Citroën C1 (Mk 1), Peugeot 107, Toyota Aygo (Mk 1) (2005-2014) - Changing the dash illumination colour.

Many many moons ago I wrote an article on changing the colour of the dash illumination on the very excellent C1,107 and Aygo cars. It's got lost in the mists of time somewhat, so I'm going to re-create it here, by popular demand! I did this some 6 years ago, so please excuse the rather bad photography (like it's ever been great anyway!!)

Best I add a disclaimer, in addition to the usual one on the right.

Don't try this at home. I will not be held responsible, under any circumstances, if you damage your car. This requires accurate soldering to delicate and tiny components. 

Right, having got that out of the way....

We'll start with the binnacle lighting first...

So, open the bonnet and disconnect the NEGATIVE pole of the battery. Done that? Good....

Close the bonnet and arm yourself with a No.1 and No.2 Philips screwdriver, and a 10mm spanner or socket, a good quality soldering iron, some solder, desoldering braid, decent tweezers, 10 off PLCC-2 type SMD LED's and 50 off 1206 SMD LEDs in the colour of your choice.... oh and a sewing needle (yes, really) and a bit of connector block.

Turn the steering wheel to the right, to reveal one of the binnacle cover screws. Remove it.....













Turn the wheel to the left, and remove the other screw...

After a bit of jiggling, remove the cover....

... to expose 2 10mm bolts....












... and remove the two bolts ...















... gently easy the instrument binnacle forward to allow better access to the wiring, and remove the connectors, starting at the top. Each connector has a little catch. Once the catch is pressed, the connector should come out of it's socket easily...

Don't worry about noting where all the connectors came from, as each one is unique and will only fit in one socket.





Now take the binnacle off, and retire to the workshop/kitchen/dining room table!














Cleanliness is the key here, so go and wash you hands, and if you have any of those nitrile or similar gloves, don a pair now. You don't want to be leaving fingerprints on the inside of your speedo once we've got it apart....

Now, gently push in the two clear plastic catches, as shown here....











And there are two similar catches at the rear of the speedo to unlatch too.













Once these are released, you can lift off the clear cover.













Next wrap a piece of wire or string round the pointer, and gently pull it off.













Now release the black plastic catches holding the surround in place, two shown here, and one on each side...













... and gently un-clip the surround....














.. remove the speedo face...














... and lift out the diffuser...















... remove the four philips screws holding the speedo electronics to the brown back & sockets.... one is a different length. Make note of which one it is!











Remove the speedo electronics from the brown back assembly. This will take some gentle prying to get the socket to release.












And place the back safely out of the way....













Now comes the tricky bit. First off, make yourself a DIY hole clearing tool, with a needle mounted in a piece of connector block..












Get yourself some good quality desoldering braid. Don't be tempted to use one of those nasty spring-loaded pump type desoldering tools. They have an unpleasant habit of removing the solder, along with the tracks on the board, which will be difficult and time consuming to repair. I personally rate Chemtronics braid. It's good stuff....

Now, desolder the 21 pins connecting the odometer and fuel gauge LCD to the board. Take your time. Breaking this would not be good news. Once they are all desoldered, check each pin can wiggle a bit in the hole with the needle, and lift off the while plastic front, along with the LCD. Clean up the holes, by warming the DIY hole clearing tool with the iron, and inserting it into the holes, pushing out any remaining solder. Clean up with the braid, and breathe a sigh of relief.




Now we come to change the LED's themselves.

PLCC-2 LEDs have a small cut out in one corner. This marks the cathode. There is a small arrow printed next to each LED and the cut-out must be at the pointed end of the arrow, as shown in the picture...







Using the desoldering braid, desolder the LEDs ringed in the picture. LEDs ringed in red are the backlight LEDS, and the LED ringed in blue is the illumination for the pointer. You can make them different colours if you wish...

Heating the LEDs for too long with the soldering iron will destroy them, which isn't much of an issue when removing the old orange ones, but you want to be careful when soldering the new ones in place.

Once this is done, carefully reunite the LCD display, still on the white surround with the electronics are resolder it in place. Replace the electronics back into the rear cover, making sure the socket seats home, and replace the 4 philips screws, ensuring the long one goes back in the right hole. Replace the diffuser, and fit the face. When pushing the pointer back on, fit it pointing towards 110 MPH, and once seated move it round to 0.
Replace the black cover and the clear cover. Done.

Ignore this bit if you haven't got a tachometer (rev counter).....

Squeeze the tacho slightly, and grip the front surround. It should pop off easily.

Remove the philips screw in the rear of the tacho, allowing it to be pulled out slightly.













Unclip the black plastic surround and remove the surround and plastic lens in one piece...

Remove the pointer with a piece of wire or string in the same manner as we removed the speedo pointer.

Undo the two screws securing the face to the speedo...






Remove the electronics and face in one piece, disconnect the plug, and remove the two screws.













Replace the two LEDs shown, remembering the arrows...











and reassemble.

Take the binnacle back out to the car and re-fit it.

Good eh??

But we're not done yet!









Radio looks awful in orange now doesn't it? ... and that heater illumination ... ugh! ... and the headlight aim switch ... horrendous!

Let's deal with the radio and heater.










Set the heater fan to the II position, and locate the small gap in the knob underneath. Gently prise it off, using a piece of cardboard to cushion the dash so we don't scratch it.

Remove the screw.









Get a credit card or something (I've got strong finger nails!) behind the cowling, and pull it off. It shouldn't put up much of a fight.












