Author Topic: Yet another DIY GPSDO - yes, another one  (Read 155335 times)

0 Members and 1 Guest are viewing this topic.

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #275 on: October 11, 2021, 07:27:10 pm »
Thanks AndrewBCN. I will proceed as you suggest. Looking forward to when I box this project, I expect the OCXO to generate a fair amount of heat in an enclosure. Are there any heat sensitive components/ modules that need to be thermally isolated so as not to compromise performance?
No, in principle everything should work fine up to 50C and even beyond that. The OCXO dissipates around 3W during the warmup phase (less than 3 minutes in general) and then around 1W. So at most the GPSDO enclosure should feel slightly warm to the touch after a few hours of operation.

Does the type of 5v power supply make any difference, ie linear vs SMPS?
enut11

The basic circuit (without the 1ns TIC) is pretty much impervious to power supply noise and ripple, as far as I could tell, but I haven't extensively tested or quantified that. I have powered one of the prototypes with a cheap 5V 18W USB power supply and it performs pretty much the same as the other two which use 5V linear regulators. Of course, YMMV.  :)

PS: I have just noticed that STM32 Core version 2.1 has just been released a few days ago. It is the version I recommend for use with the STM32 GPSDO firmware.
« Last Edit: October 11, 2021, 07:32:22 pm by AndrewBCN »
 
The following users thanked this post: enut11

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #276 on: October 11, 2021, 08:53:08 pm »


"PS: I have just noticed that STM32 Core version 2.1 has just been released a few days ago. It is the version I recommend for use with the STM32 GPSDO firmware."

Hi AndrewBCN
Would you please explain what this means?
an electronics nut from wayback...
 

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #277 on: October 11, 2021, 09:07:14 pm »


"PS: I have just noticed that STM32 Core version 2.1 has just been released a few days ago. It is the version I recommend for use with the STM32 GPSDO firmware."

Hi AndrewBCN
Would you please explain what this means?

Sure. To be able to program the STM32 chips using the Arduino IDE, you have to install the STM32 support package called Arduino Core STM32, which you can find on GitHub, with its installation instructions for Windows or Linux. Here is the link: https://github.com/stm32duino/Arduino_Core_STM32

The latest version of Arduino Core STM32 is version 2.1.

To install, follow the instructions here: https://github.com/stm32duino/wiki/wiki/Getting-Started

Once you have installed the Arduino Core STM32 version 2.1 on your PC, you can test your Arduino IDE setup using the Blinky example, by flashing the compiled Blinky code to the STM32F411CEU6 Black Pill.

« Last Edit: October 12, 2021, 03:54:40 pm by AndrewBCN »
 
The following users thanked this post: enut11

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #278 on: October 12, 2021, 12:47:31 am »
OK, thanks. I do not have a breadboard so have started to assemble the modules on perf-board.
For the micro I have used plug-in stand-offs to allow easy replacement if necessary.
an electronics nut from wayback...
 

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #279 on: October 12, 2021, 01:00:40 am »
There are lots of different STM32xx boards with V2.1 support at Github. Does it matter which one I use?
an electronics nut from wayback...
 

Offline rodpp

  • Frequent Contributor
  • **
  • Posts: 307
Re: Yet another DIY GPSDO - yes, another one
« Reply #280 on: October 12, 2021, 06:08:13 am »
Very interesting topic, thank you @AndrewBCN for sharing this project.

I'm a total noob and would like to implement this GPSDO. Searching for antennas in the Aliexpress, I found different options for use with GNSS. Will a better antenna give better results in this GPSDO application?

Considering that I'll install the antenna on the roof of my house, connected by a 10m cable, will a cheap external car antenna have a similar performance than a more expensive antenna?

