10MHz GPSDO by "BH3SAP"

[daddygo]

Although I confirmed that the operation was stable for a long time, I noticed that the number of satellites and the time were occasionally not displayed correctly, so I investigated.

I found that when communication with the GPS module was 9600bps, the NMEA sentence sometimes overran.
I tried to set the GPS module to 19200bps or higher, but it did not work properly.

After further investigation, I found that only one of the two UARTs was set, so I corrected it. In addition, I modified it so that the baud rate setting is automatically changed and saved when the GPS module is ATGM336H.
I also modified sprintf to use snprintf.
As an additional element, I added an hour offset to the GPS menu so that the local time can be displayed.
Since the year, month, and date are not displayed, it does nothing, and daylight saving time is not taken into account.

I don't know much about github, so I've attached a zip of the modified source code and the bin file.
The version name has been slightly changed to make it easier to distinguish.
I hope it will be helpful to someone.

Hi All,

@fredo_ can you merge this on github?

@2A03 how about displaying the date in day/month/year format in the UTC time row on the main display, if this data is already avail from the GPS module, you can change the time / act. date values on the display with 5s shift for example...

- if I understand correctly only the ATGM336H setup is automated?
-  ATGM336H has 2 UART int. but NEO-6M does not
- the NEO-M9N I am developing with is also built with two UARTs

the original Chinese GPSDO comes with two types of modules, NEO6M and ATGM336H

@fredo_ & @2A03 what do you think about adding a GPS module selection menu?


I start my trip on Thursday, so I can test the FW mods until then.

THX

[2A03]

Although I confirmed that the operation was stable for a long time, I noticed that the number of satellites and the time were occasionally not displayed correctly, so I investigated.

I found that when communication with the GPS module was 9600bps, the NMEA sentence sometimes overran.
I tried to set the GPS module to 19200bps or higher, but it did not work properly.

After further investigation, I found that only one of the two UARTs was set, so I corrected it. In addition, I modified it so that the baud rate setting is automatically changed and saved when the GPS module is ATGM336H.
I also modified sprintf to use snprintf.
As an additional element, I added an hour offset to the GPS menu so that the local time can be displayed.
Since the year, month, and date are not displayed, it does nothing, and daylight saving time is not taken into account.

I don't know much about github, so I've attached a zip of the modified source code and the bin file.
The version name has been slightly changed to make it easier to distinguish.
I hope it will be helpful to someone.

Hi All,

@fredo_ can you merge this on github?

@2A03 how about displaying the date in day/month/year format in the UTC time row on the main display, if this data is already avail from the GPS module, you can change the time / act. date values on the display with 5s shift for example...

- if I understand correctly only the ATGM336H setup is automated?
-  ATGM336H has 2 UART int. but NEO-6M does not
- the NEO-M9N I am developing with is also built with two UARTs

the original Chinese GPSDO comes with two types of modules, NEO6M and ATGM336H

@fredo_ & @2A03 what do you think about adding a GPS module selection menu?


I start my trip on Thursday, so I can test the FW mods until then.

THX

Hi. @daddygo

I couldn't add the date because the firmware only parsed the time part of the NMEA sentence.
Also, you probably won't have enough code flash to add the correct timezone.

The ATGM336H detection and configuration is in gps.c. It's a short code, so it shouldn't be difficult.
If the ATGM336H is not detected, the UART baud rate will simply be set as before.

The two UARTs are BluePill's UART2 and UART3.
It has nothing to do with whether the GPS module has two UARTs or not.

[daddygo]

Although I confirmed that the operation was stable for a long time, I noticed that the number of satellites and the time were occasionally not displayed correctly, so I investigated.

I found that when communication with the GPS module was 9600bps, the NMEA sentence sometimes overran.
I tried to set the GPS module to 19200bps or higher, but it did not work properly.

After further investigation, I found that only one of the two UARTs was set, so I corrected it. In addition, I modified it so that the baud rate setting is automatically changed and saved when the GPS module is ATGM336H.
I also modified sprintf to use snprintf.
As an additional element, I added an hour offset to the GPS menu so that the local time can be displayed.
Since the year, month, and date are not displayed, it does nothing, and daylight saving time is not taken into account.

I don't know much about github, so I've attached a zip of the modified source code and the bin file.
The version name has been slightly changed to make it easier to distinguish.
I hope it will be helpful to someone.

Hi All,

@fredo_ can you merge this on github?

