Design of a Data-Logger for Acceleration and Direction

[R.Duke]

I am trying to choose components and design a data logger for tracking the acceleration and direction/heading of a car.
The design needs to meet only a few requirement specifications and i don't need to concern myself with the software implementation.

The system needs to record direction of travel and acceleration every ten seconds and store it in an appropriate memory location with a timestamp.
It also needs to be able to upload this data to a PC.

I was thinking I will use a PIC18 and copy the circuit of a Ready for Pic board I already have from Mikroelektronika.
That will take care of the oscillator and power supply to the micro derived from a battery pack.

However I am looking for advice pertaining to the additional components and interfacing methods.

A 2-axis accelerometer was considered but would a 3-axis be better? I assume the heading can be extrapolated from the acceleration on each axis.
I guess that an analogue output is also desirable?

use a real time clock module seems appropriate for the timestamp, though the interfacing method is not chosen.

An SD card was considered for data storage though again, i have not looked into the interfacing method (spi or what).

Is it likley that my RTC will require i2c and the SD card will require SPI?

I was thinking of allowing serial transfer over ftdi for real time viewing, but im unsure about the ability to implement all 3 interfacing methods on the one chip simultaneously.

Any tips or assistance is greatly appreciated!

Thanks!

 

[FvM]

An accelerometer isn't suitable to derive travel and heading information. It would be even difficult to track the movement with a gyroscope. If you want this information, think about a GPS receiver.

Is it likley that my RTC will require i2c and the SD card will require SPI?

Yes. Chose a PIC with respective hardware interfaces for your convenienece, e.g. a recent 18FxxK chip. Or use hardware SPI and software I2C with an older single MSSP PIC.

 

[R.Duke]

given that i only require acceleration and direction of heating, i considered GPS to be a bit overkill.

What about an accelerometer with a magnetometer for direction?

 

[FvM]

Travel is the double integral of forward acceleration. Calculate how the error increases with time for your preferred acceleorometer device and check if it's acceptable.

 

[R.Duke]

I will not be implementing the software side of this project.

I will simply be designing the hardware.

 

[FvM]

Isn't it that any project should start with a feasibility consideration?

 

[R.Duke]

of course.
I am simply stating that any development with a hands-on approach is beyond the scope of this post.

Thanks