I am slowly working towards the second phase of my microcontroller project.  I hope to have a new proposal ready soon, but for now I have been working on my signal processing.  I am very new to this, and I am trying to teach myself as much as possible with this project.  I currently have the accelerometer from the servo arm project running on an arduino kit.  Since the arduino has an easy to use serial library I set it up to work as a data logger for the accelerometer.  I sampled the accelerometer at 50Hz, which is the same as in the previous project.  I have not decided if this will be the sample frequency I use in the next project or not, but the results were informative.

The graph below shows the actual sampled data and the results of three different filters.  The first filter, labeled Box, is the filter I used on the last project; an 8 sample rectangular rolling average.  The Triangle filter is a 5 sample triangular weighted rolling average.  The Butterworth filter is 2nd order with a cutoff of 2Hz.

Between the Box and Triangle filters, both lag the signal by similar amounts, but the triangular is much smoother.  The butterworth is my first attempt at a real filter, it is much smoother but at the expense of a little more lag.  The lag on these filters ranges from 100ms for the triangle and box, to about 140ms for the butterworth.  I am not sure if this is a deal breaker, but it seems like a lot of lag.

My original thought with the very low sample frequency was that there was no point in sampling the accelerometer faster than I could update the servo (once every 20ms).  Now, I am thinking I will investigate faster sampling times to see if I will be able to create a smooth signal with less lag than the filters presented here.

Essentially this project is a two degree of freedom robotic arm controlled by an accelerometer.  The accelerometer is mounted on a breadboard that the user can tilt in space controlling the direction of a laser pointer mounted to the arm.  The entire project is programed in assembly language on the Motorola 68HC11 microprocessor.

—

Download (PDF, 296.5KB)

]]> http://www.chadhuard.com/2010/05/final-project/feed/ 0 http://www.chadhuard.com/2010/04/switch-bounce/ http://www.chadhuard.com/2010/04/switch-bounce/#comments Tue, 13 Apr 2010 00:26:19 +0000 Chad http://www.chadhuard.com/?p=451 Most people who have dabbled in embedded design know that all mechanical switches have some amount of “bounce” where they are not making solid contact for a short time after being pressed or released.  If you manage to read the button fast enough that you make more than one reading while the button is still bouncing you can get erroneous readings.  I was working on a lab for my microcomputer class that used a flip-flop to de-bounce the rising edge of the button press, but I was still getting extra button press readings when I was releasing the button.  This was obviously a job for scopezilla.  What I found was interesting:

So, the big problem?  Well, it turns out that the glitch that can be seen in the falling edge image was really the problem, not bouncing on the rising edge.  It didn’t occur every time I pressed and released the button, but frequently enough to be a real pain (perhaps 10%).  All of the de-bouncing in the world on the leading edge could not keep the processor from interpreting the high after the glitch as a second button press.  Obviously more sophisticated de-bouncing is required.

Download (PDF, 294.11KB)

Download (PDF, 273.55KB)

Hopefully you find it useful.

EQU.txt

At any rate, I did eventually get the HC11 to control the servo motor based on the ADC input reading the pot on the evaluation board.  I still have a lot of tuning to do.  I am currently splitting up the functions and trying to work out a good way to organize things so that I can keep the PWM stuff in a separate file and link it to the main program.  I also need to calibrate the range of the servo, but that will come later.

Here is the code:

I am currently trying to decide if I will be able to make my nunchuk work with the HC11 or not.  Unfortunately the chip does not natively support the I2c protocol that the nunchuk uses.  I have found a chip that essentially allows a processor to communicate with I2C devices using SPI, but it is in a really tiny package and my soldering is probably not up to it.  We will see.

On the other front, the servo research is going much better.  I will post my progress on this soon.