PSN-L Email List Message

Subject: Re: Spike Problem
From: =?ISO-8859-1?Q?J=F3n_Fr=EDmann?= jonfr500@.........
Date: Tue, 06 Jun 2006 20:22:44 +0000


Roger wrote.

> 1.  They seem to increase in number after you have moved the
> seismometer.

The geophone him self is in the same location that it has been for the
past months. What did change was that I got a screened cable and I did
some work on it. There are two sources of spikes, human activiti and
this electronic type of spikes. The gif images show the electronics

> 2.  They all have some component below the zero line.  This could be
> the result of high pass filtering which applies a fraction of the
> initial pulse as an inverted result.  Of course there must be an
> initial pulse.

The amplifer board is set up to work with the geophone, short period. I
don't know there is a inital pulse, it problay is there, but the time
period is so short it cannot be seen.

> Is it possible that the spikes are coming from settling of the
> seismometer components, or of the seismometer base?  Are there springy
> components, such as stiff but flexible wires, that could be slowly
> reshaping over the course of several hours (or days) as they slowly
> move  or relax to a permanent configuration?   

The base of the seismonitors is stable. There are no spring components
near the seismonitor that I know of. But all concreate icelandic house
have a wire grind in it, i doubt that is moveing.

> The air shield surrounding the seismometer could also be the source of
> spikes if it settles over time.

The air around the geophone is stable, it is inside.

> I doubt that the spikes are electrically generated. If they were, they
> should be nearly equal for each channel (or be unique for each
> channel),

The pickup along the channels might be diffrent. Also, after I did
connect the ground today I haven't seen one spike for few hours. Large
or little ones.

Mauro wrote.

> 1) When use a CAT5 shielded cable be SURE to connect to
> with the shield ALL the UNUSED wires of the cable.

The shield covers all the wires inside the cat-5 cable. The shield is
not pr cable basis.

> 2) ALL MARK L15 geophones has their case (metal) connected to
> one of the signal poles. To identify what is it try it
> with a ohmmeter (before this measure is better you disconnect
> your premamplifier).
> Touch the ohmmeter in the case (be sure the probe really touch
> the metal case, metal oxide sometime doesn't allow good contact).
> L15 has been designed to be used in plastic case.
> If you use them direct in ground or in a metal case
> then your metal case and the ground itself become in contact
> with one of the 2 poles of the preamplifier input.
> This will makeup a GROUND LOOP capable of pick almost all
> consistent spikes.
> To solve this be sure the L15 are insulated from each other
> and from ground.
> when the L15 geophones are insulated you can connect their
> poles to the CAT5 twisted couples. Each geophone on a twisted
> pair.
> The CAT5 cable shield now can be connected safely to the
> (eventual) metal box of your sensor box.
> Then go to the larry board.

I still have the sensors in the plastic box. 

> 4) BE SURE to keep separates the analogue power supply and
> the digital power supply. So use a coil transformer
> for the Larrys board and DO NOT USE THAT POWER FOR ANY OTHER

The power comes from the A/D convert board. See here,

I have connected the ground and the signal looks noise free from any
electronic disturbance, I haven't seen a large electronic type of spike
since I did connect the ground around 15:20, but I usally did see two or
three spikes (smaller ones) in this range of time. I now hope that this
problem is fixed, but next 24 hours is going to tell me if the problem
is fixed or not for ceartin. But based on resaults from the past few
hours, I have my hopes up a bit. I now only see human activiti or
natrual noise (weather, etc..) in my plots.

If this problem is fixed, I want to thank everybody for the help. :-)

Jón Frímann


Public Seismic Network Mailing List (PSN-L)

[ Top ] [ Back ] [ Home Page ]