Jun

2

Pictures from Computex 2009

15 years ago, at the start of June | Leave a Comment

Two of my colleagues are in Taipei again attending Computex 2009 this week and the news is that there appear to be a whole lot of netbooks on view. As ever Nick is uploading pictures to Picasa so you can get some idea of what’s about. As the days progress Nick will update the images, but it is a slow, slow process. Keep your eyes peeled for the rest of the week – and make sure you’re not looking at last years’ pictures by mistake ;). Some of you might notice that some things like the Asus Skype phone have been around for a little while. In fact it has been around since September of last year.

Update 1: More pictures are available.

Update 2: Pictures from day 3 are available now.

Update 3: Pictures from day 4 are available now. There are also some shots of Taipei

Tagged with: computex • hardware • linux • software
June 2, 2009 10:51

Apr

25

G1 Assault on battery

15 years ago, at the end of April | Leave a Comment

I know I’m late to the party but a few days ago I received a small package containing a T-Mobile G1 phone running Android. The first thing I did was go to unlock-now to get myself a SIM unlock code. After submitting my details and paying up the 17.50 EUR it was just a case of waiting for the email, about 3 hours in my case. The email details everything you need to know when unlocking, particularly useful is the info on setting up a new APN which you’ll need to do if you aren’t using T-Mobile as your carrier. My O2 PAYG SIM from my iPhone worked fine and soon the G1 was syncing happily with gmail. One of the things that struck me about the G1 was all the negative comment on it’s size and feel that I’d read in various blogs and reviews. Personally I couldn’t see and issue with either the thickness of the device or the the look and feel of it and I wasn’t alone in that. I really don’t see what all the ‘complaints’ were about.

One of the things I had heard though, that I can confirm, is the poor battery life. Without installing any extra software the only real control over the G1’s power management seems limited to turning off the GPS or WiFi. However, there is a ray of light. After installing Power Manager things improved immensely but it still wasn’t enough. The solution I settled on was to buy an extended, 2400mAh, battery which came with a replacement back. The battery is physically 2 or 3 times the thickness of the original, which is why you get a replacement and the end result is a device that’s about as thick as a NOKIA N95.

So why didn’t I wait for a G2? Simple really. The G1 has a wonderful physical keyboard. I really can’t stand on screen keyboards, perhaps I’m a minority.

Of course the other reason that I wont be getting a G2 is that Vodafone are capping both daily and monthly usage when you sign up – 15mb / day and 500mb a month. Perhaps everyone at Vodafone still uses a 6310? Maybe they just don’t like the G2 or Android. Who knows?

Update (Sunday 10th May) : G1 Running cupcake has been running for 3 days with the extended battery and no charging!

_{»crosslinked«}

Tagged with: android • G1 • Google Phone • opensource
April 25, 2009 22:27

Mar

1

Gratifying Graphic Gizmos

15 years ago, at the start of March | 2 Comments

Yesterday the Current Cost unit actually got installed for real, something I was dreading. Dreading not because it was a difficult thing to do but simply because I expected the figure it would show would be terrible. Guess what, it wasn’t. In fact it seemed to be fairly close to that of other people who are monitoring theirs. The initial figure of 371 Watts didn’t last too long once the dishwasher and oven came on and an interesting spike appeared when the hob went on for the kettle.

Continuing with my sudden graphing fetish, as well as using a Current Cost unit I have dotted a number of little devices I’ve built with Xbee and Arduinos to measure temperatures (and other things) around the house. More information on those will come out later, but it’s an interesting project too.

Since I have the data for those it’d be rude not to graph them too, there’s a sample in the thumbnails below.

Not content with graphing them locally I picked up a Pachube invite and started graphing data there too. The process is simple really and as you can see on the right hand side of this page under ‘Graphy Stuff’, graph images can be embedded in your own or other sites. The really interesting thing about Pachube is that you can share your data with people who may be interested. The data they get to see is delayed by approximately 15 minutes. If you output your data in the eeml format then you can specify your own tags too.

