Doz' Blog

Friday, 12 August 2016

Arduino Fast GPS datalogger - with switchable units, configuration of U-blox GPS in runtime, and KML. The International Version !!

Well, here it is folks.

Ryan's Dad lives in Australia, and Ryan wanted to make a version of the fast GPS datalogger that his dad could use on is boat.

After sorting out the problems Ryan had with his U-blox GPS module here, we set about a few mods.

Firstly, was to be able to select the units of speed selectable when the unit first starts up. The sketch now looks for a button, connected between A3 and ground. This allows the unit to be set for KPH (Kilometres per hour), KTS (knots) and MPH (Miles per hour). This is selected by the new void configureUnits. It times out after a couple of seconds. The configureUnits void sets a flag, which selects which type of data is written to the Speed variable, and written to both the display and the card.

Here's Ryan's version, bread-boarded and tested on the bench.

Here's the new schematic, showing the addition of the "Select units" switch, and the extra connection to the GPS module to allow us to set the Ublox receiver's configuration.

And, of course, the new code:

/*
  Fast GPS data logger.
  (c) 12th August 2016 A.G.Doswell  Web Page http://andydoz.blogspot.co.uk/2016/07/arduino-fast-gps-datalogger-updated-now.html

License: The MIT License (See full license at the bottom of this file)

The sketch uses a U-Blox 6M GPS satellite module connected to the hardware serial interface,
  a 128x64 SDD1306 OLED display module connected as an I2C device to pins A4 (SDA)& A5 (SCL) and an SD card interface
  connected to pins 4 (CS), 11 (MOSI), 12 (MISO) & 13 (SCK) The arduino used is a Nano with 3.3v ATMEGA328P.
  Warning - Many SD card interfaces are not 5V.
  There's a toggle switch which connects pin 9 to GND for Record.
  Data is written to the card an impressive 4 times per second, in real-time.
  A pushbutton can be used, connected between pin A3 and GND to toggle through the available speed units, Knots, Kilometres per hour, or miles per hour.

The OzOled library is not quite "right" for this application, but works, and is lightweight and fast enough. Thanks Oscar!
  It is available here : http://blog.oscarliang.net/arduino-oled-display-library/
  TinyGPSPlus is available here :https://github.com/mikalhart/TinyGPSPlus

#include <SPI.h>
#include <Wire.h>
#include <OzOLED.h>
#include <TinyGPS++.h>
#include <SD.h>

int SatVal;                        // Number of satellites locked
int Speed;                         // Speed in
char SpeedA [4];                   // Speed as char
TinyGPSPlus gps;                   // Feed gps data to TinySGPSPlus
File myFile;                       // Start SD
int speedDigit;                    // Number of digits in Speedo display
boolean card;                      // Is there a card present?
boolean rec = false;               // Is the record switch low? (false = high)
int recPin = 9;                    // Record pin (Pull low to record data)
boolean header = false;            // Has the header been written?
String filename ;                  // Concaternated filename +.kml
String filename1 ;                 // Raw GPS Time filename.
char fileNameChar [14];            // file name as char array
int unitFlag = 0;                  // Flag used to set the speed units.

void setup()   {

Serial.begin(9600);// start the comms with the GPS Rx
  delay (6000); // allow the u-blox receiver to come up
  // send serial to update u-blox rate to 200mS
  Serial.write(0xB5);
  Serial.write(0x62);
  Serial.write(0x06);
  Serial.write(0x08);
  Serial.write(0x06);
  Serial.write(0x00);
  Serial.write(0xC8);
  Serial.write(0x00);
  Serial.write(0x01);
  Serial.write(0x00);
  Serial.write(0x01);
  Serial.write(0x00);
  Serial.write(0xDE);
  Serial.write(0x6A);
  Serial.write(0xB5);
  Serial.write(0x62);
  Serial.write(0x06);
  Serial.write(0x08);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x0E);
  Serial.write(0x30);
  delay (100);
  Serial.flush();
  // set 57,600 baud on u-blox
  Serial.write(0xB5);
  Serial.write(0x62);
  Serial.write(0x06);
  Serial.write(0x00);
  Serial.write(0x14);
  Serial.write(0x00);
  Serial.write(0x01);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0xD0);
  Serial.write(0x08);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0xE1);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x07);
  Serial.write(0x00);
  Serial.write(0x02);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0x00);
  Serial.write(0xDD);
  Serial.write(0xC3);
  Serial.write(0xB5);
  Serial.write(0x62);
  Serial.write(0x06);
  Serial.write(0x00);
  Serial.write(0x01);
  Serial.write(0x00);
  Serial.write(0x01);
  Serial.write(0x08);
  Serial.write(0x22);
  delay (100);
  Serial.end();// stop serial coms at 9,600 baud
  delay (100);
  Serial.begin (56700); // start serial coms at 56,700 baud.         
  pinMode (recPin, INPUT_PULLUP);  // recPin as input and pulled high
  OzOled.init();                   // initialze SDD1306 OLED display
  OzOled.sendCommand(0x8d);        // Set displays inbuilt inverter on
  OzOled.sendCommand(0x14);
  OzOled.setBrightness(0xFF);      // ... and brightness to max
  OzOled.clearDisplay();           // Clear the screen
  OzOled.setNormalDisplay();       // Set display to Normal mode
  pinMode(10, OUTPUT);             // CS for SD card, wether it likes it, or not.
  OzOled.clearDisplay();
  pinMode (A3, INPUT);             // push button connected between A£ and GND
  digitalWrite (A3, HIGH);         // Pull A3 high.

configureUnits ();               // Set the units to use.

if (!SD.begin(4)) {              //Check SD card is present
    OzOled.printString("SD card fail    ", 0, 7);
    card = false;
  }
  else {
    OzOled.printString("SD card OK      ", 0, 7);
    card = true;
  }

