Searching \ for ' [EE] Simple direction sensing' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=simple+direction
Search entire site for: 'Simple direction sensing'.

No exact or substring matches. trying for part
'[PICLIST] [EE] Simple direction sensing'
2000\09\01@053221 by

As a simple project for my local radio club I've designed a PIC
Magic Wand that flashes 7 LEDs leaving a trailing message in the
air. The flashing (ie message) is triggered by motion sensored by a
mercury switch.

Works fine if you wave right to left but moving left to right the
message is mirror imaged (of course).

Any ideas on a simple (and cheap) method of detecting whether
motion is left-right or right-left so I can invert the message as
appropriate. 2 mecury switches mounted in some way so that the
leading edge triggered first seems the obvious solution but how to
actually implement this?

Thanks

Brian
Brian E Jones
Centre for Java Technology
IBM HURSLEY

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
> Any ideas on a simple (and cheap) method of detecting whether
> motion is left-right or right-left so I can invert the message as
> appropriate. 2 mecury switches mounted in some way so that the
> leading edge triggered first seems the obvious solution but how to
> actually implement this?

Not so simple and cheap:  use an accelerometer chip.  Positive and negative
acceleration.  Everyday Practical Electronics recently ran an article using
one of these.  I liked the calibration - tip it up one way = +1g, tip it up
the other = -1g.

Pure conjecture:

Simple and cheap - use 2 mercury switches mounted 90 degrees from each other
(and 45 degrees from vertical) so that acceleration in one direction opens
one, accel in the other opens t'other...

Dodgy ASCII art: <G>

\/

Contacts at base; mercury at rest -> no motion.
Accel to right -> right hand switch closed, left hand switch open.
Accel to left -> left hand switch close, right hand switch open.
Probably with lots of debouncing to make sure that slightly off-axis
movement doesn't really upset things.

HTH

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
You could have two mercury switches in opposite directions laying flat or
better yet, slightly down, next to each other. One will switch one way
while the other will switch the other way. This will of course work
assuming the same part (LED's) face the front everytime. Which in your
case they probably will.

On Fri, Sep 01, 2000 at 10:30:50AM +0100, Brian Jones wrote:
{Quote hidden}

--
___________________________________________
Wesley Moore
RMIT - BEng/BApp.Sc. 2nd Year

wmoorecs.rmit.edu.au
http://wmoore.tsx.org/

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
Brian,

looks like a good solid state solution.
Or, you can use a spring and 4 screws...
(use a fixed font to see the schematic)

0 <-screw/ground
| <-spring
|
contact-> 0  | 0 <-contact/pulled-up
|
|
|
0 <-screw

stress or relax the spring as needed...

Hope this helps.

Ricardo Seixas

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
These are what I use in the AC-22 Performance Meter I designed.
The new smaller packages ones are cheaper.  If it's just for a one
off you can register for samples and you get a couple free in the
post.  Can't get much cheaper than that.  If you are really smart with
it you can probably get velocity information for your display and
change the display accordingly.

Regards

Martin

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
A more complicated approach, but still inexpensive, is to hinge the handle from
the display.  Use a very strong spring so that swinging it in either direction
bends the hinge a very small amount.  You could then use cheap contacts, or even
a pot, to tell which way the device is being swung by the bend in the hinge.
Use a pot, and you can tell the acceleration.

One item it note, though, most of these methods tend to equalize, after the firt
swing or so you won't be able to tel direction by acceleration forces in either
direction.

Brian Jones wrote:
{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
If you can handle a \$20 part, Analog Devices makes real accelerometers that
are very easy to interface to a PIC CCP module or an A/D port.

I have two samples of an ADXL202EX (I think the x means pre-production) on
my desk that I'm just trying to find the time to play with.  These are about
1/4" square, and are a 2-axis +-2G accelerometer with about .1% resolution.
They also come in a 14-pin DIP, which may be cheaper and easier to get.

You could get real fancy, integrate for the velocity and adjust the timing
to keep the size of the  letters constant.

> {Original Message removed}
Don Hyde wrote:

> <snip>

> You could get real fancy, integrate for the velocity and adjust the timing

> to keep the size of the  letters constant.

I like that one, it would be a killer!

/Marcus

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
The 14pin dip part is over twice the price, non are easy to get hold
of at the moment, we have just bought 300 but had to really struggle
to get them.  Looks like MOQ on the new one is 10 000.  The new
package makes them a bit of a bugger to hand solder as well.  But
they are a lot smaller and neater.  The X is pre production.

Martin

{Quote hidden}

> > {Original Message removed}
To do this you will want to mount you 2 axis 2 g accelerometer with
the axis at 45 degrees to the horizontal, you can then measure the
horizontal acceleration whilst calculating for any tilt in the unit which
would affect the acceleration readings due to gravity.  This is
something I am currently working on.

