UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Success Stories

Notices

Success Stories If you have successfully repaired or restored a piece of equipment, why not write up what you did and post details here. Particularly if it was interesting, unusual or challenging. PLEASE DO NOT POST REQUESTS FOR HELP HERE!

Closed Thread
 
Thread Tools
Old 9th Aug 2021, 9:48 pm   #1
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Finally a GPSDO that works

I am really quite pleased tonight. Thanks to many people on this forum and a smattering of experience of my own, I have a working GPSDO for my workshop.

Having gone down a dead-end route, I then found the ublox GPS receivers that are perfect for this applicaton. The antenna, receiver and RS232 driver are in a masthead enclosure outside my workshop, and the 100kHz synchronised feed is brought into the Disciplined Oscillator on a plain multi-core cable. The whole works well, please view the circuit I have posted below.

The bit that I found most difficult was to derive a 1MHz sine wave with low enough impedance to drive any load I could imagine, from the square wave that exists inside the control loop. However, with perseverance, help from this forum and LTspice the circuit developed and now I have a very nice 4v p-p sine output which doesn't seem to care if I load it quite severely.
Attached Thumbnails
Click image for larger version

Name:	1MHz output.jpg
Views:	217
Size:	42.9 KB
ID:	239159  
Attached Files
File Type: pdf ublox GOSDO.pdf (201.4 KB, 230 views)
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 10th Aug 2021, 12:31 pm   #2
Nymrod121
Nonode
 
Nymrod121's Avatar
 
Join Date: Apr 2012
Location: Worcestershire, UK.
Posts: 2,039
Default Re: Finally a GPSDO that works

Nice work, Richard The BFX85-based output amplifier is of great interest to me as I have never been able to lock my Blackst*r Meteor 600 to my GPSDO because of the inability to drive its 'external reference' properly. I have a Racal 9478 frequency distribution amplifier that is locked to the GPSDO and which in turn provides reference outputs at 10, 5 and 1 MHz. I use the latter with my HP8640B generator (amongst other items) - it was this arrangement that I used when calibrating stevehertz's frequency counter a couple of years back.

The 9478 FDA outputs are 50 ohm, however the input Z of the Meteor 'ext ref' facility is much lower ... hence the ICL7216-based circuitry doesn't 'see' enough voltage to lock it. Your BFX85 driver circuit looks like it'll do the job ... thank you for posting it!

Best regards
Guy
__________________
"What a depressingly stupid machine." [Marvin: HHGTTG]
Nymrod121 is offline  
Old 10th Aug 2021, 5:05 pm   #3
Guest
Guest
 
Posts: n/a
Default Re: Finally a GPSDO that works

Ta-Dah, as they say, I think projects with difficulties are much more fun than ones that work first time. One thing for sure, you will not forget the things learned by doing, rather than book stuff!
 
Old 11th Aug 2021, 9:28 pm   #4
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

Another pleasing test tonight - comparing my GPSDO with a borrowed commercial one (thanks Chris!) after only one hour there is no discernable drift between the two. Plugging the 10mHz outputs into the two channels of my 'scope and triggering off one of them, the other remains stationery 'as if locked'. I'll take that as a result!

I've left the two running overnight and I'll see if they still agree in the morning.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 12th Aug 2021, 6:22 pm   #5
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

Still good news, the 10MHz signals are still in sync….. Of course, I really need to borrow yet another GPSDO to achieve a triple lock and know that it really is a 10MHz standard.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 13th Aug 2021, 9:11 pm   #6
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

Okay, this is the last smug bulletin. At 20:00hrs this evening, the two 10MHz signals are still in sync. That has to mean that my new reference is locked to the GPS signals and I can rely on it.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 19th Aug 2021, 9:39 am   #7
John Caswell
Hexode
 
Join Date: Jul 2009
Location: Wokingham, Berkshire, UK.
Posts: 437
Default Re: Finally a GPSDO that works

Could someone please explain, preferably in words of one syllable what the RS232 socket on a GPSDO is used for, control in? information out?

John
John Caswell is offline  
Old 19th Aug 2021, 1:26 pm   #8
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

In this design, the RS232 is simply used to monitor and control the ublox GPS receiver. One parameter that needs setting whenever the device is switched off for a day or so is the 100kHz timing pulse output to the oscillator. Default is 1Hz and, although there is a memory back-up battery, it is quite small and the receiver forgets the settings eventually.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 8th Oct 2021, 9:27 am   #9
Keith956
Heptode
 
Keith956's Avatar
 
Join Date: Jun 2019
Location: Oxfordshire
Posts: 719
Default Re: Finally a GPSDO that works

Hi Richard, Did you give up on the Jupiter GPS module? Reason I ask is I have come by a TU35-D410 and was thinking of knocking up a GPSDO using that.
Keith956 is offline  
Old 8th Oct 2021, 10:14 am   #10
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Finally a GPSDO that works

Congratulations.
Perseverance.
Result.
Radio1950 is offline  
Old 8th Oct 2021, 11:20 am   #11
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