OzOled.printString("Sats:", 0, 0); // Set up display
  OzOled.printString("Speed:", 0, 1);
}

void loop() {
  while (Serial.available() > 0) //While GPS message received
    if (gps.encode(Serial.read())) {
      displayInfo();
    }
}

void displayInfo() {               // Display the data
  SatVal = int(gps.satellites.value());

if (unitFlag == 3) {
    unitFlag = 0;
  }
  if (unitFlag == 0) {
    Speed = int(gps.speed.kmph());
  }
  if (unitFlag == 1) {
    Speed = int(gps.speed.knots());
  }
  if (unitFlag == 2) {
    Speed = int(gps.speed.mph());
  }

itoa (Speed, SpeedA, 10);
  OzOled.printNumber ((long)SatVal, 6, 0);
  OzOled.printBigNumber (SpeedA, 7, 1);
  speedDigit = strlen(SpeedA);    // get length of Speed , and delete left behind zero if req'd
  if (speedDigit == 1) {
    OzOled.printBigNumber(" ", 10, 1);
  }

if (digitalRead(recPin) == HIGH && card == true) { //If the record switch is high and the card is OK, write card data, and update display
    if (header == false) {
      if (gps.date.year() > 2000 && gps.location.age() < 1500) { // makes sure we have (at some stage since power up) a valid time, and the current fix is valid.{ // if the header hasn't been written, create a new filename with current time HHMMSSCC.kml
        String filename1(gps.time.value());
        String filename = filename1 + ".kml";
        filename.toCharArray(fileNameChar, 13);
      }
    }
    OzOled.printString("SD card OK REC", 0, 7);
  }
  if (digitalRead(recPin) == LOW && card == true ) { // if record switch is off, set flag and update display
    OzOled.printString("SD card OK    ", 0, 7);
  }
  if (digitalRead(recPin) == LOW && header == true) { // if record switch is off, but the header flag is set, write the footer.
    writeFooter();
  }
  if (card == true && digitalRead(recPin) == HIGH && (gps.speed.kmph() > 0)) {
    writeInfo(); //write the data to the SD card
  }
  smartDelay(200);
}

// This custom version of delay() ensures that the gps object
// is being "fed".
static void smartDelay(unsigned long ms)
{
  unsigned long start = millis();
  do
  {
    while (Serial.available())
      gps.encode(Serial.read());
  } while (millis() - start < ms);
}

void writeInfo() { //Write the data to the SD card fomratted for google earth kml

myFile = SD.open(fileNameChar, FILE_WRITE);
  if (header == false) { // If the header hasn't been written, write it.
    if (myFile) {
      myFile.print(F("<?xml version=\"1.0\" encoding=\"UTF-8\"?> <kml xmlns=\"http://earth.google.com/kml/2.0\"> <document>"));
      header = true; // header flag set.
    }
  }
  if (myFile) { // write current GPS position data as KML
    myFile.println(F("<placemark>"));
    myFile.print(F("<name>"));
    myFile.print(Speed, 1);
    myFile.println(F("</name>"));
    myFile.print(F("<point><coordinates>"));
    myFile.print(gps.location.lng(), 6);
    myFile.print(F(","));
    myFile.print(gps.location.lat(), 6);
    myFile.println(F("</coordinates></Point></Placemark>"));
    myFile.close(); // close the file:
  }
}

void writeFooter() { // if the record switch is now set to off, write the kml footer, and reset the header flag
  myFile = SD.open(fileNameChar, FILE_WRITE);
  header = false;
  if (myFile) {
    myFile.println(F("</Document>"));
    myFile.println (F("</kml>"));
    myFile.close ();
  }
}
void configureUnits () {
  OzOled.printString("Set units       ", 0, 7);
  for (int i = 0; i < 30000; i++) {
    if (digitalRead (A3) == LOW) {
      unitFlag ++;
    }
    if (unitFlag == 3) {
      unitFlag = 0;
    }
    if (unitFlag == 0) {
      Speed = int(gps.speed.kmph());
      OzOled.printString("KPH", 7, 6);
    }
    if (unitFlag == 1) {
      Speed = int(gps.speed.knots());
      OzOled.printString("KTS", 7, 6);
    }
    if (unitFlag == 2) {
      Speed = int(gps.speed.mph());
      OzOled.printString("MPH", 7, 6);
    }
    delay (10); // debounce delay
  }
}

/*
   Copyright (c) 2016 Andrew Doswell

Permission is hereby granted, free of charge, to any person obtaining a copy
   of this software and associated documentation files (the "Software"), to deal
   in the Software without restriction, including without limitation the rights
   to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
   copies of the Software, and to permit persons to whom the Software is
   furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
   all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
   IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
   FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
   AUTHOR(S) OR COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
   LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
   OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
   THE SOFTWARE.
*/

Thursday, 4 August 2016

Automatic configuration of ublox M6 GPS receivers for Arduino during runtime.

My good friend Ryan wants to make a GPS tracker with KML, as detailed on this very website.

He wants to make a couple of revisions, which I'll publish here later...

But, he runs into a problem.

It doesn't work....

After some careful examination of the receiver, and checking the configuration with the U-blox U-center application, it becomes apparent that the receiver isn't storing the new configuration (57,600 baud serial, and 200mS update rate) after a power cycle... It defaults to 9,600 baud and a rate of 1s.

... damn it, a faulty receiver. So another is tried. Same results...

So, presumably the receivers aren't faulty....

This is one of the guilty receivers....

.... notice anything??

Here's one of mine....

... and there's the difference. See where the "faulty" receiver has a white circle marked with a 2?? Mine has a battery fitted. After a quick look at the datasheet for the U-blox chip, this is a RAM back-up battery. The configuration settings are stored in RAM. On Ryan's receiver these are lost as soon as the supply is removed.