Here are some examples (all from the same Aliexpress store, to compare the prices):

1- external car antenna - USD 22
https://www.aliexpress.com/item/4000058995645.html?spm=a2g0o.store_pc_groupList.8148356.8.1b17160aJTPOwE

2- "mushroom" antenna - USD 26
1295536-0
https://www.aliexpress.com/item/4000132134387.html?spm=a2g0o.store_pc_groupList.8148356.12.1b17160aJTPOwE

3- "mushroom 2" - USD 72
1295542-1
https://www.aliexpress.com/item/4000132160065.html?spm=a2g0o.store_pc_groupList.8148356.52.1b17160aJTPOwE

4- "mushroom 3" - USD 130
https://www.aliexpress.com/item/33051103562.html?spm=a2g0o.store_pc_groupList.8148356.4.1241160azKuoHa

5- "mushroom 4" - USD 150
https://www.aliexpress.com/item/4001252421244.html?spm=a2g0o.store_pc_groupList.8148356.10.1241160azKuoHa

6- "disc" - USD 280
https://www.aliexpress.com/item/33051111137.html?spm=a2g0o.store_pc_groupList.8148356.30.1241160azKuoHa

7- "dome" - USD 3500
1295560-2
https://www.aliexpress.com/item/33049894933.html?spm=a2g0o.store_pc_groupList.8148356.42.1241160azKuoHa

Of course, I'm not going to buy a 3.5K antenna. But maybe it is woth to invest a little more in the antenna if the GPSDO performance could be improved. And considering that the antenna will be installed on the roof, it can be used in a future GPSDO upgrade.

« Last Edit: October 12, 2021, 06:15:25 am by rodpp »
 

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #281 on: October 12, 2021, 09:58:06 am »
There are lots of different STM32xx boards with V2.1 support at Github. Does it matter which one I use?

You have to select the WeAct STM32F411CEU6 Black Pill in the Boards Manager menu in the Arduino IDE.
See attached screenshot.
« Last Edit: October 12, 2021, 12:05:19 pm by AndrewBCN »
 

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #282 on: October 12, 2021, 10:05:53 am »
Very interesting topic, thank you @AndrewBCN for sharing this project.

I'm a total noob and would like to implement this GPSDO. Searching for antennas in the Aliexpress, I found different options for use with GNSS. Will a better antenna give better results in this GPSDO application?

...

Yes, a better antenna will provide a more stable satellite signal reception and a lower jitter than a small "puck" antenna. If you decide to mount a GNSS antenna on the roof of your home then yes I would consider a better, more expensive antenna. Just for experimenting, a <$10 "puck" antenna with a 3 meters cable is good enough though.

I would start with a small antenna and once you have the STM32 GPSDO working and thoroughly tested, invest in a better, rooftop-mounted antenna.
« Last Edit: October 12, 2021, 11:57:09 am by AndrewBCN »
 
The following users thanked this post: rodpp

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #283 on: October 12, 2021, 10:15:09 am »
OK, thanks. I do not have a breadboard so have started to assemble the modules on perf-board.
For the micro I have used plug-in stand-offs to allow easy replacement if necessary.
Looks good. Please check the following datasheet for your OCXO: https://www.isotemp.com/wp-content/uploads/2011/03/OCXO-131.pdf
Note that the Isotemp 131-xxx was manufactured in different versions/configurations, xxx being the customer order number.
Your OCXO is possibly the 12V, sine-wave output type. If you have an oscilloscope at hand, please check it out beforehand. Power it with 5V first and check the output pin for a 10MHz signal, with an oscilloscope. If you see a CMOS-level 10MHz square wave then you indeed got a 5V, square wave version of the Isotemp OCXO, which is what we want for the STM32 GPSDO.

Also, a last minute tip: I suggest you use standoffs for the DAC and GPS modules, and even for the OCXO too. This will make it easier to "recycle" all the modules for other projects or different versions of the STM32 GPSDO.
« Last Edit: October 12, 2021, 01:56:21 pm by AndrewBCN »
 

Offline Johnny B Good

  • Frequent Contributor
  • **
  • Posts: 811
  • Country: gb
Re: Yet another DIY GPSDO - yes, another one
« Reply #284 on: October 12, 2021, 05:02:29 pm »
@rodpp

 That 22 dollar puck antenna should be more than ok for the time being even if you have to extend the feeder by another ten metres. The most important aspect of any GPS antenna being that you site it in a location that has an unobstructed 360 degree view of the horizon.

