DIY seismograph suggestions

[coppercone2]

So my DSA thread had some merit with applications including balancing and seismometers, I think I will try a seismometer first because I thought about it for a long time.

There was a suggestion to make one using a speaker.

I thought I would ask people for more specific details about good DIY experimental seismometers of low complexity and cost, and if there are any projects that avoid a lathe but have most other processes (I am fairly well equipped).

[Berni]

Easiest is probably placing a big weight onto a piezzo buzzer or the element in a piezzo igniter from a lighter.

Vibrations try to move the weight up and down, this puts varying amounts of force onto the piezzo that generates a varying voltage.

There are also piezzo vibration sensors designed for doing this: https://www.sparkfun.com/products/9199 But im guessing the weight would need to be bigger than that when sensing real sieziminc events that have components in the Hz range.

[Zipdox]

I think the best way to make a seismometer would be with an accelerometer. They are pretty precise, and measure acceleration in 3 axes. There's accelerometer shields you can buy but perhaps the easiest solution would be to use an old smartphone and write an app for it.

[Ian.M]

Scientific American magazine published quite a few Seismograph designs over the years.  You'll find some of the better ones here: http://psn.quake.net/infoequip.html
together with a lot of other amateur seismography info.

In most cases the electronics are severely outdated and redesigning them with modern components would be worthwhile.

[Gyro]

If you want to go 'old school', there was a seismograph project in Practical Electronics magazine back in August/September 1971. The electronics could, of course, be updated massively but there's lots of interesting mechanical construction stuff.

https://worldradiohistory.com/UK/Practical-Electronics/70s/Practical-Electronics-1971-08.pdf
https://worldradiohistory.com/UK/Practical-Electronics/70s/Practical-Electronics-1971-09.pdf

Ah, the nostalgia! 

[phil from seattle]

Scientific American magazine published quite a few Seismograph designs over the years.  You'll find some of the better ones here: http://psn.quake.net/infoequip.html
together with a lot of other amateur seismography info.

In most cases the electronics are severely outdated and redesigning them with modern components would be worthwhile.

This. The old SA column Amateur Scientist had some very nice seismograph designs.  The pickups were basically a magnet on large weighted boom and a coil sensor.  An aggressive LP filter followed by an op amp fed a strip chart recorder. My problem as a 16 year old geek was finding a strip chart recorder - everything else was possible. But, strip chart recorders were simply out of reach for my limited funds. These days, this is all very simple and you can use pretty bog standard datalogging to record the signals and sophisticated signal processing on them.  Redesign with modern electronics should be pretty straightforward.

Also, you can find geophones that will work for a seismograph.   Sparkfun carries one.

[coppercone2]

If you want to put something in a basement like 20-30 feet away from a road on a hill is a geophone or a seismophone a better option?

https://www.nanometrics.ca/sites/default/files/2017-11/ism-_bb_seismometer_geophone_comparison.pdf

Not familiar with this at all

[Kleinstein]

30 feet from a street one would mainly record the trafic, at least if the road conditions are not that good. For sensitive experiments like AFM one usually wants a place away from a road and at a quiet place in the building (e.g. more closer to a stiff part (e.g. corner) and not in the middel of a room).


AFAIK the main difference from geophone to seismometer is the frequency range - the seismometer is lower frequency. Lower frequencies see less daming and that way could also see events further away.
So it depends on what you are looking for.

[phil from seattle]

The longer and heavier the moment arm of the seismometer, the lower frequencies it will sense.  You will also want some dampening to reduce oscillations.  Being 30 ft from a road, you will definitely pickup trucks and such, especially if there are potholes. 

There are sources of back prints of the SA Amateur Scientist columns. Worth looking at the ones about seismometers as they detail out the physical construction.  Try this one - https://earthalabama.com/build.html

[coppercone2]

I mean its not like a busy road at night there should be nothing for a long time

this actually looks like quite a bit of work, but it looks like I have most of that crap, maybe even a turnbuckle

So the lever arms need to be pretty stiff I guess?

[Ian.M]

So the lever arms need to be pretty stiff I guess?

I assume we are discussing the near-horizontral pendulum style of seismograph.

The arm needs to be stiff enough so that any flexure resonant modes are above the highest frequency of interest, even with the mass on the end.  The frame doesn't have a large mass rigidly coupled to it lowering its resonant frequency so its stiffness requirements to get the resonance to a high enough frequency are much lower.