Tagged with: 1-Wire • arduino • development • environment • hardware • monitoring • Pachube • software
March 1, 2009 8:47

Feb

13

Current Cost Capers

15 years ago, mid-February | 4 Comments

My ~~Unbioctium~~ Current Cost CC128 arrived on Thursday and although I’m not going to actually get a chance in install it for a couple of weeks I did have a play. The first thing to do once it was all up and paired with the base was get the output into my laptop. To do this you will need to buy the data cable or make up your own…

At first I tried to be smart with reading from the serial port, as that seems to be a bit unreliable I fell back on the old tried and tested method. I also decided from the outset that I was going to parse the XML that the the CC128 spat out rather than just filter the bits needed with regexp.

You will need to make sure you have the perl modules Device::SerialPort and XML::Simple installed. If you don’t have them then as root on your linux box do the following:

perl -MCPAN -e shell

then install each module, for example Device::SerialPort,

install Device::SerialPort

and here’s the script which is also available for download.Updated to make it more reliable.

#!/usr/bin/perl
 
use Device::SerialPort qw( :PARAM :STAT 0.07 );
use XML::Simple;
 
$port = "/dev/ttyUSB0";
 
$ob = Device::SerialPort->new($port) or die "Can not open port $port\n";
$ob->baudrate(57600);
$ob->write_settings;
$ob->close;
 
# using XML::Parser speeds xml parsing up lots!
$backend = 'XML::Parser';
$ENV{XML_SIMPLE_PREFERRED_PARSER} = $backend;
 
# we use this to only do 1 iteration (or not)
$escape=0; 
 
open(SERIAL, "<$port");
 
while($escape <= 0) {
	sleep(2);
 
	while ($line = <SERIAL>) {
 
		# for debug
		#print $line;
 
		$isValid = (index($line,"<msg>") != -1);
 
 
		if (!$isValid) { last; }
 
		print "This data is".($isValid==1?"":" not")." valid\n";
 
		# force XML::Simple to see this as a string not as a file
		# since XML::Simple is stupid and needs to be shot
 
		$line = "<fakeTag>$line</fakeTag>";
 
		$isHistoric = (index($line,"<hist>") != -1);
 
		$nref = XMLin($line,forcearray => 0);
 
		$ref = $nref->{msg};
 
		# just for reference, show if data is historic or not
 
		print "This data is".($isHistoric==1?"":" not")." historic\n";
 
 
		if (!$isHistoric) {
 
	        	$dsb        = 0 + $ref->{dsb};
        		$recordTime = $ref->{time};
        		$ccname     = $ref->{src};
        		$temp       = $ref->{tmpr};
			$ch1watts   = 0 + $ref->{ch1}->{watts};
			$sensor     = 0 + $ref->{sensor};
			$id         = $ref->{id};
			$type       = 0 + $ref->{type};
 
 
			if (defined $ref->{whatever}) {
				# do something based on whatever
 
			}
 
        		print "This $ccname was born $dsb days ago as at $recordTime - temperature is: $temp :: Current Watts in use on channel 1 are $ch1watts :: Sensor is $sensor, with an id of $id and a type of $type\n";
 
			# for cacti you'd probably just want to output CC_Temperature:$temp CC_Watts1:$ch1watts 
 
			# insert data into db
 
			# if you want to exit after a 'good' iteration set this to 1 otherwise set it to 0 (or don't change it to 1 :) );
			$escape=1;
		} else {
			#process or ignore historic data
		}
	}
}
 
close(SERIAL);

and here’s some example output, remember that I haven’t actually connected the loop around my electrical cable as yet.

This data is not historic
This CC128-v0.11 was born 1 days ago as at 21:16:44 - temperature is: 23.7 :: Watts used on channel 1 are 0 :: Sensor is 0, with an id of 02835 and a type of 1

Historic data isn’t processed at the moment but it would be trivial to do. I’m not processing it since I don’t really see the point if you’re going to store the results in a database.

