ADXL202 dual axis accelerometer used as tilt sensors / spirit level :
http://ww1.microchip.com/downloads/e...tes/00715a.pdf
It says :
Static acceleration due to gravity can be used
to determine tilt and inclination provided that the sensor
is responsive to static acceleration.
wi77iam made a good example how a simple accelerometer work with spring. In fact I even made a crude accelerometer using a joystick with weight on top of it described in this forum else where. Moving it around backward/forward and left/right register accelerations the spring and heavy weight make all this work. And of course tilting it will move the stick and registering the effect of earths g
I guess this means we need to distinguish between what an accelerometer reads and actual acceleration.
But how? this is exactly what has been puzzling me since I started posting in this thread...same as measuring real cars acceleration and latteral g, those sensor simply cannot meaure directly what we need to know.
What about rotation? If you simply have x, y and z accelerometers and your car were to, say, stop and spin around the center point of the accelerometers and then start moving again, how would you know which direction you were going in?
Very unlikely you be doing a handbrake turn
Doing spinout brake turns in tunnels is basically my life...
Must...have...gyros...
I think we're all leaning toward piezoelectric gyroscope + speed sensor is our best bet...
Compass + speed sensor may work, depending on how well it works in a car (my guess is OK, since we're really just looking for something that can reckon for 1000's of feet, not multiple miles).
Speed sensor + steering wheel sensor might work, although I'm personally not going to go that route... Not plug-and-play, plus it needs calibration, plus some cars don't have linear steering (My 3-Series steered different amounts for given wheel motion at different speeds, for example)
So, the accelerometer chip I found for this seems to be in EOL, and is $40 for just the chip, sans supporting electronics:
http://search.murata.co.jp/Ceramy/ow...03J&sHnTyp=NEW
Damn, I want a radial Phidget!!!
Bush: Proof that humans really did evolve from apes.
Yep, gryo + speed sensor is how the OEM systems work...I think. Since I already have the compass though, I'm going to try that first -- If ever I get a moment. I agree, it's a rare case that you would be using only d.r. for a long period of time. GPS alone already works quite well. My only interest in d.r. is just the fun of building it factor.
Compass, steering wheel sensor, odometer, etc...the trick is, the more inputs you have, the more accurate it is. Which is good, because adding more inputs also increases the coolness factor signifigantly.
Can't we just add extra LED's to enhance the coolness factor? Blue ones?Originally Posted by GWRedDragon
Bush: Proof that humans really did evolve from apes.
Ideal configuration for GPS corrector is 100% hardware released device. Gyro + speed pulse, like OEM navigation systems. If GPS stream increase to 2 birds and GPS reported for incorrect position calculation "Gyro + speed" corrector make corrected stream. This entire device coming up to $150 bucks. How interesting to people invest extra to close "dead zone" on the road?
p.s. compass, accelerometers and ODB, slow and have low accuracy to calculate position with maximum precision.
Posting Permissions
LinkBack URL
About LinkBacks
Reply With Quote
Bookmarks