calibcant/procedure.tex: Discuss 14kHz spike in fig:calibcant:vibration

diff --git a/src/calibcant/procedure.tex b/src/calibcant/procedure.tex
index a1da8c1343e9bda1cadde60f2de6f3a090e985ac..65a45310142ba7bda878a6135156d53821796ff7 100644 (file)
--- a/src/calibcant/procedure.tex
+++ b/src/calibcant/procedure.tex
@@ -163,7 +163,11 @@ vibrations are configurable (with \hFconfig,
       (\cref{eq:peak-frequency}).  Only data in the blue region was
       used when computing the best fit.  This is the first vibration
       from the 2013-02-07T08-20-46 calibration, yielding a fitted
-      variance $\avg{V_p(t)^2}=96.90\pm0.99\U{mV$^2$}$.
+      variance $\avg{V_p(t)^2}=96.90\pm0.99\U{mV$^2$}$.  The narrow
+      spike around $14.3\U{kHz}$ is not due to the cantilever's
+      thermal vibration, and rejecting noise like this is the reason
+      we use a frequency-space fit to calculate the thermal deflection
+      variance.
       \label{fig:calibcant:vibration}}
   \end{center}
 \end{figure}