Book – “Arduino Workshop – A Hands-On Introduction with 65 Projects”

Over the last few years I’ve been writing a few Arduino tutorials, and during this time many people have mentioned that I should write a book. And now thanks to the team from No Starch Press this recommendation has morphed into my new book – “Arduino Workshop“:

Although there are seemingly endless Arduino tutorials and articles on the Internet, Arduino Workshop offers a nicely edited and curated path for the beginner to learn from and have fun. It’s a hands-on introduction to Arduino with 65 projects – from simple LED use right through to RFID, Internet connection, working with cellular communications, and much more.

Each project is explained in detail, explaining how the hardware an Arduino code works together. The reader doesn’t need any expensive tools or workspaces, and all the parts used are available from almost any electronics retailer. Furthermore all of the projects can be finished without soldering, so it’s safe for readers of all ages.

The editing team and myself have worked hard to make the book perfect for those without any electronics or Arduino experience at all, and it makes a great gift for someone to get them started. After working through the 65 projects the reader will have gained enough knowledge and confidence to create many things – and to continue researching on their own. Or if you’ve been enjoying the results of my thousands of hours of work here at tronixstuff, you can show your appreciation by ordering a copy for yourself or as a gift

You can review the table of contents, index and download a sample chapter from the Arduino Workshop website.

Arduino Workshop is available from No Starch Press in printed or ebook (PDF, Mobi, and ePub) formats. Ebooks are also included with the printed orders so you can get started immediately.

In the meanwhile have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column? And join our friendly Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

Arduino and KTM-S1201 LCD modules

Learn how to use very inexpensive KTM-S1201 LCD modules in this edition of our Arduino tutorials. This is chapter forty-nine of a series originally titled “Getting Started/Moving Forward with Arduino!” by John Boxall – A tutorial on the Arduino universe. The first chapter is here, the complete series is detailed here.

Introduction

After looking for some displays to use with another (!) clock, I came across some 12-digit numeric LCD displays. They aren’t anything flash, and don’t have a back light –  however they were one dollar each. How could you say no to that? So I ordered a dozen to try out. The purpose of this tutorial is to show you how they are used with an Arduino in the simplest manner possible.

Moving forward – the modules look like OEM modules for desktop office phones from the 1990s:

With a quick search on the Internet you will find a few sellers offering them for a dollar each. The modules (data sheet) use the NEC PD7225 controller IC (data sheet):

They aren’t difficult to use, so I’ll run through set up and operation with a few examples.

Hardware setup

First you’ll need to solder some sort of connection to the module – such as 2×5 header pins. This makes it easy to wire it up to a breadboard or a ribbon cable:

The rest of the circuitry is straight-forward. There are ten pins in two rows of five, and with the display horizontal and the pins on the right, they are numbered as such:

Now make the following connections:

• LCD pin 1 to 5V
• LCD pin 2 to GND
• LCD pin 3 to Arduino D4
• LCD pin 4 to Arduino D5
• LCD pin 5 to Arduino D6
• LCD pin 6 to Arduino D7
• LCD pin 7 – not connected
• LCD pin 8 – Arduino D8
• LCD pin 9 to the centre pin of a 10k trimpot – whose other legs connect to 5V and GND. This is used to adjust the contrast of the LCD.

The Arduino digital pins that are used can be changed – they are defined in the header file (see further on). If you were curious as to how low-current these modules are:

That’s 0.689 mA- not bad at all. Great for battery-powered operations. Now that you’ve got the module wired up, let’s get going with some demonstration sketches.

Software setup

The sketches used in this tutorial are based on work by Jeff Albertson and Robert Mech, so kudos to them – however we’ve simplified them a little to make use easier. We’ll just cover the functions required to display data on the LCD. However feel free to review the sketches and files along with the controller chip datasheet as you’ll get an idea of how the controller is driven by the Arduino.

When using the LCD module you’ll need a header file in the same folder as your sketch. You can download the header file from here. Then every time you open a sketch that uses the header file, it should appear in a tab next to the main sketch, for example (click to enlarge):

There’s also a group of functions and lines required in your sketch. We’ll run through those now – so download the first example sketch, add the header file and upload it. Your results should be the same as the video below:

So how did that work? Take a look at the sketch you uploaded.  You need all the functions between the two lines of “////////////////////////” and also the five lines in void setup(). Then you can display a string of text or numbers using

ktmWriteString();

which was used in void loop(). You can use the digits 0~9, the alphabet (well, what you can do with 7-segments), the degrees symbol (use an asterix – “*”) and a dash (use  - “-”). So if your sketch can put together the data to display in a string, then that’s taken care of.

If you want to clear the screen, use:

 ktmCommand(_ClearDsp);

Next – to individually place digits on the screen, use the function:

 ktmPrnNumb(n,p,d,l);

Where n is the number to be displayed (zero or a positive integer), p is the position on the LCD for the number’s  (the positions from left to right are 11 to 0…), d is the number of digits to the right of the decimal point (leave as zero if you don’t want a decimal point), and l is the number of digits being displayed for n. When you display digits using this function you can use more than one function to compose the number to be displayed – as this function doesn’t clear the screen.

To help get your head around it, the following example sketch (download) has a variety of examples in void loop(). You can watch this example in the following video:

Conclusion

So there you have it – an incredibly inexpensive and possibly useful LCD module. Thank you to Jeff Albertson and Robert Mech for their help and original code.

In the meanwhile have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column? And join our friendly Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

mbed and the Freescale FRDM-KL25Z development board

In this article we examine the mbed rapid prototyping platform with the Freescale FRDM-KL25Z ARM® Cortex™-M0+ development board.

Introduction

A while ago we looked at the mbed rapid prototyping environment for microcontrollers with the cloud-based IDE and the NXP LPC1768 development board, and to be honest we left it at that as I wasn’t a fan of cloud-based IDEs. Nevertheless, over the last two or so years the mbed platform has grown and developed well – however without too much news on the hardware side of things. Which was a pity as the matching development boards usually retailed for around $50 … and most likely half the reason why mbed didn’t become as popular as other rapid development platforms.

Also – a few months ago – we received the new Freescale Freedom FRDM-KL25Z development board from element14. I started to write about using the board but frankly it did my head in, as at the time the IDE was almost a one gigabyte download and the learning curve too steep for the time I had available. Which was a pity as the board is inexpensive and quite powerful. So the board went into the “miscellaneous dev kit” box graveyard. Until now. Why?

You can now use the Freedom board with mbed. 

It isn’t perfect – yet – but it’s a move in the right direction for both mbed and Freescale. It allows educators and interested persons access to a very user-friendly IDE and dirt-cheap development boards.

