PSN-L Email List Message

Subject: Re: low noise PSD
From: Brett Nordgren brett3nt@.............
Date: Tue, 18 May 2010 20:44:38 -0400


Hi Chuck,

An excellent design that Dave Nelson in California came up with from 
his satellite days can be seen at:
http://bnordgren.org/seismo/FBV/1043_bridge.pdf    note that R1 is 
now 4.7K, which works better.

It is not really a PSD design but is acting more like a 
switched-capacitor, charge-transfer circuit.  The 48pF sensor 
capacitors are substantially fully charged during each half cycle, 
leaving only short switching transients to be filtered, which is easy 
enough to do.  In the design as shown, there is a pole in the 
baseband response at about 320 Hz. and it gives a sensitivity of 
about 1200 V/m.  The driven, fixed, sensor plates are on two sides of 
a double sided PC board, resulting in a stray capacitance Cc of about 
130pF, which doesn't have much effect.  The moving, grounded, plates 
are above and below the driven plates with a spacing of about 1/32", 
making Ca and Cb somewhere around 48pF.  The fact that the moving 
plates are grounded is a significant advantage in reducing the number 
of wires bridging the pivots.

To insure gain stability we use an external clock based on a ceramic 
resonator rather than the internal oscillator in the LTC1043.

Tests show that its noise is very nearly as good as some professional 
instruments and is significantly better than any sites we have been 
able to find.

You can get an idea of its construction and performance at:
http://bnordgren.org/seismo/gif_images.htm

Best regards,
Brett

At 12:07 PM 5/16/2010, you wrote:

>Here's a technical question about seismometer electronics.  What 
>phase sensitive detector design do you recommend for low noise?
>
>I have built PSDs using an op amp and FET switches (the amp switches 
>between gain of +1 and -1).  Analog Technology and National Semi 
>describe these circuits in application notes.  And you can use a 
>switch with an instrumentation amp.  But these switches produce huge 
>high frequency noise pulses that are very difficult to shield out or 
>filter out.
>
>You could also use IC multipliers like the AD630 or a Gilbert cell 
>multiplier.  I have not tried either of these and wonder if they are 
>worth the cost and/or trouble of construction.
>
>Any suggestions?
>
>Thanks,
>
>Chuck Burch
>
>
>__________________________________________________________
>
>Public Seismic Network Mailing List (PSN-L)
>
>To leave this list email PSN-L-REQUEST@.............. with the body 
>of the message (first line only): unsubscribe
>See http://www.seismicnet.com/maillist.html for more information.


__________________________________________________________

Public Seismic Network Mailing List (PSN-L)


[ Top ] [ Back ] [ Home Page ]