PSN-L Email List Message

Subject: Re: My AS-1 Type Sensor
From: Christopher Chapman chrisatupw@.......
Date: Wed, 24 Nov 2010 07:52:11 -0500


I recently constructed an AS-1 type sensor. My station KCMA is in the=20
IRIS network.

I am using an amplifier/filter, designed by Andy Loomis, which works=20
quite well. The natural period of the sensor is 1.7 seconds. The cutoff=20
period of the filter is set at 1.4 seconds.

My questions are.... Does anyone have a amplifier/filter design that=20
will extend it to 10 to 15 seconds? I know the manufacturer of the AS-1=20
can provide the amp/filter, but it's a bit too pricey for my budget.

Hi Richard,

     The AS-1 electronics uses two single stages of a bass boost=20
amplifier in series to compensate for the f^2 sensor characteristic. It=20
then uses a 12 bit single polar ADC with an offset amplifier to=20
digitise the signal and this is directly fed to the computer.
     The specialist data recording and analysis program called Amaseis=20
is used. This has a range of possible ADC input capture programs. Check=20
them out? It also provides additional digital filtering, including a=20
period extension filter. But you can only apply the digital period=20
extension to an extracted trace, which means that you have first to=20
find the quake trace! This is a bit of a disadvantage!

     And, I believe somewhere, on the psn list, I recall someone who had=
=20
made improvement's to the AS-1 amp/filter. Any recommendation's?

     Have you visited John Lahr=E2=80=99s website? John and I did some=20
developments on the AS-1, but we failed to get Jeff Batten interested,=20
which was a great pity. The AS-1 as marketed, is not a very good=20
seismometer.  I replaced the oil damping with a copper plate + NdFeB=20
magnet damper, which greatly improved the performance. The AS-1 has a=20
compensated range only from about 1 to 5 seconds, which is OK for=20
recording microseisms, but otherwise is of limited use. I modified the=20
AS-1 electronics to extend the range to about 18 seconds, which brings=20
in the surface seismic waves very nicely. The other limitation is in=20
the high frequency response. You need 20 SPS to give a good trace of=20
local and regional P waves to 5 Hz, but the AS-1 is fixed at 6 SPS.

Additionally, my ADC is a Dataq 194RS which is only a 10 bit unit. The=20
original AS-1 "black box" operates at 12 bits. Is 10 bits satisfactory=20
or do I need a higher range?

     A 12 bit ADC is marginal for recording earthquakes. The relative=20
amplitudes of the P, S, L and R waves vary greatly. Dataq do market a=20
12 bit ADC, which they choose to call cheap. These work OK. I would=20
advise you to seriously consider buying a 16 bit ADC from Larry at PSN=20
net.

     The detected amplitude of the signal falls as f^2 below the=20
resonance point, so with a 1.7 second pendulum, a 20 second surface=20
wave will have 1/138 of their true amplitude.

    Regards,

    Chris Chapman


__________________________________________________________

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.

[ Top ] [ Back ] [ Home Page ]