 I wouldn't bother upgrading to another fancy (and over-priced) navigation grade GPS/GNSS antenna. I'd save the money for a dual band timing grade antenna to make the most of a dual band GPS receiver module such as a u-blox ZED9 when you finally cave in to the pressure to upgrade to an even better setup. :)

 I'm still using a mag mount puck antenna, similar to enut11's, that I bought three years ago for £2.99 delivered (less than 5 dollars!) from a UK based ebay seller. This had a 5 metre RG174 co-ax with male SMA plug making it an even better bargain than the more common 3 metre co-ax cabled types.

 Even so, I needed to extend this by another 5 metres of RG174 (SMA in-line female to SMA male terminated patch lead) which surprisingly, only reduced the C/No ratio by a mere 2dB rather than by the 6 to 7 dB I was expecting of 5 metre's worth of RG174. This allowed me to locate the antenna on a ballasted biscuit tin parked on the flat bay window roof of my 1st floor workroom (2nd floor in the oddball American terminology that names the ground floor as the 1st floor).

 Moving the antenna from a South West facing window ledge onto the bay window roof with a 180 degree view centred on a direct westerly view made a significant improvement in SV reception but it was completely obstructed to the East by the brickwork of the gable ended 2nd floor bedroom above which seemed to offer a 3dB boost by constructive reflection of near overhead satellite passes.

 Ideally, I wanted to mount a short mast onto this brickwork in order to raise the antenna a foot or so above the ridge tiles to finally gain a full 360 degree view of the horizon but stepping out through the 2nd floor bedroom window onto this flat bay roof in order to drill into the brickwork was, to say the least, a bit too precarious even for me. The ballasted biscuit tin arrangement had only required me to lean out of said 2nd floor bedroom window, no actual crawling out onto the bay window roof needed (although I have done so in times past even through it's a bit of a struggle with the top hinged upward opening window).

 I only needed to raise the puck antenna another 8 feet or so from the biscuit tin location to allow it to peek over the ridge tiles so when I spotted a 60 by 70cm metal drawer in the basement I'd been using to store some motorbike parts in, it came to me in a flash that I could extend the ballasted biscuit tin antenna mount setup using this with four old car batteries for ballast and a 7 or 8 foot 1 1/4" diameter aluminium (aluminum) pole clamped into one corner to which I could screw the 20cm diameter biscuit tin lid onto by which to mag mount that puck antenna, reinforced with some impact adhesive to make it somewhat secure against even hurricane level storms (an over-kill/engineered solution in most UK locations but only pennies to implement so, Why not?  :) ).

 The chosen pole was about 18 inches short of the required length and required another length of 3/4" diameter aluminium tube to be added a few months later. Despite this initial shortfall, I did see a significant improvement but the fact that the view of the horizon was not quite the full 360 degree I'd aimed for, niggled me enough to properly finish the job even though this required yet more faffing around to recover the support pole and extend its length and then relocate it back to its pride of place... about three times over!

 There were a few changes to the groundplane mount  and an "upgrade puck antenna" I'd wanted to try out. The "upgrade" antenna proved to be a washout since it seemed to lack any immunity to overloading by cellphone signals from a nearby tower so I landed up reverting to the original 3 quid puck and the 20cm biscuit tin lid setup.

 I'd had to replace the original 5 metre RG174 patch lead, that I'd used to connect to the GPSDO via a cracked open window, with an 11 metre RG58 patch lead so I could reroute it through an existing conveniently located hole in the wall that had been drilled for a now long since redundant TV antenna feeder years before we'd moved into the property. The C/No figures range between 35 to 45 dB and have done so ever since I finalised this antenna mounting arrangement some 18 months ago.

 That first image shows a navigation class antenna that Banggood had been selling for around £18 about a year ago. I think this was the version with a 16 metre RG58 co-ax and I had been tempted to buy one until reason stayed my hand. It had occurred to me that it would offer no more performance than the 3 quid puck I already had in my possession and it would be far better to invest in a dual band timing class antenna for when the pricing on dual band timing GPS receivers finally descends from out of the stratosphere.

