Triple Axis Accelerometer

[subpilot]

I just got my accelerometer board working. Now my RoboNova knows when he has taken a fall and gets back up on his own.
I used the Sparkfun Triple Axis Accelerometer http://www.sparkfun.com/commerce/produc ... cts_id=252

Here's where I put the board. I'm running it from A2D chans 5 & 6 (just using 2 axis so far)



Note: This accelerometer board runs on 3.3V and is NOT 5V tolerent. You must convert the 3024 over to run on 3.3V or provide an external 3.3V source for this board.

[Pev]

Neat - presumably you check this each time through the main routine?

Pev

[subpilot]

I just uncommented the robot_tilt line in the main loop and modded the subroutine like below. I'm just checking the one axis to see if he is laying on his back or his belly. Very preliminary.
'================================================
robot_tilt:
A = AD(6)
IF A < &h40 THEN GOTO tilt_back
IF A > &ha0 THEN GOTO tilt_front
RETURN
tilt_back:
A = AD(6)
'IF A < 30 THEN GOTO forward_standup
IF A < &h40 THEN GOTO backward_standup
RETURN
tilt_front:
A = AD(6)
'IF A > 200 THEN GOTO backward_standup
IF A > &ha0 THEN GOTO forward_standup
RETURN
'================================================

[RubikCube]

Why not use this accelerometer ADXL32x Series? ( http://www.sparkfun.com/commerce/produc ... cts_id=406). It does not require voltage change.

[subpilot]

Why not use this accelerometer ADXL32x Series?

Sure, that would work. I switched over to 3.3V in order to use some other mods I have in mind.

[onesandzeros]

Dimension Engineering Buffered ±2g Accelerometer
hey guys, is there any reason why this accelerometer wouldnt work?

i figure its 2 axis and thats all i need, and it uses 3-5v
-----
Dimensions: 10mm X 18mm
Weight: 0.9g
Operating Voltage: 3-5V
Draws less then 2mA

Applications:
Motion, tilt and slope measurement
Shock sensing

Performance:
Dual axis ±2g sensing
750mV/g sensitivity at 5VDC
500Hz bandwidth
Accurately drives up to a 3kΩ load
Based on the Analog Devices ADXL322

Protective Features:
Reverse voltage protection
14v overvoltage protection
Output short protection
----

[beermat]

Thats the one I use, onesandzeros. Works like a charm.

[onesandzeros]

hey beermat
hows it going
thanks for your reply
how did u hook it up? as it has negative,positive x and y .
and where did you put it

[beermat]

I put the accelerometer in a 14 pin DIP IC socket and mounted it inside his head, along with the remote receiver and the tri-colour eyes. I only use the Y axis, and it is wired into AD2.

[onesandzeros]

thanks alot for clarifying that for me

[engineer]

hey subpilot; I've got a two axis and couldn't get it to work. When I installed a hitec single axis, uncommented the GOSUB tilt it worked.the reason for the two axis is that sometimes RN-! ends on its side and nothing happens. I DIM x AS BYTE instead of A because A variable is used for the voltage subroutine. I've asked both hive and beermat this question and they show me code that I can't understand like yours.can you pm me with more information,please?

[onesandzeros]

hmm
i tried a Dimension Engineering Buffered ±2g Accelerometer 2x. hooked up as above, and it doesnt work
so your saying its because of the A variable?

[engineer]

this is what I've got running

Code: Select all

robot_voltage:                  ' [ 10 x Value / 256 = Voltage]
   DIM v AS BYTE

   v = AD(6)
   
   IF v <153> 250 THEN RETURN
    
   IF x <30> 200 THEN GOTO tilt_high
   
   RETURN
tilt_low:
   x = AD(2)
    'IF x < 30 THEN  GOTO forward_standup   
   IF x < 30 THEN  GOTO backward_standup
   RETURN
tilt_high:   
   x = AD(2)
   'IF x > 200 THEN GOTO backward_standup
   IF x > 200 THEN GOTO forward_standup
   RETURN
'================================================

this is only the Hitec tilt sensor and the v monitor w/two 51Kohm resistors at ad(2), ad(6) I still haven't got the two axis running yet. I thought that using x & y as byte would work but RN-! doesn't react to a fall with it. The dim A as byte is used for the remocon I believe I wired the two axis in and took v. readings and they are lower than the RN-1 tilt sensors, subpilot uses a different variable and that might take that into account. I hope he'll come back to this and explain what he did.