Quote:
Originally Posted by Keith956 View Post
Hi Richard, Did you give up on the Jupiter GPS module? Reason I ask is I have come by a TU35-D410 and was thinking of knocking up a GPSDO using that.
Hi Keith,

Yes I gave up after a lot of work, the Jupiter resets itself after a number of minutes and I wasn't able to issue the command to persuade it to run continuously. However, there are a number of published designs using the Jupiter and the reset issue is documented. I just couldn't find a way to resolve it and opted for the uBlox module - it's in current manufacture and has more configurable parameters as well as a neat configuration software package.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 8th Oct 2021, 11:50 am   #12
Keith956
Heptode
 
Keith956's Avatar
 
Join Date: Jun 2019
Location: Oxfordshire
Posts: 719
Default Re: Finally a GPSDO that works

That's good feedback, thanks Richard. I haven't looked at it closely enough yet to know what level of control is needed - a PIC maybe to initialise it may not be entirely beyond me. I'll read up a bit more about this reset issue.
Keith956 is offline  
Old 14th Oct 2021, 12:25 am   #13
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: Finally a GPSDO that works

Good to see this has worked out so well. I am preparing a GPSDO based on a eevblog thread that uses the ublox 1Hz output (so no need for a control setting change) and is set up for disciplining a 10MHz OCXO like your setup (a PIC is used to derive the OCXO control rather than your phase comparator). At that frequency I've found that the type of cmos inverter has to be fast with a high current sink/source capability (ie. 74AC04) otherwise the 50 ohm output capability naturally turns to a sinewave even with 4-5 paralleled inverters.

From what I can tell, the cheap ebay NEO-7M GPS module I have aligns very closely with your development board, and also has a micro-USB port so can be tested standalone using u-centre2 software from ublox.
trobbins is offline  
Old 14th Oct 2021, 2:26 am   #14
Terry_VK5TM
Nonode
 
Terry_VK5TM's Avatar
 
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 2,324
Default Re: Finally a GPSDO that works

Quote:
Originally Posted by trobbins View Post
From what I can tell, the cheap ebay NEO-7M GPS module I have aligns very closely with your development board, and also has a micro-USB port so can be tested standalone using u-centre2 software from ublox.

Fair warning, the cheap NEO-7's I got off Ebay don't work if the usb port is used - have a sneaking suspicion they wired the 5V line direct to the 3.3V line - everything just gets (well - got) hot.


Wired it up using the other inputs on the module and it worked fine.
__________________
Terry VK5TM
https://www.vk5tm.com/
Terry_VK5TM is online now  
Old 14th Oct 2021, 3:16 am   #15
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: Finally a GPSDO that works

Terry - that's bad luck - was it a pcb error on your module? One of the first things I checked was the NEO-7M device ID using device manager - and luckily it matched the official driver info. What I found clunky was the earlier u-centre software from ublox website - it hung my PC a few times depending on what I did - not even task manager could stop the app. The earlier u-centre looked like it could modify the NEO-7M parameters - not that my use needs the precision but the type of GPS sats and their elevation can be tweaked to achieve better timing precision.
trobbins is offline  
Old 14th Oct 2021, 6:27 am   #16
Terry_VK5TM
Nonode
 
Terry_VK5TM's Avatar
 
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 2,324
Default Re: Finally a GPSDO that works

I didn't bother following through on it, I just hooked it up to a usb-serial adapter and bypassed the modules usb socket.

I bought two of them from the same supplier and both were the same.

One day, when I finally get the new workroom/bench finished, I might actually get to finish the project.
__________________
Terry VK5TM
https://www.vk5tm.com/
Terry_VK5TM is online now  
Old 14th Oct 2021, 6:49 am   #17
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

I used a 100kHz signal from the uBlox, thinking that would offer greater precision through the phase comparator. 100kHz down the 5m multicore from the receiver head was "interesting" and I played with capacitors at the driven end and the receiving end to maintain a close approximation to a square wave. It works but how much it would need altering if I used another cable length I don't know.

My unit never used the usb port - I was already set up for RS-232 from the previous attempt to use the Jupiter receiver and the u-centre software worked well on my laptop so no need to change. The RS-232 driver sits atop the mast as well, providing robust signals down to the oscillator.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 14th Oct 2021, 8:08 am   #18
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,288
Default Re: Finally a GPSDO that works

Well done Richard, a great result.
Have you laid out a pcb by any chance, if so is there any chance of getting a copy of the Gerbers? I would love to build this.

Peter
Electronpusher0 is offline  
Old 14th Oct 2021, 9:28 am   #19
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Finally a GPSDO that works

I'm rather afraid not, Peter. The device went from breadboard to Veroboard and that was it. I would have designed a PCB but since I was both tired of the device after several months of engineering development (!) and elated to finally have a reliable accurate reference that I screwed the enclosure shut and haven't opened it since. I haven't switched it off either.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 14th Oct 2021, 4:45 pm   #20
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,288
Default Re: Finally a GPSDO that works

Thanks anyway Richard, I am still thinking of building one, an excellent design.
I love the fact that it does not use a PIC, I can and have used these but I am happier with "normal" devices.

Peter
Electronpusher0 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:26 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.