PSN-L Email List Message

Subject: Re: Earthquake monitoring in Iceland/Scotland/ and the US
From: ChrisAtUpw@.......
Date: Wed, 28 Dec 2005 15:27:42 -0500

Hi Chris- Thanks for the comment on Robert's Circuit- Equalizer Board as I have used Mauro's commercial version (SEQUA10) and found it to be quite good.

Do you believe a digital inverse filter would work just as good?
                  Jim O'Donnell     
         Geological/Geophysical Consultant

Hi Jim,

     You need a filter which has an inverse f^2 relationship from 0.5 Hz up to 4.5 Hz and that merges into a x1 gain by about 9 Hz. The gain should also drop off sharply below 0.5 Hz to remove VLF noise. There must be NO zero output pole at 4.5 Hz. I do not know enough about the current digital filter designs to give a firm opinion as to how practicable this is, but 'probably not' in answer to your question. 
     The other problem is that the additional x100 gain at 0.5 Hz has to come from somewhere. If it is not provided by an auxiliary amplifier, it will have to come from the main amplifier, probably reducing the range and increasing the noise a bit. You also have the stepwise digitisation to cope with. This can limit the accuracy of the compensation process. 

     If you had a signal of 1,000 counts at 4.5 Hz, there will be just 10 counts at 1/10 the frequency to do any compensation. Using an additional compensation amplifier is likely to give better results. Lower signal levels cannot be fully compensated - the digital signal is just not available - unless you are either using a 24 bit ADC, or are using an oversampling technique to increase the accuracy.   

     I understand that the current SEQUA10 is a significantly 'improved' version of the original. I have seen a modified version used with a 2 Hz geophone with very good results, in comparison with a Willmore. 


     Chris Chapman


Public Seismic Network Mailing List (PSN-L)

[ Top ] [ Back ] [ Home Page ]