In ways similar to your own experiences, Geoff, I am familiar with PID cont=
rollers.  I even published a paper in one of the first journal issues conce=
rned with micro-electro-mechanical systems (MEMS) about a feedback system i=
nvolving integration and differentiation that was not 'proportional'.   I  =
greatly admire the many places where PID controllers outperform anything el=
se.  But at the same time I am confident of one thing, learned like yoursel=
f through experiences in the 'school of hard knocks', by methods never taug=
ht in an institution like Mercer University-there is no single method or si=
ngle system that will ever be perfect for every application.  The reason IR=
IS held a "Broadband Conference" several years ago, concerned with improvem=
ent to seismic instrumentation-is because the best and most expensive of PI=
D-like instruments-still can't do it all.   I predict that this will always=
 remain the case; that there will never be a single-package-do-everything s=
eismograph, no matter the cost.
   You suggest that capacitive sensors are 'flawed', as compared to some ot=
her possibilities, for reason of their need for AC excitation.  I don't thi=
nk the frequencies normally used with them is going to contaminate the worl=
d very much, compared to the host of truly-radiative devices like cell-phon=
es.  Only time will tell if something else could come along and do a better=
 job.  Practicing seismologists have settled on the capacitive sensor, beli=
eving it to be the best type presently available, based on compromise betwe=
en cost and performance, relative to their need.  Much in the way of experi=
mental comparisons of a variety of different sensors were considered, befor=
e they settled on their standard.   So I don't believe Erhard Wielandt, who=
 is one of the world's most knowledgeable professional seismologists, would=
 agree with the position you seems to have taken.
    Randall













In ways similar to your own experiences, Geoff, I am
familiar with PID controllers.  I even published a paper in one of the
first journal issues concerned with micro-electro-mechanical systems (MEMS)
about a feedback system involving integration and differentiation that was =
not
‘proportional’.   I  greatly admire the many pla=
ces
where PID controllers outperform anything else.  But at the same time =
I am
confident of one thing, learned like yourself through experiences in the
‘school of hard knocks’, by methods never taught in an institut=
ion
like Mercer University—there is no single method or single system tha=
t
will ever be perfect for every application.  The reason IRIS held a
“Broadband Conference” several years ago, concerned with
improvement to seismic instrumentation—is because the best and most
expensive of PID-like instruments—still can’t do it all. &=
nbsp;
I predict that this will always remain the case; that there will never be a
single-package-do-everything seismograph, no matter the cost.

   You suggest that capacitive sensors are
‘flawed’, as compared to some other possibilities, for reason o=
f
their need for AC excitation.  I don’t think the frequencies
normally used with them is going to contaminate the world very much, compar=
ed
to the host of truly-radiative devices like cell-phones.  Only time wi=
ll
tell if something else could come along and do a better job.  Practici=
ng
seismologists have settled on the capacitive sensor, believing it to be the
best type presently available, based on compromise between cost and
performance, relative to their need.  Much in the way of experimental
comparisons of a variety of different sensors were considered, before they
settled on their standard.   So I don’t believe Erhard
Wielandt, who is one of the world’s most knowledgeable professional
seismologists, would agree with the position you seems to have taken. =
 

    Randall