Rotary encoder using magnetics / reeds

R

Rob

Guest
I'm playing around with the idea of creating quadrature encoder using:
* magnets.
* a magnetic 'shading' or 'focusing' rotor of some sort.
* reed switches.


My requirements are
* measuring slow rotation so speed is an issue.
* Very low power- hence the reeds.
High value pull ups -->state change --> wake a sleeping micro. I need to
keep the average current consumption in the 10's of microamps range to
extend battery life.

I reckon some robotics buffs or someone else must have done this but heaps
of googling has turned up nothing. I'm trying do this using readily
available materials.

Can anyone point me toward some info on this topic?

many thanks for any ideas guys,
rob
 
Hi Rob,

Dave gave you a few good pointers. Let me add one:

Reeds have a finite life no matter how carefully you select the magnet
strength. There comes a point where you find a loose piece in there.

How about coils? Then the magnets, if the shaft has any reasonable
speed, generate a small current through these coils which could be
detected and that could wake up the controller.

Regards, Joerg

http://www.analogconsultants.com
 
I'm playing around with the idea of creating quadrature encoder using:
* magnets.
* a magnetic 'shading' or 'focusing' rotor of some sort.
* reed switches.
Click to expand...
Hi, a stepper motor will give you quadrature outputs, you could use then
directly if the speed is right. Alternatavly you can dismantle 1, remove the
coils, put your reed switchs in the coil spaces.In theory it should work but
I've never tried it.
 
"Rob" <rdsfal@yahoo.com.au> wrote:

I'm playing around with the idea of creating quadrature encoder using:
* magnets.
* a magnetic 'shading' or 'focusing' rotor of some sort.
* reed switches.


My requirements are
* measuring slow rotation so speed is an issue.
Click to expand...
How slow? or more importantly how fast?

As others have said reeds will fall apart eventually. A pulled up reed will
draw current all the time it is closed.

A opto sensor only needs to be turned on for a few uS to determine its
state. You could wake up from a timer interrupt and check the sensor. The
timer interrupt rate would put a hard limit on the maximum encoder 'rate'
and power consumption would be proportional.

10mA for 5uS at a 10Hz rate averages to 0.5uA how does that compare with a
reed pull up resistor?
 
On Sat, 2 Oct 2004 11:59:48 -0700, "Rob" <rdsfal@yahoo.com.au> wroth:

My requirements are
* measuring slow rotation so speed is an issue.
* Very low power- hence the reeds.
High value pull ups -->state change --> wake a sleeping micro. I need to
keep the average current consumption in the 10's of microamps range to
extend battery life.

rob
Click to expand...
What is the maximum rotation rate? 1 RPM? 1 RPS? 1 RPDay?

Do you need to sense more than one point in every complete revolution?

Jim
 
On Sat, 2 Oct 2004 11:59:48 -0700, "Rob" <rdsfal@yahoo.com.au> wrote:

I'm playing around with the idea of creating quadrature encoder using:
* magnets.
* a magnetic 'shading' or 'focusing' rotor of some sort.
* reed switches.


My requirements are
* measuring slow rotation so speed is an issue.
* Very low power- hence the reeds.
High value pull ups -->state change --> wake a sleeping micro. I need to
keep the average current consumption in the 10's of microamps range to
extend battery life.

I reckon some robotics buffs or someone else must have done this but heaps
of googling has turned up nothing. I'm trying do this using readily
available materials.

Can anyone point me toward some info on this topic?

many thanks for any ideas guys,
rob
Click to expand...

I've had bad experiences as regards reed life, and don't really trust
them. How about using an optical encoder and pulsing the led's now and
then to read out the states? How about a mechanical (contact) encoder?
Or capacitive... that would be fun.

John
 
I've had bad experiences as regards reed life, and don't really trust
them. How about using an optical encoder and pulsing the led's now and
then to read out the states? How about a mechanical (contact) encoder?
Or capacitive... that would be fun.
Click to expand...
Under what conditions?
They are probably the most popular switch in the alarm industry.

You do need to observe ratings for wetting current and maximum voltage/break
current.

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR
 
"Dave VanHorn" <dvanhorn@cedar.net> wrote in message
news:naidnf79z66df8PcRVn-ow@comcast.com...
I've had bad experiences as regards reed life, and don't really trust
them. How about using an optical encoder and pulsing the led's now and
then to read out the states? How about a mechanical (contact) encoder?
Or capacitive... that would be fun.

Under what conditions?
They are probably the most popular switch in the alarm industry.

You do need to observe ratings for wetting current and maximum
voltage/break current.

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR
Click to expand...
Yes and one of the most comon failures in security systems are ruined
switches on doorways.

Charles
 
On Sat, 2 Oct 2004 12:14:55 -0500, "Dave VanHorn" <dvanhorn@cedar.net>
wrote:

I've had bad experiences as regards reed life, and don't really trust
them.

Under what conditions?
Click to expand...
1. Switching low-level signals.

2. Cold switching non-low-level signals.

3. Hot switching non-low-level signals.

John
 
Under what conditions?

1. Switching low-level signals.
Click to expand...
Without wetting current, I'd only use mercury wetted switches in this
application.

2. Cold switching non-low-level signals.
3. Hot switching non-low-level signals.
Click to expand...
Examples?

John
Click to expand...

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR
 
On Sat, 02 Oct 2004 11:39:38 -0700, John Larkin
<jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:

On Sat, 2 Oct 2004 12:14:55 -0500, "Dave VanHorn" <dvanhorn@cedar.net
wrote:


I've had bad experiences as regards reed life, and don't really trust
them.

Under what conditions?

1. Switching low-level signals.

2. Cold switching non-low-level signals.

3. Hot switching non-low-level signals.
Click to expand...
---
<G>

--
John Fields
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top