Martin

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
You also get the ADXL150 (not sure of the prefix) which is single axis
and comes in a metal can. Should be much cheaper.

Quentin
Martin Hill wrote:
{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
Two words:

Leaf Switch
--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
I have been told that Japanese police have wands similar to what you are
building that are used in crowd control situations.  The wands have a
short handle perpendicular to the main wand.  The wand is swung about
the handle by the operator.  An optical encoder at the junction of the
wand and handle is used to clock character (row/column - should be
azimuth/radius) pixels into the display.  The handle is kept from
rotating  by being held firmly by the operator.

Regards,
Dave Minkler

Brian Jones wrote:
{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.
----- Original Message -----
From: "Brian Jones" <bejonesHURSLEY.IBM.COM>
To: <PICLISTMITVMA.MIT.EDU>
Sent: Friday, September 01, 2000 10:30 AM
Subject: [EE] Simple direction sensing

{Quote hidden}

Yes, here's a cheap and cheerful solution. Take a set of changeover contacts
from an old, large relay and:

Option A, attach a vane of card, plastic sheet etc to the common contact.

Option B attach a weight to the common contact.

Even if you don't have any it's a pretty safe bet that senior members of
your ham club will have some old PO-3000 type relays in the junk box at
home.

If not, I'm sure you can find some suitable springy material to make contact
strips from.

.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

>As a simple project for my local radio club I've designed a PIC
>Magic Wand that flashes 7 LEDs leaving a trailing message in the
>air. The flashing (ie message) is triggered by motion sensored by a
>mercury switch.
>
>Works fine if you wave right to left but moving left to right the
>message is mirror imaged (of course).
>
>Any ideas on a simple (and cheap) method of detecting whether
>motion is left-right or right-left so I can invert the message as
>appropriate. 2 mecury switches mounted in some way so that the
>leading edge triggered first seems the obvious solution but how to
>actually implement this?

I have seen a commercial implementation of this that used a
piezo element for motion sensing. One part of the disc was
anchored to the circuit board. The center electrode had a
piece of circuit board fastened to it with a lump of solder
at one end. Presumably this acted as an inertial sensor;
generating positive volts for waving in one direction and
negative in the other.

Reg Neale

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use listservmitvma.mit.edu?body=SET%20PICList%20DIGEST>
A sprung magnet in a coil of wire?

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

{Original Message removed}
Measurement Specialties (formerly AMP) makes a piezo polymer vibration
sensor (LDTM-028K) with an attached proof mass that functions as a cheap
accelerometer:

http://www.measurementspecialties.com/vibration_dynamic_sensors.htm

These are \$.86 at Digi-Key.  They put out several volts of reversing
polarity as you wave them back and forth.  This may be adapable as a sensor
for this application.  They are 1" long by .5" wide, and need some wiggle
room, so they are a bit larger than an SOIC or DIP.

{Quote hidden}

Jim Vannes
Instrumentation Specialist
Information Processing Consultant
School of Education

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use listservmitvma.mit.edu?body=SET%20PICList%20DIGEST>
Everyday Practical Electronics http://www.epemag.wimborne.co.uk/ (or one of
its predecessors) had a design some time ago which Magenta Electronics still
sell as a kit http://www.magenta2000.co.uk/kits/864.htm .
If I can find the time I will try and find the original article. It would
appear to be pre 1996 from the quick look I have taken.

Regards
Chris

{Original Message removed}
Silicon Chip ("Spacewriter" 5/97) has a wand which uses a mercury
tilt switch. The switch goes to a 4093 gate with an adjustable RC to
optimise the delay

Electronics Today International, UK ("Latent Image Display, 2/94)
has a wand that's slightly different. It's a bench-mounted unit, driven
by a rotating cam, like a steam train wheel. The sensor is an opto-
coupler through a slot in the cam

The ETI News section in that issue has an article about large
LCD displays (a post from a fortnight ago). A US company called
FP Displays was at the time making 18" high LCD characters (under
the trade name Visilight) by a new dry-film process to give flexible (!!)
160 deg view angle, light-weight display modules, intended for
customer information units. The contact details given were FP
Displays AG, Bristol (UK) 0272 251 125

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use listservmitvma.mit.edu?body=SET%20PICList%20DIGEST>
Can somebody give me the link to the spinning clock that displays the
time like this?

Thanks
Quentin

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestmitvma.mit.edu>
>Can somebody give me the link to the spinning clock that displays the
>time like this?

http://www.bobblick.com/bob/projects/mclock/index.html

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestmitvma.mit.edu>
Why not just use switches at each end of travel, a nice digital on/off tells
you which way it will move next ...

{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...