Ryan looked into getting the correct battery and fitting it, but it's got one helluva specification, and was about £7... which is a touch expensive considering the GPS module was £9!

How about sending the configuration data from the sketch every time it starts up?

Why not....

I did a few experiments with two FTDI boards connected to a receiver...

In this picture you can see I've connected the lower FTDI to the receiver, RX to TX, TX to RX. This FTDI is used to communicate to the U-Center application. The top FTDI is used to "sniff" the serial data on the TX line on the first FTDI. Only it's ground and the RX pin are connected. This is so I can run a terminal program and grab the configuration data when I click "send" on the u-center application.

After a bit of buggering around with Real term (download from here.) I grabbed the data, in hex, being sent to the receiver from u-center.

This is what it looks like:

Set to 200mS rate

B5 62 06 08 06 00 C8 00 01 00 01 00 DE 6A B5 62 06 08 00 00 0E 30

set 57,600 baudrate

B5 62 06 00 14 00 01 00 00 00 D0 08 00 00 00 E1 00 00 07 00 02 00 00 00
00 00 DD C3 B5 62 06 00 01 00 01 08 22

So, now we just need to send that each time we power up...

You'll need to add a wire from the Arduino's TX pin to the receiver's RX pin, so the receiver can receive the configuration commands.

Here's the code snippet, you'll just need to add it in the void setup ()

.... and so on. I'll post up the whole thing , and the modifications we made to Ryan's tracker soon!

Saturday, 30 July 2016

The Mirfield Electronics ME-64270 Quad-band 6m, 4m, 2m and 70cm antenna review.

Now I'm not usually one to review things, but this is a bit unusual...

It's a 4 band ham radio antenna. It covers 6m (50 MHz), 4m (70MHz), 2m (145 MHz) and 70cm (430 MHz)

The reason I want to give it some publicity, is the fact that it's small, at only 1.2m in length, covers 4m, and I had some good service from the supplier....

I'd been contemplating sticking something on the roof for a while, to get better access into the local repeaters for a natter on 2m and 70cm every so often. It's got to be fairly small so as not to aggravate the neighbours or the wife....

Then I spotted a small ad in the back of RadCom magazine for a quad-band co-linear antenna. It wasn't too pricey either, having 4m is really good too. Most quad-bands are 10m, 6m,2m and 70cm. I have a magnetic loop that works well on 10m, so that doesn't really interest me.

I placed an order on the website at www.mirfield-electronics.co.uk, and very soon a cardboard tube arrived.

So, it's Sunday. The weather was looking promising, I had half a reel of RG-8 co-ax cable that would do the job, only to be thwarted. One of the radial elements didn't have a thread, and the mounting brackets seemed to be missing... :(

I telephoned the number on Mirfield's website and left a message...

... and a super chap called Martin rang me back about half an hour later (on a Sunday too!). I explained the situation.

He was great, "What WAS I thinking, must have had a bad day!" He promised to post next day, and good to his word, a nice new radial, and the missing mounting brackets turned up on Tuesday.

So, It's Saturday, and once again the weather looks favourable.

I gather the parts and assemble the antenna....

The antenna is fitted with a SO259 "UHF" connector at it's base. Quite why anyone would call it UHF is beyond me. It's a pretty poor connector at VHF, let alone UHF.... anyway, it seemed good enough for Icom to put on the back of their UHF radios, so maybe it's just me... an N-type connector would have been preferable, but at this price point, I'm more than happy.

Having soldered a PL259 on to the RG-8 co-ax, slide the coax and connector through the support tube, making sure the screw hole is pointing towards the antenna. I also slid a piece of heat-shrink tubing on too... Don't do what I did, and forget to slide the mounting clamps onto the mounting tube first!

A couple of laps of self-amalgamating tape are used to give the connector some protection against the elements, and the heat-shrink makes a tidy job ....

... slide the mounting tube onto the bottom of the antenna, and do up the screw. The mounting tube is steel, and I wonder if it could have been made a little thicker for strength. Seems to work OK though...

When doing up the mounting clamps, I found it helpful to keep them flat against the floor, as when you do up the screws, they tend to move about a bit....

I've left the radials off until I get the thing up on the roof...

So, a couple of hours later, and it's up....

(Glad I did it too, the TV co-ax was in an awful state!)

So how's it perform?

Gain figures are quoted as 2.5dBi on 6m, 2.5dBi on 4m, 3.5dBi on 2m, and 5.5dBi on 70cm. I've got no way of checking those, but the antenna opens up repeaters on 2m I haven't heard for years, and fully quietening on local repeaters at just 5 watts. The antenna is rated to 60 watts.

VSWR is better than 1.5:1 on all the frequencies I'm interested in.

Good points...

The service from Mirfield was superb.
It's short.
It's cheap.
It's largely well made, and looks like it will stand the test of time.
70 MHz!

Minor grumbles...

SO259 "good buddy" connector.
Mounting tube could be more robust.

Friday, 29 July 2016

CCT811 Video modulator massive failure.

Remember my warnings and concern about the CCT811 video modulator?

This is the offending model in question. Apparently it's also being sold under various guises, one model quoted to me is the RF9000. If it looks like this, I'd be seriously concerned....

I wrote about it here. Please take a moment to read this, if you haven't seen it before.

Well, settling down for an afternoon of vintage TV fun, I switched the video rack on....

Crack, crack, crack, bang! Uh-oh...

"It's bound to be the modulator" I thought.

I was right. It had opened the fuse I'd retro-fitted. (If you have one of these awful modulators, time to stop using it!) Damned glad I fitted one... lord knows what would have happened if it had tried to short out the mains without the protection of a fuse... Fire would have been a REAL possibility.

I opened it up and expected to find a mass of blackened bits, but no. Now I've giving up on this I thought... but it's so useful!