What is mbed anyway?

mbed is a completely online development environment. That is, in a manner very similar to cloud computing services such as Google Docs or Zoho Office. However there are some pros and cons of this method. The pros include not having to install any software on the PC – as long as you have a web browser and a USB port you should be fine; any new libraries or IDE updates are handled on the server leaving you to not worry about staying up to date; and the online environment can monitor and update your MCU firmware if necessary. However the cons are that you cannot work with your code off-line, and there may be some possible privacy issues. Here’s an example of the environment (click to enlarge):

As you can see the IDE is quite straight-forward. All your projects can be found on the left column, the editor in the main window and compiler and other messages in the bottom window. There’s also an online support forum, an official mbed library and user-submitted library database, help files and so on – so there’s plenty of support. Code is written in C/C++ style and doesn’t present any major hurdles. When it comes time to run the code, the online compiler creates a downloadable binary file which is copied over to the hardware via USB.

And what’s a Freedom board?

It’s a very inexpensive development board based on the Freescale ARM® Cortex™-M0+ MKL25Z128VLK4 microcontroller. How inexpensive? In Australia it’s $9 plus GST and delivery.

Features include  (from the product website):

• MKL25Z128VLK4 MCU – 48 MHz, 128 KB flash, 16 KB SRAM, USB OTG (FS), 80LQFP
• Capacitive touch “slider,” MMA8451Q accelerometer, tri-color LED
• Easy access to MCU I/O
• Sophisticated OpenSDA debug interface
• Mass storage device flash programming interface (default) – no tool installation required to evaluate demo apps
• P&E Multilink interface provides run-control debugging and compatibility with IDE tools
• Open-source data logging application provides an example for customer, partner and enthusiast development on the OpenSDA circuit

And here it is:

In a lot of literature about the board it’s mentioned as being “Arduino compatible”. This is due to the layout of the GPIO pins – so if you have a 3.3 V-compatible Arduino shield you may be able to use it – but note that the I/O pins can only sink or source 3 mA (from what I can tell) – so be careful with the GPIO . However on a positive side the board has the accelerometer and an RGB LED which are handy for various uses. Note that the board ships without any stacking header sockets, but element14 have a starter pack with those and a USB cable for $16.38++.

Getting started

Now we”ll run through the process of getting a Freedom board working with mbed and creating a first program. You’ll need a computer (any OS) with USB, an Internet connection and a web browser, a USB cable (mini-A to A) and a Freedom board. The procedure is simple:

1. Download and install the USB drivers for Windows or Linux from here.
2. Visit mbed.org and create a user account. Check your email for the confirmation link and follow the instructions within.
3. Plug in your Freedom board – using the USB socket labelled “OpenSDA”. It will appear as a disk called “bootloader”
4. Download this file and copy it onto the “bootloader” drive
5. Unplug the Freedom board, wait a moment – then plug it back in. It should now appear as a disk called “MBED”, for example (click to enlarge):

There will be a file called ‘mbed’ on the mbed drive – double-click this to open it in a web browser. This process activates the board on your mbed account – as shown below (click to enlarge):

Now you’re ready to write your code and upload it to the Freedom board. Click “Compiler” at the top-right to enter the IDE.

Creating and uploading code

Now to create a simple program to check all is well. When you entered the IDE in the previous step, it should have presented you with the “Guide to mbed Online Compiler”. Have a read, then click “New” at the top left. Give your program a name and click OK. You will then be presented with a basic “hello world” program that blinks the blue LED in the RGB module. Adjust the delays to your liking then click “Compile” in the toolbar.

If all is well, your web browser will present you with a .bin file that has been downloaded to the default download directory. (If not, see the error messages in the area below the editor pane). Now copy this .bin file to the mbed drive, then press the reset button (between the USB sockets) on the Freedom board. Your blue LED should now be blinking.

Moving forward

You can find some code examples that demonstrate the use of the accelerometer, RGB LED and touch sensor here. Here’s a quick video of the touch sensor in action:

So which pin is what on the Freedom board with respect to the mbed IDE? Review the following map:

All the pins in blue – such as PTxx can be referred to in your code. For example, to pulse PTA13 on and off every second, use:

#include "mbed.h"

DigitalOut pulsepin(PTA13);

int main() {
 while(1) {
 pulsepin = 1;
 wait(1);
 pulsepin = 0;
 wait(1);
 }
}

The pin reference is inserted in the DigitalOut assignment and thus “pulsepin” refers to PTA13. If you don’t have the map handy, just turn the board over for a quick-reference (click to enlarge):

Just add “PT” to the pin number. Note that the LEDs are connected to existing GPIO pins: green – PTB19, red – PTB18 and blue – PTB.

Where to from here? 

It’s up to you. Review the Freedom board manual (from here) and the documentation on the mbed website, create new things and possibly share them with others via the mbed environment. For more technical details review the MCU data sheet.

Conclusion

The Freedom board offers a very low cost way to get into microcontrollers and programming. You don’t have to worry about IDE or firmware revisions, installing software on locked-down computers, or losing files. You could teach a classroom full of children embedded programming for around $20 a head (a board and some basic components). Hopefully this short tutorial was of interest. We haven’t explored every minute detail – but you now have the basic understanding to move forward with your own explorations.

The Freescale Freedom FRDM-KL25Z development board used in this article was a promotional consideration supplied by element14.

Have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column, or join our Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

Tutorial: Arduino and the MSGEQ7 Spectrum Analyzer

This is a tutorial on using the MSGEQ7 Spectrum Analyser with Arduino, and chapter forty-eight of a series originally titled “Getting Started/Moving Forward with Arduino!” by John Boxall – A tutorial on the Arduino universe. The first chapter is here, the complete series is detailed here.

Updated 30/01/2013

In this article we’re going to explain how to make simple spectrum analysers with an Arduino-style board. (Analyser? Analyzer? Take your pick).

First of all, what is a spectrum analyser? Good question. Do you remember what  this is?

It’s a mixed graphic equaliser/spectrum analyser deck for a hi-fi system. The display in the middle is the spectrum analyser, and roughly-speaking it shows the strength of  different frequencies in the music being listened to – and looked pretty awesome doing it. We can recreate displays similar to this for entertainment and also as a base for creative lighting effects. By working through this tutorial you’ll have the base knowledge to recreate these yourself.

We’ll be using the MSGEQ7 “seven band graphic equaliser IC” from Mixed Signal Integration. Here’s the MSGEQ7 data sheet (.pdf).  This little IC can accept a single audio source, analyse seven frequency bands of the audio, and output a DC representation of each frequency band. This isn’t super-accurate or calibrated in any way, but it works. You can get the IC separately, for example:

and then build your own circuit around it… or like most things in the Arduino world – get a shield. In this case, a derivative of the original Bliptronics shield by Sparkfun. It’s designed to pass through stereo audio via 3.5mm audio sockets and contains two MSGEQ7s, so we can do a stereo analyser:

As usual Sparkfun have saved a few cents by not including the stackable header sockets, so you’ll need to buy and solder those in yourself. There is also space for three header pins for direct audio input (left, right and common), which are useful – so if you can add those as well.