 Those fancy looking antennas will almost certainly be using the same 1 inch square patch antenna (pretty well the best choice for GNSS service) as that 22 dollar puck antenna. The radome comes into its own where you have to endure heavy persistent snowfall cursed winters - not usually a problem south of the Scottish border here in the UK.

 Unless you have to endure such winter conditions or need to use a very long run of feeder, I'd save my money for a more effective dual band antenna upgrade and make do with a well sited cheap puck antenna for now.
« Last Edit: October 12, 2021, 06:10:59 pm by Johnny B Good »
John
 
The following users thanked this post: rodpp

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #285 on: October 12, 2021, 05:43:41 pm »
From AndrewBCN:
Looks good. Please check the following datasheet for your OCXO: https://www.isotemp.com/wp-content/uploads/2011/03/OCXO-131.pdf
Note that the Isotemp 131-xxx was manufactured in different versions/configurations, xxx being the customer order number.
Your OCXO is possibly the 12V, sine-wave output type. If you have an oscilloscope at hand, please check it out beforehand. Power it with 5V first and check the output pin for a 10MHz signal, with an oscilloscope. If you see a CMOS-level 10MHz square wave then you indeed got a 5V, square wave version of the Isotemp OCXO, which is what we want for the STM32 GPSDO.

Also, a last minute tip: I suggest you use standoffs for the DAC and GPS modules, and even for the OCXO too. This will make it easier to "recycle" all the modules for other projects or different versions of the STM32 GPSDO.
[/quote]

The Isotemp OCXO was purchased as a 12v unit but I was able to run it at 5v, albeit with increased current.
The shape of the 10MHz square wave was the same at both voltages. The DSO was set to unlimited bandwidth.
« Last Edit: October 12, 2021, 05:46:11 pm by enut11 »
an electronics nut from wayback...
 

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #286 on: October 12, 2021, 06:05:31 pm »
I am not familiar with OCXO specs so cannot determine if mine is suitable for use in your GPSDO project.
an electronics nut from wayback...
 

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #287 on: October 12, 2021, 08:54:26 pm »
I am not familiar with OCXO specs so cannot determine if mine is suitable for use in your GPSDO project.

Well, we know it's a square wave version, now we just have to check the control voltage range. We'll do that later. First thing is flashing the firmware to the Black Pill.
 

Offline rodpp

  • Frequent Contributor
  • **
  • Posts: 307
Re: Yet another DIY GPSDO - yes, another one
« Reply #288 on: October 12, 2021, 10:05:13 pm »
Very interesting topic, thank you @AndrewBCN for sharing this project.

I'm a total noob and would like to implement this GPSDO. Searching for antennas in the Aliexpress, I found different options for use with GNSS. Will a better antenna give better results in this GPSDO application?

...


Yes, a better antenna will provide a more stable satellite signal reception and a lower jitter than a small "puck" antenna. If you decide to mount a GNSS antenna on the roof of your home then yes I would consider a better, more expensive antenna. Just for experimenting, a <$10 "puck" antenna with a 3 meters cable is good enough though.

I would start with a small antenna and once you have the STM32 GPSDO working and thoroughly tested, invest in a better, rooftop-mounted antenna.


Ok, I'll start with the small puck antenna. Thank you!

@rodpp

 That 22 dollar puck antenna should be more than ok for the time being even if you have to extend the feeder by another ten metres. The most important aspect of any GPS antenna being that you site it in a location that has an unobstructed 360 degree view of the horizon.

 I wouldn't bother upgrading to another fancy (and over-priced) navigation grade GPS/GNSS antenna. I'd save the money for a dual band timing grade antenna to make the most of a dual band GPS receiver module such as a u-blox ZED9 when you finally cave in to the pressure to upgrade to an even better setup. :)

 I'm still using a mag mount puck antenna, similar to enut11's, that I bought three years ago for £2.99 delivered (less than 5 dollars!) from a UK based ebay seller. This had a 5 metre RG174 co-ax with male SMA plug making it an even better bargain than the more common 3 metre co-ax cabled types.