It looks as though the insulation had failed on the transformer, and destroyed the semi-conductors in the drive-side of the supply.

I thought I'd remove the ghastly existing "switched-mode" blocking oscillator, and replace it with something ...

1) Electrically safe.

2) Reliable.

OK. First things first. To find out if it still functions.

I removed the transformer, and connected the workshop power supply between the end of the rectifier diode, and ground....

It seems to want about 6.5 VDC to operate, and has even remembered the settings it had before the power supply failed... good.

It's drawing about 165mA at that voltage, so a supply is not too challenging!

I decided to have a look in the drawer of redundant wall-warts to see if I could find anything suitable, and I spy one of my favourites, an old Nokia phone charger!

Now I always pick these up from boot-sales for a few pence, and are very useful. This one is rated at 3.7 Volts, at 350mA...

"But Andy, you said the modulator needs 6.5 volts to work" ... and so it does, but this particular Nokia charger is unregulated. Off-load it makes about 10 Volts.... Will it make our 6.5 volts at 165mA?

I solder the white +VE lead to the anode of D107, and the black lead to the far right hole (viewed from the rear of the modulator) left behind when I removed the transformer...

... and switch on....

Eureka! It works.

So it's now electrically safe, and I can sit back, and watch some "proper" TV....

Friday, 22 July 2016

Arduino GPS master clock with 433/315 MHz transmitter

I've got a few clock projects in the back of my mind, and this is a bit of an aside to them all. I also wish to upgrade the Astronomical clock project, as the DS1307 often drifts about, losing and gaining seconds as it feel like it...

The idea is to have a master clock, GPS locked with built in 433 MHz transmitter, which will send an accurate time signal to the other clocks to bring them into sync. The transmitter will be triggered randomly, at least once every 24 hours, or by a push button connected to pin 12.

I'll be re-using the code from the quick and dirty 7-segment GPS clock here, and some of the code for the transmit function from the dehumidifier project.

I've added a 4x20 LCD display, with I2C piggyback interface board, and the u-blox GPS module we've used in the GPS logger.

The GPS module will be configured by the sketch itself, to 200mS refresh rate.

There's two sketches, the GPS clock itself, and a simple test receiver sketch to check everything's working as it should.

The I2C display is a generic display, purchased from eBay, and uses the PCF8574 I2C receiver IC. I tried a couple of different libraries, but eventually settled with the library from https://bitbucket.org/fmalpartida/new-liquidcrystal/downloads which works well with my display.
Well also use the virtual wire library to handle our comms, https://www.pjrc.com/teensy/td_libs_VirtualWire.html and the most excellent TinyGPS++ http://arduiniana.org/libraries/tinygpsplus/

There's an LED connected to pin 9, which blinks on and off every time the seconds are updated.

Here's the schematic...

... and the hardware lashed up for testing.

The second arduino is acting as our test receiver. There's no RF module in there as yet.. It's coupled to the clock arduino, GND is coupled, and pin 10 on the clock is connected to pin 9 on the receiver board. It's also supplying 5V to the other board.

Here's the data being displayed. Date (in UK format DD/MM/YYYY) , UTC time, latitude, longitude, no of satellites in use, horizontal dilution of precision and the number of seconds until the next transmission.

Here's the output from our terminal window, connected to the receiver board, showing the time and date are being correctly received when A3 is pulled to ground.

So, the final build looks like this...

You can see the antenna is simply a straight piece of wire, 173mm long (for 433 MHz modules, other frequencies will need a different length)

And assembled. Managed to cut the right sized hole for the display for once!

Here's the GPS receiver, which I'm going to mount remotely on the roof, so it get's good reception. I've stuck it into of a water-proof IP rated box, and fitted a Buccaneer connector.

OK, so to the code....

This is the transmitter code:

/*
  GPS Master clock
  (C) Andy Doswell 22nd July 2016
  License: The MIT License (See full license at the bottom of this file)

Schematic can be found at www.andydoz.blogspot.com/

Uses the very excellent TinyGPS++ library from http://arduiniana.org/libraries/tinygpsplus/
  and also the fabulous Virtualwire library from https://www.pjrc.com/teensy/td_libs_VirtualWire.html
  and the most superb I2C LCD library from https://bitbucket.org/fmalpartida/new-liquidcrystal/downloads

pin 0 (RX)  is connected to the GPS TX pin.
  A3 is connected to a "push to transmit" button.
  Pin 10 is connected to the transmitter (433MHz or whatever's legal in your area)
  LCD is a 20 x 4 I2C display fitted with PCF8574 IC

Sketch transmits a clock signal at a random time (never more than 24 hours) to sync up other clocks.

*/
#include <Wire.h>
#include <LiquidCrystal_I2C.h>
#include <TinyGPS++.h>
#include <VirtualWire.h>
#include <VirtualWire_Config.h>

LiquidCrystal_I2C  lcd(0x3f, 2, 1, 0, 4, 5, 6, 7); // I2C addreSerial 0x3f, 20x4 display
TinyGPSPlus gps; //TinyGPS++ claSerial
static const uint32_t GPSBaud = 9600; //Our GPS baudrate.
int hourTen; //tens of hours
int hourUnit; //units of hours
int minTen; // you get the idea..
int minUnit;
int secTen;
int secUnit;
unsigned long transmitTimer;
unsigned long secsPastMidnight;
unsigned long SerialTimeout = 0;
unsigned long secsToMidnight;
unsigned long timerPrint;
boolean displayLoop = false;

struct TXData
{
  int TX_ID;
  int Hours;
  int Mins;
  int Secs;
  int Day;
  int Month;
  int Year;
};

