Junior Seminar/Rouvelle/Spring ‘06

Week 2- 3

Torsion Magnetometer construction part 1.

Here is a link to other magnetometers, and SAS articles on geo and electro magnetism.

Let’s first build the Torsion Balance:

The Diverdi article was a bit hard to follow, so I read through both Diverdi and Baker, and came up with the significant difference between the two, described by Diverdi in this excerpt, found on http://xtrsystems.com/magnetometer/coil:

“The original Balance itself was built according to the published directions but suffered from considerable under-damping which resulted in poor transient response of the control loop and a strong tendency for the control loop to oscillate. On the suggestion of another experimentalist, Ed Phillips of San Gabriel, CA, a silver half-dollar is used instead of a high-purity copper penny (pre 1982) for magnetic damping. A mounting is included which allows fine-tuning of the damping factor.”

Diverdi includes some photos of this and uses a flat, brass screw attached to the half-dollar (or is it two half-dollars? If you look at the side view its hard to tell) through a hole in the glass enclosure. The screw is covered by a nylon spacer - I used a plastic housing for a banana plug - and is weighted at the opposite end by a brass nut.

As an aside, Diverdi mentions how interesting it is to observe the oscillations of the magnet when a Function Generator is attached to the Helmholtz coils – this aside leads me to mention two asides of my own:

General Aside: Alvin Lucier – check out I am sitting in a room, and some of the Slow-Sweep Oscillator pieces

Specific Aside: Experimentalist David S. Walonick

When we get to the Helmholtz coils I will bring in my own function generator and we can observe the effects.

Velleman Function Generator from Electronix Express, an excellent store, inexpensive but I’ve heard reports that they don’t work with some laptops… be careful. Then there is always Ebay.
Low Cost Function Generator - a kit

An important thing to consider when buying any device is its resolution. With David Walonick’s experiment in mind, what sort of resolution do we need from the function generator? Will any function generator work?

On to the construction:

As we spoke about last week, the first thing we need to do is make a parts list, looking at both the initial Shawn Carlson article on Roger Baker’s Torsion Magnetometer and Diverdi’s site I came up with the following parts list – bearing in mind that what Diverdi refers to as the “Balance” is virtually identical to Roger Baker’s original design. Fortunately the first Carlson article does a good job of describing how to build the Balance:

Parts List for the Balance: This is how I did it, you might have a better solution – I did this in Excel – but any spreadsheet program will work. A column for pricing is also essential…

*Balance*	*Qty.*	*Where I found it*	*RadioShack part #*	*Price Per*	*Total*
silky, multifilamented nylon twine	1	home depot		1.89	1.89
window glass - 5"x8"	1	JoAnn's crafts		0.99	0.99
silicon cement	1	home depot		2.79	2.79
electrical tape	1	home depot		1.59	1.59
epoxy	1	home depot		2.49	2.49
rare earth magnets	2	RadioShack	64-1895	1.89	3.78
MAGNETS methyl ehtyl ketone (MEK)	1	home depot		3.09	3.09
small mirror squares - 1.5cm	2	99 Cent Store		0.99	1.98
donut shaped magnets	4	RadioShack	64-1888	2.79	11.16
*Coils*					0
two heavy cardboard rings - 6.5 cms	1	JoAnn's crafts		2.79	2.79
three thin wood spacers	3	around			0
nylon spacer	1	RadioShack		1.5	1.5
#30 enamel coated magnet wire	1	RadioShack		3.5	3.5
screw and nut that fits into spacer	1	home depot		0.79	0.79
*Electronics*					0
cds photo cells - 2	2	RadioShack	???		0
lm324	1	RadioShack	276-1711	2.09	2.09
various components		RadioShack			0
*Sensor Enclosure*					0
Foam Core - large sheet	1	Wal-Mart		2.5	2.5
red cellophane - roll	1	Wal-Mart		2.29	2.29
wooden dowel .5"	1	Home depot		0.89	0.89
heavy wooden clothespin - 1 bag	1	Home depot		1.99	1.99
two position switch	1	RadioShack			0
three wood screws for leveling	1	Home depot		1.09	1.09
*Laser Pointer*					0
laser	1	Friend			0
three spring loaded adjustment screws	1	around			0
*Misc*					0
wood for base and spacers					0
wood blocks to support coils					0
				*Total*	49.19

Now that we have the list we need to insure that we have all the tools we need. It looks like we need:

Glass Cutter

Bit to drill into glass

Clamps

Tweezers

Measure that has both inches and cm.

Convert Inches to cm

OK, now we will follow the build instructions from the first Shawn Carlson article and, while bearing in mind the Diverdi improvement, put the Torsion Balance and Nulling Magnet assembly together. We will be building three of these things over the next few weeks.

Prepare the Nylon Fiber:

1. Cut a 30 cm (one foot) length of nylon twine.

2. Gently unravel it and use tweezers to select the finest strands, which should be about 25 microns (0.001 inch) thick.

Build the Case:

1. Cut two strips five centimeters wide and 15 centimeters long using a glass cutter

2. Cut eight glass strips one centimeter wide by five centimeters long and use silicone cement to glue pairs together back to back.

3. Glue one pair of these small glass strips to the top and one to the bottom of each of the longer glass walls. The smaller pieces will act as spacers between the walls.

