re my 14425 aLOG Monday, this morning I indeed found the STS2 B misoriented; this is the unit in the beer garden.  Just as I suspected given the individual HEPI sensors' coherence with the X & Y signals, the X axis of the STS2 was ~-45degrees from our desired X axis.  I say ~ because we don't have a mess of X Y grid lines all over the floor.

I also found the level of the STS2 to be quite poor--the bubble was railed.  This might have been leveled upon installation, not sure when that was, but maybe the heavy instruments (55lbs) slowly differentially sinks into the vinyl floor.  Maybe the floor should be accessed to put the feet directly onto the concrete.  At least, I should look at it again in a week or something to check for tilting.

So I rotated the STS2 to align its X alignment point to be parallel with the X arm and then leveled the bubble.

I 'autoz'd the machine now several times but some axes still have relatively large values.  It may be suffering now from some thermal drift as the igloo must be removed to do the level and alignment.  We may need to rezero daily for a few or even several days.

Attached is a repeat of the measurement I posted log 14425 and as expected, now the X & Y conputed signals are strangly coherent with the STS2 X & Y outputs.  And, I think it makes sense, the 4 local HEPI sensors all have about the same marginal coherence to the ground sensor signals.

Now back where I started Monday, looking at the coherence between the ground STS2 and the HEPI L4Cs.  It is very strong between 0.15 and 4hz.  It falls pretty rapidly starting about 0.15hz and above 4hz.