4) Remove the pulgs and sockets from the radio (one white, on black (have little release levers on one side) and the aerial plug (just pulls off)) , and the heated rear window switch, hazard warning switch and (if you're lucky) the air con switch (have release buttons on the top or bottom).








These are green white and black respectively. Cart the whole assembly inside to the operating theatre.

Remove the radio from the dash. Two sprags holding the radio in each side. Two on the top and two on the bottom.












This requires use of 14 fingers, or some choice expletives, or both.

Put the cowel to one side, and remove the front from the radio. Two small philips screws each side, and one clip each side, two on the top and bottom.












12 fingers only required this time, also fewer expletives.












Remove 8 screws from the display PCB and remove. It will unplug from the push button sub-panel, leaving that in place.













Remove 4 screws from sub panel and remove.












Right. Here are the LEDs we're going to change. They are 1206 SMD LEDs, you'll need 50 of them (you'll have a few left over)





























1206 LEDs are tiny!










There are 17 on the sub panel and 27 on the display PCB. I know I told you to get 50 , but you will lose some ;) And anyway , it's a minimum order quantity :D

Remove a new diode from it's packaging to examine it. Marked on it somewhere is it's polarity. The mark will indicate the cathode. In the case of my diodes, there was a green mark on the underside. This pointed to the cathode.








The cathode of the diode on the PCB is marked with a continuous line around the end of the diode. In the case of the picture below, it's to the right. All of the diodes we are changing have there cathode to the right, or to the top if they are vertically orientated.









Desolder the first diode (I actually desoldered the whole lot, it saves time, but if you're new to this, take your time). Use the solder wick to suck off the solder, then you should find a slight push with the iron tip, will remove the diode from it's lands. Use a nice fat tip for this job. Clean up with solder wick. Add a touch of fresh solder onto each land. The PCB's in this radio are really well made and will tolerate quite some abuse before the print becomes damaged.





Grab your new diode in some tweezers, making sure you know which end is the cathode. Line the new diode up to it's solder lands, and just melt the solder with the small iron, so as to solder the one end of the diode in place. Remove the tweezers and melt the solder on the other end.





Repeat for the other 16 diodes on this board.


Put the board to one side , it's time to tackle the display board .... "What's so tricky And?" I hear you say .. it looks similar to the other PCB .. unfortunately there are 12 diodes UNDER the LCD crystal display, so it's got to come out.

Twist the 4 tags holding the display screening can to the board, and desolder the two points.













Desolder the two multipin connectors to the display, just underneath the CD slot. TAKE YOUR TIME. The display crystal is very fragile. One crack, and it's game over. I chose to flood the connectors with solder and run the iron up and down the whole length until every joint was wet, the display will then gently come out. Solder wick is another option. Once the display is out , put it somewhere safe.

Using the hole clearing tool (OK , it's a pin and a bit of connector block) and the solder braid, clean up the connectors.

Change the diodes in the same manner as you did on the first board.

CAREFULLY re-fit the display. Don't forget to twist the tags and solder the can.







If you have a bench supply, plug the boards in and power up to check it all lights up nicely. You'll need to connect +12 Permanent, and +12 lights.

Damn, that looks good....








Right. That AUX socket on the front... what a pain. Lead from your phone dangling all over the place looking untidy... Fancy moving it somewhere sensible? Read on.....

Move back to the sub-board, and looking at the aux connector, cut the track as shown...

It's uncut here...











...and after the incision is made. This is to fool the electronics into thinking there is a plug in the socket.

Turn the PCB over, and connect a piece of twin and screen cable to the points as shown.












Right channel is red, left black and the screen is ground.



Replace the sub-panel, and the main display PCB, and carefully route your new cable out of the front... it's tight , but mine fitted.

Refit the radio in the dash, again routing the new cable. I chose to route mine to the rear of the clutter box...  You'll need to solder a 3.5mm stereo jack to the end, tip is left, ring is right, and sleeve is ground.

OK... it's going well... now the heater... You can buy replacement LED lamps for the bulbs in there now. I chose to use a couple of ordinary 5mm blue LEDs, and solder them in.. It's easier if you get the right LED bulbs.

I cut the tops of the LEDs in an attempt to diffuse the light a bit, otherwise you'll end up with two spots. Not a good look. Note the 780 ohm resistors.

Undo the panel mounting screws...













 ...both of them...

...and hinge the panel forward. You will see the old bulbs in there holders, just pull 'em out, and fit the new ones or solder in your LEDs. If they don't light, you may have them in the wrong way round. If the light is still focussing as two spots, just bend the LEDs so they face away from the front. This is easier if you have soldered separate LEDs, as I did. Once it's nice and even, screw the panel back in place, and refit the radio... Give it a try.... Nice....





... except for the headlamp aim switch ... still orange... nope, it'll have to go....




Remove the philips screw holding the outer dash on....













Pull the dash down, and pop the switch out, un-plug it and back to the operating theatre....

Pull off the cover.....

Cut off the orange LED.













... and solder in a PLCC-2 diode in it's place, with the cut-off corner facing the outside edge.

Push the cover back on, and refit the switch and dash.....








There. It's done. In all it's blueness....

... or pink if you prefer ;)












Thanks to Hanna for the pink pic. White looks good too!

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. */

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 ()

  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.


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

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.

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....

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 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 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 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 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. */

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

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) 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. 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> 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(); 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); 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); } // 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(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: } } 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 (); } } /* 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. */

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...