@2A03 how about displaying the date in day/month/year format in the UTC time row on the main display, if this data is already avail from the GPS module, you can change the time / act. date values on the display with 5s shift for example...

- if I understand correctly only the ATGM336H setup is automated?
-  ATGM336H has 2 UART int. but NEO-6M does not
- the NEO-M9N I am developing with is also built with two UARTs

the original Chinese GPSDO comes with two types of modules, NEO6M and ATGM336H

@fredo_ & @2A03 what do you think about adding a GPS module selection menu?


I start my trip on Thursday, so I can test the FW mods until then.

THX

Hi. @daddygo

I couldn't add the date because the firmware only parsed the time part of the NMEA sentence.
Also, you probably won't have enough code flash to add the correct timezone.

The ATGM336H detection and configuration is in gps.c. It's a short code, so it shouldn't be difficult.
If the ATGM336H is not detected, the UART baud rate will simply be set as before.

The two UARTs are BluePill's UART2 and UART3.
It has nothing to do with whether the GPS module has two UARTs or not.

Hi @2A03,

Thanks for clarifying that.

The most STM32F103C8T6 nowadays has 128K FLASH (not ST???), there is a lot of discussion/info about this, but the original one does not 128, i.e. there is a hidden 64 + 64K version...

-not sure what this means in terms of implementation "correct timezone" ? 128K
(I'm more into HW stuff)


UART(s), so that's ok, but if I'm right UART3 is for the GPS module and UART2 is for the pass-through function???

[2A03]

Although I confirmed that the operation was stable for a long time, I noticed that the number of satellites and the time were occasionally not displayed correctly, so I investigated.

I found that when communication with the GPS module was 9600bps, the NMEA sentence sometimes overran.
I tried to set the GPS module to 19200bps or higher, but it did not work properly.

After further investigation, I found that only one of the two UARTs was set, so I corrected it. In addition, I modified it so that the baud rate setting is automatically changed and saved when the GPS module is ATGM336H.
I also modified sprintf to use snprintf.
As an additional element, I added an hour offset to the GPS menu so that the local time can be displayed.
Since the year, month, and date are not displayed, it does nothing, and daylight saving time is not taken into account.

I don't know much about github, so I've attached a zip of the modified source code and the bin file.
The version name has been slightly changed to make it easier to distinguish.
I hope it will be helpful to someone.

Hi All,

@fredo_ can you merge this on github?

@2A03 how about displaying the date in day/month/year format in the UTC time row on the main display, if this data is already avail from the GPS module, you can change the time / act. date values on the display with 5s shift for example...

- if I understand correctly only the ATGM336H setup is automated?
-  ATGM336H has 2 UART int. but NEO-6M does not
- the NEO-M9N I am developing with is also built with two UARTs

the original Chinese GPSDO comes with two types of modules, NEO6M and ATGM336H

@fredo_ & @2A03 what do you think about adding a GPS module selection menu?


I start my trip on Thursday, so I can test the FW mods until then.

THX

Hi. @daddygo

I couldn't add the date because the firmware only parsed the time part of the NMEA sentence.
Also, you probably won't have enough code flash to add the correct timezone.

The ATGM336H detection and configuration is in gps.c. It's a short code, so it shouldn't be difficult.
If the ATGM336H is not detected, the UART baud rate will simply be set as before.

The two UARTs are BluePill's UART2 and UART3.
It has nothing to do with whether the GPS module has two UARTs or not.

Hi @2A03,

Thanks for clarifying that.

The most STM32F103C8T6 nowadays has 128K FLASH (not ST???), there is a lot of discussion/info about this, but the original one does not 128, i.e. there is a hidden 64 + 64K version...

-not sure what this means in terms of implementation "correct timezone" ? 128K
(I'm more into HW stuff)


UART(s), so that's ok, but if I'm right UART3 is for the GPS module and UART2 is for the pass-through function???

Hey @daddygo

The STM32F103C8T6 is 64KB, the STM32F103CBT6 (B instead of is 128KB. The fonts are similar, so it's confusing.

I think this is a good reference for how to handle timezones correctly.
https://www.ryanthomson.net/articles/practical-guide-timezones/

That's right. UART3 is for GPS, and UART2 is for pass-through.

[daddygo]

Hi 2A03,

@2A03 "The STM32F103C8T6 is 64KB, the STM32F103CBT6 (B instead of is 128KB. The fonts are similar, so it's confusing."


