Need Advice on this project

[cybertooth]

Need Advice regarding this project

Please advice on the project and how to go about it. Here is the detail of the project



Project diagram


















The basic concept is that I need to record the position of ‘B’ in X,Y,Z axis when ‘A’ is exactly horizontal.

The two are interdependent but can not have a wire connection.

My Idea



- For (a2) .Have an arduino board with a triple axis accelerometer with blue tooth capability( only receiver will do) and an OLED display.

- For (a1) . have an arduino board with A single/two axis gyroscope with blue tooth capability.(only transmission) and a green LED.

How It should Work

1) Both (a1) and (a2) are attached to ‘A’ and ‘B’.

2) When ‘A’ moves and becomes exactly horizontal( as recorded by the gyroscope) the current output should do the following two things

i) Light the green LED

ii) Send a signal through blue tooth to (a2)

3) Since ‘A’ and ‘B’ are both in motion (albeit very slow) The OLED display is continuously showing the X,Y,Z axis values of (a2), but as ‘A’ becomes horizontal the relative values of X,Y,Z displayed on (a2) should freeze.

4) A reset button on (a2) in case the readings need to be repeated.




1) Both (a1) and (a2) should be as small as possible.

2) Power both with button cells.

3) In case of (a2) { as it has to have a OLED screen} the accelerometer can be separate from the rest of the unit and may be we can have a thin 2/3 core wire running from the base unit.



What I think



For (a2)

– Use Adrofruit feather 32UL blue fruit LE.

- Triple axis Accelerometer –ADXL335

- Adafruit OLED feather wing.

Doubts

1) How big and bulky will it get?

2) Can I use button cells to power the whole setup?

3) What steps to be followed for programming? ( Not very experienced )

For (a1)

-Use Adrafruit Trinket MO

Doubts

1) Which gyroscope to use? Or an accelerometer would be better?

2) How to connect a blue tooth transmitter module and power it with button cells?

3) Programming steps.

 

[(*steve*)]

What are A, B, a1, and a2?

What is the "position of B"? I'd it relative to A?

What happens when A is not perfectly horizontal?

What about B's orientation?

What are you actually trying to do?

 

[cybertooth]

What are A, B, a1, and a2?

What is the "position of B"? I'd it relative to A?

What happens when A is not perfectly horizontal?

What about B's orientation?

What are you actually trying to do?

A and B are two moving platforms. a1 and a2 are proposed devices.
Nothing happens if A is not horizontal. It's just required to get the position of B in 3 axis when A is exactly horizontal. The two are interlinked. B is at a lower level than A as shown in the diagram.

 

[(*steve*)]

What is the "position of B"? Is it relative to A, or absolute?

What about B's orientation?

What are you actually trying to do?

 

[cybertooth]

What is the "position of B"? Is it relative to A, or absolute?

What about B's orientation?

What are you actually trying to do?

B is not directly attached to A. It's a machine table where we need to record the position of B in 3 axis when A is horizontal. At present we use a very complicated method to do so. I am trying to make a device which would make it easier.

 

[(*steve*)]

What is the "position of B"? Is it relative to A, or absolute?

How accurate does the measurement need to be? Is within a few metres (gps) enough.

 

[cybertooth]

What is the "position of B"? Is it relative to A, or absolute?

How accurate does the measurement need to be? Is within a few metres (gps) enough.

The position of 'B' is absolute. and the measurement is basically the angulation. The distance between 'A' and 'B' is 10-14 inches.

 

[(*steve*)]

What is the "position of B"? Is it relative to A, or absolute? Do you want to know the coordinates of B independent of the position of A whenever A is level, or do you want to know that B is 3 inches to the south an inch west and 4 inches lower in altitude whenever A is level?

How accurate does the measurement need to be? Is within a few metres (gps) enough.

GPS is unlikely to be accurate enough. Differential gps might get you down to a quarter inch or so if you're prepared to shell out some big bucks, AND this device had clear visibility to a significant number of satellites.

 

[cybertooth]

What is the "position of B"? Is it relative to A, or absolute? Do you want to know the coordinates of B independent of the position of A whenever A is level, or do you want to know that B is 3 inches to the south an inch west and 4 inches lower in altitude whenever A is level?

How accurate does the measurement need to be? Is within a few metres (gps) enough.

GPS is unlikely to be accurate enough. Differential gps might get you down to a quarter inch or so if you're prepared to shell out some big bucks, AND this device had clear visibility to a significant number of satellites.

'B' is absolute and yes I want to know the coordinates of B independent of A whenever A is level.

 

[(*steve*)]

How accurate does the measurement need to be?

I hate repeating myself.

 

[cybertooth]

Sorry for missing the accuracy point. Yes the reading needs to be quite accurate .Max error tolerance would be less than 1degree.

 

[kellys_eye]

It's a machine table

To try to do this to the accuracy a machine table requires using an accelerometer is futile. Forget it. Use capacitive coupling or capacitive scales - even an optical strip arrangement would be better than anything an accelerometer could deliver.

You're setting yourself up to fail.

 

[(*steve*)]

Hang on a second. He's talking absolute position.

1 degree is about 69 miles at the equator. Easy peasy. A GPS can get you within a couple of metres, not so good in the Z axis, but a few tens of metres should be fine.

:-D

 

[kellys_eye]

Since ‘A’ and ‘B’ are both in motion (albeit very slow)

I understand that accelerometers only give an output when they are 'accelerating' - once a steady speed has been reached the output falls to zero.