Ideas for magnetometer torsion thread

Quartz fibers are not very fragile. Another engineer and I did this many years ago. We bought quartz rod from General Electric, but instead of doing that you could probably use burnt out bulbs from 500 watt halogen lamps. They use a quartz tube to take the high temperatures. We used a oxy/acet torch. Never tried a propane torch. For short fibers just heat the rod ( or tube ) in the middle while rotating it using one hand at each end. When you have some quartz melted just lift the rod out of the flame and extend your arms.

The catapult sounds like a good way. We got longish fibers by having one person hold one end and the other person hold the other end and run. The fiber cools quickly so the runner has to run and then stop as you feel the fiber is no longer elongating. I fell on my ass more than once. This will only get you fibers about thirty feet long. For longer fibers you can chuck a quartz rod in a small lathe and rotate the rod, heat the end of the rod, use another piece of quarts rod to contact the melted end, pull a fiber out by hand and then use a rubber coated wheel on a slow drive to pull the fiber at a constant speed. The torch should be attached to the lathe carriage and slowly moving toward the headstock.

You can make quartz coil springs by chucking a quartz rod in a small lathe and getting a flame just right so it softens the fiber right as it winds onto the rod.

Dan

Maybe even just buy a quartz torsion spring (Leeds and Northrup used to make galvanometers and such that used these as replacement parts).

A glassblower told me quartz is NOT easy to work with; it's got a really high working temperature/softening point, and that usually isn't done with a natural-gas flame but with something hotter (air-acetylen= e, maybe?). Quartz melts at circa 1700C. Glass melts (softens, actually) a= t much lower temperatures. Propane flame (1980 C) isn't hot enough, but acetylene/air (2500 C) or hydrogen/oxygen (3200 C) or acetylene/oxygen (348=

0 C) can get you there quickly.

Glass doesn't age as gracefully as pure quartz.

From "Scientific American Supplement 6"

--'The apparatus consists of a small crossbow, and an arrow made of straw w= ith a needle point. To the tail of the arrow is attached a fine rod of quar= tz which has been melted and drawn out in the oxyhydrogen jet. I have a pie= ce of the same material in my hand, and now after melting their ends and jo= ining them together, an operation which produces a beautiful and dazzling l= ight, all I have to do is to liberate the string of the bow by pulling the = trigger with one foot, and then if all is well a fiber will have been drawn= by the arrow'

The traditional venue for this operation is the crawlspace/basement of any = building named "Physics Hall".

I've used a miniature hyrogen-oxygen torch and a plasma arc. An inexpensive MAPP+oxygen kit (if you can still buy them) should get you there too.

a needle point. To the tail of the arrow is attached a fine rod of quartz which has been melted and drawn out in the oxyhydrogen jet. I have a piece of the same material in my hand, and now after melting their ends and joining them together, an operation which produces a beautiful and dazzling light, all I have to do is to liberate the string of the bow by pulling the trigger with one foot, and then if all is well a fiber will have been drawn by the arrow'

building named "Physics Hall".

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Monofilament glass fibre is probably the most easily available weak torsion spring material. But you probably meed to encase the whole thing in a plastic tube to prevent air currents from interfering.

Regards, Martin Brown

On a sunny day (Mon, 12 Sep 2011 15:20:10 -0700 (PDT)) it happened Alexander Avtanski wrote in :

Actually quite cheap, they have good application info too on the site.

t l

If I remember rightly, one of the techniques was firing an anchored blob of molten quartz out of a cross-bow.

&f=3Dfalse

The quote talks about using an oxy-hydrogen blow-torch to get the quartz hot enough. An ordinary glass-blowers torch got my quartz hot enough to work, back when I was a graduate student. Apaprently I didn't blow enough quartz glass to get silicosis, though my lungs have a couple of small patches of scar tissue that encapsulate my exposre to asbestos at roughly the same time.

-- Bill Sloman, Nijmegen

Hi Alex, You've already gotten lots of good suggestions. Let me quote a bit from RV Jones, "Instuments and experiences", chap.

