Detect red 'hand'

[Tor Tveitane]

Hi,

I have an 'analog' water flow meter with small red rotating hands against a
white background. The hands are about 0,5mm thick. I need to make a
detector which senses the rotation of one of these hands. I.e. detect the
passing of a tiny red 'needle' against a white background. The detector can
be mounted around a half inch above the white background.

What is the best solution to solve this task? A non-red small laser diode
and a narrow-beam sensitive photodetector?

Thanks for hints and tips on this issue

regards

Tor

 

[Ian Stirling]

Tor Tveitane <epost@bushsoft.com> wrote:

Hi,

I have an 'analog' water flow meter with small red rotating hands against a
white background. The hands are about 0,5mm thick. I need to make a
detector which senses the rotation of one of these hands. I.e. detect the
passing of a tiny red 'needle' against a white background. The detector can
be mounted around a half inch above the white background.

What is the best solution to solve this task? A non-red small laser diode
and a narrow-beam sensitive photodetector?

Thanks for hints and tips on this issue

Laser is a bit of an overkill.
I'd start by taking a black cup.
Now, put a bright blue or green LED in pointing at the dial.
And end up with a couple of small photodiodes with integrated lenses, and
'narrow beam' pattern (in black tubes to prevent light scatter).
Now, point them at two adjacent points on the dial, not shadowing.
Job done.

 

[Joerg]

Hi Tor,

A really simple method might be to buy a very cheap webcam and get some
simple imaging software that sometimes even comes with it. Train it on
that rotating hand and do a pattern detection. Of course this requires
that a PC is running which may not always be so practical.

A pair of infrared diodes (rec and xmit) well shielded from each other
might also do to detect revolutions if that is the goal. Such as those
used in TV remotes.

Regards, Joerg

http://www.analogconsultants.com

 

[Rich Grise]

On Fri, 29 Oct 2004 20:21:18 +0000, Joerg wrote:

Hi Tor,

A really simple method might be to buy a very cheap webcam and get some
simple imaging software that sometimes even comes with it. Train it on
that rotating hand and do a pattern detection. Of course this requires
that a PC is running which may not always be so practical.

A pair of infrared diodes (rec and xmit) well shielded from each other
might also do to detect revolutions if that is the goal. Such as those
used in TV remotes.

It sounds to me like he's looking for something from here:
http://www.google.com/search?q=reflective+optical+sensor

I've seen them before - check a dead printer.

If he can hack the meter, he could use an optical interruptor.

Cheers!
Rich

 

[Ken Moffett]

Tor Tveitane wrote:

Hi,

I have an 'analog' water flow meter with small red rotating hands against a
white background. The hands are about 0,5mm thick. I need to make a
detector which senses the rotation of one of these hands. I.e. detect the
passing of a tiny red 'needle' against a white background. The detector can
be mounted around a half inch above the white background.

What is the best solution to solve this task? A non-red small laser diode
and a narrow-beam sensitive photodetector?

Thanks for hints and tips on this issue

regards

Tor Check OPTEK's reflective switches/sensors OPB704, OPB706A, OPB707A,
OPB606A, OPB716, and OPB770T in Digikey. One of these, with Spehro

Pefhany's mask, might do it. The white background and the contrasting
red pointer, and the narrow field of view, might give you enough signal
level shift to detect the pointer passage.

 

[Joerg]

Hi Rich,

It sounds to me like he's looking for something from here:
http://www.google.com/search?q=reflective+optical+sensor


That's the high-tech solution.

If he can hack the meter, he could use an optical interruptor.


Hacking a water meter can lead to serious consequences. At first some

leaking, then flooding, then a chat with a judge ;-)

Regards, Joerg

http://www.analogconsultants.com

 

[Rich Grise]

On Fri, 29 Oct 2004 20:54:00 +0000, Joerg wrote:

Hi Rich,

It sounds to me like he's looking for something from here:
http://www.google.com/search?q=reflective+optical+sensor


That's the high-tech solution.

This might look high-tech at first glance,
http://www.newark.com/NewarkWebCommerce/newark/en_US/endecaSearch/partDetail.jsp?SKU=95B4223&N=0
but look at the part that this page is actually about, at $1.03 per:
http://www.vishay.com/docs/83751/83751.pdf

And this is only one example - they come in various form factors.

You point it at the spot you want to detect the needle at, and put
a comparator on the transistor output. Voila!

Cheers!
Rich

 

[Frithiof Andreas Jensen]

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:1Dxgd.36577$QJ3.22105@newssvr21.news.prodigy.com...

A really simple method might be to buy a very cheap webcam and get some
simple imaging software that sometimes even comes with it.

Actually -- One can get all that in one package: The motion sensor device in
an optical mouse.

(They seem to have more uses, I read an IEEE spectrum article about NASA
using the very same commercial chips, with different optics, to provide
motion feedback for a robot plane).

 

[Joerg]

Hi Frithiof Andreas,

Actually -- One can get all that in one package: The motion sensor device in
an optical mouse.


That, plus a lens, sounds like a great idea. It would not surprise me if

it would be cheaper to buy optical mice and part them out instead of
purchasing the sensors alone, at least for prototype quantities.

Regards, Joerg

http://www.analogconsultants.com

 

[Spehro Pefhany]

On Fri, 29 Oct 2004 08:34:44 +0200, the renowned "Tor Tveitane"
<epost@bushsoft.com> wrote:

Hi,

I have an 'analog' water flow meter with small red rotating hands against a
white background. The hands are about 0,5mm thick. I need to make a
detector which senses the rotation of one of these hands. I.e. detect the
passing of a tiny red 'needle' against a white background. The detector can
be mounted around a half inch above the white background.

What is the best solution to solve this task? A non-red small laser diode
and a narrow-beam sensitive photodetector?

Thanks for hints and tips on this issue

regards

Tor

Maybe shield the face from light. Illuminate with one or two
super-bright green LEDs. Put a phototransistor in a tube well behind
(a multiple of the distance to the needle) a mask with a 0.5mm line
width.

Here's artwork for a suitable mask: http://www.speff.com/mask.pdf
Print on film (preferably) or transparency.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com