If only it were that easy to tell the difference between these MCu's, ST forums are full of descriptions of "fake" BP boards and MCUs 
I have quite a few on the shelf, marked STM32F103C8T6, but with 128K flash, most work fine, but some are only good for blink LED 

show you some CubeProgrammer "prtsc" with 64K & 128K flash

or f.e. from the ST & EEV forum:

https://community.st.com/t5/stm32-mcus-products/stm32f103c8t6-with-128kb-of-flash-instead-of-64-fake-or-genuine/td-p/267742
https://www.eevblog.com/forum/microcontrollers/cheap-bluepill-very-likely-it-has-fake-stm32-right/150/

[2A03]

Hello @daddygo

I opened the case and took a photo.
Is this okay?

[ME]

I tried reprograming my gpsdo blupill with stlink v2 and cube programer, It throws a msg error no     stm32 target found, do i need to change the blupill jumper settings?.

[daddygo]

Hello @daddygo

I opened the case and took a photo.
Is this okay?

Is this okay???

I do not want to disappoint you, but the BPs I have seen and I mean by that in this GPSDO are all equipped with fake MCUs, but they work...

I've already suggested this link to someone in this topic, you should try it too...

"
There's a good tool you can test to see if it's genuine ST:
https://mecrisp-stellaris-folkdoc.sourceforge.io/bluepill-diagnostics-v1.6.html
"

The market is also full of BP boards with 128K flash and they work well, so I thought there would be more headroom for further development of FW 

[daddygo]

Hi @ME,


nope, in this case you don't have to (SWD header), just follow this:

https://github.com/fredzo/gpsdo-fw?tab=readme-ov-file#flashing-the-firmware

+++edit1:
in the unfortunate event that ST-Link V2 is fake and the MCU is fake, well you know....

buy one of these, they are only pennies and later you will know that at least "programmer" is original, I use it too, and/or V3
https://eu.mouser.com/ProductDetail/STMicroelectronics/ST-LINK-V2?qs=H4BOwPtf9MC1sDQ8j3cy4w%3D%3D

The CUBE Programmer runs with a fake ST-Link V2 (in most cases), but CubeIDE, for example, will not

[2A03]

Hello @daddygo

I opened the case and took a photo.
Is this okay?

Is this okay???

I do not want to disappoint you, but the BPs I have seen and I mean by that in this GPSDO are all equipped with fake MCUs, but they work...

I've already suggested this link to someone in this topic, you should try it too...

"
There's a good tool you can test to see if it's genuine ST:
https://mecrisp-stellaris-folkdoc.sourceforge.io/bluepill-diagnostics-v1.6.html
"

The market is also full of BP boards with 128K flash and they work well, so I thought there would be more headroom for further development of FW 

Hi @daddygo

I'm using a translation tool, so my intention may not be conveyed well.

I understand that there are many fakes from Chinese clones. I tried the tool you told me about, and it seems that the STM chip is genuine and there is 128KB of flash. I've attached a screenshot.

Well, I'm happy with the firmware I modified, so I don't plan to make any further improvements even if the flash increases.

[daddygo]

Hello @daddygo

I opened the case and took a photo.
Is this okay?

Is this okay???

I do not want to disappoint you, but the BPs I have seen and I mean by that in this GPSDO are all equipped with fake MCUs, but they work...

I've already suggested this link to someone in this topic, you should try it too...

"
There's a good tool you can test to see if it's genuine ST:
https://mecrisp-stellaris-folkdoc.sourceforge.io/bluepill-diagnostics-v1.6.html
"

The market is also full of BP boards with 128K flash and they work well, so I thought there would be more headroom for further development of FW 

Hi @daddygo

I'm using a translation tool, so my intention may not be conveyed well.

I understand that there are many fakes from Chinese clones. I tried the tool you told me about, and it seems that the STM chip is genuine and there is 128KB of flash. I've attached a screenshot.

Well, I'm happy with the firmware I modified, so I don't plan to make any further improvements even if the flash increases.

Hi 2A03,

Yes, most visual "originality" check say if the PIN1 marking on the MCU is not concave inwards, it is original, like yours.
There are still plenty of fakes on the market, but at least yours is OK 

128K question, - it just came up when you wrote that flash would be low for date display, but 128 would be fine, if still need room for the code.

@fredo_ did a huge and concrete job (THX), I think FW is fine in this state, - yes we agre on it...

I just add the NEO-M9N and a 50ohm driver and that's the end of it for me also 

[ME]

well this is odd, My bp from my gpsdo gives errors and wont connect using st-link v2 but my spare bp connected and seemed to program ok!.

