» Fixtures "The prefix" Tester of magnets "

The prefix "Tester of magnets"

Hello, dear authors, journalists, readers!

This prefix to the multimeter allows you to determine and compare the strength of magnets, the direction of the magnetic field and the shielding (anti-magnetic) effect of various materials.

Applied Tools:

Wood hacksaw
Hacksaw
Drill with a drill with a diameter of 1 mm
Flat file
The file is flat
Nippers
Pliers
Round pliers
Ruler
Marker
Soldering Iron 25 W
Scissors
Sandpaper
Brush

Materials:

Pine beam
Fiberglass
Scotch
Glue
Solder
Rosin
Solvent 646
Rags
Stranded wire
Reed switches
Plastic line

I used a magnetically controlled sealed contact (reed switch) as a magnetic field sensor, and a multimeter as an indicator.

Is that simple? Let's take a closer look at my homemade. Here is her schematic diagram.

Two parallel-connected reed switches are connected to a multimeter turned on in semiconductor continuity mode.

This mode can be found even in the most inexpensive multimeters. In this mode, the device gives a sound signal at a low resistance of the measured circuit, of course, and when it is closed.

The reed switch used in this homemade product has a normally open contact. This means that the contact is open in the absence of a magnetic field. When the field appears, the contact closes.

Why did I use two reed switches and not one? The fact is that the reed switch, due to its design features, reacts to the magnetic field differently from different directions. Good, the more reed switches connected in parallel, the better. But, guided by my experience with magnetically controlled contacts and the principle of reasonable sufficiency, I settled on two, placing them in a special way.

This photo shows the details from which I assembled this homemade product. There are very few of them.

Two identical reed switches from burglar alarm sensors, a board, a ruler, wires, a piece of fiberglass.

I placed the reed switches one above the other in the form of the letter X. With this arrangement, when one reed switch stops responding to a magnetic field of a certain direction, the other reed switch starts to respond.

To do this, I cut a small piece of fiberglass, marked out and drilled holes. I inserted the reed switch legs into the holes, bent them on the back of the fiberglass. Glued.

The photo shows that the installation of the reed switches, when bending the terminals, broke off part of the glass case of one of them. However, I checked the serviceability of the reed switch and continued to work. I soldered the reed contacts and the output wires of the console.


I made the ends of the wires in the form of tinned loops, since my multimeter can use crocodile clips on the probes.

I inserted a board with reed switches in a slot made in advance in the board - at the base, adding a little glue for reliability.
The prefix

I glued the ruler.

Fastened the wires with tape.

The prefix is ​​ready.

A small addition. Assembled this design as follows.

I took a pine beam suitable for the width for the manufacture of the base.
I tried on it a ruler, which in the design serves to determine the distance to the magnet under study.
I figured out how much space is needed to fix the board with reed switches and fasten the wires.

Summarized, gave a margin up, marked out and sawed off the desired piece of timber with a hacksaw on a tree.
He made a transverse cut in the timber with a hacksaw and a flat file to install a board with reed switches.
I used a hacksaw to cut a piece of fiberglass that was necessary in size. Marked and drilled four holes in it for the conclusions of the reed switches.

I filed the edges of the bar, cleaned it with sandpaper and covered the bar with dark lacquer for refinement.
Measured two pieces of a multicore installation wire. In the end he stripped the insulation and tinned.
The rest is described above.

Now that the prefix can and how to work with it.

Turn on the multimeter in semiconductor dialing mode. The magnet under investigation is slowly brought closer to the circuit board with reed switches until a sound signal appears. We read the distance to the magnet on a scale (ruler). We write the result in a notebook. We take the magnet back until the sound signal disappears. Expand the magnet on the other side. Repeat the operation as the magnet approaches. We write in the notebook a new result. Similarly, we obtain a lot of data on the effect of a magnet on an object, depending on the position of the magnet relative to a stationary object. Difficult, huh? But it’s clear.

Next, take another magnet and repeat these operations. Now we have the opportunity to compare two magnets in the same conditions.