4. When the glue sets, select one of the wall pairs and calculate its center point.

5. Carefully (use a bit designed to drill glass) drill a hole large enough for the screw that will hold the silver half-dollar that will act as a magnetic damper. This wall will be the back wall of the glass enclosure.

6. Cover the horizontal spacers of the front wall (the wall without the hole) with a layer of stretchy, black vinyl electrical tape (The tape prevents the glass from cuffing the fiber).

7. Lay one end of the fiber across the center of the top spacer of the front wall and tack it in place with a small dollop of epoxy; secure it with another strip of tape (the epoxy will keep the thread from slipping over time).

Generate the necessary tension in the fiber:

1. Temporarily attach four nickels to the other, free end of the fiber.

2. Epoxy and tape the end of the fiber into place against the bottom spacer, locking in the tension. The nickels will be below the bottom spacer.

3. When the glue sets, release the nickels.

Attach the rare earth magnet to the fiber:

1. Deposit a thin smear of silicone cement on one rare-earth magnet and sandwich the fiber between the two of them. Be sure that the magnets overlap each other completely and are perfectly centered on the fiber as the glue hardens.

Attach the mirrors to the fiber:

1. With a glass cutter, cut 1.5mm square chips from a small, vanity mirror.

2. Remove the lacquer that covers the reflective coating on the mirror squares by rubbing a little MEK (methyl ethyl ketone) on the back surface with a Q-tip.

3. Cement the mirror squares, back to back, centered on the fiber just above and in contact with the rare-earth magnets.

4. The mirrors and magnets will then rotate as a well-balanced unit.

Attach the silver half-dollar to the brass screw:

1. Thread the brass screw through the brass nut until the brass nut is near the head of the screw.

2. Thread the brass screw through the nylon spacer.

3. Thread the brass screw through the hole in the rear glass wall.

4. Glue the end of the brass screw to the center of the silver half-dollar so that the nut and nylon spacer are on one side of the rear glass wall, and the silver half-dollar is on the other side.

5. Let the glue set.

Complete the glass enclosure:

1. Encase the sensor and silver half-dollar by gluing the spacers of the two glass walls together.

2. When the glue sets, seal off the open sides with electrical tape to protect the magnetometer from air currents.

3. Mount the entire assembly vertically to a smooth, flat base.

4. The sensor is now an accurate compass. As you walk around, the magnets should align to magnetic north and display little oscillation.

Nulling Magnet Assembly:

1. Attach four doughnut shaped magnet side by side to a small piece of glass or wood using silicone cement (you’ll need to use small clams to hold them in place against their mutual magnetic repulsion until the glue sets.

2. Turn the array of doughnut magnets upright and cement it to a free-standing base so that the center of the assembly aligns with the rare-earth magnets in the sensor.

3. As the nulling magnets are brought close to the magnetometer (approximately 30 centimeters), the sensor will begin to wobble and then rotate quite freely when the combined forces of the earth’s magnetic field and those of the doughnut magnets almost cancel each other our. The period of the oscillations should lengthen to a second or more when all the forces are nearly balanced. In this configuration the magnetometer will be the most sensitive.

The Laser:

1. Position the laser so that the beam shines through the glass case and bounces off the mirror and onto a distant wall. Ripples in the earth’s magnetic field will show up as deflections of the beam.

This completes the first part of the Torsion Magnetometer Assembly. We can detect the laser beam on a photo-cell or Solar Cell, amplify the signal, and output to a speaker for a sound work that sonifies geomagnetic activity. We will continue next week by adding Helmholtz coils and control circuitry so that changes in the geomagnetic field of the earth will be digitized and fed into a desktop PC. Our guide will be the Diverdi article listed on the main page of the syllabus.

What we need for next week:

Electrical Components
Coil stuff

With this in mind, here is:

How to get electrical components for free:

Google the part # - the FULL part #, as in LM342
Make sure you know what package you want to work with, e.g., DIP is “Dual-inline-Package”, meaning that the silicon has legs that you can solder onto. Many other packages exist, most of which are Surface Mount – which are very hard to solder by hand.
Select a top link – as those are usually to the manufacturer.

Here is something great about National’s site

Scroll down to see if they offer Free Samples.
Usually free samples include from shipping, if the company doesn’t offer free shipping, try another company.

Chip designs, like any other patent-able invention, eventually become public domain. This means that any chip manufacturer can produce them.

In the very near future we are going to be digitizing the output from our sensors and creating some visualizations/sonifications/???tions with them. In order to do this we will have to use an Analog to Digital Converter. The higher the sampling rate, the better the resolution, so I will use the ADS7825P. These are awesome chips that have four bi-polar, differential inputs that can accept analog voltages in the +/-10V range. This is very useful! And each of you can get one for free (they retail for $29.95 each)!

As if all of the above wasn’t enough. We will also be creating custom PCB’s for this project using: expressPCB – please download the software and look it over before next class. We will be taking the schematic, drawing it in expressSCH, then linking it to expressPCB and ordering.

Assignment:

Whoever is dealing with the circuit needs to create a parts/price list, and
make sure that all the stuff we need is here.

IMPORTANT: we will need a socket for the LM324.

Each of you should order some free chips ASAP so that they are here.
Can someone look into 3-D plotting in Processing? We will be inputting a serial stream of data that will be composed of between three and eight sensor readings. We want to map each stream to a specific axis to create a 3-D plot. Ideally, we want to have live video feed underneath the 3-D plotting.