 Even so, I needed to extend this by another 5 metres of RG174 (SMA in-line female to SMA male terminated patch lead) which surprisingly, only reduced the C/No ratio by a mere 2dB rather than by the 6 to 7 dB I was expecting of 5 metre's worth of RG174. This allowed me to locate the antenna on a ballasted biscuit tin parked on the flat bay window roof of my 1st floor workroom (2nd floor in the oddball American terminology that names the ground floor as the 1st floor).

 Moving the antenna from a South West facing window ledge onto the bay window roof with a 180 degree view centred on a direct westerly view made a significant improvement in SV reception but it was completely obstructed to the East by the brickwork of the gable ended 2nd floor bedroom above which seemed to offer a 3dB boost by constructive reflection of near overhead satellite passes.

 Ideally, I wanted to mount a short mast onto this brickwork in order to raise the antenna a foot or so above the ridge tiles to finally gain a full 360 degree view of the horizon but stepping out through the 2nd floor bedroom window onto this flat bay roof in order to drill into the brickwork was, to say the least, a bit too precarious even for me. The ballasted biscuit tin arrangement had only required me to lean out of said 2nd floor bedroom window, no actual crawling out onto the bay window roof needed (although I have done so in times past even through it's a bit of a struggle with the top hinged upward opening window).

 I only needed to raise the puck antenna another 8 feet or so from the biscuit tin location to allow it to peek over the ridge tiles so when I spotted a 60 by 70cm metal drawer in the basement I'd been using to store some motorbike parts in, it came to me in a flash that I could extend the ballasted biscuit tin antenna mount setup using this with four old car batteries for ballast and a 7 or 8 foot 1 1/4" diameter aluminium (aluminum) pole clamped into one corner to which I could screw the 20cm diameter biscuit tin lid onto by which to mag mount that puck antenna, reinforced with some impact adhesive to make it somewhat secure against even hurricane level storms (an over-kill/engineered solution in most UK locations but only pennies to implement so, Why not?  :) ).

 The chosen pole was about 18 inches short of the required length and required another length of 3/4" diameter aluminium tube to be added a few months later. Despite this initial shortfall, I did see a significant improvement but the fact that the view of the horizon was not quite the full 360 degree I'd aimed for, niggled me enough to properly finish the job even though this required yet more faffing around to recover the support pole and extend its length and then relocate it back to its pride of place... about three times over!

 There were a few changes to the groundplane mount  and an "upgrade puck antenna" I'd wanted to try out. The "upgrade" antenna proved to be a washout since it seemed to lack any immunity to overloading by cellphone signals from a nearby tower so I landed up reverting to the original 3 quid puck and the 20cm biscuit tin lid setup.

 I'd had to replace the original 5 metre RG174 patch lead, that I'd used to connect to the GPSDO via a cracked open window, with an 11 metre RG58 patch lead so I could reroute it through an existing conveniently located hole in the wall that had been drilled for a now long since redundant TV antenna feeder years before we'd moved into the property. The C/No figures range between 35 to 45 dB and have done so ever since I finalised this antenna mounting arrangement some 18 months ago.

 That first image shows a navigation class antenna that Banggood had been selling for around £18 about a year ago. I think this was the version with a 16 metre RG58 co-ax and I had been tempted to buy one until reason stayed my hand. It had occurred to me that it would offer no more performance than the 3 quid puck I already had in my possession and it would be far better to invest in a dual band timing class antenna for when the pricing on dual band timing GPS receivers finally descends from out of the stratosphere.

 Those fancy looking antennas will almost certainly be using the same 1 inch square patch antenna (pretty well the best choice for GNSS service) as that 22 dollar puck antenna. The radome comes into its own where you have to endure heavy persistent snowfall cursed winters - not usually a problem south of the Scottish border here in the UK.