Tagged with: hardware • monitoring • opensource • software
February 13, 2009 9:38

Feb

4

One Wireduino made easy

15 years ago, at the start of February | 12 Comments

For a very long time now I’ve been playing with the Arduino boards but it wasn’t until I recently discovered some DS1820 1-wire chips in my parts box that I started even thinking about 1-wire stuff. Connecting 1-wire devices to the Arduino is, well, amazingly simple. Using the sample circuit from the datasheet we can see the connections are few and far between.

Please note that this is the corrected circuit, previously the VDD was not tied to ground. While the circuit would work it would only give reliable results on very short runs. This circuit allows longer runs to the sensor.

Coding this up is also made extremely easy with the OneWire library. I’ve put together a little application below that will read all of the 1-wire devices on a bus. If you press L in the Arduino IDE serial monitor it will list all the 1-Wire device id’s and if you press T you’ll get the id and temperature reading.

/*
Simple 1-wire device reading
 
Pass T over serial to get temperatures
Pass L to list 1-wire devices
 
*/
#include <OneWire.h>
#include <stdio.h>
 
#define CONVERT     0x44
#define READSCRATCH 0xBE
#define SKIP_ROM    0xCC
#define MATCH_ROM   0x55
 
OneWire  ds(10);  // The DS18S20 is connected on pin 10
int ledPin = 13;  // flash an led on 13 - we all like flashing lights
 
void setup(void) {
  // initialize inputs/outputs
  // start serial port
  Serial.begin(9600);
  pinMode(ledPin, OUTPUT);  //we'll use the led to output a heartbeat
}
 
void loop(void) {
  int incomingByte = 0;
  if (Serial.available() &gt; 0) {
		// read the incoming byte:
		incomingByte = Serial.read();
 
		// say what you got:
		//Serial.print("I received: ");
		//Serial.println(incomingByte, DEC);
                if (incomingByte == 84) {
                  getStuff();
                  Serial.print("\r\n^");
                } else if (incomingByte == 76) {
                   listDevices();
                }
	}
 
}
 
void listDevices(void) {
 
  byte addr[8];
 
  ds.reset_search();
 
  while(ds.search(addr)) {
 
     digitalWrite(ledPin, HIGH);   // sets the LED on
 
     if ( OneWire::crc8( addr, 7) == addr[7])   {
         if ( addr[0] == 0x10) {
            // Make sure it is a DS18S20 device
            char buffer[512];
            sprintf(buffer,"Id:%02x%02x%02x%02x%02x%02x%02x%02x\r\n",
                       addr[0], addr[1], addr[2], addr[3], addr[4],
                       addr[5], addr[6], addr[7]);
              Serial.print(buffer);
          }
      }
  }
 
    Serial.print('^');
}
 
void getStuff(void) {
  byte i;
  byte present = 0;
  byte data[12];
  byte addr[8];
  char buff[15];
  float real_temp;
 
  float temp_count;
  float read_temp; 
 
  ds.reset_search();
 
  while(ds.search(addr))  {
 
     digitalWrite(ledPin, HIGH);   // sets the LED on
 
    if ( OneWire::crc8( addr, 7) != addr[7])  { // Check CRC is valid
        Serial.print("CRC is not valid");
    }
 
    if ( addr[0] != 0x10) {   // Make sure it is a DS18x20 device
      Serial.print("Device is not a DS18x20 family device.");
    }
 
    ds.reset();    // Reset device
    ds.select(addr);     // Select device
    ds.write(CONVERT,1);   // Issue Convert command 
 
    delay(1000);     // maybe 750ms is enough, maybe not
    digitalWrite(ledPin, LOW);    // sets the LED off
 
    present = ds.reset();  // Reset device
    ds.select(addr);  // Select device
    ds.write(READSCRATCH);  // Read Scratchpad
 
