How to measure contactlessly position of a spinning miniature cylindrical magnet?

Can you paint marks on it?

Jerry

What's the minimum speed at which you need to monitor the position?

Can you use a pickup coil to make an AC generator and watch the phase?

Are there fast enough hall effect devices, that you could use one or several of and interpolate position under an assumption of no abrupt changes in speed?

Use two sensors a Hall effect sensor for angular data and use a photo sensor and a spot of white paint to index reference the rotor. I have seen Hall Sensors used to monitor Centrifuges to 36,000 RPM !

Yukio YANO

---

1. Is the rotor spinning about its cylindical axis or radially, somewhere along its length?

1. What is the strength of the rotor's magnetic field?

2. How closely can the sensor approach the spinning cylinder?

1. I don't understand the +/-45° requirement. Do you mean that if, in one instance, the sensor is located +45° off axis and, in the next it's located -45° off axis it should still report the absolute position of the rotor with respect to some arbitrary reference?

2. Assuming that you want to know the angular position of the rotor relative to some reference, how accurately do you need that position to be known?

-- John Fields Professional Circuit Designer

What's making it spin? Can you monitor that or do you expect some slip?

Modern brushless DC permanant magnet motor controllers monitor the position of the rotor using unpowered windings as a sensor coil.

If it's in a motor, use the windings. If not, use a hall effect sensor. Phase will tell you angle; intensity will tell you distance. If you don't know distance in 2 dimensions, you can use a second hall sensor. If you don't know alignment, you can use another hall sensor. A BASIC stamp can do any math you need if you don't have a computer connected.

Hello Jerry,

No I can't as this rotor is a part of an implanted biosensor.

Typically, the measurement is performed between from 300 to 1 Hz.

The rotor spins around its cylindrical axis.

The rotor is made of SmCo and the typical energy product is around 200 kJ/m3.

No, the capsule enclosing the rotor is implanted and after implnattion in the body, it is not supposed to move. The question is the tolerance of the position sensor to a misalignment between the rotor and the sensor.

It would be great if the position sensor output could toggle every

The rotor is excited by an external rotating magnetic field generator composed of a 3-Phi coil and 3 PWM generators.

...

...

Implanted, magnetically driven rotor! Implanted into what? Wow!

Jerry

It seems that you can know the position of the magnetic field by measuring the coil currents. A knowledge of the viscosity on the medium that embeds the rotor and the velocity of the field should allow you to calculate the power angle between the field and the magnet. If accelerations are small enough, a static calculation will do. Otherwise, the rotor's inertia will need to be accounted for.

What you have is essentially a synchronous motor. It must either spin at the same rate as the magnetic field or stall. When spinning, the angle between magnet and field must be less than 90 degrees.

Jerry

I think Jerry's on to a good idea. If you're using external coils to excite it then you should have everything you need to measure it as well. Whether there are other objects within the fields which may interfere with the measurements may be an issue, but you may be able to calibrate or sense those and work around them.

Depending on the amount of power involved I think this may preclude the use of a Hall effect sensor, anyway, which I think was the next best idea. If this thing is small, though, and you can't get closer than 10mm a Hall effect device may not have worked, anyway.

Eric Jacobsen Minister of Algorithms, Intel Corp. My opinions may not be Intel's opinions.

If the rotor is a two pole magnet, perhaps you could surround the rotor with say 64 hall effect sensors. Scan and convert the analog value of each sensor many times a second and "plot" a graph. Do averaging to improve S/N ratio and curve fitting to look for the maximium and minimium flux positions (indicating where the north and south poles are).

However I think I would investigate using something else in the rotor - perhaps a directional antenna/coil that's externally excited? It's usually better to start with an AC signal rather than a DC signal if you think you are going to have S/N ratio problems.

with

sensor

I forgot to add...

You will need to deal with noise produced by whatever is spinning the rotor. Hope you don't have an electric motor too close to it.

Questions:

- The position measurement is relative to what? Displacement between the cylinder and it's container?

- Which axis do you want to measure displacement along, the axis of rotation or either of the 2-axis at 90 deg to axis of rotation? Or some angular displacement from the rotation axis (i.e. precession).

- Is the container transparent (i.e. could you use a light pipe combined with a light source and light sensor)?

- can you post a diagram online?

---------------------- I think you are on to the best bet so far. However, the problem of the power angle is still there and there seems to be a demand for a position reading in the order of 5 degrees. If the motor is always lightly loaded then ignoring the power angle may be sufficient. It does appear that the problem is still not well specified beyond 1-300Hz and implanted as well as the type of drive. What is missing is the purpose of the device, the location of the device, the necessity for accurate position sensing and all in all the minimum satisfactory criteria. Depending on these factors, this may well be something for a biomedical engineering (assuming implantation is in a carbon based life form) grad student to sink his/her teeth into. There are places which have good biomedical engineering programs - the University of Alberta, Edmonton, Alberta, Canada has (or at least had) such a program. With due respect to this group, you need someone with the necessary skills who can spend time and effort on this problem. Time and effort on the definition of the problem is also required and this has, so far, been minimal.

What's more, if the plane of the rotating cylinder is not normal to the axis of the rotating field, it will lope like a canted Hooke coupling (the standard -- not CV -- universal joint).

Jerry