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...

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...