PSN-L Message - Subject: Re: Seismometer Questions

=20

In a message dated 30/05/2010, jzambory@......... writes:

I am new=20 to this hobby of seismology. I have photos of everything I will refer to= =20 at:
www.flick= r.com/photos/darksky123/

I=20 changed a few things. I thought it would be very difficult for me to mad= e a=20 perfect knife edge
for the pivot point so I epoxyed a ball bearing= to the=20 end of my steel rod. I sanded smooth the
surface that it rests on,= (not=20 sure what it is called). Seems to work?

Hi Jeff,

Good choice. Knife edges and point in a cup= =20 'bearings' are almost guaranteed to fail, often sooner than later. They ar= e both=20 seriously inadequate.

You need Stainless Steel or Tungsten Carbide= ball=20 bearings and a stainless steel or carbide counterface. See www.smallparts.com I bought a large=20 SS scalpel blade and cut a square from it. Ordinary steel bearings corrode= quite=20 quickly - don't use them. Two part acrylic glue is much better than= epoxy,=20 which dries brittle and does not give it's best performance on polished=20 surfaces.

See http://www.jclahr.com/sci= ence/psn/chapman/2008%20lehman/lehman_prototype.jpg I=20 checked the maximum period and was able to get 60 seconds quite OK, so I= know=20 that this construction works well. I normally set the seismometer to opera= te at=20 20 to 25 seconds with the 22" arm. Note that the mass is fitted to th= e end=20 of the arm to get the maximum period.

For my=20 pickup magnet, I hung it down off the rod. I could not get it to hang pr= operly=20 when I
tried it on the side of the rod. Made things twist.=20

Do NOT mount the magnet on the arm. It gives= noisy=20 results sensing changes in the earth's magnetic field and surge curre= nts in=20 the house wiring. Put the coil on the arm.

See http://www.jclahr.com/science/psn/chapman/= lehman/index.html for=20 the damping and sensor magnet block designs.

You are far better off with a rectangular coi= l and=20 NdFeB rectangular magnets. They are cheaper and more powerful than Alnico= =20 magnets.

What damping is your arm it set to? Can= you=20 set it to 0.7 critical quite easily? If you displace the mass 10mm and rel= ease=20 it, it should cross over the balance position by 0.5 mm. This is not diffi= cult=20 to measure with a ruler and a magnifying glass.

You need a V suspension connected to a crossb= ar on=20 the arm to prevent it from rotating about it's axis due to the offset drag= from=20 the damping plate. I use 30 lb 7 core nylon coated SS fishing trace with= crimped=20 loop ends. If you don't do this, you get odd looking signals produced by= the=20 coil sensing the rotation of the arm as well as the=20 seismic displacement.

I got a=20 bolt for the thing that the wire attaches to on the top piece. To vary= the=20 period of the
arm, I just adjusted the length of the=20 bolt.

I think that this may be your problem when tr= ying=20 to get a long period. What thickness of wire are you using? About 10 to 12= thou=20 diameter wire works well. See http:/= /www.daddario.com Ask for wire for a=20 mandolin. The suspension needs to be rigid with sharp edges on the cl= amp. I=20 suggest that you use a 3/8" bolt. File off the thread for about 1/2" at th= e end.=20 Then file off 1/3 the thickness of the bolt on the other side. Drill an 1/= 8"=20 hole and use an 1/8" bolt and two washers to firmly clamp the suspens= ion=20 wire.

So my question. I read that I should be able to detect magnitude 6=20 and greater for earthquake
around the world.

M 6.5 world wide.

What really prompted me to write was= not detecting the=20 Vanuatu 7.2 quake from last week. I got a
small squiggle, but not en= ough=20 to put WinSDR into alarm and the time was a couple of
minutes early= from=20 when the waves should have arrived according to the USGS. Compared
t= o what=20 I got with the Baja quake, I thought I should have gotten something. I= get=20 nothing. I am hoping the people will be able to point out improveme= nts or=20 mistakes that I am making.

For my pick up coil, I originally had= a coil=20 from a relay. It was about an inch in diameter. It
seemed to work fo= r=20 starting out. That was the pickup coil I was using when I got the Baja= =20
quake. I knew I had to get one with more turns of wire and a higher= =20 resistance, that one was
only about 300 ohms. A local electronics sh= op had=20 some 40AWG magnet wire, a 1/4 pound
roll. I bought it and made a thi= ng to=20 roll it onto.

