Hi Chris,

Yes; the commercial available springs are a problem (unless one is satisfied
with a 1 to ~ 6 sec period vertical); and/or
use a form of feedback.  Even with a zero length spring (such as in my old
Sprengnether), temperature changes made
it tough to keep roughly centered within a series of 3 ~ enclosures with a
set period of 15 seconds.

Just noted a interesting web article this morning on the Physics Forums web
site about "Designing and making springs
with music wire" (Circa May-June 2009).  Its at:

http://www.physicsforums.com/showthread.php?t=315723

Although the wire was twisted prior to spring forming; it doesn't really
sound like a "real" zero length spring too
me....but....such might be ~ better ~ than a common hardware spring.  I
don't have the expertise to evaluate the spring discussion technical
implications to be frank.

Take care, Meredith Lamb


On Wed, Jul 15, 2009 at 4:58 PM,  wrote:

>  In a message dated 15/07/2009, paleoartifact@......... writes:
>
> Outside of buying anything.... versus home building,  the old Georgia Tech
> vertical (Classroom
>  Demonstration model) still remains as a quite interesting experimental
> setup; especially because they claim it is capable of much longer
> periods. It is rather strange that after some 11 years of its existence on
> the web that there is still no known private individual that has garnered up
> a web site on such with their opinion of merit.
> Of course the original design was of wood and some metal, and lacked a
> basic viable sensor and dampening; but the spring and pivot arrangement was
> the main very interesting approach idea for a long period seismometer.
> For readers it is at: http://quake.eas.gatech.edu/Instruments/LPVERT0.htm
>
>  Hi Meredith,
>
>     I agree that this is a good design. The problem with all amateur
> vertical sensors is the spring. Steel coil springs have quite a high
> temperature coefficient of the modulus, so you are limited to periods of
> less than about 6 seconds for simple systems, unless you add a full force
> feedback loop. If you try to make a longer natural period, the system will
> simply collapse as the room temperature increases.
>
>     The second limitation is in the extension springs that you can buy.
> They all have a relatively low initial tension, whereas seismometer springs
> require quite a high initial tension. You can wind springs with a much
> higher initial tensions by either twisting the wire as it is wound onto a
> mandrel, or by pre-bending the wire in the direction of the spring axis, as
> LaCoste did.
>
>     This design can also be compensated for atmospheric pressure noise -
> the principle source of noise by a large factor in uncompensated vertical
> systems.
>
>     Regards,
>
>     Chris Chapman
>
Hi Chris,
=A0
Yes; the commercial available springs are a problem (unless one is sat=
isfied with a 1 to ~ 6=A0sec period vertical); and/or
use a form of feedback.=A0 Even with a zero length spring (such as in =
my old Sprengnether), temperature changes made
it tough to keep roughly centered within a series of=A03 ~ enclosures =
with a set period of 15 seconds.
=A0
Just noted a interesting web article this morning on the Physics Forum=
s web site about "Designing and making springs
with music wire" (Circa May-June 2009).=A0 Its at:
=A0
http://www.physicsforums.com/showthread.php?t=3D315723
=A0
Although the wire was twisted prior to spring forming; it doesn't=
=A0really sound like a "real" zero length spring too
me....but....such might be ~ better ~ than a=A0common hardware spring.=
=A0 I don't have the expertise to=A0evaluate the=A0spring discussion te=
chnical implications to be frank.=A0 
=A0
Take care, Meredith Lamb
=A0
=A0
On Wed, Jul 15, 2009 at 4:58 PM, <ChrisAtUpw@a=
ol.com> wrote:




In a message dated 15/07/2009, paleoartifact@......... writes:

Outside of buying anything.... versus home building,=A0 the old Georgi=
a Tech vertical (Classroom

Demonstration model) still remains as a quite interesting experimental=
 setup; especially because they claim it is capable of much longer periods.=
=A0It is rather=A0strange that after some 11 years of its existence on the =
web that there is still no known private individual that has garnered up a =
web site on such with their opinion of merit.

Of course the original design was of wood and some metal, and lacked a=
 basic viable sensor and dampening; but the spring and pivot arrangement wa=
s the main=A0very interesting approach idea for a long period seismometer.<=
/div>

For readers it is at: http://quake.eas.gatech.edu/Instruments/LPVERT0.htm



Hi Meredith,
=A0
=A0=A0=A0=A0I agree that this is a good design. The problem with all a=
mateur vertical sensors is the spring. Steel coil springs have quite a high=
 temperature coefficient of the modulus, so you are limited to periods of l=
ess than about 6 seconds for simple systems, unless you add a full force fe=
edback loop. If you try to make a longer natural period, the system will si=
mply collapse as the room temperature increases.

=A0
=A0=A0=A0=A0The second limitation is in the extension springs that you=
 can buy. They all have a relatively low initial tension, whereas seismomet=
er springs require quite a high initial tension. You can wind springs with =
a much higher initial tensions by either twisting the wire as it is wound o=
nto a mandrel, or by pre-bending the wire in the direction of the spring ax=
is, as LaCoste=A0did.

=A0
=A0=A0=A0=A0This design can also be compensated for atmospheric pressu=
re noise - the principle source of noise by a large factor=A0in uncompensat=
ed vertical systems.
=A0
=A0=A0=A0=A0Regards,
=A0
=A0=A0=A0=A0Chris Chapman