So now you have a shield that’s ready for use. Before moving forward let’s examine how the MSGEQ7 works for us. As mentioned earlier, it analyses seven frequency bands. These are illustrated in the following graph from the data sheet:

It will return the strengths of the audio at seven points – 63 Hz, 160 Hz, 400 Hz, 1 kHz, 2.5 kHz, 6.25 kHz and 16 kHz – and as you can see there is some overlap between the bands. The strength is returned as a DC voltage – which we can then simply measure with the Arduino’s analogue input and create a display of some sort. At this point audio purists, Sheldonites and RF people might get a little cranky, so once again – this is more for visual indication than any sort of calibration device.

However as an 8-pin IC a different approach is required to get the different levels. The IC will sequentially give out the levels for each band on pin 3- e.g. 63 Hz then 160 Hz then 400 Hz then 1 kHz then 2.5 kHz then 6.25 kHz  then 16 kHz then back to 63 Hz and so on. To start this sequence we first reset the IC by pulsing the RESET pin HIGH then low. This tells the IC to start at the first band. Next, we set the STROBE pin to LOW, take the DC reading from pin 3 with analogue input, store the value in a variable (an array), then set the STROBE pin HIGH. We repeat the strobe-measure sequence six more times to get the rest of the data, then RESET the IC and start all over again. For the visual learners consider the diagram below from the data sheet:

To demonstrate this process, consider the function

readMSGEQ7()

in the following example sketch (download):

// Example 48.1 - tronixstuff.com/tutorials > chapter 48 - 30 Jan 2013 
// MSGEQ7 spectrum analyser shield - basic demonstration

int strobe = 4; // strobe pins on digital 4
int res = 5; // reset pins on digital 5

int left[7]; // store band values in these arrays
int right[7];

int band;

void setup()
{
 Serial.begin(115200);
 pinMode(res, OUTPUT); // reset
 pinMode(strobe, OUTPUT); // strobe
 digitalWrite(res,LOW); // reset low
 digitalWrite(strobe,HIGH); //pin 5 is RESET on the shield
}

void readMSGEQ7()
// Function to read 7 band equalizers
{
 digitalWrite(res, HIGH);
 digitalWrite(res, LOW);
 for(band=0; band <7; band++)
 {
 digitalWrite(strobe,LOW); // strobe pin on the shield - kicks the IC up to the next band 
 delayMicroseconds(30); // 
 left[band] = analogRead(0); // store left band reading
 right[band] = analogRead(1); // ... and the right
 digitalWrite(strobe,HIGH); 
 }
}

void loop()
{
 readMSGEQ7();
 // display values of left channel on serial monitor
 for (band = 0; band < 7; band++)
 {
 Serial.print(left[band]);
 Serial.print(" ");
 }
 Serial.println();

// display values of right channel on serial monitor
 for (band = 0; band < 7; band++)
 {
 Serial.print(right[band]);
 Serial.print(" ");
 }
 Serial.println();
}

If you follow through the sketch, you can see that it reads both left- and right-channel values from the two MSGEQ7s on the shield, then stores each value in the arrays left[] and right[]. These values are then sent to the serial monitor for display – for example:

If you have a function generator, connect the output to one of the channels and GND – then adjust the frequency and amplitude to see how the values change. The following video clip is a short demonstration of this – we set the generator to 1 kHz and adjust the amplitude of the signal. To make things easier to read we only measure and display the left channel:

Keep an eye on the fourth column of data – this is the analogRead() value returned by the Arduino when reading the 1khz frequency band. You can also see the affect on the other bands around 1 kHz as we increase and decrease the frequency. However that wasn’t really visually appealing – so now we’ll create a small and large graphical version.

First we’ll use an inexpensive LCD, the I2C model from akafugu reviewed previously. To save repeating myself, also review how to create custom LCD characters from here.

With the LCD with have two rows of sixteen characters. The plan is to use the top row for the levels, the left-channel’s on … the left, and the right on the right. Each character will be a little bar graph for the level. The bottom row can be for a label. We don’t have too many pixels to work with, but it’s a compact example:

We have eight rows for each character, and the results from an analogueRead() fall between 0 and 1023. So that’s 1024 possible values spread over eight sections. Thus each row of pixels in each character will represent 128 “units of analogue read” or around 0.63 V if the Arduino is running from true 5 V (remember your AREF notes?). The sketch will again read the values from the MSGEQ7, feed them into two arrays – then display the required character in each band space  on the LCD.

Here’s the resulting sketch (download):

// Example 48.2 - tronixstuff.com/tutorials > chapter 48 - 30 Jan 2013 
// MSGEQ7 spectrum analyser shield and I2C LCD from akafugu

// for akafugu I2C LCD
#include 
#include "TWILiquidCrystal.h"
LiquidCrystal lcd(50);

// create custom characters for LCD
byte level0[8] = { 0b00000, 0b00000, 0b00000, 0b00000, 0b00000, 0b00000, 0b00000, 0b11111};
byte level1[8] = { 0b00000, 0b00000, 0b00000, 0b00000, 0b00000, 0b00000, 0b11111, 0b11111};
byte level2[8] = { 0b00000, 0b00000, 0b00000, 0b00000, 0b00000, 0b11111, 0b11111, 0b11111};
byte level3[8] = { 0b00000, 0b00000, 0b00000, 0b00000, 0b11111, 0b11111, 0b11111, 0b11111};
byte level4[8] = { 0b00000, 0b00000, 0b00000, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111};
byte level5[8] = { 0b00000, 0b00000, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111};
byte level6[8] = { 0b00000, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111};
byte level7[8] = { 0b11111, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111, 0b11111};

int strobe = 4; // strobe pins on digital 4
int res = 5; // reset pins on digital 5

int left[7]; // store band values in these arrays
int right[7];

int band;

void setup()
{
 Serial.begin(9600);
 // setup LCD and custom characters
 lcd.begin(16, 2);
 lcd.setContrast(24);
 lcd.clear();

lcd.createChar(0,level0);
 lcd.createChar(1,level1);
 lcd.createChar(2,level2);
 lcd.createChar(3,level3);
 lcd.createChar(4,level4);
 lcd.createChar(5,level5);
 lcd.createChar(6,level6);
 lcd.createChar(7,level7);

 lcd.setCursor(0,1);
 lcd.print("Left");
 lcd.setCursor(11,1);
 lcd.print("Right");

 pinMode(res, OUTPUT); // reset
 pinMode(strobe, OUTPUT); // strobe
 digitalWrite(res,LOW); // reset low
 digitalWrite(strobe,HIGH); //pin 5 is RESET on the shield
}

void readMSGEQ7()
// Function to read 7 band equalizers
{
 digitalWrite(res, HIGH);
 digitalWrite(res, LOW);
 for( band = 0; band < 7; band++ )
 {
 digitalWrite(strobe,LOW); // strobe pin on the shield - kicks the IC up to the next band 
 delayMicroseconds(30); // 
 left[band] = analogRead(0); // store left band reading
 right[band] = analogRead(1); // ... and the right
 digitalWrite(strobe,HIGH); 
 }
}

void loop()
{
 readMSGEQ7();