Now we examine the antimagnetic properties of materials, how much they weaken the effect of the magnetic field. To do this, take any magnet, preferably more powerful. According to the method described above, we determine and record the distance to the magnet at which the signal begins to sound. Without changing the position of the magnet, we remove it on a scale - a ruler until the signal stops. Immediately in front of the reed switches we place the investigated antimagnetic material. The area of ​​the material sample should be such as to completely close the reed switches from the magnet. Zoom in the magnet. When a beep appears, stop. We read and write the result. The distance (result) should decrease. From this we conclude how much this material weakens the magnetic field. This is similar to how materials attenuate radioactive radiation. It was very interesting to compare the properties of tin, brass, permalloy tape, transformer screens and more. Now think about why I used wood for the base of the console, and a plastic ruler for the scale.

Recently, I was engaged in research work "on the influence of a magnetic field on water meters." Thanks to this prefix, I was able to explain the “phenomenon” why a vaunted neodymium magnet cannot stop some counters, but a regular, ferrite one, from the speaker can.

You can make not a prefix, but a functionally complete device. In this case, you can replace the multimeter with just two parts. Battery and "tweeter" with a built-in generator, collecting such a circuit.

A power switch is not required; in the absence of a magnetic field, the circuit consumes nothing.

Or by replacing the multimeter with three parts. Battery, resistor and LED, as in the diagram.

A power switch is also not required, in the absence of a magnetic field, the circuit consumes nothing.

In conclusion, I want to add. Magnets, if they are large, can be approximated from the sides of the console, also read the result on a ruler. For this reason, I took a block rather than a flat board as the basis. The ruler has a scale on both sides, which makes it convenient to work with different positions of the console relative to the experimenter.

I hope this article was interesting and useful to you.

I will be glad to your comments and suggestions.

Sincerely, author.
9.7
9.3
7.7

Add a comment

    • smilesmilesxaxaokdontknowyahoonea
      bossscratchfoolyesyes-yesaggressivesecret
      sorrydancedance2dance3pardonhelpdrinks
      stopfriendsgoodgoodgoodwhistleswoontongue
      smokeclappingcraydeclarederisivedon-t_mentiondownload
      heatirefullaugh1mdameetingmoskingnegative
      not_ipopcornpunishreadscarescaressearch
      tauntthank_youthisto_clueumnikacuteagree
      badbeeeblack_eyeblum3blushboastboredom
      censoredpleasantrysecret2threatenvictoryyusun_bespectacled
      shokrespektlolprevedwelcomekrutoyya_za
      ya_dobryihelperne_huliganne_othodifludbanclose
155 comments
Well and still, magnetic fields - they are heterogeneous. If you turn your round magnet, the force of gravity will already change.
From revision - you need a magnetic circuit through which the magnetic field from the magnet will not go through the air to the sensors.

Then the type of magnet will not play a big role, and the size. The magnetic core of each magnet will take the same area. As a magnetic circuit, I propose a piece of a bolt, but rather a core from a motor of silt, another piece of transformer steel.
Well, maybe, depending on the type, you will have to make corrections.
A huge plus for enthusiasm and a lively brain ... BUT ...

They did not take into account the area of ​​the magnet, on which its power will depend. Also a type of magnet, in neodymium the magnetic fields are short, but very strong, and in the black field long, but weaker. Therefore, your machine will show that the black magnet is stronger than neodymium.

Well, to check the shielding ... maybe, but the magnetic field is shielded only by the materials that absorb it, magnetized (steel, ferromagnets ...)
Author
Well, yes, you can also go to a circle, one of the applications of the console. Why compare magnets. Let's say you have a bunch of magnets. And you do some home-made work where you need a strong magnet. The prefix will help you quickly select it. It is inconvenient to choose with a nail or other magnet, long and inaccurate.
It would be good to use this prefix as a visual aid in the radio circle, to demonstrate the principle of the operation of reed switches. To show that they really react differently to different positions of magnets (magnetic field), and the effect of magnetic screens can be explained. And so, just to compare the strength of magnets, why? These are not transistors that need to be selected in pairs. The tester can only show that it is time to throw out the magnet, or so far it will fit.

We advise you to read:

Hand it for the smartphone ...