EEVblog Electronics Community Forum
Electronics => Open Source Hardware => Topic started by: Bounty on September 26, 2013, 11:33:13 pm
-
Hello evryone, I would like to build an OpenHardware project here:
a DIY Magnetometer Gussmeter using an Hard Disk Head.
https://en.wikipedia.org/wiki/Magnetometer (https://en.wikipedia.org/wiki/Magnetometer)
https://en.wikipedia.org/wiki/Disk_read-and-write_head (https://en.wikipedia.org/wiki/Disk_read-and-write_head)
(https://upload.wikimedia.org/wikipedia/commons/thumb/e/ec/Seagate_ST33232A_hard_disk_head_and_platters_detail.jpg/800px-Seagate_ST33232A_hard_disk_head_and_platters_detail.jpg)
(https://upload.wikimedia.org/wikipedia/commons/thumb/7/75/Hard_disk_head.jpg/660px-Hard_disk_head.jpg)
(https://upload.wikimedia.org/wikipedia/commons/thumb/5/54/Kopftraeger_WD2500JS-00MHB0.jpg/800px-Kopftraeger_WD2500JS-00MHB0.jpg)
Why ? Because I need a magnetometer to measure from 1 to 20 Tesla to be put in a gap of 0.5 - 0.25 millimeters
like in the gap of the following picture:
(http://i.stack.imgur.com/9Ak7B.png)
I have asked for the same help here:
http://forums.hackaday.com/viewtopic.php?f=4&t=3751 (http://forums.hackaday.com/viewtopic.php?f=4&t=3751)
Thank you in advance
-
Just buy one of the shell, with I2C or SPI interface, that will be the easiest way to do it
If you make your own, how are you going to make sure you are reading the correct Tesla ?
Remember you need to compensate for earth magnetic field as well on any case
David.
-
Just buy one of the shell, with I2C or SPI interface, that will be the easiest way to do it
I need a magnetometer to measure from 1 to 20 Tesla to be put in a gap of 0.5 - 0.25 millimeters
The problem is the gap size, most Hall sensors seem to have a package with a thickness of 1 millimeter
which is too thick. I'will look better on the internet later today.
Thank you for the following advices.
If you make your own, how are you going to make sure you are reading the correct Tesla ?
Remember you need to compensate for earth magnetic field as well on any case
David.
-
The following idea occurred to me after reading your last post:
File down the ceramic/plastic package from an off-the-shelf sensor until it fits? ^^'
Might be a s**t idea, but it might work.
-
Just buy one of the shell, with I2C or SPI interface, that will be the easiest way to do it
If you make your own, how are you going to make sure you are reading the correct Tesla ?
Remember you need to compensate for earth magnetic field as well on any case
David.
Idea 2: use an off-the-shelf sensor to calibrate/characterise your hacked one?
-
Remember you need to compensate for earth magnetic field as well on any case
The field ranges between approximately 25,000 and 65,000 nT (0.25–0.65 G).
http://en.wikipedia.org/wiki/Earth%27s_magnetic_field#Description (http://en.wikipedia.org/wiki/Earth%27s_magnetic_field#Description)
I was just wake up, I didn't noticed that earth magnetic field ranging from 25*10^-6 Tesla to 65*10^-6 Tesla
won't bother me at all when, triyng to measure something between 1 Tesla and 20 Tesla
Bye
-
Have you considered other means of measuring the magnetic field - like proton resonance, or the Faraday/Kerr effect ?
Are you trying to measure the whole field or are you trying to map the field locally?
-
Hi Bounty
For starters, the current harddisk heads use use Giant magnetoresistance sensors. I don't think you will be able to get any sort of linear measurement on such a wide range with these, since they are designed to give a digital (0 or 1) response for very small values of magnetic flux density.
There are several different ways to measure a magnetic field that are comonly employed as we speak.
Since I'm guessing you are looking for an analog response, you will want to try building your own Hall effect Sensor or Magnetoresistance(not the giant kind) sensor.
I say "build your own" because it doesn't look like you will find off the shelf components that small.
What kind of environments do you work in? Temperature variations can have quite an effect on these types of measurements and off the shelf parts are usually compensated for that...
-
For starters, the current harddisk heads use use Giant magnetoresistance sensors. I don't think you will be able to get any sort of linear measurement on such a wide range with these, since they are designed to give a digital (0 or 1) response for very small values of magnetic flux density.
Fail....
NO they are not, they are materials that CHANGE their electrical resistance based on a magnetic field, there are no '1 & 0' involved until much...much... later in the process.
Modern disk-drives DO NOT record values of One's and Zero's, and in many cases the actual recovery of values is based on PROBABILITY.,disk-drives are a powerful collection of analog signal and DSP hardware combined together with software, find the RIGHT place to tap in and you can recover the data you are interested in.
-
For starters, the current harddisk heads use use Giant magnetoresistance sensors. I don't think you will be able to get any sort of linear measurement on such a wide range with these, since they are designed to give a digital (0 or 1) response for very small values of magnetic flux density.
Fail....
NO they are not, they are materials that CHANGE their electrical resistance based on a magnetic field, there are no '1 & 0' involved until much...much... later in the process.
Modern disk-drives DO NOT record values of One's and Zero's, and in many cases the actual recovery of values is based on PROBABILITY.,disk-drives are a powerful collection of analog signal and DSP hardware combined together with software, find the RIGHT place to tap in and you can recover the data you are interested in.
Doesnt mean they are linear or even non-linear upto 20T though ;)
http://en.wikipedia.org/wiki/Giant_magnetoresistance (http://en.wikipedia.org/wiki/Giant_magnetoresistance)
shows a definite saturation, and this (http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&ved=0CDwQFjAC&url=http%3A%2F%2Fusers.ece.cmu.edu%2F~jzhu%2Fpublications%2Fnewheights.pdf&ei=iwlgUo-FI5DD4AONtYGoDg&usg=AFQjCNHFt8LjB6nzsGOHmjauPWLwst88pQ&bvm=bv.54176721,d.dmg&cad=rja) seems to indicate that the write heads have fields upto ~2.45T. I wouldnt expect the read heads to take 10x that an not be saturated.
-
Monotonic is all you need, you can use software to linearise the response; you just need a magnetic references to calibrate against (maybe a simple air coil driven by a current source?)
-
What about a piece of foil as a homemade Hall sensor? Sensitivity would be poor but you are working with a large magnetic field.
-
What ever probe you make you could calibrate it with a Helmholz coil made out of some pvc drain pipe, drilling a hole in the pipe between the two coils to put the probe in.
http://en.wikipedia.org/wiki/Helmholtz_coil (http://en.wikipedia.org/wiki/Helmholtz_coil)
-
What about a piece of foil as a homemade Hall sensor? Sensitivity would be poor but you are working with a large magnetic field.
Yes Hall Effect is the only what to do this, build your own from some foil material (you would get bad sensitivity by using metal, but for these very high fields it might work).
You can get an old hard-drive, the type before magneto resistance, where they used small coils in the tip, not sure if you can use it to measure static magnetic fields though.
How did you get your hands on a 20 Tesla magnet by the way? What is the pole-piece made from, Holmium or something similar exotic i can imagine? :)