[MikeK]

Something I've had bookmarked for a while.  Only on web archive it seems:

https://web.archive.org/web/20180325052841/http://science.uniserve.edu.au/school/Seismograph/menu.html

[coppercone2]

Oh man water.. I thought they used a electromagnetic damper?

I don't want liquids in this.

[Ian.M]

The electromagnetic damper from the 1979 S.A. Lehman Seismometer is probably the best option if you want to avoid liquids.  In its original form its an aluminum vane moving transversely between two horseshoe magnets arranged with their poles aligned on opposite sides of the gap so they attract each other, and is adjustable by increasing the gap to weaken the field.  A modern implementation could use four Neodymium magnets epoxied to two mild steel bars in place of the horseshoe magnets, or could even use the magnet assembly from a scrap hard disk drive, with the vane in the gap the voice coil actuator normally goes in, adjusted by varying the depth of engagement of the vane with the gap.

[coppercone2]

What can be improved in this device?

I know for one, the knife edge is imporant in scales. My best scale has a sapphire knife edge.

I might have a piece of titanium tubing that I have been saving for something, maybe this would make a good lever arm?

I am trying to think of where high quality materials can be used and if I have any high quality materials that can be used for this.

The proper mechanism to seat the lever when it is not in use for transport would be sapphire jewels that spikes go into (that are on the lever), which lower and let it rest without wear when not in use.

I do wonder if I can put a piece of metal in the bed of my scale, put damper magnets in there and use a laser to measure the deflection on the scale (it is optical). It already visible functions as a seismograph as far as I can tell lol

But that device has the advantage of a longer lever, something that may out benefit even the most precise construction.

The lever arm in the scale is in the form of a + , I guess this shape is more favorable over a tube for some reason.

[Dunckx]

Try looking at the seismic projects here: http://www.vaxman.de/projects/projects.html , they're down towards the bottom of the page.  His negative resistance damping is a great idea, and can be used to reduce the natural frequency of oscillation of a geophone.  These things are fairly cheap.

Loads of other interesting stuff here too.  He has the world's biggest collection of fully functional analogue computers.

If you intend building something which is sensitive to higher frequencies, and want to digitise using a PC soundcard, ARTA could be handy: https://www.artalabs.hr/

There's loads of practical stuff here: http://www.seismicnet.com/infoequip.html

You can also download a free ebook: https://ardbark.com/fundamentals-seismic-wave-propagation-chris-chapman/

If you really want to roll your own in a serious way and you've got machine tools available, there's the Yuma design here: http://www.earthmode.org/

There's some genuine old school 100% mechanical seismometers here: https://www.erdbebenwarte.de/en/ some over 100 years old and they all work, including one with a 17 tonne pendulum.

Quake net has some DIY links and live records: http://psn.quake.net/ and you can also try searching US patents on seismometers / seismographs.  There's gazillions of them.

I've attached a few of the references I've collected over the years.

That lot should keep you out of mischief for a while.

[Kleinstein]

The material of the beam of the seismometer is not crital. One wants quite some mass anyway - though it sould be non magnetic if mangnetic detection is used.  For the knife edge, one could consider a solid state joint instead, so a piece of thin high strengh material - would likely need to be under tension and thus some extra holder. The joint is not that critical, as different from a clock pendulum high dampling is wanted.

There are now better permanent magnets to make the magentic picup more compact / more sensitive.  Still old style AlNiCo can get a strong field it is just much heavier and often more expensive.

[Dunckx]

The material of the beam of the seismometer is not crital. One wants quite some mass anyway - though it sould be non magnetic if mangnetic detection is used.  For the knife edge, one could consider a solid state joint instead, so a piece of thin high strengh material - would likely need to be under tension and thus some extra holder. The joint is not that critical, as different from a clock pendulum high dampling is wanted.

There are now better permanent magnets to make the magentic picup more compact / more sensitive.  Still old style AlNiCo can get a strong field it is just much heavier and often more expensive.

The problem with modern neodymium magnets is that their field is quite temperature-sensitive, so you get long term drift problems unless the magnet part is temperature-controlled or compensated as in the Sprengnether S-7000, but unless you are after pro levels of reproducibility and stability it probably is OK.  Alnico is much better in that regard and AFAIK is the most thermally stable permanent magnet material.  Capacitive sensors are frequently used in seismometers for that reason.  Another good choice would be an LVDT.  See attached.

See also the Theremino-based projects on PhysicsOpenLab http://physicsopenlab.org/category/geophysics/ .