When Hall Effect Sensors aren't sensitive enough...

Thread Starter

Holes Flow

Joined Oct 26, 2017
14
All i need to do is check for PM (neodymium) proximity/existence at as much as 12" away.

I've tried linear bipolar HEs, and i get 1-2" max. Digging outside my familiarity zone, I've found 3axis magnetometers. But at their price point i need to know which specs to look for to ensure that level of sensitivity.

Adding to this, the neo magnets arent huge, some as small as 5mm.

Mostly for a feeding system tho avoid one piggy cat from eating all the "snack throughout the day" cat's food, i habe other ideas of i can get a group of parts together that work well together...

TIA
 

Thread Starter

Holes Flow

Joined Oct 26, 2017
14
Is it actually a cat you want to detect 12" away?
That might be easier to solve.
Funny! But like i said, i have plans for a PM/sensor pairing, and this is an ideal. Rube Goldberg PoC for that.

So what should one use to detect the presence [linearly] of small PMs? AmI barking yup the wrong tree with magnetometers?

Thx!
 

AlbertHall

Joined Jun 4, 2014
12,345
See if you can work out how the field from the magnet compares with the Earth's magnetic field. Whatever you use it is going to have decide what magnetism is due to the magnet and what to the Earth.
 

Thread Starter

Holes Flow

Joined Oct 26, 2017
14
See if you can work out how the field from the magnet compares with the Earth's magnetic field. Whatever you use it is going to have decide what magnetism is due to the magnet and what to the Earth.
Albert,
Thanks for writing. I can calibrate for earth's magnetic field in code, and with a 3-axis accelerometer keep track of the vector of it, then anything in a 3-axis magnetometer that makes it past that should be something detected (is my theory)...

Since the idea is to sense a neo PM at 12" distant, and then fire up some EM coils to aim at & grab it, as long as the sensor can detect linearly, I can use multiple sensors to determine rough vector to PM. The real roadblock has been in being able to sense a PM from 12". I had high hopes for HE sensors, but they seem to be for detecting at very close range, which is what you'd need in manufacturing.

I asked about magnetometers because they appear to be made in much more sensitive versions, but at their price point I don't want to buy a bunch I'll never use. That is why I was asking about the specs to look for.
 

AlbertHall

Joined Jun 4, 2014
12,345
See: https://www.supermagnete.de/eng/faq/How-do-you-calculate-the-magnetic-flux-density
Neodymium seems to be between 10000 and 15000 Gauss.
For a neodymium cylindrical magnet 1 inch long and 1 inch diameter the field 12 inches away from its pole face would be 0.96 Gauss. The Earth's field is 0.25 to 0.65 Gauss so should be detectable by anything that can measure the Earth's field like an HMC5883.
Play with the attached spreadsheet for your magnet (I picked the shape with the simplest formula).
 

Attachments

DickCappels

Joined Aug 21, 2008
10,152
The engineer behind Apple's first 21" high resolution color CRT (actually CDT) display was concerned that if the monitor were rotated after degaussing there would be beam landing errors that would result in color splotches. His creative solution was to add a fluxgate magnetomiter to the display so that every time the monitor was moved it would initiate a degauss cycle. It worked very well. So well in fact that things like opening a file cabinet anywhere nearby would trigger the circuit. Eventually the feature was dropped.

Fluxgate Magnetometer <== This is a link to a pdf file
 

ebeowulf17

Joined Aug 12, 2014
3,307
Albert,
Thanks for writing. I can calibrate for earth's magnetic field in code, and with a 3-axis accelerometer keep track of the vector of it, then anything in a 3-axis magnetometer that makes it past that should be something detected (is my theory)...

Since the idea is to sense a neo PM at 12" distant, and then fire up some EM coils to aim at & grab it, as long as the sensor can detect linearly, I can use multiple sensors to determine rough vector to PM. The real roadblock has been in being able to sense a PM from 12". I had high hopes for HE sensors, but they seem to be for detecting at very close range, which is what you'd need in manufacturing.

I asked about magnetometers because they appear to be made in much more sensitive versions, but at their price point I don't want to buy a bunch I'll never use. That is why I was asking about the specs to look for.
One issue with using linear hall effect sensors is that direction/polarity matters. If the null plane of the magnet is aimed at the sensor, the sensor won't detect it no matter how close. Were you paying attention to orientation when you tried hall effects before?

Also, which hall effect sensors did you try? They come in many different sensitivity ranges.
 

Thread Starter

Holes Flow

Joined Oct 26, 2017
14
@crutschow,

I thought RFID was super-short distance, but it appears to be able to do distance. I'm now taking your lead to determine if RFID (at any wavelength) can yield a relative, linear signal based on proximity.

The goal for me is to detect proximity in a linear fashion-whether electromagnetic or RFID radio. There isn't a need for precise absolute measurements, but approximately repeatable proximity. If RFID can yield linear readings that increase as an object gets closer, I'm all for it!
 

Thread Starter

Holes Flow

Joined Oct 26, 2017
14
Also, which hall effect sensors did you try? They come in many different sensitivity ranges.
@ebeowulf17 Thanks for writing.

I've tried the 3503s, which ended up being latched, 3144s which were linear. I understand the polarity constraints. Anywhere past a few cm the signal was lost in the noise floor, and I was using an STM32 with a 12-bit ADC... It didn't look promising... :(

If magnetometers are more sensitive in general than HEs, all the better. I have never worked with a magnetometer before, though, but I have a variety coming in the mail that are already broken out on PCBs...
 

ebeowulf17

Joined Aug 12, 2014
3,307
@ebeowulf17 Thanks for writing.

I've tried the 3503s, which ended up being latched, 3144s which were linear. I understand the polarity constraints. Anywhere past a few cm the signal was lost in the noise floor, and I was using an STM32 with a 12-bit ADC... It didn't look promising... :(

If magnetometers are more sensitive in general than HEs, all the better. I have never worked with a magnetometer before, though, but I have a variety coming in the mail that are already broken out on PCBs...
The more I think about it, the more I suspect that other people's comments are right about the earth's magnetic field being the limiting factor. In Albert's example above, a 1" diameter, 1" long magnet produces a field which should be detectable from a foot away, but not with huge amounts of leeway. If you're talking about 5mm magnets, they may just not be strong enough to detect at that distance, regardless of the sensor.

One easy way to check how strong your magnets are, compared to the earth's magnetic field, would be to use a compass (most smartphones have a compass function built in, which you could use if you don't have a real compass.) Just experiment and see how close your magnet has to get to the compass (or phone) before it starts changing the compass reading. That should show you a ballpark estimate of what detection range is possible for any given magnet size.

Either way, I'm sure it'll be fun and informative to play with the new sensors. Who doesn't love playing with magnets?! If love to solve every engineering problem with either magnets or lasers. Doesn't always work out that way, but that's the dream!

In the meantime, the RFID idea sounds promising.
 

DNA Robotics

Joined Jun 13, 2014
647
Just some ideas.

I got a magnetometer chip one time to play with and they said it could be used to detect cars at an intersection.
Another time, curious about vehicle detection loops, they said it was part of a Wheatstone Bridge to be very sensitive.
Vehicle detection
https://en.wikipedia.org/wiki/Induction_loop
Vehicle detection loops, called inductive-loop traffic detectors, can detect vehicles passing or arriving at a certain point, for instance approaching a traffic light or in motorway traffic. An insulated, electrically conducting loop is installed in the pavement. The electronics unit transmits energy into the wire loops at frequencies between 10 kHz to 200 kHz, depending on the model. The inductive-loop system behaves as a tuned electrical circuit in which the loop wire and lead-in cable are the inductive elements. When a vehicle passes over the loop or is stopped within the loop, the vehicle induces eddy currents in the wire loops, which decrease their inductance. The decreased inductance actuates the electronics unit output relay or solid-state optically isolated output, which sends a pulse to the traffic signal controller signifying the passage or presence of a vehicle.
 

Thread Starter

Holes Flow

Joined Oct 26, 2017
14
One easy way to check how strong your magnets are, compared to the earth's magnetic field, would be to use a compass (most smartphones have a compass function built in, which you could use if you don't have a real compass.) Just experiment and see how close your magnet has to get to the compass (or phone) before it starts changing the compass reading. That should show you a ballpark estimate of what detection range is possible for any given magnet size.
To show how daft I am, aren't flux gate compasses different than 'normal' ones? I can't wait to publish out some results!
 
Top