Accelerometer

Hi, 

I am controlling a single axis DC motor to follow the elevation of the Sun using two light sensors. The motor moves 180 degrees in a day. 

The control circuitry of the motor ( solar tracker) will be mounted on the moving solar panel mechanical assembly. I want to know the changing tilt / position. 


I choose the following accelerometer to calculate the tilt from dimension engineering 

https://www.dimensionengineering.com/datasheets/DE-ACCM3D.pdf

The accelerometer part number is ADXL335B. The accelerometer is 3 axis. It operates at 3.3V with sensitivity of 333mV/g. Its data sheet is as follows 

http://www.analog.com/media/en/technical-documentation/data-sheets/ADXL335.pdf


My questions are 

1. How to choose the accelerometer for such an application? 
2. Did I choose the right accelerometer? Can it follow 180 degrees? How should I proceed with the software?

jess

On 2015-11-29, jsscshaw88@gmail.com <jsscshaw88@gmail.com> wrote:
> Hi, 
>
> I am controlling a single axis DC motor to follow the elevation of the Sun using two light sensors. The motor moves 180 degrees in a day. 
>
> The control circuitry of the motor ( solar tracker) will be mounted on the moving solar panel mechanical assembly. I want to know the changing tilt / position. 
>
>
> I choose the following accelerometer to calculate the tilt from dimension engineering 
>
> https://www.dimensionengineering.com/datasheets/DE-ACCM3D.pdf
>
> The accelerometer part number is ADXL335B. The accelerometer is 3 axis. It operates at 3.3V with sensitivity of 333mV/g. Its data sheet is as follows 
>
> http://www.analog.com/media/en/technical-documentation/data-sheets/ADXL335.pdf
>
>
> My questions are 
>
> 1. How to choose the accelerometer for such an application? 

it seems ok. 

> 2. Did I choose the right accelerometer? Can it follow 180 degrees?
> How should I proceed with the software? 

I would have chosen with with built-in DAC

180 degrees is no probelem

atan2()

-- 
  \_(&#12484;)_

Hi,=20

The thing is that I measured single  x axis tilt with respect to earth. The=
 accelerometer was sitting flat on the table. I used arc sin to calculate t=
he tilt. I used protractor to confirm the calculated tilt. It was ok.

Then I measured single y axis tilt and used arc sin and the result was good=
.=20

I get confused when I tried to measure the tilt with x and y axes. Someone =
advised me to measure the tilt using "arc tan" and with respect to z axis. =
How does this work?=20

I will mount the accelerometer on the box and the box will move with the pa=
nel from its resting position. There will be some values of x, y and z. I w=
ill subtract these values from the future values of the x, y  and z and cal=
culate the tilt or current position using arc tan.=20

I sound confused and horrible. Please help what direction should I take.=20
jess

On Mon, 30 Nov 2015 07:08:35 -0800, jsscshaw88 wrote:

> Hi,
> 
> The thing is that I measured single  x axis tilt with respect to earth.
> The accelerometer was sitting flat on the table. I used arc sin to
> calculate the tilt. I used protractor to confirm the calculated tilt. It
> was ok.
> 
> Then I measured single y axis tilt and used arc sin and the result was
> good.
> 
> I get confused when I tried to measure the tilt with x and y axes.
> Someone advised me to measure the tilt using "arc tan" and with respect
> to z axis. How does this work?
> 
> I will mount the accelerometer on the box and the box will move with the
> panel from its resting position. There will be some values of x, y and
> z. I will subtract these values from the future values of the x, y  and
> z and calculate the tilt or current position using arc tan.
> 
> I sound confused and horrible. Please help what direction should I take.
> jess

You haven't defined your axes.

Each axis should respond to the acceleration due to gravity as cos
(theta), where theta is the angle of _that axis_ from vertical (so you 
really have theta_x, theta_y, theta_z).

You ought to be able to work backwards from there, after coiling yourself 
in much trigonometry.  It's best to do the math yourself -- that way you 
know that it applies to your situation, and you're not depending on 
someone else's competence or assumptions.

-- 
www.wescottdesign.com