 Unless you have to endure such winter conditions or need to use a very long run of feeder, I'd save my money for a more effective dual band antenna upgrade and make do with a well sited cheap puck antenna for now.



So it seems that antenna is the last piece of the kit that I'll need to upgrade. I have no severe winter where I live, no snowfall. I'll go with the puck antenna. Thanks for sharing your experiences!
 

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #289 on: October 12, 2021, 10:12:56 pm »
I am not familiar with OCXO specs so cannot determine if mine is suitable for use in your GPSDO project.

Well, we know it's a square wave version, now we just have to check the control voltage range. We'll do that later. First thing is flashing the firmware to the Black Pill.

When flashing the firmware using the Arduino software, does the sketch have to be compiled first?
an electronics nut from wayback...
 

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 954
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Yet another DIY GPSDO - yes, another one
« Reply #290 on: October 12, 2021, 11:42:46 pm »
Progress on the hardware so far. All modules are plug-in.
an electronics nut from wayback...
 

Offline Johnny B Good

  • Frequent Contributor
  • **
  • Posts: 811
  • Country: gb
Re: Yet another DIY GPSDO - yes, another one
« Reply #291 on: October 13, 2021, 01:17:36 am »
I am not familiar with OCXO specs so cannot determine if mine is suitable for use in your GPSDO project.

Well, we know it's a square wave version, now we just have to check the control voltage range. We'll do that later. First thing is flashing the firmware to the Black Pill.

When flashing the firmware using the Arduino software, does the sketch have to be compiled first?

 In my limited experience of the Arduino IDE with nano MCUs (and one RPi Pico), the answer is yes. There's not much time to be saved by using the generated code from a previous compile time run so I guess there was little to no incentive to provide such a "time saving" option considering the hobbyist users to whom this was aimed at. Far less chance of the end user coming to grief by always forcing a recompile prior to every upload to the MCU.
John
 
The following users thanked this post: enut11

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Yet another DIY GPSDO - yes, another one
« Reply #292 on: October 13, 2021, 05:05:22 am »
I am totally lost here.  I guess I don't have what it takes to follow along, much less to accomplish anything.

Flashing, compiling, all these terms aren't real to me.  I wish someone would build this for me and I can be done with it.  I have all the parts, I think.  I just don't know how to proceed without screwing up.  Maybe I have to accept being too old.
 

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #293 on: October 13, 2021, 05:12:42 am »
I am not familiar with OCXO specs so cannot determine if mine is suitable for use in your GPSDO project.

Well, we know it's a square wave version, now we just have to check the control voltage range. We'll do that later. First thing is flashing the firmware to the Black Pill.

When flashing the firmware using the Arduino software, does the sketch have to be compiled first?

It's a one-click operation: click on the "Upload" button in the Arduino IDE, this will compile the sketch and upload (flash) the firmware to the Black Pill.
 
The following users thanked this post: enut11

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #294 on: October 13, 2021, 05:25:37 am »
Progress on the hardware so far. All modules are plug-in.

Well done. Two comments:

1. I would add a yellow/red LED + 330R resistor as per the schematic. This is the main status LED, it provides some useful indication of what's going on.

2. Don't use 12V for your OCXO! Power it with 5V for now. You'll damage the Black Pill if you apply a > 5.5V signal to any of its pins (as per the datasheet).
 
The following users thanked this post: enut11

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #295 on: October 13, 2021, 05:37:31 am »
I am totally lost here.  I guess I don't have what it takes to follow along, much less to accomplish anything.

Flashing, compiling, all these terms aren't real to me.  I wish someone would build this for me and I can be done with it.  I have all the parts, I think.  I just don't know how to proceed without screwing up.  Maybe I have to accept being too old.

Bob, as the title of this thread says, this is a DIY project. If you feel like you can't build it yourself, there is no shame in getting somebody to help you (a friend, your grandson, another electronics DIYer, etc), specially in regards to your age. But most importantly, I would say it's not worth getting frustrated over it.

Start with testing your OCXO with an oscilloscope, that's already one step that you can accomplish.
 