38 or 36 AWG wire is easier to handle. Make= the=20 rectangular former out of glass circuit board.

&nb= sp;Not really feeling like playing with that tiny wire again, I=20 looked through an electronics supply catalogue and found a 110Vac power= relay=20 that stated a coil resistance of 11K=20 ohms.

Aim for a coil resistance of less than 500 Oh= ms. 10=20 k Ohms is far too high and it will be noisy.

The=20 question I have about it is, does size matter? The coil itself is
on= ly=20 about 1/2' in diameter. I look at other amateur seismometers that I can= find=20 photos for
and the coils are a lot larger. Should I get some larger= gauge=20 wire (36AWG?) and make a
new coil? Or is the coil I have now=20 OK?

No. This coil is far too small to allow the= mass to=20 drift by up to +/-10mm.

Size matters a lot.

I=20 believe the seismometer works. When I walk up to it, WinSDR shows moveme= nt and=20 it
usually goes into alarm unless I walk up very slowly.
    You should be able to see the ~6 second= ocean=20 microseisms at 100 to 200 counts.

The=20 period of my system is about ten seconds. That is about as long as I am= able=20 to get it. I
can't seem to get the 14+ I read about for some systems= ..=20 Should I fight with it to get a longer
period, or is ten seconds lon= g=20 enough?

    10 seconds is NOT long enough. You need 15 to= =20 25 seconds.

Is a=20 ball bearing OK for a pivot point?

    Yes, provided that it is Stainless Steel= and=20 over 1/4" diameter. Mount the ball on the FRAME and the counterface on the= end=20 of the ARM - NOT the other way around!

I sent a=20 letter and photos to Mr. Cochrane and he told me to get rid of the wood= I was=20 using
on my system. I had made a mount for the pick up coil magnet= out of=20 wood at first (it is now
metal). It is easier for me to work with wo= od=20 than metal. I have the copper plate for my
dampening attached to the= rod=20 with a piece of wood. Is this OK? The original diagram I was
working= with=20 says use wood. Should I get rid of it too and use=20 metal?

    Wood tends to warp and the dimensions change= with=20 moisture levels. It needs to be thoroughly varnished to be reasonably stab= le -=20 particularly the end grain.

I have my seismometer orientated basically north-south. Is that OK=20 for North America or
should I point it east-west?
    Should be OK with the arm N/S

Should I=20 be getting more earthquakes?. Or for where I live, are things working as= best=20 as I can
expect?

    Depends on where you are located. In a low se= ismic=20 signal region, I get one or two teleseismic quakes per week as well= as=20 regional quakes. Check the drumplots daily?

I am=20 looking forward to my new hobby of seismology. If things work well I hop= e to=20 make
another one for the other direction and maybe even a vertical= =20 seismometer. I bought a three
channel amp from Mr. Cochrane. Just in= =20 case... :-)

    Next time I suggest that you make a T fr= ame=20 out of 3" x 1" Aluminum U channel with 6" triangular x 1/8" thick corner= plates.=20 Use all Stainless Steel bolts, nuts, washers and fittings. I buy mine= from=20 a boat chandler. This construction is not only easier but much better. It= allows=20 you to set up all the working parts and clearances in sequence and you jus= t tilt=20 the whole frame to set the period.

    What metal is used for your arm? It should no= t be=20 heavy in comparison to the mass. What mass do you use and how is it secure= d?=20

    Check http://www.mutr.co.= uk/images/Seismometer.pdf?osCsid=3D8cl12vb4vnhq2l65723jqa9em6 for= =20 the SEPUK seismometer manual. We have over 400 of these now operating in= the UK=20 and they seem to give excellent results with AmaSeis. Also check http://www.bgs.ac.uk/schoolsei= smology/ for=20 the school seismology project.

Thanks=20 for all your time. Kind of a long letter...
Jeff=20 Zambory

    Let me know if you need more help?

<= FONT style=3D"BACKGROUND-COLOR: transparent" color=3D#000000 size=3D2 fac= e=3DArial>    Chris Chapman

<= /DIV>

PSN-L Email List Message