed leveling I moved the Z axis down to 0 and measured the distance from the glass to the hot end, it was .09mm. When I set up the ABL I did so with a distance of .02 mm between the glass to the hot end.

So the question is, if I lower the hot end to zero should it be:


1 on the glass
2 .02mm above the glass
3 some other value


Below are the results of a G29 and M48 test, to me they look OK, would anyone with more experience than me comment on the values please.


Apart from the first layer, the rest of the printing seems to be pretty good.
---------------------------------------------------------------------------------
Send: G29 n 5 T
Recv: Eqn coefficients: a: -0.00 b: 0.00 d: 17.67
Recv:
Recv: Bed Height Topography:
Recv: +0.04015 +0.03390 +0.01740 --0.02935 --0.07585
Recv: +0.03590 +0.03990 +0.03090 --0.00910 --0.05760
Recv: +0.03715 +0.03765 +0.03215 --0.00760 --0.05460
Recv: +0.04715 +0.03240 +0.01890 --0.02260 --0.05985
Recv: +0.02765 +0.00065 --0.02160 --0.04960 --0.04410
Recv:
Recv: planeNormal x: 0.00 y: -0.00 z: 1.00
Recv:
Recv:
Recv: Bed Level Correction Matrix:
Recv: 1.000000 0.000000 -0.000582
Recv: 0.000000 1.000000 0.000073
Recv: 0.000582 -0.000073 1.000000
Recv: ok
---------------------------------------------------------------------------
Send: M48
Recv: M48 Z-Probe Repeatability test. Version 2.00
Recv: Full support at: http://3dprintboard.com/forum.php
Recv: ........


Recv: Mean: -4.997275
Recv: Standard Deviation: 0.002461
Recv:
Recv: ok
Recv: echo:endstops hit: Z:-5.00
Recv: Mean: -4.997275
Recv: Standard Deviation: 0.002461
Recv:
Recv: ok
Recv: echo:endstops hit: Z:-5.00