    for ( i = 0; i &lt; 9; i++) {  // we need 9 bytes
          data[i] = ds.read();
    }
 
    if(OneWire::crc8( data, 8) == data[8]) {  // Check CRC is valid
 
      // CRC is ok
      // Divide the temperature by 2 - note the » and
      // « need to be replaced with double < and >
      // wordpress kills the server when you try to 
      // save otherwise.
      read_temp=((data[1]«8) | data[0]) » 1 ;
     // Convert to real temperature
      temp_count=float(data[7] - data[6])/(float)data[7];
      real_temp = ((float)read_temp-0.25)+temp_count;
 
       // Convert float to ascii
      tempToAscii(real_temp,buff);
      char buffer[512];
      sprintf(buffer,"%02x%02x%02x%02x%02x%02x%02x%02x %s\r\n",
                  addr[0], addr[1], addr[2], addr[3], addr[4],
                  addr[5], addr[6], addr[7], buff);
 
      Serial.print(buffer);
 
    } else    {
       Serial.println("CRC Failed");
    }
  }    
 
void tempToAscii(double temp, char *buff) {
  int frac;
  //get three numbers to the right of the decimal point
  frac=(unsigned int)(temp*1000)%1000;
  itoa((int)temp,buff,10);
 
  strcat(buff,".");
  //put the frac after the decimal
  itoa(frac,&amp;buff[strlen(buff)],10);
}

There’s also a copy of the .pde file that you can download if you feel more inclined. It’s not an exact copy, I removed the commented out xml for the above version. It’s worth pointing out that this circuit is using what is called parasitic power, which means your cable run should be shortish. If you want to use an external power supply then pin 3 (VDD) on the DS18S20 should go to your external supply. This would allow for much longer runs of cable. Below is an example graph generated by cacti with 2 DS1820’s connected the the arduino. As you can see the temperatures aren’t exactly the same but this is to be expected since the accuracy of the DS1820 is +/- 0.5 ° C

The breaks in the graph were caused my me removing the USB plug from the PC it was connected to so that I could make some minor alterations to the circuit.

Josh asked how I was getting the input for Cacti, here’s the perl script I use

#!/usr/bin/perl
 
$output = `echo -en "T" >>/dev/ttyUSB0 && cat /dev/ttyUSB0`;
 
@lines = split(/\r\n/,$output);
$t=1;
 
foreach $line (@lines) {
 
       @spl = split(/\s+/,$line);
 
	$lc = @spl;
 
	if ($lc gt 1) {	
                $node=$spl[0];
		$val=$spl[1];
 
		print "Temperature_$node:$val ";
	}
$t++;
}

This outputs Temperature_x:n where x is the 1-Wire id and n is the temperature. It could probably do with tidying up and if you wanted to you could pass the 1-Wire id to the script to and only output the temperature for that particular 1-Wire device. A copy of the exported Cacti template can be downloaded, I’m hoping that you just need to import that into your Cacti after placing the script in the scripts directory. Just remember, under linux, if you unplug and replug in your Arduino you may need to set the tty port speed to 9600 (or whatever you put in your Serial.begin(); ). You can use the following command for that;

1
2
3

 
 
stty -F /dev/ttyUSB0 cs8 9600 ignbrk -brkint -icrnl -imaxbel -opost -onlcr -isig -icanon -iexten -echo -echoe -echok -echoctl -echoke noflsh -ixon -crtscts

Make sure that ttyUSB0 is what your Arduino is connected to first. If not you will need to change it in the perl script and on the line above.

Tagged with: 1-Wire • arduino • hardware • opensource • software
February 4, 2009 21:46

keep looking »

Current Electricity Use (15min)

iPhone/Webkit RSS Reader

Twitpic

Graphy Stuff

Nasty Spam Monkeys

56,158

Recently Written

Categories

Archives

Admin

Search

Jun

2