Offline Renate

  • Super Contributor
  • ***
  • Posts: 1460
  • Country: us
Re: Yet another DIY GPSDO - yes, another one
« Reply #296 on: October 13, 2021, 11:30:06 am »
I am totally lost here.
C'mon, Bob, we're all rooting for you! You can get it done.

I don't want to derail this, but just a quick question:
For somebody who doesn't want to run a GPSDO all the time, wouldn't a good PPS do most of some minimal need?
Use it to gate a counter and you have ~1 Hz measurement in a second, ~0.001 Hz in 17 minutes?
Ok, I know that there is jitter in there too, but 1 ppm of 24 MHz is 24 Hz (xtal calibration on PSoC4) which should be easy.
What's the best receiver with a high res PPS for under ~$50? Thanks!
 
The following users thanked this post: thinkfat

Offline AndrewBCNTopic starter

  • Frequent Contributor
  • **
  • Posts: 571
  • Country: fr
Re: Yet another DIY GPSDO - yes, another one
« Reply #297 on: October 13, 2021, 12:45:21 pm »
...
For somebody who doesn't want to run a GPSDO all the time, wouldn't a good PPS do most of some minimal need?

You are confusing a GPSDO with a frequency meter; these are two distinct instruments with different purposes. A GPSDO is a high accuracy low phase noise reference frequency generator, as the name indicates, where the frequency generated is either the fundamental or some harmonic of the oscillator that is "disciplined" (in the STM32 GPSDO, 10MHz). It may or may not output a 1Hz signal, and it may or may not have that 1Hz aligned with UTC/GPS time, these are optional auxiliary functions. And a GPSDO will continue to generate the reference frequency even without a GPS signal (that's called holdover mode).

Use it to gate a counter and you have ~1 Hz measurement in a second, ~0.001 Hz in 17 minutes?
Ok, I know that there is jitter in there too, but 1 ppm of 24 MHz is 24 Hz (xtal calibration on PSoC4) which should be easy.
What's the best receiver with a high res PPS for under ~$50? Thanks!

The best GPS receiver with a high res PPS is one where you have a good rooftop GNSS antenna with a clear view of the sky in 360 degrees. If you can get that for under ~$50, congratulations! ;)

Check a few posts back, I have posted a link to a comparison of various GPS receiver modules.
« Last Edit: October 13, 2021, 01:30:20 pm by AndrewBCN »
 

Offline Renate

  • Super Contributor
  • ***
  • Posts: 1460
  • Country: us
Re: Yet another DIY GPSDO - yes, another one
« Reply #298 on: October 13, 2021, 01:48:00 pm »
Wouldn't a good PPS do most of some minimal need?

You are confusing a GPSDO with a frequency meter...
Um, no I'm not. I could use a calibrated frequency meter. A 10 MHz carrier (no matter how pure) won't do anything for me as my hearing has deteriorated.

Many USB "hockey pucks" go for under $50. I run a BU-353S4 24/7. I'd rather not tear it apart to go find the PPS.
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Yet another DIY GPSDO - yes, another one
« Reply #299 on: October 13, 2021, 04:38:37 pm »
I have already tested the OCXO.  Waveform looks good with the scope, the amplitude is fine, the frequency can be adjusted nicely with the control voltage (1.38V seems about right) and so all is well there.

Where I am having trouble is 1) mounting the parts properly (the OCXO has oddball pin spacing) and 2) the software/firmware aspects.

There seems to be an intellectual/experience gap I can't cross.  I dip my toe into it and am immediately flummoxed.  My grandson can be no help, as he has returned home to another state.  My friends are mostly worse off than I, never having gone into this territory.

If I decide to rely on my rubidium standard, I can be done with this stuff and forget about GPSDO.  But I really want to have a means of getting a solid, reliable frequency standard that coincides (as closely as possible) with the NIST value.  Yes, I am an analog guy and can't seem to get on board with the digital stuff.  You'd think that in the small town where I live (Los Angeles California) there ought to be a neighbor who would be willing to help but I know of no such person.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf