[EE] Picking omnipolar Hall effect sensors

[mike brown]

Hello all,

I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.



The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¾” in diameter and about 3/32” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.



I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.

[Bob Blick]

Use two hall effect sensors mounted back-to-back?

________________________________________
From: piclist-***@mit.edu <piclist-***@mit.edu> on behalf of mike brown
Sent: Saturday, October 28, 2017 10:48 AM
To: Microcontroller discussion list - Public.
Subject: [EE] Picking omnipolar Hall effect sensors

Hello all,

I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.



The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¾” in diameter and about 3/32” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.



I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.

[mike brown]

Thanks to everyone for your input. This is what I really like about
asking here, five responses every one is unique.

Bob - I have to admit that never crossed my mind, but it sounds like a
workable solution. I'd probably put them next to each other, one face up
and the other face down. I'll do some experiments with that.

Mike B - I'm glad it's not just me. Good idea on making a ring of reed
switches. Since these will be embedded in tiles, positioning accuracy
will not be an issue.

Clint - The magnet must be sensed without regards to which pole is
presented, though I suppose I could make a sorter circuit for the guy
building the tiles so that he always puts the magnets in the right way.

Rodolfo - I'm not sure I understand what you mean.

Russell - Thanks, I'll check availability of those here. So, the "brain"
in the Hall sensor sleeps most of the time and polls for a field every
100mS or so. Does it keep the output latched until the next poll cycle,
or would I be seeing 100uS pulses every 100mS? The prop will likely be
battery operated and sleeping for most of the time. I'd planned to just
poll all of the sensors after sleep and logically AND them all. If there
are pulses, I could manage but it would add a lot of complication
(relatively speaking) to the code.

[RussellMc]

Post by mike brown
Russell - Thanks, I'll check availability of those here.

​Where is "here"?
​

Post by mike brown
So, the "brain"
in the Hall sensor sleeps most of the time and polls for a field every
100mS or so. Does it keep the output latched until the next poll cycle,
or would I be seeing 100uS pulses every 100mS?

​Output is latched and changes if required at wakeup time.
See diagrams in datasheet at bottom of pages 2 & 3.


​Datasheet

https://www.diodes.com/assets/Datasheets/AH180.pdf


​Russell

​

<https://www.diodes.com/assets/Datasheets/AH180.pdf>

Post by mike brown
The prop will likely be
battery operated and sleeping for most of the time. I'd planned to just
poll all of the sensors after sleep and logically AND them all. If there
are pulses, I could manage but it would add a lot of complication
(relatively speaking) to the code.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Mike B]

A reed relay not closing when centered on the magnet is a common problem.
One solution is to surround the target with 4 sensors wired in series. All
need to be closed for a valid detection.

On Oct 28, 2017 12:50, "mike brown" <***@n5qmg.com> wrote:

Hello all,

I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.



The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¾” in diameter and about 3/32” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.



I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Rodolfo]

XY coils to sense magnetic field when neo magnets are over two coils, one
over the other? Like a magnetic digitizer?

Rodolfo

Post by Mike B
A reed relay not closing when centered on the magnet is a common problem.
One solution is to surround the target with 4 sensors wired in series. All
need to be closed for a valid detection.
Hello all,
I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.
The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¾” in diameter and about 3/32” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.
I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Clint Jay]

Why unipolar? I had a cute little circuit using an opamp and a comparator
which would light one led or another depending on magnetic polarity, would
be trivial to diode OR the outputs together.

Post by mike brown
Hello all,
I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.
The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¾” in diameter and about 3/32” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.
I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

Post by mike brown
Hello all,
I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive.

AH180
Much used in Asia.

Bipolar (2 Hall plates) with chopper stabilisation.
8/24 uA average typ/max
2.5 - 5 VDC
~= +/- 5 mT operate
~= +/- 1 mT release

Low current achieved by using sampling - 0.1% on time.
100 +/- 10 ms sampling interval - so MAY not be suitable in your application

Diodes inc make them so they must be good :-)

https://www.diodes.com/assets/Datasheets/AH180.pdf


Digikey 50,000+ in stock
$US0.54/1
$US0.18x / 3000

https://www.digikey.com/products/en/sensors-transducers/magnetic-sensors-switches-solid-state/565?k=ah180

Russell

​

[Justin Richards]

I had a quick play with cheap switching hall effect sensors from ebay.
They had similar sensitivity to a reed. I was not aware they may respond
differently to the poles. Will test again when I get home. This may help
solve a problem I have been having to uniquely identify 4 out of 8 foam
balls.

I use reeds end on and find the are very sensitive used this way. The
effective pattern is definitely a figure 8. Typically I cut the legs
short, bend one leg all the way back (with needle nose pliers supporting
the leg at the glass end because as you say they are very delicate) so
effectively forming a U shape with one leg running beside the glass. The
shortened lead are then soldered.

To install a small hole is drilled underneath the table or prop. Push reed
into the hole and pot with hot glue encapsulating the solder joints etc.
They are very robust once potted and you can get the sensitive end of the
reed very close to the surface with no visible signs (unless you happen to
drill all the way thru).

I have recently experimented with reeds in an attempt to improve the
precision. I tried various things to shield one end so it would only be
sensitive to a magnets approach at the other end but it proved to be
futile. It maybe doable if I had a lathe etc. This was to help solve the
uniquely identify 4 out of 8 small foam balls mentioned above.

Justin

Post by mike brown
Hello all,
I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.
The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¾” in diameter and about 3/32” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.
I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

Post by Justin Richards
I use reeds end on and find the are very sensitive used this way. The
effective pattern is definitely a figure 8. Typically I cut the legs
short, bend one leg all the way back (with needle nose pliers supporting
the leg at the glass end because as you say they are very delicate) so
effectively forming a U shape with one leg running beside the glass. The
shortened lead are then soldered.

1. As well as clamping the wire between reed and bend when forming the
wire.​ you should also

2. *When cutting reed-leads you should (must) clamp the wire between reed
and cutting point *so the shock from the cutting is not transmitted into
the reed assembly. As unlikely as it may seem, such shocks are said to
damage the wire/glass seal and may lead to 'walking wounded' early field
failures.

I designed a product that used a reed switch and 100,000+ were made. We
very carefully specified bending and cutting procedures and jigs were made
to ensure that the assemblers would have no difficulty in meeting the
requirements. We had very little trouble with the reeds - suggesting either
that they are more robust than claimed, or that our procedures were
adequate.

I found that the reed survivability under high-g impact conditions appeared
to be significantly better than the manufacturer's claims.
The product (a small solar flashlight) could be subject to c=virtually any
degree of abuse without (apparent) reed damage.

A big advantage of reed switches over even sampling Hall sensors is the
essentially zero quiescent current.
1 uA drain requires about 9 mAh of battery capacity per year.
So eg the 8 / 24 uA typical/max drain of the AH180 Hall sensor requires 70
/ 200 mAh / year to power.

Russell

[Justin Richards]

Post by RussellMc
2. *When cutting reed-leads you should (must) clamp the wire between reed
and cutting point *so the shock from the cutting is not transmitted into

Great tip. Will do so from now on.

I designed a product that used a reed switch and 100,000+ were made. We

Post by RussellMc
very carefully specified bending and cutting procedures and jigs were made

100,000+ That is a lot of reed switches/solar flashlights. What was the
application if you care to elaborate.

Never considered the quiescent current. Good point.

Justin

[RussellMc]

Post by RussellMc
​>​
I designed a product that used a reed switch and 100,000+ were made. We

Post by RussellMc
very carefully specified bending and cutting procedures and jigs were

made
100,000+ That is a lot of reed switches/solar flashlights.

​Total numbers uncertain without checking.
30,000 in Haiti. Various 10,000 - 20,000 batches elsewhere.

Post by RussellMc
What was the
application if you care to elaborate.
​Small flashlight in photos here

http://public.fotki.com/RussellMc/atw/bogo/sl2africa01/?view=roll#19

Roll with cursor keys or PgUp/PgDown or ...
"Enter" drops into a more normal / slower display mode.

Small lights have reed switch.
100% waterproof (up to some thousands of psi :-) ).
No quiescent current.
Impact limit (notionally)(the carabiner breaks off when thrown 15' in air
to land on concrete Reed survives.)

Larger lights have mechanical switch ​with attendant sealing issues.
Zero quiescent current.

Hall
​can be waterproof but has quiescent current issues.


​ Russell​
.

[Dwayne Reid]

I really Bob Blick's answer.

But a different approach could use a linear Hall
Effect sensor. No magnetic field has the output
sitting at half the rail voltage. Increasing
magnetic field has the output move away from the
half-rail value: towards either Vdd or Vss, depending upon polarity.

Seems like it would be trivial to detect the field.

dwayne

Content-Transfer-Encoding: base64Hello all,
I need some cheap, easily available Hall effect sensors that are switches,
omnipolar and sensitive. I’m doing a game room project and need to sense
some magnets that are embedded within “tiles” made from three pieces of
thin plywood. The sensors will be under a table top approximately ¾” thick
(18mm). I can carve away some of the wood to get the sensor near the
surface. Normally I would use reed switches, but I just can’t stand
working with their frailty.
The builder of the tiles can’t be bothered with sorting the polarity of the
neodymium magnets that are ¸ 'H[ˆX[Y]\ˆ[™X›Ý]ËÌ̀” thick. These are
fairly powerful magnets that can activate a reed switch several inches
away, but not reliably when centered precisely over the reed switch. I’m
guessing that has to do with the flux field being flattened by the large
diameter and relative thinness of the magnets.
I can get qty 10 A3144 sensors for under $5 including fast shipping, but
these are unipolar. Is there another part number, in this price range,
that would be at least as sensitive and in a omnipolar form? I don’t want
latches or linear devices.

--
Dwayne Reid <***@planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
780-489-3199 voice 780-487-6397 fax [phone] Toll Free
www.trinity-electronics.com
Custom Electronics Design and Manufacturing

[RussellMc]

A number of possibly useful files (+ some less relevant but 'interesting'
Hall effect related files) that I have downloaded are (perhaps temporarily)
in a Dropbox folder here

http://bit.ly/rusl_hall_sensors

An index of the files is in INDEX.TXT

Post by Dwayne Reid
But a different approach could use a linear Hall
Effect sensor. No magnetic field has the output
sitting at half the rail voltage. Increasing
magnetic field has the output move away from the
half-rail value: towards either Vdd or Vss, depending upon polarity.

​An advantage of a linear sensor is that hysteresis can be altered to suit.

The AH180 ​that I mentioned appears to have models with several
operate/release ratios and ranges of field strengths, and each magnet
polarity has somewhat different operate/release characteristics.
Such variations in a linear device could be accommodate by external
circuitry if desired.

It would be worth having a look at *Digikey's AH180 product page
<https://www.digikey.com/products/en/sensors-transducers/magnetic-sensors-switches-solid-state/565?k=ah180>*,
regardless of which Hall sensor you end up using.

My experience with reed switches suggests that they should be easily robust
enough and would easily meet your application.
Reed orientation relative to the moving magnet has significant effect.


Russell

[smplx]

Post by RussellMc
A number of possibly useful files (+ some less relevant but 'interesting'
Hall effect related files) that I have downloaded are (perhaps temporarily)
in a Dropbox folder here
http://bit.ly/rusl_hall_sensors
An index of the files is in INDEX.TXT

Post by Dwayne Reid
But a different approach could use a linear Hall
Effect sensor. No magnetic field has the output
sitting at half the rail voltage. Increasing
magnetic field has the output move away from the
half-rail value: towards either Vdd or Vss, depending upon polarity.

​An advantage of a linear sensor is that hysteresis can be altered to suit.
The AH180 ​that I mentioned appears to have models with several
operate/release ratios and ranges of field strengths, and each magnet
polarity has somewhat different operate/release characteristics.
Such variations in a linear device could be accommodate by external
circuitry if desired.
It would be worth having a look at *Digikey's AH180 product page
<https://www.digikey.com/products/en/sensors-transducers/magnetic-sensors-switches-solid-state/565?k=ah180>*,
regardless of which Hall sensor you end up using.
My experience with reed switches suggests that they should be easily robust
enough and would easily meet your application.
Reed orientation relative to the moving magnet has significant effect.

Just curious but would the operator not experience feedback on the moving
magnet? And if so could this not be used by the operator to locate hidden
reed switches?

Regards
Sergio Masci

[RussellMc]

Post by smplx
Just curious but would the operator not experience feedback on the moving
magnet? And if so could this not be used by the operator to locate hidden
reed switches?

Using force, potentially yes. But the force is extremely small. ​
Scattering a few iron nails around the place would mask the ability.

Using sound, maybe. The reeds make a small but distinctive sound when they
operate.
​Unlikely to be audible through typical material coverings - unless you had
eg a conduction microphone ​adjacent. Adding FFT and filtering may help.

M61 is liable to be a surer way of getting out (but less compact).


Russell