Randall,

I am interested in "digging deeper" and in better=20
understanding the practical implications of=20
mesoscopic material characteristics on the performance of our designs.

I'll apologize to the List in advance for going=20
into some fairly exotic stuff here which isn't=20
likely to be of interest to most of us, but I do=20
know of 10 List members who have built or are=20
building force-balance verticals who will=20
definitely be interested.  So please let me know=20
if I'm cluttering up the List too much.


We have some fairly powerful tools for analyzing=20
the performance of our designs which assume that=20
they are linear systems to a first=20
approximation.  A simple engineering technique=20
for dealing with nonlinear systems is to consider=20
that some parameters which are normally taken as=20
constants are actually variables, either with=20
time or with some other parameter.  Then the=20
system is analyzed with the linear tools, using a=20
range of assumed values for the 'variable=20
constants'.  Though not rigorous, that approach=20
will usually give an excellent idea as to what=20
the real-world performance might be when you include the nonlinearities.

In the case of mesoscale elasticity in springs, I=20
would think that a linear damping factor or=20
possibly the spring constant, df/dx or dx/df as=20
you prefer, or something else, might be assumed=20
to vary with strain, or with time, which raises=20
the question: from looking at decay curves or=20
other data you have obtained, is it possible to=20
crudely describe the mesoscale properties of our=20
springs as variations in their assumed linear=20
damping or variations in stiffness, and what=20
would be an appropriate range of values for me to=20
assume that they could take under various=20
conditions?  I realize that I could undoubtedly=20
derive these from data in your various=20
publications, but thought it would be safer for me to simply ask.

At 03:39 PM 8/13/2011, you wrote:
>Also to All:
>       Nonlinearity (better called  mechanical=20
> complexity) is what ultimately limits, at low=20
> frequencies, the performance of every=20
> seismograph; but it is not the kind of=20
> nonlinearity that feedback overcomes in=20
> remarkable fashion, as is commonly well=20
> known.  The kind that is not accommodated  is=20
> related to the very reason materials=20
> creep=ADbecause of defect structures.  At the=20
> mesoscopic level, these defect structures cause=20
> the potential energy well to be other than=20
> smooth.  In other words, the force required to=20
> accomplish feedback (standard electrical=20
> engineering) is not able, at very low levels,=20
> to operate on an error signal that is=20
> consistent with simple-minded theoretical=20
> expectations.   As one of my astute students=20
> years ago said, =93physics is easy if you don=92t dig too deep=94.
>        If the force feedback approach were as=20
> perfect as some want to believe, then there=20
> would have been no reason to hold the IRIS=20
> sponsored =93broadband conference=94 several years=20
> ago, which I attended.   A poster session that=20
> I presented at that conference is online at=20
>=
 http://www.iri=
s.edu/stations/seisWorkshop04/PDF/tahoeI1.pdf

It was my understanding that the conference was=20
inspired by the shock to the long-period=20
community when Streckeisen decided to stop making=20
his STS-1, and was an effort to see if any new=20
technologies might have arisen since the original=20
STS-1 design which ought to be considered when=20
designing its replacement.  In fact, I believe=20
partly as a result of that conference, the=20
decision was made that the best technology for=20
that replacement should again be a force-balance=20
design, which is now nearing commercial availability.
http://www.brtt.com/events/eaug2011/talks/ogie/STS-1_Sensor_Replacement_Marc=
h_21_2011.pdf



Thanks,

Brett=20


__________________________________________________________

Public Seismic Network Mailing List (PSNLIST)

To leave this list email PSNLIST-REQUEST@.............. with 
the body of the message (first line only): unsubscribe
See http://www.seismicnet.com/maillist.html for more information.