void setup() {

Serial.begin(GPSBaud); // start the comms with the GPS Rx
  transmitTimer = random(0, 86399); // transmit sometime in the next 24 hours
  lcd.begin(20, 4); // set up the LCD as 20 x 4 display
  lcd.setBacklightPin(3, POSITIVE);
  lcd.setBacklight(HIGH);
  lcd.clear();
  // Virtualwire setup
  vw_set_tx_pin(10); // TX pin set to 10
  vw_set_rx_pin(9); // RX pin set to a spare pin, even though we don't use it.
  vw_setup(1200); //sets virtualwire for a tx rate of 1200 bits per second
  pinMode (A3, INPUT); // push to transmit connected to A3 and GND
  digitalWrite (A3, HIGH); // A3 pulled high
  randomSeed(analogRead(A0));
  transmitTimer = random(0, 86399); // set a random transmit time.
}

void displayData() {

SerialTimeout = millis();
  hourUnit = (gps.time.hour() % 10);
  hourTen = ((gps.time.hour() / 10) % 10);
  minUnit = (gps.time.minute() % 10);
  minTen = ((gps.time.minute() / 10) % 10);
  secUnit = (gps.time.second() % 10);
  secTen = ((gps.time.second() / 10) % 10);
  secUnit = (gps.time.second() % 10);

if (displayLoop == false) { // this bit of code is to overcome a bit of an odd bug, and I don't know why it happens,
    lcd.setCursor (0, 1);   // but Satellites in view and HDOP return 0 if you get them along with location, regardless of order. So we only update the position every 100 cycles.
    lcd.print("  ");
    lcd.print (gps.location.lat(), 4);
    lcd.print (" ");
    lcd.print (gps.location.lng(), 4);
    lcd.print ("   ");
    displayLoop = true;
  }
  else {
    lcd.setCursor (0, 2);
    lcd.print ("  Sat:");
    lcd.print (gps.satellites.value());
    lcd.print (" HDOP:");
    lcd.print (gps.hdop.value());
    lcd.print (" ");
    displayLoop = false;
  }
  lcd.setCursor(0, 0);
  lcd.print (" ");
  lcd.print(gps.date.day());
  lcd.print("/");
  lcd.print(gps.date.month());
  lcd.print("/");
  lcd.print(gps.date.year());
  lcd.print(" ");
  lcd.print(hourTen);
  lcd.print(hourUnit);
  lcd.print(":");
  lcd.print(minTen);
  lcd.print (minUnit);
  lcd.print (":");
  lcd.print (secTen);
  lcd.print (secUnit);
  lcd.setCursor (0, 3);
  lcd.print ("    TX in:");

lcd.print (timerPrint);
  lcd.print (" ");

}

void displayNoSerial() {
  lcd.clear();
  lcd.print ("ERROR");
  lcd.setCursor (0, 1);
  lcd.print ("Serial data comms");
  lcd.setCursor (0, 2);
  lcd.print ("Check cable");
  delay (2000);
}

void displayNoGPS() {
  lcd.clear();
  lcd.print("Warning");
  lcd.setCursor(0, 1);
  lcd.print("No or poor GPS data");
  lcd.setCursor(0, 2);
  lcd.print("Sats in view=");
  lcd.print (gps.satellites.value());
}

void transmitTime () { // transmits the current time and date.

struct TXData payload; //Loads the Data struct with the payload data
  lcd.clear ();
  lcd.print ("Transmitting clock sync");
  lcd.setCursor (0, 1);
  payload.TX_ID = 1;
  payload.Hours = gps.time.hour();
  payload.Mins = gps.time.minute();
  payload.Secs = gps.time.second();
  payload.Day = gps.date.day();
  payload.Month = gps.date.month();
  payload.Year = gps.date.year();
  vw_send((uint8_t *)&payload, sizeof(payload)); //Transmits the struct
  vw_wait_tx(); //waits for the TX to complete

transmitTimer = random(0, 86399);
  lcd.clear();
}

static void smartDelay(unsigned long ms)
{
  unsigned long start = millis();
  do
  {
    while (Serial.available())
      gps.encode(Serial.read());
  } while (millis() - start < ms);
}

void loop()
{

smartDelay (0); // get the gps data

if (millis() > SerialTimeout + 1500 ) { // detects if the Serial comms has failed. // if the position hasn't become valid in 1.5 seconds then display No serial.

displayNoSerial();

SerialTimeout = millis (); // reset the serial time out
  }
  secsPastMidnight =  gps.time.second() + (gps.time.minute() * 60) + (gps.time.hour() * 3600UL); // number of seconds past midnight
  secsToMidnight = 86399UL - secsPastMidnight ; // number of second to midnight.

if (transmitTimer >= secsPastMidnight) { // ensures the seconds to TX display is correct.
    timerPrint = transmitTimer - secsPastMidnight;
  }
  else {
    timerPrint = transmitTimer + secsToMidnight;
  }

if (secsPastMidnight == transmitTimer) { //if it's time to transmit, do it.
    transmitTime();
  }

if (digitalRead (A3) == LOW) { // Checks the "push to transmit" button, and transmits if it's low.
    transmitTime();
  }

if (gps.location.isUpdated() ) { // if there's new data, display it.
    displayData();
  }

}

/*
   Copyright (c) 2016 Andrew Doswell

The above copyright notice and this permiSerialion notice shall be included in
   all copies or substantial portions of the Software.

Any commercial use is prohibited without prior arrangement.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
   IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
   FITNESerial FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
   AUTHOR(S) OR COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
   LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
   OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
   THE SOFTWARE.
*/

and the test receiver code...

