PSN-L Email List Message

Subject: Re: Seismic Sensors
From: Brett Nordgren brett3nt@.............
Date: Sat, 30 Mar 2013 14:50:22 -0400

Hi Geoff,

Actually in our feedback instruments we don't care at all about the 
sensor not being linear, just that it gives a large signal change for 
very small motions, is very stable and inserts very little noise.  In 
Dave's designs the sensor plate area is limited by the intended 
instrument size and the spacing is made as small as can be reliably 
constructed.  For the same plate area, if I make a few assumptions, 
variable-spacing capacitor sensors are 40x more sensitive than ones 
in which the plate motion is parallel to the gap.  And to match the 
performance of the feedback verticals, in a 'keyboard' sensor the 
keys would have to be just a few atoms wide with plate spacing maybe 
1/5 of that.

In our case we are happy to trade linearity for sensitivity.  In 
practice, when moved small distances either side of center the 
variable-gap sensors do act reasonably linear, and with the feedback 
they never are allowed to move far from center.

I was just thinking of one fairly basic problem with some of the 
designs being discussed, regarding dynamic range, i.e. the ratio 
between the minimum detectable motion and the maximum tolerable 
motion.  To match the dynamic range of an analog instrument feeding a 
16-bit A/D where the clipping level will be +/- 32,768 x the minimum 
detectable signal level (one count), wouldn't you need to be working 
with an optical array which had 32768 rows each side of center, or am 
I not understanding.


At 01:22 PM 3/30/2013, you wrote:
>To go a bit further, you might make such a thing
>be linear by the geometry of the plates meshing together.
>Also, increasing area you can increase the space.
>Also, this is an analog and not a digital solution.
>To get rid of noise like they did with Video I think
>some kind of digital sensor is in order.
>Direct reading of the X/Y axis of a single illuminated pixel
>might possibly provide this digital solution.
>You are looking at jumps of discrete distances
>and nothing in between on a basic algebra graph
>which is in fact the sensor itself.
>Such a capacitive sensor would work like a capacitive
>keyboard with a moving mass pressing only one key at a time
>according to relative motion.
>We got to think micro switches here and not
>those purely analog amplitude oriented things.
>If you get my drift (chuckle).
>I feel really stupid saying these things even tho I know what I'm looking for.


Public Seismic Network Mailing List (PSNLIST)

To leave this list email PSNLIST-REQUEST@.............. with 
the body of the message (first line only): unsubscribe
See for more information.

[ Top ] [ Back ] [ Home Page ]