// display values of left channel on LCD
 for( band = 0; band < 7; band++ )
 {
 lcd.setCursor(band,0);
 if (left[band]>=895) { lcd.write(7); } else
 if (left[band]>=767) { lcd.write(6); } else
 if (left[band]>=639) { lcd.write(5); } else
 if (left[band]>=511) { lcd.write(4); } else
 if (left[band]>=383) { lcd.write(3); } else
 if (left[band]>=255) { lcd.write(2); } else
 if (left[band]>=127) { lcd.write(1); } else
 if (left[band]>=0) { lcd.write(0); }
 }
 // display values of right channel on LCD
 for( band = 0; band < 7; band++ )
 {
 lcd.setCursor(band+9,0);
 if (right[band]>=895) { lcd.write(7); } else
 if (right[band]>=767) { lcd.write(6); } else
 if (right[band]>=639) { lcd.write(5); } else
 if (right[band]>=511) { lcd.write(4); } else
 if (right[band]>=383) { lcd.write(3); } else
 if (right[band]>=255) { lcd.write(2); } else
 if (right[band]>=127) { lcd.write(1); } else
 if (right[band]>=0) { lcd.write(0); }
 }
}

If you’ve been reading through my tutorials there isn’t anything new to worry about. And now for the demo, with sound -

That would look great on the side of a Walkman, however it’s a bit small. Let’s scale it up by using a Freetronics Dot Matrix Display - you may recall these from Clock One. For some background knowledge check the review here.  Don’t forget to use a suitable power supply for the DMD – 5 V at 4 A will do nicely. The DMD contains 16 rows of 32 LEDs. This gives us twice the “resolution” to display each band level if desired. The display style is subjective, so for this example we’ll use a single column of LEDs for each frequency band, with a blank column between each one.

We use a lot of line-drawing statements to display the levels, and clear the DMD after each display. With this and the previous sketches, there could be room for efficiency – however I write these with the beginner in mind. Here’s the sketch (download):

// Example 48.3 - tronixstuff.com/tutorials > chapter 48 - 30 Jan 2013 
// MSGEQ7 spectrum analyser shield with a Freetronics DMD

// for DMD
#include  // for DMD
#include  // SPI.h must be included as DMD is written by SPI (the IDE complains otherwise)
#include  
#include "SystemFont5x7.h" // keep next two lines if you want to add some text
#include "Arial_black_16.h"
DMD dmd(1, 1); // creates instance of DMD to refer to in sketch

void ScanDMD() // necessary interrupt handler for refresh scanning of DMD
{ 
 dmd.scanDisplayBySPI();
}

int strobe = 4; // strobe pins on digital 4
int res = 5; // reset pins on digital 5

int left[7]; // store band values in these arrays
int right[7];

int band;

void setup()
{
 // for DMD
 //initialize TimerOne's interrupt/CPU usage used to scan and refresh the display
 Timer1.initialize( 5000 ); //period in microseconds to call ScanDMD. Anything longer than 5000 (5ms) and you can see flicker.
 Timer1.attachInterrupt( ScanDMD ); //attach the Timer1 interrupt to ScanDMD which goes to dmd.scanDisplayBySPI() 
 dmd.clearScreen( true ); //true is normal (all pixels off), false is negative (all pixels on)

 // for MSGEQ7
 pinMode(res, OUTPUT); // reset
 pinMode(strobe, OUTPUT); // strobe
 digitalWrite(res,LOW); // reset low
 digitalWrite(strobe,HIGH); //pin 5 is RESET on the shield
}

void readMSGEQ7()
// Function to read 7 band equalizers
{
 digitalWrite(res, HIGH);
 digitalWrite(res, LOW);
 for( band = 0; band < 7; band++ )
 {
 digitalWrite(strobe,LOW); // strobe pin on the shield - kicks the IC up to the next band 
 delayMicroseconds(30); // 
 left[band] = analogRead(0); // store left band reading
 right[band] = analogRead(1); // ... and the right
 digitalWrite(strobe,HIGH); 
 }
}

void loop()
{
 int xpos;
 readMSGEQ7();
 dmd.clearScreen( true ); 
 // display values of left channel on DMD
 for( band = 0; band < 7; band++ )
 {
 xpos = (band*2)+1;
 if (left[band]>=895) { dmd.drawLine( xpos, 15, xpos, 1, GRAPHICS_NORMAL ); } else
 if (left[band]>=767) { dmd.drawLine( xpos, 15, xpos, 3, GRAPHICS_NORMAL ); } else
 if (left[band]>=639) { dmd.drawLine( xpos, 15, xpos, 5, GRAPHICS_NORMAL ); } else
 if (left[band]>=511) { dmd.drawLine( xpos, 15, xpos, 7, GRAPHICS_NORMAL ); } else
 if (left[band]>=383) { dmd.drawLine( xpos, 15, xpos, 9, GRAPHICS_NORMAL ); } else
 if (left[band]>=255) { dmd.drawLine( xpos, 15, xpos, 11, GRAPHICS_NORMAL ); } else
 if (left[band]>=127) { dmd.drawLine( xpos, 15, xpos, 13, GRAPHICS_NORMAL ); } else
 if (left[band]>=0) { dmd.drawLine( xpos, 15, xpos, 15, GRAPHICS_NORMAL ); }
 }

 // display values of right channel on DMD
 for( band = 0; band < 7; band++ )
 {
 xpos = (band*2)+18;
 if (right[band]>=895) { dmd.drawLine( xpos, 15, xpos, 1, GRAPHICS_NORMAL ); } else
 if (right[band]>=767) { dmd.drawLine( xpos, 15, xpos, 3, GRAPHICS_NORMAL ); } else
 if (right[band]>=639) { dmd.drawLine( xpos, 15, xpos, 5, GRAPHICS_NORMAL ); } else
 if (right[band]>=511) { dmd.drawLine( xpos, 15, xpos, 7, GRAPHICS_NORMAL ); } else
 if (right[band]>=383) { dmd.drawLine( xpos, 15, xpos, 9, GRAPHICS_NORMAL ); } else
 if (right[band]>=255) { dmd.drawLine( xpos, 15, xpos, 11, GRAPHICS_NORMAL ); } else
 if (right[band]>=127) { dmd.drawLine( xpos, 15, xpos, 13, GRAPHICS_NORMAL ); } else
 if (right[band]>=0) { dmd.drawLine( xpos, 15, xpos, 15, GRAPHICS_NORMAL ); }
 }
}

… and here it is in action:

Conclusion

At this point you have the knowledge to use the MSGEQ7 ICs to create some interesting spectrum analysers for entertainment and visual appeal – now you just choose the type of display enjoy the results.

