position sensors

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
rocdoc
Posts: 1
Joined: Wed Apr 05, 2006 3:37 am
Location: Northbrook, IL
Contact:

position sensors

Post by rocdoc » Wed Apr 05, 2006 3:50 am

I have a 2 foot square surface with a moving post above it. I need a way to sense and graph its x - y position in motion so that its accelerations can be calculated. I have considered a mems gyro-accelerometer but they would be very expensive, and complicated, and require wiring to the post (bad) or RF circuit add on. This would also be too heavy.
Is there a way to do this with the sensor element off the surface and a "target" on the post, ie. a light, a magnet or ?.
It should also be cheap and reliable.
Please help
Dan,
Python Engineering

Colinr
Posts: 33
Joined: Wed Oct 23, 2002 1:01 am
Location: UK
Contact:

Post by Colinr » Wed Apr 05, 2006 5:40 am

firstaly what accuracy do you whis the position to be detected ?

my first thought is to attach an optical mouse sencor to the bottom of the rod. this will detect the movement over the base board. it will not provide actual position but reletive movement and speed form this allelleration can be calcuated. the only problem is the need to wire up the rod so why not use a wireless optical mouse.

if the resolution is not to high say 1" then you could fit a magnet to the rod and then fit reed switches on a 1" matrix on the base board.
this may be to expencive as for a 2' squar the number of read switches is large. (576)

I am thnking of another option but need to know some more information can the rod rotare about its access?

don't know when i will be able to get back to you as i will be away form my computer for a fiew days

Colin

Ravenex
Posts: 1
Joined: Wed Apr 05, 2006 7:11 am
Contact:

Post by Ravenex » Wed Apr 05, 2006 7:19 am

I am working on this project with rocdoc, one of the difficulties we are facing, and the reason we can't simply use a cheap two axis accelerometer, is that the post will rotate throughout it's movement cycle. We are hoping to measure it's behavior relative to the plane of the base and as it rotates the orientation of the accelerometer would change. As for accuracy, I'm hoping for 1/100th of an inch if possible, 1/10 would be acceptable. I was actually concidering using a brightly colored post and a camera above the setup with object tracking software, but I am concerned about the cost.

User avatar
Sambuchi
Posts: 366
Joined: Tue Jan 18, 2005 1:01 am
Location: Orlando FL
Contact:

Post by Sambuchi » Wed Apr 05, 2006 7:32 am

Helllo rocdoc

I am not sure how your setup looks like but I'll give it a try... So you have a post that can move in two directions in a 2 square foot area. ..

Picture this ... I once had to make a computer screen a touch screen .. to track the user finger on the screen (two directions in a 1 square foot area) I setup IR emitters on two sides and phototransistors on the other. "stack sensors close as you want... the more the more accurate response"

Look out for ambient light.. IR and ambient light are near the same frequency. Solution would be filters or a modulated signal.

Good Luck

User avatar
haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup » Wed Apr 05, 2006 12:05 pm

Can the surface be reflective. If so I might put a colored mark on the end of the post and use the camera pointed at the reflection to record its position.

If the color of the mark were selected so that the contrast/brightness of the camera could be adjusted so that it is the only thing it sees then the result wold be a dot moving across the screen. Now you could digitally extrapolate the location of the post from the location of the mark on the screen.

Somewhat more complicated would be two camers recording the X and Y positions seperately.

These methods are non contact and do not add any mechanical response to your system and can be done with some web cams, a video codec and some software. Spacial resolution would be a function of the lens and camera resolution. Hardware cost would be nominal , you might already have what you need but the S/W is not familiar to me. Search on "Machine Vision software" or "Machine Vision software open source"

http://sourceforge.net/projects/opencvlibrary/

For an all electronic solution, I would need to know more about how the post is suspended. A sensor could be concieved at the pivot point to measure the X and Y angle directly using a potentiometer style thing.

Ultrasonic comes to mind also but seems more complicated to extract position since they are bettrer at finding range.

If cost weren't a concern a laser scanner would be the best means.

User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Post by Edd » Wed Apr 05, 2006 3:37 pm

I readily can see the 2 ft square static base plate, but now you have an apparent suspended animation “postâ€

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Post by philba » Thu Apr 06, 2006 10:03 am

Depending on how large your target is, you could use ultrasonic ranging. You could position the sensors at the corners and then each sensor would give you a single axis distance which you would translate into X/Y location. you might need more than 2 transmitters/receivers because the beam angle is often not very wide.

You could also look at IR distancing using phase shift since the distances are highly constrained. An IR target could be made small. Modulate the IR beam and detect phase difference between the outgoing and reflected beams.

User avatar
haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup » Thu Apr 06, 2006 12:59 pm

I was picturing an electronic version of a Foucault Pendulum but I don't know if you can make one that small.

cato
Posts: 366
Joined: Sun Sep 07, 2003 1:01 am
Contact:

Post by cato » Sat Apr 08, 2006 3:42 am

Two line scan cameras (not cheap 18 years ago but available with 1024, 2048 or 4096 ccd based pixels) should, if I understand your project, give you what you need.

JPKNHTP
Posts: 488
Joined: Wed Jun 29, 2005 1:01 am
Location: Missouri
Contact:

Post by JPKNHTP » Sat Apr 08, 2006 3:57 am

-JPKNHTP
-God Bless

cato
Posts: 366
Joined: Sun Sep 07, 2003 1:01 am
Contact:

Post by cato » Sat Apr 08, 2006 4:25 am

JPKNHTP wrote:Whats moving the post? What kind of accuracy are you after? What would you like to use if you had a choice? What kind of budget do you have? Someone else suggested this a few psts up but beam break would probably be the the cheapest and most simple. There are a few tricks you could use but I would have to know the answers to the above questions. Got pictures? or Milk? Hope we can help you.
Ya know, its not really fair to alter the question.

Anybody could help them with all THAT info..... the trick is to solve the problem with the information that's given. :razz:

JPKNHTP
Posts: 488
Joined: Wed Jun 29, 2005 1:01 am
Location: Missouri
Contact:

Post by JPKNHTP » Sat Apr 08, 2006 4:35 am

-JPKNHTP
-God Bless

Post Reply

Who is online

Users browsing this forum: CeaSaR and 24 guests