[ME]

Well i downloaded and flashed the attached bin , All i get is a blank display now, Any ideas?., Had to change the file extension to .doc to attach here so needs swapping back to bin ok.

[ME]

heres the log file after erasing then programing the new bin, Still a blank display tho.well it says i cant upload the file even tho its a .txt file!.

[daddygo]

heres the log file after erasing then programing the new bin, Still a blank display tho.well it says i cant upload the file even tho its a .txt file!.

Hi @ME,

I'll try to outline a sequence for you to do (watch the screen shots, pls.):

1. download the gpsdo.bin file from github, do not modify the file name (to gpsdo fredzo.bin, leave so gpsdo.bin)
2. make sure that the CubeProgrammer program and the FW of ST-Link are the latest, I currently have a new version for CUBE Prg. (v2.19.0)
3. if you want(?), but I suggest you do a full chip erase
4. then open the downloaded (github) gpsdo.bin under the pencil icon menu (top left) + Open File sign (task bar)
5. press the Download button close to the top left
6. go to the Device memory tab again and Read the uploaded FW for check only...
7. notice, check where it starts, - it should be 0x0800000000

Well, if you have a good BP and good ST-Link, even if not original, STM32 is still programmable....
if not buy a good ST-Link V2, or replace the BP 
(that's why I said it's good to buy your ST-Link from a safe place, because then you know it's OK - so much shorter troubleshooting time)

more prtsc in the next post

[daddygo]

more prtsc.....

- the last picture is the programmer settings!

Cheers DG

[ME]

Hi @ME,


nope, in this case you don't have to (SWD header), just follow this:

https://github.com/fredzo/gpsdo-fw?tab=readme-ov-file#flashing-the-firmware

+++edit1:
in the unfortunate event that ST-Link V2 is fake and the MCU is fake, well you know....

buy one of these, they are only pennies and later you will know that at least "programmer" is original, I use it too, and/or V3
https://eu.mouser.com/ProductDetail/STMicroelectronics/ST-LINK-V2?qs=H4BOwPtf9MC1sDQ8j3cy4w%3D%3D

The CUBE Programmer runs with a fake ST-Link V2 (in most cases), but CubeIDE, for example, will not

can you link the cube programer sw, is there a version of it for ubuntu?. Thank you.

[daddygo]

Hi @ME,


nope, in this case you don't have to (SWD header), just follow this:

https://github.com/fredzo/gpsdo-fw?tab=readme-ov-file#flashing-the-firmware

+++edit1:
in the unfortunate event that ST-Link V2 is fake and the MCU is fake, well you know....

buy one of these, they are only pennies and later you will know that at least "programmer" is original, I use it too, and/or V3
https://eu.mouser.com/ProductDetail/STMicroelectronics/ST-LINK-V2?qs=H4BOwPtf9MC1sDQ8j3cy4w%3D%3D

The CUBE Programmer runs with a fake ST-Link V2 (in most cases), but CubeIDE, for example, will not

can you link the cube programer sw, is there a version of it for ubuntu?. Thank you.

Hi @ME,

Find all OS versions here (ST reg. account required):
https://www.st.com/en/development-tools/stm32cubeprog.html#overview

I suggest if you are not experienced enough in Linux, you should use the Win version instead, there will be less headaches (works smoothly).

Yes, the linux installer runs on Ubuntu (I use Ubuntu Desktop 22.04LTS), so with some configuration tricks CUBE Programmer works fine under Linux.
I use it f.e. when I use IDE and MX at the same time, or say... need to do CMAKE, which is not a wizard under Windows.
(for a smooth FW upload via SWD, I stay under Win)

-as you can see in the screenshot the Linux ver. is now updated to v2.19.0 also, like the Win version yesterday

BTW:
if you are running Ubuntu on a VM env., as I am, make sure you have the correct USB pass to Linux VM for ST-Link!!!

[flash2b]

Sorry it has been a long time since I posted into this thread.

I see there is a lot of progress made with the alternative firmware, which I also want to try. I have some experience with BPs (and the fakes they sell on AliExpress) and also a working ST Link from WeActStudio (I have used it couple of times with good success). I have the STM32Cube programmer software up-to-date on my Win PC and I know to operate it.