11 some uses of Elasticity in Insturment design "...The original torsion members were wires or threads of circular cross section; Ohm (1926) appears to be the first to realize that a strip would be much weaker torsionally for the same cross section,...." So you might consider a strip of metal. He goes further and bends the strip into an "L" shape. See RV Jones, Journal of sci. Instrum. vol 32, page 336. In which he uses a thin BeCu strip and angle.

Personally I wouldn't bother with quartz, try some nice stiff metal.

George H.

A taut hard-copper or brass band might not be bad--flatten out some EDM wire? That's common-ish, non-magnetic, and comes as small as

0.005".

-- Cheers, James Arthur

I was thinking about using Phosphor Bronze shim stock. McMaster-C has flat pieces. I really like the L idea, it's only flexible about the one degree of freedom that you care about. I was thinking you could make a cross ( X ) also. I looks more symmetrical. Though I don't know how to do the joint down the middle... maybe it doesn't matter too much? (Someday I'd like to make a better Cavendish balance.)

George H.

ds

.

opps.

There was a SciAm experiment to measure g, using a torsional pendelum, mirrors, and movable weights. They used a piece of tape-recorder (mylar) tape.

I like the fiberoptic suggestion. You can buy a duplex glass (not plastic!) fiber cable jumper at Fry's, or from Digikey, and pull out the glass. Multimode fiber is usually 125 microns OD, pretty strong.

John

.

At some point it's all about time constants.... like everything else. The gravitation time constant is like an hour or two.... the same as if you drop yourself through the center of the earth.

George H.

Will Larkin volunteer? What a stroke of luck! Now how about Slowman ?:-) ...Jim Thompson

--
                  [On the Road, in New York]

| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Will Thompson ever volunteer so say something non-smarmy, maybe even on topic?

Unlikely.

John

Thanks for all the ideas that keep coming! I think I owe you a status update, so here it comes:

- I tried the magnet wire idea. With the magnet wire I have here it didn't work - way too tough and too much torque needed to swing.

- Next I tried flattening a piece of the magnet wire. This didn't work out because I don't have the right tools to make a nice flat band out of the wire. However, a very generous person - Boris Mohar - proposed to send me some flat torsion spring for testing. Thanks Boris!

- I did not have time yet to try drawing a glass/quartz fiber. I think Lowes sells some small torches I can use for the purpose. Also, I haven't tried fiber from a fiber-optic cable yet.

- Meanwhile, I'm testing the idea to use a simple hair. After a couple of hours adjustment I think it kind of works - not perfectly but the drift is much smaller than anything I've tried so far.

...

If somebody is interested, I set up a [very rickety] system to automatically download the data from the computer that is monitoring the magnetometer, rebuild the graph, and upload the image to the web. You can get the latest data in near real time (updated every 10 minutes or so) here:

The magnetometer is currently set up at the table in my kitchen where it is subject to all kind of disturbances, so the data is probably going to be a mess. The gray line is the direct reading, the red line

- the data after an automated attempt to remove any sudden fluctuations caused by opening the fridge door, somebody bumping into the magnetometer, etc. Still, so far I think that for first time I see the daily magnetic field fluctuation!

Reload the image to get the latest data. Keep in mind that the current system relies on a bunch of hastily written scripts that transfer data this and that way between three computers and since one of the computers is my laptop it will not work during the time when I'm commuting to and from work. Also, when my wife gets really fed up with a magnetometer blocking most of the kitchen, the live feed will probably cease, :-) Enjoy it while it lasts!

Thanks,

- Alex

... and forgot to mention - in the same folder I'm periodically downloading magnetograms from the NOAA magnetometer in Boulder, CO for comparison - otherwise I don't know what is real data and what some problem with the device. You can check the Boulder images in the same folder as the latest.png file.

Quartz melted and rapidly cooled becomes silica glass, not quartz fiber, so this would not likely be significantly different than optical fiber except in diameter. (Silica has to cool *very* slowly in order to crystallize into quartz.)

Carbon fiber (depending on type) can be stronger and stiffer than silica fiber and also extremely stable. Availability in small quantities could be a problem, and diameter is typically under 10 microns which might be too small for your application, but possibly worth a try if you can get a sample.

On the other hand fused quartz ( or silica glass ) has all the good qualities that one would want for a magnetometer.

Dan

Consider a fluxgate magnetometer or, as has been suggested, NMR. Much more electronic!

John