Have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column, or join our Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

Australian Electronics – David Jones interviews Colin Mitchell

Welcome back

In this post I would like to share a series of interviews conducted by Dave Jones from eevblog.com. Dave interviews Colin Mitchell from Talking Electronics. Throughout the 1980s and onwards, Colin published a range of electronics magazines, tutorials and a plethora of electronics kits – of which many are still available today. Personally I was a great fan of the TE products, and sold many of his books through my past retail career with DSE. I hope you enjoy these interviews, and if not – stay tuned for upcoming articles.

Once again, thanks to Dave Jones and of course Colin Mitchell from Talking Electronics for their interview and various insights.

In the meanwhile have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column? And join our friendly Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

SMD Soldering made easier

Hooray – we’re back…

SMD (surface mount device) soldering to some people can seem scary and dangerous. And if done incorrectly, or in the wrong state of mind, and/or with the wrong equipment – it can be. Or like myself, you could be pretty bad at it. To make things easier, I’d like to point you in a few directions to find help and guidance if this technique is new to you. Furthermore, if you find any more resources, leave a comment below and we will investigate them further.

First up we have a new comic from Greg Peek and Dave Roberts from siliconfarmers.com, (written in a similar vein to the “Soldering is Easy” comic that was released in 2010) that is easy to read and makes sense. Here is the cover:

As you can see from the CC logo on the title page, the comic is in the public domain, so please print it out, email it, and generally distribute it far and wide. For more information about the authors see their website at siliconfarmers.com. I have also placed the file here at tronixstuff for you to download.

Next we have a detailed and nicely illustrated tutorial by Jon Oxer from freetronics.

Jon runs through the process of soldering with a toaster over, with great success. So head over and have a read.

For the first video tutorial we have the SMD episide of the series by David L. Jones at eevblog, well worth the time:

Next, the people from Sparky’s Widgets doing some drag soldering:

That’s all we have for now, so if you find any more that are worthwhile leave a comment below.

In the meanwhile have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column? And join our friendly Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

RF Wireless Data with the Seeedstudio RFbee

Introduction

In this article we examine the Seeedstudio RFbee Wireless Data Transceiver nodes. An RFbee is a small wireless data transceiver that can be used as a wireless data bridge in pairs, as well as a node in mesh networking or data broadcasting. Here is an example of an RFbee:

You may have noticed that the RFbee looks similar to the Xbee-style data transceivers – and it is, in physical size and some pinouts, for example:

However this is where the similarity ends. The RFbee is in fact a small Arduino-compatible development board based on the Atmel ATmega168 microprocessor (3.3V at 8MHz – more on this later) and uses a Texas Instruments CC1101 low-power sub1-GHz RF transceiver chip for wireless transfer. Turning over an RFbee reveals this and more:

But don’t let all this worry you, the RFbee is very simple to use once connected. As a transceiver the following specifications apply:

• Data rate – 9600, 19200, 38400 or 115200bps
• Adjustable transmission power in stages between -30dBm and 10 dBm
• Operating frequency switchable between 868MHz and 915MHz
• Data transmission can be point-to-point, or broadcast point-to-many
• Maximum of 256 RFbees can operate in one mesh network
• draws only 19.3mA whilst transmitting at full power

The pinout for the RFbee are similar to those of an Xbee for power and data, for example:

There is also the ICSP pins if you need to reprogram the ATmega168 direcly with an AVRISP-type programmer.

Getting Started

Getting started is simple – RFbees ship with firmware which allows them to simply send and receive data at 9600bps with full power. You are going to need two or more RFbees, as they can only communicate with their own kind. However any microcontroller with a UART can be used with RFbees – just connect 3.3V, GND, and the microcontroller’s UART TX and RX to the RFbee and you’re away. For our examples we will be using Arduino-compatible boards. If Arduino is new to you, consider our tutorials first.

If you ever need to update the firmware, or reset the RFbee to factory default after some wayward experimenting – download the firmware which is in the form of an Arduino sketch (RFBee_v1_1.pde) which can be downloaded from the repository. (This has been tested with Arduino v23). In the Arduino IDE, set the board type to “Arduino Pro or Pro Mini (3.3V, 8MHz) w/ATmega168″. From a hardware perspective, the easiest way to update the firmware is via a 3.3V FTDI cable or an UartSBee board, such as:

You will also find a USB interface useful for controlling your RFbee via a PC or configuration (see below). In order to do this,  you will need some basic terminal software. A favourite and simple example is called … “Terminal“. (Please donate to the author for their efforts).

Initial Testing

After connecting your RFbee to a PC, run your terminal software and set it for 9600 bps – 8 – None – no handshaking, and click the check box next to “+CR”. For example (click to enlarge):

Select your COM: port (or click “ReScan” to find it) and then “Connect”. After a moment “OK” should appear in the received text area. Now, get yourself an Arduino or compatible board of some sort that has the LED on D13 (or substitute your own) and upload the following sketch:

// RFbee demonstration sketch

int ledPin = 13;
byte incoming=0;

void setup()
{
 Serial.begin(9600);
 pinMode(ledPin, OUTPUT);
}

void blinkLED(int i)
{
  for (int a=0; a<i; a++)
  {
    digitalWrite(ledPin, HIGH);
    delay(250);
    digitalWrite(ledPin, LOW);
    delay(250);
  }
}

void loop()
{
  if (Serial.available() > 0)
  {
    incoming = Serial.read();
    switch(incoming)
    {
      case 'A':
      blinkLED(1);
      break;
      case 'B':
      blinkLED(2);
      break;
      case 'C':
      blinkLED(3);
      break;
      default:
      blinkLED(5);

   }

  Serial.println("Blinking completed!");
  delay(2000);
  Serial.flush();
  }
}

Finally, connect the Arduino board to an RFbee in this manner:

• Arduino D0 to RFbee TX
• Arduino D1 to RFbee RX
• Arduino 3.3V to RFbee Vcc
• Arduino GND to RFbee GND

and the other RFbee to your PC and check it is connected using the terminal software described earlier. Now check the terminal is communicating with the PC-end RFbee, and then send the character ‘A’, ‘B’ or ‘C’. Note that the LED on the Arduino board will blink one, two or three times respectively – or five times if another character is received. It then reports back “Blinking completed!” to the host PC. For example (click to enlarge):

Although that was a very simple demonstration, in doing so you can prove that your RFbees are working and can send and receive serial data. If you need more than basic data transmission, it would be wise to get a pair of RFbees to experiment with before committing to a project, to ensure you are confident they will solve your problem.