I have the following questions:
- If I want to backup the existing firmware, do I need to downloaded this from the original BP with the GPSDO switched on ?
- What size the original firmware should be ?
- If I want to write the 2A03 version (I have an ATGM336H in my GPSDO) of the firmware, v0.1.7.1 into the existing BP, should the GPSDO be switched on ?
- If I want to compile a version myself, what software do I need additionally ?

I do not want want to end up with a bricked GPSDO, so hence my questions.

[daddygo]

Sorry it has been a long time since I posted into this thread.

I see there is a lot of progress made with the alternative firmware, which I also want to try. I have some experience with BPs (and the fakes they sell on AliExpress) and also a working ST Link from WeAcct (I have used it couple of times with good success). I have the STM32Cube programmer software up-to-date on my Win PC and I know to operate it.

I have the following questions:
- If I want to backup the existing firmware, do I need to downloaded this from the original BP with the GPSDO switched on ?
- What size the original firmware should be ?
- If I want to write the 2A03 version (I have an ATGM336H in my GPSDO) of the firmware, v0.1.7.1 into the existing BP, should the GPSDO be switched on ?
- If I want to compile a version myself, what software do I need additionally ?

I do want want to end up with a bricked GPSDO, so hence my questions.

Hi @flash2b

1.- If I want to backup the existing firmware, do I need to downloaded this from the original BP with the GPSDO switched on ?
2.- What size the original firmware should be ?
3.- If I want to write the 2A03 version (I have an ATGM336H in my GPSDO) of the firmware, v0.1.7.1 into the existing BP, should the GPSDO be switched on ?
4.- If I want to compile a version myself, what software do I need additionally ?

I will try to answer point by point:

1. you can download the original FW from BP if you are connected to it, but it is also uploaded to this thread by @fredo_ on third page in the thread
2. original-gpsdo.zip (16.06 kB - downloaded 23 times
3. no, don't turn on the GPSDO PSU, the BP through SWD header is powered by ST-Link (3.3V)
4. the fredo's github page describes how to edit the code, everyone does it differently, but the VSCode might be good for you under win
https://github.com/fredzo/gpsdo-fw?tab=readme-ov-file#building

if you also want HW changes, CubeMX first

only dependency: arm-none-eabi
https://developer.arm.com/downloads/-/gnu-rm

[ME]

Well i found another bp, I flashed it ok but when fitted to my unit ii get just a backlight with a blanc display,  See the bin as read after flashing fredzo 1.7 fw Any  ideas.

[ME]

Well i finally managed to flash the fredo 1.7 fw, Since loading the fw it seems i get no sat rx , sats listed as 0 ,Is the baud rate different for the gps on the new fw?, I programed a new bp and fitted it , My gpsdo works fine if i refit the original bp!, any ideas?, Thanks in advance.

[fredo_]

Hi ME,
Is the time displayed correctly ? If seconds are changing, it means that the serial connection with the GPS is OK.
You need to wait long enough for the GPS to get a fix, especially if you have bad reception.
Maybe check that the GPS antenna is well connected to the GPS board too, it might have moved while changing the bluepill board.
Also did you make the necessary modifications on the new bluepill board ? I never did that, but I think you need to remove a capacitor or a resistor around the crystal and you need to connect the wire from the TCXO to the bluepill board.
That should not affect satellite reception though.
Is the PPS led on the GPS module blinking ? If not, that means that your GPS was not able to get a fix.

[ME]

Ok in answer to your questions fredo, Firstly the time displays a row of zeros with an alternating minus and equal displayed top lh side of the display , The bp resistor value was ok so didnt require swapping, The gps led flashes,at about 1 hz, All connections are good on the gps board, Also the display doesent always come on, The backlight illuminates with nothing displayed, There are 2 led's on the bp red and green, When the display shows info the red led illuminates and the green one flashes, When it does that it seems to do it in a sequence sort of with the gps pcb led, The gps led flashes about 10 times with the bp green led on, Then the green bp led goes off for another 10 flashes of the gps board and the cycle continues, If i power cycle the whole unit it shows no display and the bp green led stays off, If i leave it powered off for 10 mins and power it on the  bp green comes on and flashes and the display shows info ,The wire from the main pcb to the bp xtal is connected ok, Thanks for taking the time to reply 73.   

[ME]

Just to add, When the unit first powers on when the bp green led comes on and the display works , It displays an exclamation mark followed by a flashing question mark on the lh side of the display, after about 10 seconds it changed and displays
-/= flashing alternating 00  0.00
                               ----
                           -         .----
Hope that helps.