No 'butt-in' offense to me by your email, GM. (I hope it is acceptable to=
address you in this manner.) We in physics also 'speak our peace' on subj=
ects important to us.
You are absolutely correct that periodic signals with a frequency hig=
her than the Nyquist frequency (1/2 the sampling rate of an ADC) will show =
up as an alias. The key element to whether or not you get such a high freq=
uency signal masquerading as one of lower frequency-depends on the coherenc=
e time of the offending signal. It must exist through a large enough numbe=
r of cycles that what is admitted to the record saved is coherent enough to=
show up as a spectral line.
In the case of the VolksMeter, there is no analog signal to begin with.=
Stated differently, there is no ADC; rather the device is a 'direct' capa=
citance to digital converter (CDC). Because of the award winning AD7745 ch=
ip of Analog Devices, that is the heart of the instrument, the VM is more t=
han just a "State of the art digital seismometer", as described in the link=
on Larry's page.
http://seismicnet.com/volksmeter/State-of-the-art_Digital_Seismograph.pdf
It is also the worlds' first fully digital seismograph'. It is true that =
some detractors from this claim would say that there is no such thing as a =
fully-digital anything that interacts with our world which is itself analog=
.. But my point is the following. Unlike other commericial instruments tha=
t use analog electronics to monitor a capacitive sensor (and then finally d=
igitize the analog signal), the VolksMeter has no analog components whatsoe=
ver, except for internal 'workings' of the chip-for which one would have gr=
eat debate as to whether the 'pieces' are digital or analog.
There is one other significant 'fully'-part to the instrument. The capacit=
ive sensor used by the VolksMeter is "fully differential".
There are a host of advantages associated with an architecture that =
is fully digital in the electronics with a fully symmetric (differential) s=
ensor; I don't have time now to go into detail.
Incidently, some of you amateurs might want to 'tackle' the home-brew c=
onstruction of such an instrument yourself; maybe even extend what I've don=
e to the world of vertical seismographs. As Chris Chapman has indicated, t=
hey're much tougher because of material problems (not my expertise). If y=
ou want to build something that could make even the seismology pro's saliva=
te over, then talk with Larry about the electronics package. He will proba=
bly sell you the all the necessary things 'for a song', compared to what th=
e big companies want for anything.
Randall
No ‘butt-in’ offense to me by your email,
GM. (I hope it is acceptable to address you in this manner.)&nb=
sp; We
in physics also ‘speak our peace’ on subjects important to us.<=
o:p>
You are absolutely corr=
ect
that periodic signals with a frequency higher than the Nyquist frequency (1=
/2
the sampling rate of an ADC) will show up as an alias. The key elemen=
t to
whether or not you get such a high frequency signal masquerading as one of
lower frequency—depends on the coherence time of the offending
signal. It must exist through a large enough number of cycles that wh=
at
is admitted to the record saved is coherent enough to show up as a spectral
line.
In the case of the VolksMeter, ther=
e is
no analog signal to begin with. Stated differently, there is no ADC;
rather the device is a ‘direct’ capacitance to digital converte=
r
(CDC). Because of the award winning AD7745 chip of Analog Devices, th=
at
is the heart of the instrument, the VM is more than just a “State of =
the
art digital seismometer”, as described in the link on Larry’s p=
age.
http://seismicnet.com/volksmeter/State-of-the-art_Digital_Seismogra=
ph.pdf
It is also the worlds’ first fully digital
seismograph’. It is true that some detractors from this claim w=
ould
say that there is no such thing as a fully-digital anything that interacts =
with
our world which is itself analog. But my point is the following. =
;
Unlike other commericial instruments that use analog electronics to monitor=
a
capacitive sensor (and then finally digitize the analog signal), the VolksM=
eter
has no analog components whatsoever, except for internal ‘workings=
217;
of the chip—for which one would have great debate as to whether the &=
#8216;pieces’
are digital or analog.
There is one other significant ‘fully’-par=
t to the
instrument. The capacitive sensor used by the VolksMeter is “fu=
lly
differential”.
There are a host =
of
advantages associated with an architecture that is fully digital in the
electronics with a fully symmetric (differential) sensor; I don’t hav=
e
time now to go into detail.
Incidently, some of you amateurs mi=
ght
want to ‘tackle’ the home-brew construction of such an instrume=
nt
yourself; maybe even extend what I’ve done to the world of vertical
seismographs. As Chris Chapman has indicated, they’re much toug=
her because
of material problems (not my expertise). If you want to build somethi=
ng that
could make even the seismology pro’s salivate over, then talk with La=
rry
about the electronics package. He will probably sell you the all the
necessary things ‘for a song’, compared to what the big compani=
es
want for anything.
Randall