More Control

If you are looking to use your RFbees in a more detailed way than just sending data at 9600 bps at full power, you will need to  control and alter the parameters of your RFbees using the terminal software and simple AT-style commands. If you have not already done so, download and review the RFbee data sheet downloadable from the “Resources” section of this page. You can use the AT commands to easily change the data speed, power output (to reduce current draw), change the frequency, set transmission mode (one way or transceive) and more.

Reading and writing AT commands is simple, however at first you need to switch the RFbee into ‘command mode’ by sending +++ to it. (When sending +++ or AT commands, each must be followed with a carriage return (ASCII 13)). Then you can send commands or read parameter status. To send a command, just send AT then the command then the parameter. For example, to set the data rate (page ten of the data sheet) to 115200 bps, send ATBD3 and the RFbee will respond with OK.

You can again use the terminal software to easily send and receive the commands. To switch the RFbee from command mode back to normal data mode, use ATO0 (that’s AT then the letter O then zero) or power-cycle the RFbee.

RFbee as an Arduino-compatible board with inbuilt wireless

As mentioned previously the RFbee is based around an Atmel ATmega168 running at 8MHz with the Arduino bootloader. In other words, we have a tiny Arduino-compatible board in there to do our bidding. If you are unfamiliar with the Arduino system please see the tutorials listed here. However there are a couple of limitations to note – you will need an external USB-serial interface (as noted in Getting Started above), and not all the standard Arduino-type pins are available. Please review page four of the data sheet to see which RFbee pins match up to which Arduino pins.

If for some reason you just want to use your RFbee as an Arduino-compatible board, you can do so. However if you upload your own sketch you will lose the wireless capability. To restore your RFbee follow the instructions in Getting Started above.

The firmware that allows data transmission is also an Arduino sketch. So if you need to include RF operation in your sketch, first use a copy of the RFBee_v1_1.pde included in the repository – with all the included files. Then save this somewhere else under a different name, then work your code into the main sketch. To save you the effort you can download a fresh set of files which are used for our demonstration. But before moving forward, we need to learn about controlling data flow and device addresses…

Controlling data flow

As mentioned previously, each RFbee can have it’s own numerical address which falls between zero and 255. Giving each RFbee an address allows you to select which RFbee to exchange data with when there is more than two in the area. This is ideal for remote control and sensing applications, or to create a group of autonomous robots that can poll each other for status and so on.

To enable this method of communication in a simple form several things need to be done. First, you set the address of each RFbee with the AT command ATMAx (x=address). Then set each RFbee with ATOF2. This causes data transmitted to be formatted in a certain method – you send a byte which is the address of the transmitting RFbee, then another byte which is the address of the intended receipient RFbee, then follow with the data to send. Finally send command ATAC2 – which enables address checking between RFbees. Data is then sent using the command

transmitData(*byte data, byte length, byte sourceAddress, byte destinationAddress)

Where data is … the data to send. You can send a single byte, or an array of bytes. length is the number of bytes you are sending. sourceAddress and destinationAddress are relevant to the RFbees being used – you set these addresses using the ATMAx described earlier in this section.

If you open the file rfbeewireless.pde in the download bundle, scroll to the end of the sketch which contains the following code:

byte testData[4] = {'A','B','C','D'};

void sendTestData()
{
 // send the four bytes of data in the byte testData[] from address 1 to address 2
 transmitData(testData,4,1,2);
 delay(1000);
}

This is a simple example of sending data out from the RFbee. The RFbee with this sketch (address 1) sends the array of bytes (testdata[]) to another RFbee with address 2.  You can disable address checking by a receiving RFbee with ATAC0 – then it will receive any data send by other RFbees.

To receive data use the following function:

result=receiveData(rxData, &len, &sourceAddress, &destinationAddress, (byte *)&rssi , &lqi);

The variable result will hold the incoming data, len is the number of bytes to expect, sourceAddress and destinationAddress are the source (transmitting RFbee) and destination addresses (receiving RFbee). rssi and lqi are the signal strength and link quality indicator – see the TI CC1101 datasheet for more information about these. By using more than two RFbees set with addresses you can selectively send and receive data between devices or control them remotely. Finally, please note that RFbees are still capable of sending and receiving data via the TX/RX pins as long as the sketch is not executing the sendTestData() loop.

I hope you found this introduction interesting and useful. The RFbees are an inexpensive and useful alternative to the popular Xbee modules and with the addition of the Arduino-compatible board certainly useful for portable devices, remote sensor applications or other data-gathering exercises.

For more information and product support, visit the Seeedstudio product pages.

RFbees are available from Seeedstudio and their network of distributors.

Disclaimer - RFbee products used in this article are promotional considerations made available by Seeedstudio.

In the meanwhile have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column? And join our friendly Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

Tutorial: Analog input for multiple buttons – Part Two

This is chapter forty-six of a series originally titled “Getting Started/Moving Forward with Arduino!” by John Boxall – A tutorial on the Arduino universe. The first chapter is here, the complete series is detailed here.

[Updated 19/01/2013]

A while back I described how to read multiple buttons using only one analogue input pin. However we could only read one button at a time. In this instalment we revisit this topic and examine an improved method of doing so which allows for detecting more than one button being pressed at the same time. This method is being demonstrated as it is inexpensive and very easy to configure.

(For a more exact and expensive method please consider the use of the Microchip MCP23017 which allows for sixteen inputs via the I2C bus).

As you know the analogue input pins of the Arduino can read a voltage of between zero and five volts DC and return this measurement as an integer between zero and 1023. Using a small external circuit called a “R-2R ladder”, we can alter the voltage being measured by the analogue pin by diverting the current through one or more resistors by our multiple buttons. Each combination of buttons theoretically will cause a unique voltage to be measured, which we can then interpret in our Arduino sketch and make decisions based on the button(s) pressed.

First the circuit containing four buttons:

Can you see why this is called an R-2R circuit? When building your circuit – use 1% tolerance resistors – and check them with a multimeter to be sure. As always, test and experiment before committing to anything permanent.