/*
  GPS Master clock receiver
  (C) Andy Doswell 22nd July 2016
  License: The MIT License (See full license at the bottom of this file)

Schematic can be found at www.andydoz.blogspot.com/

#include <VirtualWire.h>
#include <VirtualWire_Config.h>

int TX_ID = 1;
int Hours;
int Mins;
int Secs;
int Day;
int Month;
int Year;
boolean Output;
char buffer[5];
int led = 13;

void setup() {
  pinMode(led, OUTPUT);
  //Comms set up
  Serial.begin(9600);
  // Virtualwire setup
  vw_set_tx_pin(10); // Even though it's not used, we'll define it so the default doesn't interfere with the LCD library.
  vw_set_rx_pin(9); // RX pin set to pin 9
  vw_setup(1200); //sets virtualwire for a tx rate of 300 bits per second, nice and slow!
  vw_rx_start();
}
void loop()
{
  typedef struct rxRemoteData //Defines the received data, this is the same as the TX struct
  {
    int TX_ID;
    int Hours;
    int Mins;
    int Secs;
    int Day;
    int Month;
    int Year;

};

struct rxRemoteData receivedData;
  uint8_t rcvdSize = sizeof(receivedData);

if (vw_get_message((uint8_t *)&receivedData, &rcvdSize))
  {
    if (receivedData.TX_ID == 1)     { //Only if the TX_ID=1 do we process the data.
      int TX_ID = receivedData.TX_ID;
      int Hours = receivedData.Hours;
      int Mins = receivedData.Mins;
      int Secs = receivedData.Secs;
      int Day =  receivedData.Day;
      int Month =  receivedData.Month;
      int Year =  receivedData.Year;
      digitalWrite(led, HIGH);

// writes the received data to the serial interface

Serial.print(itoa((Hours), buffer, 10));
      Serial.print(":");
      Serial.print(itoa((Mins), buffer, 10));
      Serial.print(":");
      Serial.print(itoa((Secs), buffer, 10));
      Serial.print(" ");
      Serial.print(itoa((Day), buffer, 10));
      Serial.print("/");
      Serial.print(itoa((Month), buffer, 10));
      Serial.print("/");
      Serial.println(itoa((Year), buffer, 10));
      digitalWrite(led, LOW);

}

}
}
/*
   Copyright (c) 2016 Andrew Doswell

The above copyright notice and this permiSerialion notice shall be included in
   all copies or substantial portions of the Software.

Any commercial use is prohibited without prior arrangement.

Now all that remains is to box it up, and create some clocks!

Wednesday, 13 July 2016

Arduino Fast GPS datalogger updated. Now with KML!

Now Updated (again) with selectable speed units, and configuration of U-blox GPS in runtime, avoiding the need to use U-Center to configure the receiver. Click Here!

The fast GPS datalogger project is the most popular project on here. Link.

It's been working well on the race car since it's inception, and has provided some really useful data to improve the performance of the driver!

What isn't so great is how the data is stored. It stores everything, regardless of fix, number of races etc, so it's a headache to filter out each race, and remove useless invalid data, either without a time stamp or valid position or speed, and then turn it into something that could be used. I have been painstakingly converting the data into KML, which we can directly read with mapping software, such as Google Earth.

I've made some modifications to the software to sort out some of these issues... here's the new features:

Each time the record switch is pressed, valid GPS data is required, and once obtained, a new file name is created from the current time, HHMMSSCC.kml (in UTC, CC representing centiseconds here), and the KML header file is written.

Now the file is built as in the previous version, except this time in KML format.

Speeds below 5 MPH are not recorded.

When record is switched off, the KML file is completed and closed.

So once the card is removed from the unit, simply read the kml file from the card, and open in google earth.

So I went out for a quick jaunt in England's green and pleasant....

... and loaded the KML file straight into google earth.

No more tedious editing required!

No changes to the hardware are required. Just remember to disconnect the RX line from the GPS module when attempting to load the new sketch.

Here's the code:

/*
  Fast GPS data logger.
  (c) 13th July 2016 A.G.Doswell

License: The MIT License (See full license at the bottom of this file)

#include <SPI.h>
#include <Wire.h>
#include <OzOLED.h>
#include <TinyGPS++.h>
#include <SD.h>

static const uint32_t GPSBaud = 57600;
int SatVal;                        // Number of satellites locked
int Speed;                         // Speed in MPH
char SpeedA [4];                   // Speed as char
TinyGPSPlus gps;                   // Feed gps data to TinySGPSPlus
File myFile;                       // Start SD
int speedDigit;                    // Number of digits in Speedo display
boolean card;                      // Is there a card present?
boolean rec = false;               // Is the record switch low? (false = high)
int recPin = 9;                    // Record pin (Pull low to record data)
boolean header = false;            // Has the header been written?
String filename ;                  // Concaternated filename +.kml
String filename1 ;                 // Raw GPS Time filename.
char fileNameChar [14];            // file name as char array

void setup()   {

Serial.begin(GPSBaud);           // Start GPS serial comms
  pinMode (recPin, INPUT_PULLUP);  // recPin as input and pulled high
  OzOled.init();                   // initialze SDD1306 OLED display
  OzOled.sendCommand(0x8d);        // Set displays inbuilt inverter on
  OzOled.sendCommand(0x14);
  OzOled.setBrightness(0xFF);      // ... and brightness to max
  OzOled.clearDisplay();           // Clear the screen
  OzOled.setNormalDisplay();       // Set display to Normal mode
  pinMode(10, OUTPUT);             // CS for SD card, wether it likes it, or not.
  OzOled.clearDisplay();