Now to determine a method for detecting each button pressed, and also combinations. When each button is closed, the voltage applied to analogue pin zero will be different. And if two buttons are pressed at once, the voltage again will be different. Therefore the value returned by the function analogRead() will vary for each button-press combination. To determine these, I connected a numeric display to my Arduino-compatible board, then simply sent the analogRead() value to the display. You can see some of the results of this in the following video:

The analogRead() results of pressing every combination of button can be found in the following table:

After this experiment we now have the values returned by analogRead() and can use them in a switch… case function or other decision-making functions in our sketches to read button(s) and make decisions based on the user input. Unfortunately there was some overlap with the returned values and therefore in some cases not every possible combination of press will be available.

However, we’re still doing well and you can get at least eleven or twelve combinations still with only one analog input pin. You can add delay() functions in your sketch if necessary to take care of switch debouncing or do it with hardware if you feel it is necessary.

So now you have a more useful method for receiving input via buttons without wasting many digital input pins. I hope you found this article useful or at least interesting. This series of tutorials has been going for almost two years now, and may soon start to wind down – it’s time to move forward to the next series of tutorials

So if you have any suggestions for further articles (and not thinly-veiled methods of asking me to do your work for you…) – email them to john at tronixstuff dot com.

Have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column, or join our Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

Experimenting with Surface-Mount Component Prototyping

Experimenting with hand-soldering SMT components.

Updated 18/03/2013

Now and again I have looked at SMT (surface-mount technology) components and thought to myself “I should try that one day”. But not wanting to fork out for a toaster oven and a bunch of special tools I did it on the cheap – so in this article you can follow along and see the results. Recently I ordered some ElecFreaks SOIC Arduino Mega-style protoshields which apart from being a normal double-sided protoshield, also have a SOIC SMT pad as shown below:

First up I soldered in two SOIC format ICs – a 555 and a 4017:

These were not that difficult – you need a steady hand, a clean soldering iron tip and some blu-tac. To start, stick down the IC as such:

… then you can … very carefully … hand-solder in a few legs, remove the blu tac and take care of the rest …

The 4017 went in easily as well…

…however it can be easier to flood the pins with solder, then use solder-wick to soak up the excess – which in theory will remove the bridges between pins caused by the excess solder. And some PCB cleaner to get rid of the excess flux is a good idea as well.

Now to some smaller components – some LEDs and a resistor. These were 0805 package types, which measure 2.0 × 1.3 mm – for example a resistor:

The LEDs were also the same size. Unlike normal LEDs, determining the anode and cathode can be difficult – however my examples had a small arrow determining current flow (anode to cathode) on the bottom:

Another way is to use the continuity function of a multimeter – if their output voltage is less than the rating of the LED, you can probe it to determine the pins. When it glows, the positive lead is the anode. Handling such small components requires the use of anti-magnetic tweezers – highly recommended…

… and make holding down the components with one hand whilst soldering with the other much, much easier. Unlike normal veroboard, protoshield or other prototyping PCBs the protoshield’s holes are surrounded with a “clover” style of solder pad, for example:

These solder pads can make hand-soldering SMT parts a little easier. After some experimenting, I found the easiest way was to first flood the hold with solder:

… then hold down the component with the tweezers with one hand while heating the solder with the other – then moving and holding one end of the component into the molten solder:

The first time (above) was a little messy, but one improves with practice. The clover-style of the solder pads makes it easy to connect two components, for example:

With some practice the procedure can become quite manageable:

As the protoshields are double-sided you can make connections between components on the other side to keep things neat for observers. To complete the experiment the six LEDs were wired underneath (except for one) to matching Arduino Mega digital output pins, and a simple demonstration sketch used to illuminate the LEDs, as shown below:

For one-off or very low-volume SMD work these shields from elecfreaks are quite useful. You will need a steady hand and quite a lot of patience, but if the need calls it would be handy to have some of these boards around just in case. For a more involved and professional method of working with SMT, check out this guide by Jon Oxer.

In the meanwhile have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column? And join our friendly Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.

Tutorial: Arduino Port Manipulation

This is chapter forty-three of a series originally titled “Getting Started/Moving Forward with Arduino!” by John Boxall – a series of articles on the Arduino universe. The first chapter is here, the complete series is detailed here. Any files from tutorials will be found here.

[Updated 19/01/13]

In this article we are going to revisit the I/O pins, and use what is called “Port Manipulation” to control them in a much faster manner than using digitalWrite()/digitalRead().

Why?

Speed! Using this method allows for much faster I/O control, and we can control or read groups of I/O pins simultaneously, not one at a time;

Memory! Using this method reduces the amount of memory your sketch will use.

Once again I will try and keep things as simple as possible. This article is written for Arduino boards that use the ATmega168 or ATmega328 microcontrollers (used in Arduino Duemilanove/Uno, Freetronics Eleven/EtherTen, etc). My Arduino Mega is out at the moment, so I will update the tutorial for Mega users when it is replaced. This tutorial is not applicable to the Arduino DUE.

First, we’ll use the I/O as outputs. There are three port registers that we can alter to set the status of the digital and analogue I/O pins. A port register can be thought of  as a special byte variable that we can change which is read by the microcontroller, therefore controlling the state of various I/O ports. We have three port registers to work with:

• D – for digital pins seven to zero (bank D)
• B – for digital pins thirteen to eight (bank B)
• C – for analogue pins five to zero (bank … C!)

Register C can control analogue pins seven to zero if using an Arduino with the TQFP style of ATmega328, such as the Nano or Freetronics EtherTen). For example:

It is very simple to do so. In void setup(), we use

DDRy = Bxxxxxxxx

where y is the register type (B/C/D) and xxxxxxxx are eight bits that determine if a pin is to be an input or output. Use 0 for input, and 1 for output. The LSB (least-significant bit [the one on the right!]) is the lowest pin number for that register. Next, to control a bank of pins, use

PORTy = Bxxxxxxxx

where y is the register type (B/C/D) and xxxxxxxx are eight status bits – 1 for HIGH, 0 for LOW. This is demonstrated in the following example:

// Example 43.1
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011
// Digital 0~7 set to outputs, then on/off using port manipulation

void setup()
{
  DDRD = B11111111; // set PORTD (digital 7~0) to outputs
}

void loop()
{
  PORTD = B11110000; // digital 4~7 HIGH, digital 3~0 LOW
  delay(1000);
  PORTD = B00001111; // digital 4~7 LOW, digital 3~0 HIGH
  delay(1000);
}

It sets digital pins 7~0 to output in void setup(). Then it alternates turning on and off alternating halves of digital pins 0~7.