OzOled.printString("Sats:", 0, 0); // Set up display
  OzOled.printString("Speed:", 0, 1);
  OzOled.printString("MPH", 7, 6);
}

void loop() {
  while (Serial.available() > 0) //While GPS message received
    if (gps.encode(Serial.read()))
      displayInfo();
}

void displayInfo() {               // Display the data
  SatVal = int(gps.satellites.value());
  Speed = int(gps.speed.mph());
  itoa (Speed, SpeedA, 10);
  OzOled.printNumber ((long)SatVal, 6, 0);
  OzOled.printBigNumber (SpeedA, 7, 1);
  speedDigit = strlen(SpeedA);    // get length of Speed , and delete left behind zero if req'd
  if (speedDigit == 1) OzOled.printBigNumber(" ", 10, 1);
  
  if (digitalRead(recPin) == HIGH && card == true) { //If the record switch is high and the card is OK, write card data, and update display
    if (header == false) {
      if (gps.date.year() > 2000 && gps.location.age() < 1500) { // makes sure we have (at some stage since power up) a valid time, and the current fix is valid.{ // if the header hasn't been written, create a new filename with current time HHMMSSCC.kml
        String filename1(gps.time.value());
        String filename = filename1 + ".kml";
        filename.toCharArray(fileNameChar, 13);

}
    }
    OzOled.printString("SD card OK REC", 0, 7);
  }
  if (digitalRead(recPin) == LOW && card == true ) { // if record switch is off, set flag and update display
    OzOled.printString("SD card OK    ", 0, 7);
  }

if (digitalRead(recPin) == LOW && header == true) { // if record switch is off, but the header flag is set, write the footer.
    writeFooter();
  }

if (card == true && digitalRead(recPin) == HIGH && (gps.speed.mph() >= 5)) {
    writeInfo(); //write the data to the SD card
  }
  smartDelay(200);

}

void writeInfo() { //Write the data to the SD card fomratted for google earth kml

myFile = SD.open(fileNameChar, FILE_WRITE);
  if (header == false) { // If the header hasn't been written, write it.
    if (myFile) {
      myFile.print(F("<?xml version=\"1.0\" encoding=\"UTF-8\"?> <kml xmlns=\"http://earth.google.com/kml/2.0\"> <document>"));
      header = true; // header flag set.
    }
  }
  if (myFile) { // write current GPS position data as KML
    myFile.println(F("<placemark>"));
    myFile.print(F("<name>"));
    myFile.print(gps.speed.mph(), 1);
    myFile.println(F("</name>"));
    myFile.print(F("<point><coordinates>"));
    myFile.print(gps.location.lng(), 6);
    myFile.print(F(","));
    myFile.print(gps.location.lat(), 6);
    myFile.println(F("</coordinates></Point></Placemark>"));
    myFile.close(); // close the file:
  }
}

/*
   Copyright (c) 2016 Andrew Doswell

The above copyright notice and this permission notice shall be included in
   all copies or substantial portions of the Software.

Sunday, 10 July 2016

Grundig TK120 repair

Gareth (a fine chap from the motherland) walks into the workshop with a bin bag under his arm....

"My Father-in-law's dug this tape recorder out of the attic. It's got a recording of him on it, but he says there's no sound. Can you have a look?"

Yeah ... why not?

It's a Grundig TK120, in distressed condition. It looks like it's been left rather too close to the fire!

... and after removing the tape and the top cover, the immediate issue is all the rubber parts have disintegrated....

Now the belts are no real problem... but the drive idler tyre is going to be...

After some head scratching, and some looking about for suitable parts, an O-ring is super-glued to the remaining idler wheel....

Now to replace the belts. So I unscrewed the four screws with blobs of paint on, to remove the top plate. DO NOT DO THIS!

FATANG! All the spring loaded control levers fired out of their mounts.... Thankfully their positions are quite simple to work out, but very fiddly.... several hours later and the thing is back together....

Here's how it should have been done!..

Underneath the deck is a small plate, with two screws... undo those to gain access to the bottom of the capstan flywheel....

... and work the belts about the bottom of the flywheel, through the small hole. It's still fiddly, but much less agro than re-aligning the top mechanism!

You'll need to remove the take-up reel and clutch to replace the take-up drive belt. The belt itself is straight forward. The clutch reassembly is a bit tricky...

This is the bottom of the deck, showing the new tape counter belt... Remove the little pulley and it's clip from the take up reel shaft. There's another clip underneath and an antifriction washer. Remove those.

Turn the machine up the right way, and remove the take up spool...

This will reveal the take up clutch, and it's little three pads... if they've fallen out, don't panic... they'll be there somewhere!

Fit the belt and reassemble the clutch. Some adjustment of the clip above the small pulley we removed earlier may be required to get the right amount of friction on those pads, but it's not difficult.

The motor was a little sticky on this machine. I stripped the lower bearing off and re-lubricated it. It now spins like a top....

... there's still no sound.

This is rapidly traced back to an open-circuit HT fuse, it's replaced and there's some audio briefly. The HT (and heaters) is provided from two extra windings on the motor. There's no transformer. Clever stuff. Garrard and BSR used a similar trick on their record changers in the 60's. The electronics are a simple design with just two valves, an ECC83 and an ECL86.

The main smoothing capacitor is getting warm. A sure sign that it's had it day. It's a two section 50uF+50uF 350V part. I replace it with two suitable caps.

Switching back on and there's audio again... but it rapidly fades.... A check round the other caps proves their innocence... I'm not liking the look of the getter flash on the top of the ECL86 triode-pentode valve though, it looks like it's had a hard life.... The getter flash should be sliver or black looking, and should have nice sharp edges...

This is sort of brown looking....

... and checks on the avo valve tester show the triode is very low at 0.6 mA/V (It should be 1.5 when new) and the pentode section won't read at all! When testing the pentode section, there's a blue glow inside the anode structure, a sure sign there's some gas in there. A replacement restores operation, and the dulcet tones of a young man playing the guitar once again issue forth from the speaker... recorded some 50 years ago!

Audio quality isn't fantastic. These machines were a bit of a domestic novelty really. After transcribing the contents of the tape, Gareth described the quality as "Like a walkman with flat batteries", which isn't a bad description at all...

Friday, 24 June 2016

Mitsubish MC8000 audio system repairs.

Sadly, my friend Richard's dad passed away a while ago...

"There's some stuff here Andy, would you like it? I'd rather it went to a good home."

So it's now my duty to the man, to get it all going.....

This interesting audio unit was amongst the gear. It's a Mitsubishi MC-8000, originally sold by my old employer, B.V.Harrod (sadly long gone), and has it's original speakers and cabinet.

... and yes, that's the right way up, it has a vertical linear tracking turntable.

After some cursory checks, I give it some mains. There's a motor running somewhere, the turntable doesn't work, neither does the cassette, but the radio works fine!

Off with the back, and it's immediately obvious why the turntable doesn't work .... the belt's missing!

Actually it's not... Some of it is in the bottom of the cabinet, and some has stuck to the turntable flywheel. I quickly clean it up, with some wipes and a bit of IPA. Sizing of these things can be a pain, but it's easy really. Get a piece of string, and wrap it round the motor shaft and flywheel, where the belt would run, and cut the string. This will be the approximate circumference of the belt. Now belts are sold in dimensions of diameter, width and thickness, so divide the length of your string by Pi... in this case it's 200mm, and about 5mm wide.

The arm is driven by two other motors too. One raises and lowers the arm, and the other is used to track the arm across the record.

The belts that drive these functions are also in a bad way... just touching them leaves a nasty sticky black mess. A bit of IPA to clean up, and some replacement belts see the turntable working....

... now the motor continuously running is the motor in the cassette deck.... probably another belt! It's buried right in the bottom of the cabinet. See that brown, spotted thing lying across the circuit board? That's a bit of turntable belt!

... the motor nicely dates the unit from 1980...

Getting the transport out requires a lot of disassembly, and quite a considerable amount of cussing....

... but eventually yields ... I will not be beaten by mere machinery!

... sure enough the capstan belt has gone the same way as all the others! I'll change the lot...

The tape transport is carefully disassembled to gain access to the belts..

..which involved removal of the record switch actuator lever. Capstan flywheel is removed to clean the remains of the belt off...

... and some suitable replacements found from the belt box. The capstan belt measures in at 71mm. There's a 2.8mm wide belt which will do nicely.

The deck is repaired, and time to contemplate re-uniting it with the rest ... look at all those screws!

Finally back into the lower part of the cabinet!

... and that reunited with the top...

Thankfully, all the plugs are nicely labelled...

... and once the transformer is re-fitted, speakers connected, it's tested...

A new stylus is fitted, and the turntable works a treat. Sounds far better than I had expected... unfortunately there's nothing but a nasty hum from the cassette deck. It appears either the wiring to the tape head, or the head itself is open circuit.... I dread having to remove the transport again. Closer inspection reveals the wiring to the head is broken, and it's been repaired before.. there's some evidence of a poor soldering job, and the wires have been left very short. Thankfully, access, although difficult, is possible through the front, after removal of the door, and the repair is completed.

Here's a quick video of the turntable in action....

And, finally, a few photo's of the original dealership who supplied the unit. These were taken on the last day of trading, back in November 2005.. a sad day.

Shop front...

Interior... looking sadly empty :(

and the workshop...

Disclaimer.

Many pages on this blog describe working with high voltages, and vintage equipment. Sadly in this day of litigation, and common sense being absent, I must publish a disclaimer.

WARNING. Do NOT try these things at home. Most of the equipment I deal with in these pages uses high voltage, it will floor you or kill you if you come into contact with it. It will not warn you, blow a fuse, nor pull your safety trip. It may not give you a second chance. RESPECT IT'S AUTHORITY.

Many pages describe working with mains, or dealing with mains powered equipment which may have exposed mains connections inside the enclosure. Pull the plug and isolate before removing covers! Some equipment, in particular valve (tube) amplifiers, television sets and switched mode power supplies, can store high voltages for a considerable period of time after the power has been removed, especially if a fault condition exists. Always check and safely discharge capacitors. Safety is paramount.

There is risk of exposure to chemicals and physical debris. Precautions I take include, but are not limited to, use of rubber gloves, fume extraction and safety glasses.

All material on these pages is provided 'as is' without warranty of any kind, either expressed or implied. I will not, under any circumstances, accept liability should you injure, or kill yourself repeating any of the content of this website.

Always seek professional advice if in ANY doubt.

By entering and using this website, you agree to the above disclaimer. Play safely. I need all the readers I can get ;)

All of the delightful prose, imagery and video are copyright © 2014-2024 by A.G.Doswell unless otherwise stated, and may not be copied, reprinted or used elsewhere without the express written permission of A.G.Doswell. Just ask!

Not a business.

I repeatedly get requests, via messaging etc, asking me to fix their kit for them. This is just a hobby for me. I don't take on repairs on a commercial basis. The idea is to give my readers an insight into what's involved, and enable them to fix their own gear, (within the realms of safety of course), and I'll always help if I can. It's really for my own enjoyment and entertainment. It all started out with a simple website to actually document what I was doing, rather than making copious notes on paper, which I'd been doing since the late 80's. It was a friend that suggested I should make it all public, so I did. I never dreamed anyone else would find it interesting, and yet, I get around 200 people visiting everyday, which astounds me even to this day. So, sadly I can't fix your gear for you. Take it to a local firm. There are still a few TV repair places dotted about, a good few repair cafe's operate up and down the country, and see if you can find a tame radio ham who may share some of their knowledge and experience. There are a few "Makerspaces" dotted around the country, and they may be willing to help. I currently work as an engineer in the broadcast industry, and, until that changes, this will just remain a hobby. Thanks. Doz.

Translate

Friday, 12 August 2016

Thursday, 4 August 2016

Saturday, 30 July 2016

Friday, 29 July 2016

Friday, 22 July 2016

Wednesday, 13 July 2016

Sunday, 10 July 2016

Friday, 24 June 2016

log