At the start I mentioned that using port manipulation was a lot faster than using regular Arduino I/O functions. How fast? To test the speed of port manipulation vs. using digitalWrite(), we will use the following circuit:

… and analyse the output at digital pins zero and seven using a digital storage oscilloscope. Our first test sketch turns on and off digital pins 0~7 without any delay between PORTD commands – in other words, as fast as possible. The sketch:

// Example 43.1.1
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011
// Digital 0~7 set to outputs, then on/off using port manipulation

void setup()
{
  DDRD = B11111111; // set PORTD (digital 7~0) to outputs
}

void loop()
{
  PORTD = B11111111;
  PORTD = B00000000;
}

In the image below, digital zero is channel one, and digital seven is channel three:

Wow – check the frequency measurements – 1.1432 MHz! Interesting to note the longer duration of time when the pins are low vs. high.

[Update] Well it turns out that the extra time in LOW includes the time for the Arduino to go back to the top of void loop(). This can be demonstrated in the following sketch. We turn the pins on and off five times instead of once:

// Example 43.1.2
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011

void setup()
{
  DDRD = B11111111; // set PORTD (digital 7~0) to outputs
}

void loop()
{
  PORTD = B11111111;
  PORTD = B00000000;
  PORTD = B11111111;
  PORTD = B00000000;
  PORTD = B11111111;
  PORTD = B00000000;
  PORTD = B11111111;
  PORTD = B00000000;
  PORTD = B11111111;
  PORTD = B00000000;
}

And the results from the MSO. You can see the duty cycle is much closer to 50% until the end of the sketch, at which point around 660 nanoseconds is the time used between the end of the last LOW period and the start of the next HIGH:

Next we do it the normal way, using this sketch:

// Example 43.2
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011
// Digital 0~7 set to outputs, then on/off using digitalWrite()

void setup()
{
  for (int a=0; a<8; a++)
  {
    pinMode(a, OUTPUT);
  }
}

void loop()
{
  for (int a=0; a<8; a++)
  {
    digitalWrite(a, HIGH);
  }
  for (int a=0; a<8; a++)
  {
    digitalWrite(a, LOW);
  }
}

And the results:

That was a lot slower – we’re down to 14.085 kHz, with a much neater square-wave output. Could some CPU time be saved by not using the for loop? We tested once more with the following sketch:

// Example 43.3
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011
// Digital 0~7 set to outputs, then on/off using individual digitalWrite() 

void setup()
{
  for (int a=0; a<8; a++)
  {
    pinMode(a, OUTPUT);
  }
}

void loop()
{
  digitalWrite(0, HIGH);
  digitalWrite(1, HIGH);
  digitalWrite(2, HIGH);
  digitalWrite(3, HIGH);
  digitalWrite(4, HIGH);
  digitalWrite(5, HIGH);
  digitalWrite(6, HIGH);
  digitalWrite(7, HIGH);
  digitalWrite(0, LOW);
  digitalWrite(1, LOW);
  digitalWrite(2, LOW);
  digitalWrite(3, LOW);
  digitalWrite(4, LOW);
  digitalWrite(5, LOW);
  digitalWrite(6, LOW);
  digitalWrite(7, LOW);
}

and the results:

A small speed boost, the frequency has increased to 14.983 kHz. Hopefully you can now understand the benefits of using port manipulation. However there are a few things to take note of:

• You can’t control digital pins 0 and 1 (in bank D) and use the serial monitor/port. For example if you set pin zero to output, it can’t receive data!
• Always document your sketch – take pity on others who may need to review it later on and become puzzled about wchich bits are controlling or reading what!

Now to waste some electron flows by blinking LEDs. Using the circuit described earlier, the following sketch will create various effects for someone’s enjoyment (download):

// Example 43.4
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011
// Fun with 8 LEDs on digital 7~0

void setup()
{
  DDRD = B11111111; // set PORTD (digital 7~0)
  //  to output
}

byte a = B11111111;
byte b = B00000001;
byte c = B10000000;
byte e = B10101010;

void krider()
{
  for (int k=0; k<5; k++)
  {
    for (int z=0; z<8; z++)
    {
      PORTD = b << z;
      delay(100);
    }

    for (int z=0; z<8; z++)
    {
      PORTD = c >> z;
      delay(100);
    }
  }
}

void onOff()
{
  for (int k=0; k<10; k++)
  {
    PORTD = a;
    delay(100);
    PORTD = 0;
    delay(100);
  }
}

void invBlink()
{
  for (int z=0; z<10; z++)
  {
    PORTD = e;
    delay(100);
    PORTD = ~e;
    delay(100);
  }
}

void binaryCount()
{
  for (int z=0; z<256; z++)
  {
    PORTD = z;
    delay(100);
  }
  PORTD=0;
}

void loop()
{
  invBlink();
  delay(500);
  binaryCount();
  delay(500);
  krider();
  delay(500);
  onOff();
}

And here it is in real life:

Now to use the I/O pins as inputs. Again, it is very simple to do so. In void setup(), we use

DDRy = Bxxxxxxxx

where y is the register type (B/C/D) and xxxxxxxx are eight bits that determine if a pin is to be an input or output. Use 0 for input. The LSB (least-significant bit [the one on the right!]) is the lowest pin number for that register. Next, to read the status of the pins we simply read the byte:

PINy

where y is the register type (B/C/D).

So if you were using port B as inputs, and digital pins 8~10 were high, and 11~13 were low, PINB would be equal to B00000111. Really, that’s it!

Now for another demonstration using both inputs and outputs. We will use a push-wheel switch from Chapter 40 on our inputs (digital pins 8~11), and a seven segment LED display for output (on digtal pins 7~0 – segments dp then a~f). The following sketch reads the input from the switch, which returns 0~9 in binary-coded decimal. This value is then used in the function void disp() to retrieve the matching byte from the array “segments”, which contains the appropriate outputs to drive the seven segment LED display unit. Here is the sketch (download):

// Example 43.5
// tronixstuff.wordpress.com/tutorials > chapter 43
// John Boxall - October 2011
// inputs and outputs

byte segments[] = {
  B01111110, B00110000, B01101101, B01111001, B00110011, B01011011, B01011111, B01110000, B01111111, B01111011};
// digital pins 7~0 connected to display pins dp,a~g

void setup()
{
  DDRB = B00000000; // set PORTB (digital 13~8) to inputs
  DDRD = B11111111; // set PORTD (digital 7~0) to outputs
}

void disp(int z)
{
  PORTD = segments[z];
}

void loop()
{
    disp(PINB);
    delay(100);
}

And the ubiquitous demonstration video:

By now I hope you have an understanding of using port manipulation for your benefit. With a little effort your sketches can be more efficient in terms of speed and memory space, and also allow nifty simultaneous reading of input pins.

Have fun and keep checking into tronixstuff.com. Why not follow things on twitter, Google+, subscribe  for email updates or RSS using the links on the right-hand column, or join our Google Group – dedicated to the projects and related items on this website. Sign up – it’s free, helpful to each other –  and we can all learn something.